university of groningen the role of abc-transporters in ... · university of groningen the role of...

University of Groningen

The role of ABC-transporters in childhood and adult acute lymphoblastic leukemiaPlasschaert, Sabine Louise Anne

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2005

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Plasschaert, S. L. A. (2005). The role of ABC-transporters in childhood and adult acute lymphoblasticleukemia. [S.n.].

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 06-05-2020

https://www.rug.nl/research/portal/en/publications/the-role-of-abctransporters-in-childhood-and-adult-acute-lymphoblastic-leukemia(fbcf6e5a-6346-4207-b78a-943780eb68b8).html



The role of ABC-transporters in childhood and adult acute lymphoblastic leukemia

Publication of this thesis was financially supported by : Stichting Kinderoncologie Groningen Stichting Hemat ol og ie Stichting Werkgr oep I nte rne O nc ol og ie Groningen U nivers ity Inst itu te for Drug Exploration (GUIDE)

© 2005 by S.L.A. Plasschaert. All rights served. N o part of this b ook ma y be repr oduced or tra nsmitted i n a ny f orm or b y any means with out written permiss ion of the auth or a nd the publisher h old ing the c opyr ight of the publ ished art icles.

C over : Illustrati ons D ick Bru na c opyright Mercis bv, 1 967. Met da nk aa n D ick Bru na

ISBN: 90-367-2226-8

Page lay out: P. va n der S ijde, Gr on inge n, The Netherla nds Pri nted b y: P onse n & L ooije n B.V., Wage ni nge n, The Netherla nds

Stellingen behorende bij het proefschrift

"The role of ABC-transporters in childhood and adult acute lymphoblastic leukemia"

I. Het verschil in prognose van acute lymfatische leukemie (ALL) tussen kinderen en volwassenen kan mede verklaard worden door verschil in origine: ALL op de volwassen leeftijd ontstaat in de multipotente stamcel terwijl het op de kinderleeftijd ontstaat uit de rijpere lymfolde voorlopercel.

MGreaves, dit proefschrift

2. Functionele activiteit van P-glycoprotelne wordt voomamelijk gezien in volwass�nen met T-cell acute lymfatische leukemie en heeft een ongunstige invloed op prognose.

Dit proefschrift

3. Genetische varianten in het MDR-1 gen, dat codeert voor P-glycoprotelne, kunnen de grate variabiliteit in vincristine fannacokinetiek onvoldoende verklaren.

Dit proefschrift

4. Het borstkankerresistentie-eiwit is functioneel aanwezig in voorloper-B-ALL en in mindere mate in T-cell ALL.

Dit proefschrift

5. Leukemiecellen van patienten met acute lymfatische leukemie die een recidief krijgen, hebben een hogere expressie van MRP I, MRP2, MRP3, MRP5 en MRP6 dan patienten die in complete remissie blijven.

Dit proefschrift

6. In toekomstige studies naar ABC-transporters in acute lymfatische leukemiecellen moet niet aileen naar ABC-transporters in de grate meerderheid van leukemische blasten gekeken worden, maar ook in de leukemische stamcellen.

Dit proefschrift

7. Pubers willen het liefst als volwassenen 'behandeld' worden, maar krijgen zij acute lymfatische leukemie, dan kunnen ze beter 'als kind' worden behandeld.

8. Door 'fishing expeditions', zoals bet gebruik van DNA microarrays, voor de identificatie van genen om inzicht te krijgen in de pathogenese van ziekten, wordt de gebruikelijke wetenschappelijke volgorde van hypothese en toetsing omgedraaid.

9. Het feit dat de kans op emotionele uitputting bij medisch specialisten nog grater is dan bij specialisten in opleiding vraagt om des te meer aandacht hiervoor tijdens de opleiding.

I 0. Liefhebben is meer lief dan hebben. Toon Hermans

I I . Een mens lijdt dikwijls het meest, door het lijden dat hij vreest. Nico/aas Beets

Sabine Plasschaert, 16 maart 2005

RIJKSUNIVERSITEIT GRONINGEN

The role of ABC-transporters in childhood and adult acute lymphoblastic leukemia

Proefschrift

ter verkrijging van het doctoraat in de Medische Wetenschappen

aan de Rijksuniversiteit Groningen op gezag van de

Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

woensdag 16 maart 2005 om 16.15 uur

door

Sabine Louise Anne Plasschaert

geboren op 18 juni 1973 te Nijmegen

Promotores:

Co-promotor

Beoordelingscommissie:

Prof. dr. W.A. Kamps Prof. dr. E.G.E. de Vries Prof. dr. E. Vellenga

Dr. E.S.J.M. de Bont

Prof. dr. P.M. Kluin Prof. dr. R. Pieters Prof. dr. R.J. Scheper

Paranimfen: Dorina van der Kolk Saskia Wolt-Plompen

Contents

Scope of the thesis 9

Chapter 1

Chapter 2

Chapter3

Chapter4

Chapter 5

Chapter 6

Chapter?

Prognosis in childhood and adult acute lymphoblastic 1 1 leukemia: a question of maturation? Cancer Treatment Reviews 2004; 30: 37-51

High functional P-glycoprotein activity is more often 39 present in T-cell acute lymphoblastic leukemia cells in adults than in children. Leukemia & Lymphoma 2003; 44: 85-95

Influence of the functional polymorphism of the human 59 multidrug-resistance gene on vincristine pharmacokinetics in childhood acute lymphoblastic leukemia Clinical Pharmacology & Therapeutics 2004; 76: 220-229

Breast cancer resistance protein (BCRP) in acute leukemia 79 Leukemia & Lymphoma 2004; 45: 649-654

The role of breast cancer resistance protein (BCRP) in acute 93 lymphoblastic leukemia Clinical Cancer Research 2003; 9: 5171-7

Expression of multidrug resistance-associated proteins 1 13 predicts prognosis in childhood and adult acute lymphoblastic leukemia Submitted

Summary, general discussion and future perspectives 1 3 1

Nederlandse samenvatting 1 39

Dankwoord 1 45

Scope of the thesis

SCOPE OF THE THESIS

Acute lymphoblastic leukemia (ALL) is a disease that occurs in adults and children,

but the incidence changes with age. The highest incidence is in children between the

age of two and five years old, followed by the five to nine years old. However,

ALL has a bimodal distribution with a low and steady rise from the age of 40.

Progress in the treatment of ALL has led to better survival rates, however, children

have benefited more from improved treatment modalities than adults. The disease

free survival rates for children with ALL are approximately 80 percent and 40

percent for adults. The incidence of ALL and differences in prognosis and biological

characteristics between children and adults suggest that ALL in childhood and

adulthood behaves like different diseases. A possible explanation for the worse

prognosis in adults, is a difference in drug sensitivity of leukemic blasts between

children and adults. One of the mechanisms of cancer cells to develop drug

resistance is the overexpression of proteins involved in multidrug resistance, which

is resistance to structurally and functionally unrelated natural product drugs. The

scope of this thesis is to investigate the role of ATP-binding cassette (ABC)

transporters, potentially involved in multidrug resistance, in ALL in children and

adults. Therefore, the expression and/or functional activity of the ABC-transporters

P-glycoprotein (P-gp ), the multi drug resistance protein 1 (MRP 1 ) and other members

of the MRP family (MRP2-6), and the Breast Cancer Resistance protein (BCRP) was

investigated in leukemic blasts from children and adults and related to prognostic

factors and clinical outcome.

Chapter 1 describes the current knowledge on the difference in prognosis

between children and adults in ALL. It compares patient characteristics, the extent

of the disease, leukemic cell characteristics, treatment and other prognostic factors

between childhood and adult ALL. This is discussed in relation to the hypothesis

that the maturation stage of the cells, from which the leukemia arises, might play an

important role in the differential behaviour of adult and childhood ALL.

In chapter 2 the functional activity of P-gp and MRP 1 is investigated in ALL.

P-gp activity is determined flow cytometrically in leukemic blasts by measuring the

modulation of rhodamine 1 23 fluorescence by PSC833. For assessing MRP activity,

the modulation of carboxyfluorescein by MK-571 is measured. The impact of P-gp

and MRP functional activity on prognosis is investigated, and the functional activity

is compared between children and adults.

The role of P-gp in cancer cells, i.e. leukemic blasts, is described in chapter

2. However, the efflux pump P-gp is not only present in leukemic blasts, but also

in normal tissues including the gastro-intestinal tract, kidney and liver, affecting

9

the distribution of many drugs . Vincristine is a P-gp substrate, and therefore

P-gp, encoded by the MDRJ gene, may influence the exposure of cancer cells

to vincristine, measured by a pharmacokinetic profile of vincristine. To date, 29

single nucleotide polymorphisms (SNPs) have been identified in the entire human

MDRJ gene, of which the SNP C3435T and G2677T showed an effect on substrate

pharmacokinetics. In chapter 3 the relation between germline MDRJ genetic

variants, vincristine pharmacokinetics and/or vincristine side effects is described in

childhood ALL.

Another ABC transporter has been identified in a multidrug resistant

human breast cancer cell line MCF-7/AdrVp. This transporter, first described

in 1 998, received several names, including BCRP, ABCG2, placental transporter

or mitoxantrone resistance protein. Overexpression of this efflux pump confers

resistance to mitoxantrone, doxorubicin and daunorubicin, drugs which are used in

the treatment of ALL. Chapter 4 gives an overview of the current knowledge on the

role of BCRP in acute leukemia. In chapter 5 we describe our study investigating

the functional activity of BCRP in ALL in childen and adults. The functional activity

is determined by measuring mitoxantrone accumulation in the absence or presence

of the BCRP inhibitor fumitremorgin C (FTC), and FTC in combination with the

inhibitors of the other transporters P-gp and MRPl, which might obscure the BCRP

function.

To date, the MRP-family consists of 8 members. Of these MRPs, MRPl -6 are known

to transport chemotherapeutic drugs. Of these MRPs, MRP4 and MRP5 can transport

6-mercaptopurine and methotrexate, which are important chemotherapeutic agents

in the maintenance phase of ALL treatment. In chapter 6 we determined the mRNA

expression of these six MRP genes by real-time quantitative RT-PCR in children

and adults with ALL. Since numerous transporters can be active in leukemic blasts,

it is unlikely that only one transporter plays a role in multidrug resistance. The

transporters appear to act in concert and therefore we studied the connection between

the six transporters and investigated their correlation with known prognostic factors

and clinical outcome.

The results presented in this thesis are summarized in Chapter 7. In addition,

conclusions from our studies are formulated and future perspectives are described.

10

Chapter 1

Prognosis in childhood and adult acute lymphoblastic

leukaemia: a question of maturation?

Sabine L.A. Plasschaert1, Willem A. Kamps1, Edo Vellenga2,

Elisabeth G.E de Vries3, Eveline S.J.M. de Bont1

Departments of1Paediatric Haematology and Oncology, 2Haematology, 3 Medical Oncology, University Hospital Groningen,

Groningen, The Netherlands

Cancer Treatment Reviews 2004; 30: 37-51

Chapter 1

ABSTRACT

Acute lymphoblastic leukaemia (ALL) is a disease diagnosed in children as well as

adults. Progress in the treatment of ALL has led to better survival rates, however,

children have benefited more from improved treatment modalities than adults.

Recent evidence has underscored that the difference in characteristics and biology of

adult versus childhood ALL might be the result of a different origin. According

to the two-hit paradigm of Knudson, to develop cancer two genetic events are

necessary. It has been suggested, that in childhood ALL the first genetic event

happens in the more mature lymphoid committed progenitor cells, whereas in adult

ALL the first hit occurs in multipotent stem cells. This review compares patient

characteristics, the extent of the disease, leukaemic cell characteristics and treatment

between childhood and adult ALL. This is discussed in relation to the hypothesis that

the maturation stage of the cells, from which the leukaemia arises, is responsible for

the differential behaviour of adult and childhood ALL.

1 2

Differences in childhood and adult ALL

INTRODUCTION

Acute lymphoblastic leukaemia (ALL) is a disease characterized by an uncontrolled

proliferation and maturation arrest of lymphoid progenitor cells in the bone marrow

resulting in an excess of malignant cells. ALL occurs in children as well as in adults,

but the incidence changes with age. The highest incidence is in children under the

age of five years old (5.7 per 1 00,000 person years), followed by the 5- to 9- years

old (2. 7) and the 1 0- to 14-years old ( 1 .6). However, ALL has a bimodal distribution,

with a low and steady rise from the age of 40 with an incidence of 0.4 per 1 00,000

person years to an incidence of 0.8 per 1 00,000 person years at an age higher than

75-years old1•

Progress in the treatment of patients with ALL has led to better survival rates, but

children have benefited more from improved treatment modalities than adults. The

prognosis in children with ALL is one of the most favourable of all disseminated

cancers: more than 95% achieve complete remission (CR) and disease-free survival

rates (5 years) are 63 to 83%2• Adults with ALL have a worse prognosis, with CR

rates of 75 to 89% and long-term disease-free survival rates of 28 to 39% (3 to 5

years)3•

From the incidence of ALL and the difference in characteristics and prognosis

between children and adults, it could be concluded that ALL in childhood and

adulthood are distinct diseases.

It has been suggested that paediatric cancer results from genetic defects related to

growth and development of the different organs or tissues during risk periods of

proliferative stress4• This suggestion is based on the observation of different risk

periods for paediatric tumours depending on the age of the child. For example,

osteosarcoma occurs mainly during adolescence when the growth spurt of the long

bones takes place5• The peak incidence of ALL is between the age of two and

five years, when B-lymphocyte progenitor cells are extensively proliferating and

rearranging their immunoglobulin genes6•

According to the two-hit model from Knudson, to develop cancer, two genetic or

transforming events are necessary7;8• In this model, the occurrence of a translocation

is considered to be a genetic event9• In childhood, a period of proliferative stress such

as extensive immunoglobulin recombinations in B-cells induces a higher chance for

a second hit resulting in ALL. The first genetic hit might occur as early as during

fetal haematopoiesis in utero10• This is supported by studies, demonstrating the same

mutation in monozygotic twins, in which the 'pre-leukaemic' clone arises in one

fetus and disseminates to the other by the placenta11 ; 12 • Moreover, the chromosomal

translocation TEL-AML 1 , which occurs frequently in childhood ALL, could be

13

Chapter 1

detected in neonatal blood spots from children who later on developed TEL-AMLl

positive ALL1 0:1 3-1 5• Leukaemia associated with MLL gene fusions often develops at

a very early age and has a very high concordance rate in identical twins, suggesting

that in this type of leukaemia a second genetic event may not even be necessary1 3•

When a pre-leukaemic clone is present, immune-mediated stress can induce a second

genetic event, leading to leukaemia. The immune response, for example elicited by

a viral infection, may stimulate the proliferation of pre-leukaemic cells and increase

the risk for a second event1 6•

Differences in characteristics and prognosis of ALL in adults compared to children

might result from the fact that the first hit in childhood ALL takes place in more

mature lymphoid committed progenitor cells, whereas in adult ALL the first hit

occurs in multipotent stem cells4:9• In children with TEL-AMLl positive ALL,

the translocation is not found in the stem cell compartment (i.e. CD34+/CD 1 9-

subpopulation), but in subpopulations with the phenotype of the more committed

progenitor cells (CD34+/CD19+ subpopulation)1 7:1 8• In contrast, in Philadelphia

positive ALL, which is observed in a much higher frequency in adult ALL than

childhood ALL, the translocation is already present in the stem cell compartment1 8•

Therefore, Philadelphia-positive leukaemia is thought to arise from multipotent stem

cells rather than committed progenitors1 9:20•

The difference between childhood and adult ALL could be explained by assuming

that the haematopoietic cells, in which the transformation takes place, share

some characteristics with their normal haematopoietic counterpart. The committed

progenitors have a limited self-renewal capacity and are dependent on cytokines

and growth factors for their proliferation and differentiation. These cells are

more sensitive to apoptosis induced by chemotherapy or cytokine withdrawal.

Multipotent stem cells have a greater self-renewal capacity and are more resistant to

chemotherapy than committed progenitors. Therefore, ALL arising from multi potent

stem cells might give rise to a more aggressive leukaemic clone. However, both in

childhood and adult ALL differentiation of the leukaemic stem cells can take place,

giving their progeny a more sensitive, differentiated phenotype4•

Studies of leukaemia have focussed on four major aspects of the disease: patient

characteristics, the extent of the disease, leukaemic cell characteristics and treatment.

This review will describe these various aspects in children and adult ALL in relation

to the hypothesis described above. Hereby, we hope to gain further insight into the

differences between childhood and adult ALL, which might allow optimization of

treatment for patients with ALL.

14


PATIENTS CHARACTERISTICS

Age

The influence of age on clinical outcome is very important. In one clinical trial,

children and adults were treated according to similar protocols, and the impact of age

was studied. A 20-year-old had double the risk of treatment failure of a 1 0-year-old;

and a 44-year-old had a four-fold higher risk of treatment failure, when corrected for

all possible influencing factors21 • Infants were not included in this study, since infant

leukaemia is considered to be a different type of disease with adverse prognostic

features.

No explanation has been found for the impact of age itself on prognosis. A possible

explanation could be differences in telomere length and telomerase activity between

blasts from adults and children. The length of telomeres serves as a useful indicator

for the replicative history of a cell as well as for estimating its residual replicative

potential, since each cell division results in a loss of terminal telomere DNA.

Telomerase is a telomere-synthesizing reverse transcriptase that can compensate for

the loss of telomeres associated with cell divisions. Haematopoietic stem cells

show telomerase activity and early progenitors exhibit a significant upregulation

of this enzyme activity22• It is thought that primitive stem cells may not have

telomerase activity since they are quiescent, and cells in GO phase do not express

telomerase activity. When a subpopulation enters a proliferative phase, telomere

erosion is initiated and telomerase activity may be upregulated. In ALL blasts

elevated telomerase activity has been found in comparison with normal peripheral

blood. It has been suggested that this elevated telomerase activity might parallel

disease progression and be of prognostic relevance23;24•

Until now, no studies have been performed comparing the telomere length and

telomerase activity between childhood and adult leukaemic blasts. It is unlikely that

the difference in telomerase activity in connection with the 'cells of origin' may

account for the different behaviour of ALL in children and adults, since telomerase

activity seems mainly to depend on the level of proliferation. However, the fact that

the cells of origin from adult leukaemia have a longer replicative history, could be

a determining factor. Telomere length could be shorter, causing genomic instability

or the emergence of immortal clones with increased telomerase activity21 ;25;26• Thus,

the impact of age on prognosis could partly be explained by differences in telomere

length and/or telomerase activity between blasts from children and adults, but these

differences need to be subject of future studies.

15

Chapter 1

Gender

The incidence of ALL is higher in males than in females, regardless of age category.

Overall, males have a 1 .4 higher risk of ALL than females27, however, the cause of

the higher risk for males is unclear. There are suggestions that a higher susceptibility

for infections in males might play a role in their higher incidence of ALU8.

Not only is there a difference in incidence between males and females, also the

prognosis of ALL seems to depend on gender. Most studies in children with ALL

show a better treatment outcome in girls than boys29-32, but this difference is not

always observed33;34• The difference between the sexes becomes even more apparent

after the first two years of treatment. It has been suggested that the testis functions

as a so-called "sanctuary site" with the blood-testis barrier protecting leukaemic cells

from anti-cancer drugs. Subsequently, dissemination to the bone marrow occurs,

resulting in a relapse. However, a clinical study investigating testicular irradiation

showed a decrease in testicular relapse following irradiation, but no reduction in

bone marrow relapse35;36• Therefore, this sanctuary site can not sufficiently explain

the poorer prognosis in males. Boys do not have a higher risk of central nervous

system (CNS) relapse, which suggests that the poorer prognosis is not a consequence

of inadequate induction therapy, but systemic maintenance therapy. This is also

supported by the observation that differences between the sexes become more

apparent after the first two years of treatment. The explanation for inadequate

systemic maintenance therapy may lie in biological differences in metabolizing

the drugs used in maintenance therapy, particularly 6-mercaptopurine and

methotrexate. Boys tolerate higher cumulative doses of 6-mercaptopurine

than girls, and it has been suggested that this tolerance may be due to

differences in metabolism of the drug, perhaps due to variable inheritance of

sex -linked enzymes36-38• In contrast, in most studies in adult ALL, gender does not influence clinical

outcome39-43• A single study reported increased survival rates in females than

males44• An explanation for not observing this difference in survival rates in

adults, could be that adults with ALL more often relapse than children. As a

consequence, the possible additive effect of differences in drug metabolism in

males might therefore, not be observed. In summary, ALL occurs more often

in males than in females, and an effect of gender on prognosis is observed in children.

Constitutional genetic defects

Evidence for genetic influences on the development of ALL is the higher

incidence of ALL among children with constitutional genetic defects, such

16


as trisomy 21, Fanconi's anaemia and Bloom syndrome45-47• Both Fanconi's

anaemia and Bloom syndrome have defects in DNA-repair. DNA double

strand breaks can be repaired by homologous recombination and non

homologous end-joining. In both syndromes homologous recombination is defective. This leads to the activation of the error-prone non-homologous

end-joining, resulting in increased genomic instability and predisposition

to malignancies. The role of DNA damage repair in the development of

leukaemia is subscribed by the occurrence of secondary leukaemia. Treatment

with topoisomerase II inhibitors results in increased DNA double strand breaks, which can lead to chromosomal translocations by non-homologous

end joining and secondary leukaemia48•

EXTENT OF THE DISEASE

White blood cell count at diagnosis

The initial white blood cell count (WBC) at diagnosis is a reflection of the biological

behaviour of ALL cells. Its prognostic significance in both childhood and adult ALL

has been demonstrated by many studies49.5 1 • Although there is no threshold value of

WBC count, there is a clear continuous relation between high WBC count and unfa

vourable prognosis. A uniform risk classification has been proposed in which a WBC

count > 50 x 1 09/L amongst others was considered high-risk52• In children with ALL,

1 9 to 27% had a high WBC count (> 50 x 1 09/L) at diagnosis (Table 1 )29:32:34:53:54• To

improve the prognosis in high risk ALL patients, these patients receive augmented

therapy compared to standard risk patients. In adults, few studies have used the uni

form risk classification and thus can be compared; 1 9 to 30% of patients had a high

WBC count at diagnosis (Table 1 )39;44:55:56 • A study from Chessells et a!. compared

WBC count between different age groups. They found increased incidences of high

WBC count in 1 0-39 year olds. Although patients older than 40 years of age have

a poorer prognosis than younger patients, they were found to have WBC counts

comparable with children younger than 1 0 years of age21 • Another study compared

the distribution of WBC counts between children and adults, and showed a greater

proportion of high WBC counts in adults compared with children. However, in this

study the majority of the adult patients were aged between 1 1 - and 39-years old, with

only a limited group older than 40-years old. Overall, it can be said that there is no

difference in WBC count between children and adults.

In the determination ofWBC count, we look mainly at the progeny of the leukaemic

cells. This does not reveal anything concerning the size of the leukaemic stem cell

pool. Thus, the WBC count can only be interpreted as a reflection of the total leukae-

17

Chapter 1

Table 1 Comp ariso n ch ar acteristics be twee n childre n and adults

Childre n* Adults Re fere nces

WBCt (>50xl 09/L) 1 9 to 27% 1 9 to 30 % 29,32,34,39 44,53-56

Hep atosple nomeg aly 50 to 65% 50 to 65% 2 1 Me ni nge al leukemi a 3 % ( 1 to 7%) 6 % ( l to 1 0%) 6 1 -63 at di ag nosis

FAB-type L l 7 6 to 89 % 3 1 to 43 % 34,39,66-69 L2 14 to 22 % 49 to 60 %

lmmu nophe notype 30,32-34,49,74 T-li ne age 1 3 to 1 5 % 26 % M ature B-l i ne age 1 to 2 % 3 to 4 %

Co-expressio n myeloid m arkers 25 % 33 % 76-80

Cytoge netics 2 1 ,67,72,84,85, 140 Hyperploidy (>50) 20 to 32 % 5 to 12 % t (9;22) 3 to 5 % 1 5 to 25 % t (4; 1 1 ) 2 % 8 % t ( 12;2 1) 26 % 3 to 4 %

* I nfants (p atie nts < 1 ye ar of age) are not i ncluded. t WBC: white blood cell cou nt at di ag nosis

mic burden and no difference can be observed in WBC count between children and

adults.

Hepatosplenomegaly and lymphadenopathy

Besides the central nervous system, other organs can be susceptible to infiltration by

leukaemic blasts, such as the liver, spleen and lymph nodes. Nodal enlargement is an

indirect measure of tumour burden and is associated with prognosis. Also the extent

of hepatic and splenal enlargement (> 3 em below the costal margin) is correlated

with prognosis. Since all these factors reflect the leukaemic cell mass, a risk factor

(BFM risk factor)57;ss has been defined in childhood ALL that can be calculated

from the WBC count at diagnosis, and liver and spleen enlargement. The risk

factor allows stratification of patients for different risk of relapse and therefore,

different clinical treatment protocols34•54;57;59• In both children and adults, one-half to

two-thirds of patients have asymptomatic enlargement of the liver and spleen and

lymphadenopathy (Table 1 ). In one study however, the presence of hepatomegaly,

18


splenomegaly and lymph node enlargement was compared in different age groups,

and this study showed that these features decreased significantly with increasing

age21• The presence of hepatosplenomegaly and lymph node enlargement also

depends on the ability of the leukaemic cells to infiltrate organs. It is unknown

whether stem cells or committed progenitors differ in this aspect. Since the

development of novel, improved prognostic factors, such as immunophenotyping

and cytogenetic analysis, the extent of hepatosplenomegaly and/or lymphadenopathy

is of limited use in the treatment of children and adults with ALL.

Meningeal leukaemia

One of the most frequent sites of extramedullary involvement of leukaemia is the

eNS. The usual clinical definition of meningeal leukaemia requires the finding of

more than 5 cells/f.1L60 in a cytocentrifuge preparation of the cerebrospinal fluid

(eSF), which can morphologically be identified as leukaemic blasts. The incidence

of eNS involvement at diagnosis in adults is about 6% of cases (ranging from

1-10%)61• In children, eNS leukaemia at diagnosis occurs in approximately 3% of

cases (ranging from 1-7%) (Table 1)62;63•

Although a higher incidence of eNS involvement in adults has been found in some

studies, it is suggested that young children are more susceptible for eNS leukaemia

than adolescents and adults, possibly because they have a higher proportion of their

vascularity in the leptomeninges63;64•

Patients with eNS-involvement at diagnosis, whether children or adults, require

additional eNS-directed treatment. A common treatment strategy is the application

of extra intrathecal therapy, and in some regimens additional radiotherapy, until blast

cells are cleared from the eSF and regression of possible clinical neurological signs

occurs.

Although eNS involvement is not often observed, prophylactic eNS therapy is

necessary for higher survival rates. Prophylactic treatment of the eNS is based

on the assumption that the eNS acts as a sanctuary site for leukaemic cells that

are protected from cytotoxic concentrations of drugs by the blood-brain barrier.

For children, the general treatment consists of regular intrathecal methotrexate for

low-risk cases, with additional intravenous methotrexate for higher risk groups.

Because of the severe side effects of cranial irradiation in children, including

growth disturbances, endocrine dysregulation, learning difficulties and secondary

brain tumours, fewer patients undergo radiotherapy but instead receive increased

intensity systemic and intrathecal chemotherapy"6• In adults there are fewer trials

of eNS directed therapy and the overall priority remains the achievement of

durable bone marrow remission, besides the prevention of eNS-involvement. The

19

Chapter 1

most favourable results are obtained with high dose chemotherapy combined with

intrathecal therapy and/or CNS irradiation. The majority of treatment regimens in

adult ALL already include high dose chemotherapy in order to reduce the risk of

bone marrow relapse63•

Although the relapse rate is higher in adults than in children, the chance of having

an isolated CNS relapse, instead of an isolated or combined marrow relapse, is

actually higher in children than in adults21• This finding is in agreement with the

previously described higher susceptibility in children for CNS leukaemia. This is

possible due to the greater proportion of vascularity in the meninges in children or

other differences in the blood-brain barrier.

CNS relapse in adult patients often occurs during therapy, is frequently followed

by a bone marrow relapse and the survival is poor61• In children, patients with

a relatively late CNS relapse can often be successfully re-treated with intensified

therapy and additional irradiation. However, children who relapse during or shortly

after treatment have a worse prognosis65.

In conclusion, no clear difference in the incidence of meningeal leukaemia at time

of diagnosis can be observed between children and adults. It has been suggested that

children might be more susceptible for CNS involvement and this could be due to

differences in the composition of the blood-brain barrier. Moreover, leukaemic cells

from children and adults might differ in their ability to infiltrate the CNS, but until

now this has not been studied.

LEUKAEMIC CELL CHARACTERISTICS

FAD-classification

The French-American-British (FAB) cell classification system distinguishes ALL

among three subtypes, according to morphologic microscopic features66•

The L1 subtype is observed more frequently in children (76 to 89%)34;66·69 than

in adults (31 to 43%), whereas the L2 subtype is observed more often in adults

(adults 49 to 60% vs. children 14 to 22%) (Table 1)39;66;67• The L3 subtype, which

is associated with mature B-cell ALL, is rarely seen, but may be observed in adults

slightly more often than in children.

The prognostic impact of the different FAB subtypes remains unclear. In children

and adults, some studies have shown a better prognosis for L 1 patients and a higher

relapse rate for L2 patients39;68;70• However this remains controversial since other

studies have not confirmed a prognostic impact for the L1 and L2 FAB subtypes55;71•

Although historically the FAB classification in ALL was utilised, its clinical use

20


is disputed nowadays. This is due to the limited heterogeneity of leukaemic

blasts in light microscopy and the degree of subjectivity in subclassifying

affecting the reproducibility of the FAB classification. The current availability of

immunophenotyping and cytogenetic analysis in combination with the dubious

prognostic impact of the different FAB subtypes, have made FAB classification of

limited use in the clinical management of ALL.

lmmunophenotype

With the development of monoclonal antibodies, characterization of normal lymphoid

progenitors and leukaemic cells can be performed. Differentiation of B-lineage

and T-lineage progenitors to mature B- and T-cells is characterized by specific

antigen expression during the different stages of development. According to the

differentiation stages of lymphoid progenitors in normal haematopoiesis, leukaemic

cells can be classified in null ALL, pro-B, common pre-B, mature-B, early-T and

T-ALL. Although several stages can be subclassified, only the B-cell precursor,

mature B-cell and T-cell precursor ALL have been defined as subgroups with a

differential prognostic impact12• The mature B-ALL occurs in only 1 to 2% of

children and 3 to 4% of adults with ALL. When treated according to lymphoma

protocols, the prognosis of mature B-ALL patients is very good, especially in

children73•

Of all children with ALL, 13 to 15 % has a T-lineage immunophenotype, and

in adults around 26% has this immunophenotype (Table 1)49• Children with the

T-cell immunophenotype are usually considered high-risk patients, and are treated

according to high-risk protocols. This immunophenotype is often associated with

a high WBC count, a mediastinal mass and it is more frequently found in older

children. Comparing the standard risk of T-ALL childhood patients with the standard

risk B-lineage patients, some studies have found no difference in prognosis30;33;34.

However, others studies have found a more favourable prognosis in standard

B-lineage patients32;74• In adults, the T-cell immunophenotype is also associated with

high WBC count and a mediastinal mass, and therefore adult patients are often

treated according to high-risk protocols49• However, contradictory results have also

been published with regard to the prognostic impact of the T-cell immunophenotype

in adult T-ALU;3943;75• The co-expression of myeloid markers occurs more often in

adults than in children (33 vs. 25%)(Table 1 ), but the prognostic significance of this

observation remains uncertain 76·80•

In conclusion, the distribution of immunophenotypes in childhood ALL differs from

adult ALL. With improved treatment modalities, the prognostic impact of the different

immunophenotypes has diminished. When determining the immunophenotype, the

21

Chapter 1

progeny of the leukaemic cells is examined. The small percentage of cells in the

original leukaemic stem cell pool from which the leukaemia arises cannot be observed

in the determined immunophenotype. The co-expression of myeloid-associated

antigens suggests that the leukaemic cells originate from early stem cells, which still

have the potential to develop into myeloid or lymphoid committed progenitors, and

not from the already committed lymphoid progenitor cells. Furthermore, the higher

co-expression of myeloid markers in adult ALL therefore, supports this hypothesis.

Cytogenetics

The karyotype (chromosomal makeup) of the leukaemic cells is another variable

diagnostic feature by which patients can be classified. Both the chromosome number

(ploidy) and chromosomal structural abnormalities have been identified as powerful

prognostic indicators for therapeutic outcome.

The number of chromosomes in a leukaemic clone can be subdivided in high

hyperdiploid with more than 50 chromosomes, low hyperdiploid with 47 to

50 chromosomes, pseudodiploid ( 46 chromosomes with structural or numerical

abnormalities), diploid (normal 46 chromosomes), and hypodiploid (less than 46

chromosomes). Clinically, hyperdiploidy with > 50 chromosomes is the most

important, since it is strongly associated with good-risk features, such as a low

WBC count and the common-ALL phenotype81, and therefore, a good prognosis.

The prognosis for other ploidy groups shows variation between different studies.

Hyperploidy with > 50 chromosomes has a higher incidence in children than adults

(20 to 32 % vs. 5 to 12%). Leukaemic cells from high hyperdiploid ALL have been

found to selectively accumulate methotrexate polyglutamates82 (see drug resistance)

and to be more sensitive to other cytotoxic drugs83•

The t(9;22), 11q23 rearrangements and t(12;21) are the translocations with known

prognostic impact. The Philadelphia translocation t(9;22)(q34;q 11) occurs in

leukaemic cells of 3 to 5% of children and 15 to 25% of adults (Table 1)72• Several

studies have shown an increasing incidence of the Philadelphia translocation with

increasing age. The translocation is associated with unfavourable prognostic factors

and a poor prognosis84.

The incidence of t(4;11)(q21;q23) and other 11q23 rearrangements is very different

in various age groups (Table 1 ). In infants an incidence of 70 to 80% has been

observed85;86, whereas in children less than 15 years of age the incidence is 2%. The

incidence in adults appears to be nearer to 8%. Patients with t( 4; 11) have the worst

prognosis of all known chromosomal subgroups.

The translocation t(12;21) is the most common genetic lesion in childhood ALL.

Up to 26% of the children with ALL have this translocation and it is associated

22


with an exceptionally positive outcome. In adults the incidence is much lower (3 to

4%) (Table 1)72• In summary, the distribution of chromosomal abnormalities differs

between children and adults, with a higher incidence of unfavourable abnormalities

t( 4; 11) and t(9;22) in adults, and a higher incidence of the favourable abnormalities

t( 12;21) and hyperploidy in children.

Besides the association of particular genetic defects with certain age groups, there

is also a connection with leukaemia sub-types as defined by immunophenotype.

Infant ALL with 11 q23 rearrangements often show markers that are typical for the

B- and monocytic bilineal origin, suggesting committed progenitor cells to be

'the cells of origin'45• The t(12;21) translocation seems to originate in an early

B-lineage restricted stem cell, since this translocation cannot be observed in the

stem cell compartment17;18• In contrast, the BCR-ABL fusion in t(9;22) is present in

the stem cell compartment and therefore BCR-ABL positive leukaemia is thought

to arise from lympho-myeloid stem cells9;19;20• Both associations, with age and

immunophenotype, support the assumption that tumourigenesis is lineage-specific

and could therefore suggest that adult and childhood leukaemia originate from

different haematopoietic populations9;87• Other differences in the distribution of

chromosomal translocations between children and adults can thus be explained by

the fact that some rearrangements occur especially in the immature stem cells,

whereas others are typical for the committed lymphoid progenitors.

Drug resistance

The poorer clinical outcome of adults compared with children is reflected in lower

CR rates and higher relapse rates in adults. Indeed, with similar treatment protocols,

adults show lower CR rates21• A possible explanation for this could be the difference

in drug sensitivity of leukaemic blasts between children and adults. In vitro cellular

drug resistance, as determined by the methyl-thiazol-tetrazolium (MTT) assay, is

a strong independent adverse risk factor in childhood ALL88;90• In addition, factors

such as DNA-ploidy, age and immunophenotype are also related to differences

in drug resistance83;91;92• Several studies, using an in vitro drug sensitivity assay,

have shown that adult ALL blasts are more resistant than those from children to a

number of drugs, including prednisolone, dexamethasone, cytarabine, daunorubicin

and methotrexate93·95•

One of the mechanisms for cancer cells to develop resistance to chemotherapeutic

agents is the overexpression of transmembrane drug-efflux pumps96• The transporter

proteins P-glycoprotein (P-gp), encoded by the M DR-1 gene, and multidrug

resistance associated proteins (MRP) are the most studied transporters in ALU7;98•

They act as ATP-dependent membrane efflux pumps and actively extrude cytostatic

23

Chapter 1

compounds from the cell. Protein expression and functional activity of P-gp has

been mostly studied in childhood ALL, but also to a limited extent in adults.

Contradictory results have been obtained with respect to the clinical significance

of P-gp levels99•106• We have recently investigated P-gp activity in both children

and adults and observed an increased incidence of high P-gp activity in adults,

but restricted to T-ALLI07• MRP 1 protein expression and function did not have a

prognostic impact in ALL101;107;108. To briefly summarize, a difference is found in in

vitro drug resistance between children and adults, which cannot be explained by the

expression of the membrane transporters P-gp and MRP. Over the past few years

additional transporter proteins have been identified, including MRP4109, MRP5110

and the breast cancer resistance protein (BCRP)111, which can also play a role in drug

resistance. The role of these transporters in ALL has not yet been determined and is

currently under investigation.

The expression of several transporter proteins, including P-gp and BCRP, is also

associated with an immature phenotype in normal haematopoietic stem cells112•115.

Assuming that adult ALL arises from stem cells and childhood ALL from committed

progenitors, adult ALL could have a higher expression of these transporters in

the leukaemic stem cells, making them more resistant to chemotherapy. However,

until now, no difference in the expression of transporters has been observed in the

leukaemic progeny of ALL between children and adults.

Another mechanism of drug resistance is the metabolism of anti-cancer drugs.

Methotrexate (MTX) is commonly used in the treatment of ALL in children and

adults, and can be polyglutamated intracellularly. These polyglutamated forms of

MTX are retained longer in the cell than the parent drug. MTX and polyglutamated

MTX are similar in their ability to inhibit dihydrofolate reductase. However, the

polyglutamates are potent inhibitors of several enzymes in de novo purine synthesis,

which are not inhibited by MTX itself. In paediatric B-lineage leukaemia, increased

formation of MTX polyglutamates in vitro was shown to correlate with a better

prognosis116• B-lineage and T-lineage blasts in adults accumulate significantly lower

levels of MTX polyglutamates compared to B-lineage blasts from children. A

difference in methotrexate metabolism between children and adults might therefore

contribute to the shorter duration of remission in adults117• Perhaps the enzymes

responsible for methotrexate metabolism are differentially transcribed in stem cells

and committed progenitors, but this has not yet been investigated.

Apoptosis

Programmed cell death, also known as apoptosis, is a physiological mechanism

by which cells can be eliminated from tissues without inducing an inflammatory

24


response. Two different signaling pathways have been detailed that can induce

apoptosis. In the first "extrinsic" pathway, apoptosis is induced by the ligation

of transmembrane death receptors, leading to the activation of caspases, cysteine

proteases responsible for controlled cellular destruction. In the second "intrinsic"

pathway, a variety of extracellular and intracellular death stimuli trigger the release

of cytochrome c from mitochondria, subsequently leading to the activation of

caspases and apoptosis.

Sensitivity and resistance of tumour cells to cytotoxic drugs seems, apart from

several factors as addressed in the previous paragraph, finally to depend on activation

of apoptosis pathways. As mentioned above, one mechanism of inducing apoptosis

involves interactions of death receptors with their cognate ligands. These include

tumour necrosis factor (TNF-a), tumour necrosis factor-related apoptosis-inducing

ligand (TRAIL) and Pas-ligand (PasL). Through expression of Pas receptors,

leukaemic cells can be killed by cellular immune effectors that express PasU18;119•

Moreover, death receptors could be used as a target for immuno-therapy by

developing specific antibodies that bind to these receptors. On the other hand,

leukaemic cells expressing PasL can escape from the immune response by killing

Pas-bearing cytotoxic T-lymphocytes120• Treatment with cytotoxic drugs can induce

death receptor expression in leukaemic cells. Several studies have shown that Pas

receptor expression is upregulated by cytotoxic drugs, and ligation of PasL to the Pas

receptor on the leukaemic cells can lead to increased apoptosis of these neoplastic

cells121;122• However, other studies have reported no effect of chemotherapy on Pas

induced apoptosis123;124• PasL expression on leukaemic cells can also be upregulated

by cytotoxic drugs thereby decreasing cellular cytotoxicity of natural effectors and

facilitating immune escape.

The role of death receptors has been investigated in ALL, mainly in children125;127•

Most patients showed very low sensitivity to Pas- or TRAIL-mediated apoptosis and

the sensitivity did not correlate with expression levels of the various death receptors.

Mutational analyses of death receptors have also been performed125;128;129, but only

in T-lineage ALL few mutations were found that might be of functional relevance.

Therefore, the mechanism of resistance to death receptor-mediated apoptosis cannot

be explained simply by receptor mutations. Thus it has been suggested that this

resistance is maintained by disruption of Pas-mediated death signaling130• Currently

it is unknown whether differences exist between adults and children regarding the

death receptor-system, since no comparative studies have been performed.

Besides death receptors, apoptosis can be induced by the release of cytochrome c in response to various death stimuli. This leads to caspase activation and eventually

apoptosis, and is regulated by a group of mitochondria-associated proteins, called

25

Chapter 1

the Bcl-2 family. The Bcl-2 family consists of anti-apoptotic members, such as

Bcl-2, Mcl-1 and Bcl-Xl, and death promoting members, such as Bax and Bak. The

balance of pro- and anti-apoptotic family members acts as an intracellular "rheostat",

determining whether a particular cell lives or dies. Since these proteins are critical

in regulating apoptosis signaling pathways, they may also play an important role

in the development of cytotoxic drug resistance. A high Bcl-2 expression has been

observed in ALL blasts from both children and adults, but this did not appear

to correlate with either prognosis or with in vitro drug resistance131-137• Since

the ability of Bcl-2 to modulate the apoptotic threshold can be affected by the

relative expression of Bax, several studies in children have attempted to assess the

relationship between the Bcl-2/Bax ratio and clinical outcome. However, conflicting

data have been reported concerning the prognostic impact of the Bax expression

level and Bax/Bcl-2 ratios134:135:138• The expression levels of other members of the

Bcl-2 family, such as Bcl-xs, Bcl-xL and Mcl-1, have also been studied in children

and did not correlate with chemosensitivity and clinical outcome134:136• In conclusion, high levels of Bcl-2 expression are found in children and adults, but

do not seem to have a prognostic impact. The expression of other Bcl-2 family

members has only been investigated in childhood ALL and its role in predicting

prognosis remains unclear. The difficulty in interpreting expression levels of Bcl-2

family member is that, since these proteins interact with each other, analysis of only

one protein is insufficient. In general, it seems unlikely that the Bcl-2 family

will attribute to the difference in chemosensitivity observed between children

and adults. The use of DNA micro-arrays, allowing analysis of expression

of multiple genes potentially involved in regulating apoptosis in ALL, will

enable a detailed comparison between children and adults. In normal haematopoiesis a correlation between the maturation stage of

haematopoietic cells and apoptosis is unclear, although it has been suggested

that stem cells are less prone to apoptosis-inducing factors. If this is true, it

could also contribute to the assumed lower apoptosis sensitivity of leukaemic

stem cells in adult ALL, in view of the hypothesis that adult ALL derives from immature stem cells.

TREATMENT

It is not always straightforward to compare treatment protocols of children and adults, since even within these two groups there are some differences. However, in general protocols are comparable and start with induction therapy, using multiple chemotherapeutic drugs, followed by consolidation and maintenance.

26


Both groups receive CNS prophylaxis, usually intrathecal chemotherapy

and medium- or high dose MTX intravenously. The chemotherapeutic

drugs utilised are also similar, and include vincristine, dexamethasone,

daunorubicin, doxorubicin, MTX and 6-mercaptopurine. Some drugs are

more specifically used in adults, such as etoposide and mitoxantrone.

Older patients have a lower tolerability towards the applied chemotherapy

than younger patients, due to comorbidity and increased haematological and

non-haematological toxicity139• One study compared the rate of death in remission,

an indicator of treatment related toxicity, in different age groups and observed

an increase of deaths in remission with age21 • Non-haematological toxicity,

such as hepatic and cardiotoxicity leads to higher morbidity and mortality in

adults. Increased haematological toxicity causes a higher incidence of life

threatening infections in adults and treatment delays49;140• When comparing

protocol compliance between children and adults, it has been noticed that

there is a stricter compliance in paediatric patients than in adults21 ;44• Not

only treatments have to be postponed, sometimes the dose has to be reduced

as a result of these toxicities. There is evidence that dose reduction leads

to inadequate drug delivery, which might contribute to the higher relapse

rate in adults1 39; 14 1 ; 142• Studies in adolescents, who were treated according to

childhood or adult protocols, observed that adolescent patients benefit more

from childhood protocols than adult protocols as shown by higher event-free

survival rates 143; 144• In brief, the lower tolerance of chemotherapy and, as a consequence, less

strict protocol compliance in adults might enhance the worse clinical outcome

of adult patients in comparison with children.

Minimal residual disease (MRD)

The ability to detect and quantitate malignant cells in patients in complete

remission has permitted analysis of the effectiveness of various treatments

and the biology of the disease during and after treatment. PCR methods

based on identification of clonal rearrangements of the Ig heavy chain or

T-cell receptor offer an extremely sensitive technique by which one tumour

cell can be detected among 105-6 normal cells145 • The MRD level provides

a measure of in vivo resistance of the overall leukaemic population to the

chemotherapeutic agents, and therefore MRD levels are associated with

relapse and can strongly predict outcome. Several studies have shown that a

single time point for MRD detection is sufficient for identifying patients with

a poor prognosis 146-148• However, another study demonstrated that monitoring

27

Chapter 1

patients at consecutive time points gives greater insight into the recognition of

patients at risk for relapse149• Although most experience with longitudinal and

quantitative detection ofMRD exists in childhood ALU46;149, two studies have

compared the MRD levels between children and adults 150;1 5 1 • As expected by

the therapeutic outcome, a higher level ofMRD was observed in adults versus

children, also after stratifying for known risk factors. The higher MRD levels

imply that adults with ALL are more resistant to therapy than children.

DISCUSSION

This review has compared the various characteristics of childhood and adult ALL in order to find an explanation for the difference in prognosis. In view of

the hypothesis as described by Greaves4;9, the characteristics were compared

with those of the stem cells and committed progenitors. Adolescents form an

interesting group since they constitute a mixture and their prognosis is indeed

intermediate between that of children and adults.

Assuming that the primitive leukaemic stem cells in adult ALL share features with stem cells, it is possible that they can only be eliminated

by chemotherapy at a level that also kills normal haematopoietic stem cell

populations. Moreover, their self-renewing capacity results in an increased

likelihood that drug resistant mutants will exist by the time that treatment is

initiated. Both factors can explain the worse clinical outcome of adults with

ALL in comparison with children. Most of the differences in characteristics found between adults and children

can fit the hypothesis that adult ALL originates from stem cells and childhood

ALL from committed lymphoid progenitors. However, it should be noted that

one could reason from the perspective of clonogenic leukaemic stem cells and

from the perspective of their more mature progeny. The bulk of the leukaemic cells (the mature progeny) may share the same characteristics in children and

adults, whereas the clonogenic leukaemic stem cells may differ.

From a physiological perspective, it is possible that stem cells possess

mechanisms, to protect themselves from toxic substances or other potentially

dangerous influences. Stem cells are indispensable for normal haematopoiesis, whereas committed lymphoid progenitors can develop from stem cells. What kind of clinical implications does this hypothesis have? Intensification

of adult ALL treatment protocols was assumed to improve clinical outcome.

Unfortunately an increase in treatment-related toxicity and toxic deaths

has been observed, which underscores the detrimental effect of intensified

28


chemotherapy on normal haematopoietic stem cells. Thus further studies are

required to identify the specific characteristics of leukaemic stem cells, which

are not displayed by normal stem cells. These characteristics can possibly

be identified by performing gene expression profiling. The development

of oligonucleotide microarrays has made it possible to readily analyse the

patterns of genes expressed, and compare childhood and adult ALL samples.

This could allow novel treatment strategies to be developed, making use

of these typical characteristics. The BCR-ABL tyrosine kinase inhibitor

STI571 is an important example of how targeting a specific characteristic of

leukaemic cells can have a dramatic impact on treatment strategies152•

29

Chapter 1

REFERENCES

1 . v an Dijck JAAM, Coebergh JWW, Siesling S, Visser OE. Trends of c ancer in the Netherl ands 1 989-1 998. Utrecht: Vereniging v an Integr ale K ankercentr a, 2002.

2 . Schr appe M, C amitt a B, Pui CH et a/. Long-term results of l arge prospective tri als in childhood acute lymphobl astic leukemi a. Leukemia 2000: 14: 2 1 93-2 1 94_

3 . Gokbuget N, Hoelzer D. Recent appro aches in acute lymphobl astic leukemi a in adults. Rev Clin Exp Hematol 2002: 6: 1 14- 1 4 1 .

4. Gre aves MF. Stem cell origins of leuk aemi a and cur ability. Br J Cancer 1 993: 67: 413-423.

5. Meyers PA. M align ant bone tumors in children: osteos arcom a. In: Pochedly C ed, Hem atology/Oncology Clinics of North Americ a: C ancer in Children. Phil adelphi a:W.B.S aunders, 2003, 655-666.

6. Gre aves MF. Differenti ation-linked leukemogenesis in lymphocytes. Science 1 986: 234: 697-704.

7. Knudson AG. Two genetic hits (more or less) to c ancer. Nat Rev Cancer 200 1 : 1 : 1 57- 1 62.

8. Knudson AG, Jr. Genetics and etiology of hum an c ancer. Adv Hum Genet 1 977: 8: 1 -66.

9. Gre aves M. Molecul ar genetics, n atur al history and the demise of childhood leuk aemi a. Eur J Cancer 1 999: 35: 1 94 1 - 1953.

1 0. Mori H, Colm an SM, Xi ao Z et a!. Chromosome tr ansloc ations and covert leukemic clones are gener ated during norm al fet al development. Proc Nat! Acad Sci USA 2002: 99: 8242-8247.

1 1 . Ford AM, Bennett CA, Price CM, Bruin MC, v an Wering ER, Gre aves M. Fet al origins of the TEL-AMLl fusion gene in identic al twins with leukemi a. Proc Nat! Acad Sci USA 1 998: 95: 4584-4588.

1 2. M ai a AT, Ford AM, J al ali GR et a!. Molecul ar tr acking of leukemogenesis in a triplet pregn ancy. Blood 200 1 : 98: 478-482.

13 . G ale KB, Ford AM, Repp R et a!. B acktr acking leukemi a to birth: identific ation of clonotypic gene fusion sequences in neon at al blood spots. Proc Nat! Acad Sci U S A 1 997: 94: 13950-13954.

14. Wiemels JL, C azz anig a G, D aniotti M et a!. Pren at al origin of acute lymphobl astic leuk aemi a in children. Lancet 1999: 354: 1499- 1 503 .

1 5. Wiemels JL, Ford AM, v an Wering ER, Postm a A, Gre aves M. Protr acted and v ari able l atency of acute lymphobl astic leukemi a after TEL-AMLl gene fusion in utero. Blood 1 999: 94: 1 057- 1062.

1 6. T aylor GM, De arden S, R avetto P et a!. Genetic susceptibility to childhood common acute lymphobl astic leuk aemi a is associ ated with polymorphic peptide-binding pocket profiles in HLA-DPB l *020 1 . Hum Mol Genet 2002: 1 1 : 1 585-1 597.

1 7. Hotfilder M, Rottgers S, Rosem ann A, Jurgens H, H arbott J, Vormoor J. Imm ature CD34+CD 1 9- progenitor/stem cells in TEL/ AML l -positive acute lymphobl astic leukemi a are genetic ally and function ally norm al. Blood 2002: 100: 640-646 .

1 8 . C astor A, Nilsson L, Astr and-Grundstrom I et a!. The hem atopoietic stem cell pool is norm al in size and function in children with acute lymphobl astic leukemi a and 12;2 1 tr ansloc ation, but not in Phil adelphi a positive p atients. Blood 2002: 100: 1 52 a (Abstr act 571) .

30


1 9. Kasprzyk A, Harriso n CJ, Seeker-Walker LM. I nvestigatio n o f clo nal i nvolveme nt o f myeloid cells i n Philadelphia- positive a nd high hyperdiploid acute lymphoblastic leukemia. Leukemia 1 999: 13: 2000-2006.

20. Sche nk TM, Keyha ni A, Bottcher S et a/. Multili neage i nvolveme nt o f Philadelphia chromosome positive acute lymphoblastic leukemia. Leukemia 1 998: 12 : 666-674.

2 1 . Chessells JM, Hall E, Pre ntice HG, Durra nt J, Bailey CC, Richards SM. The impact o f age o n outcome i n lymphoblastic leukaemia; MRC UKALL X a nd XA compared: a report from the MRC Paediatric a nd Adult Worki ng Parties. Leukemia 1 998: 12 : 463-473.

22. Norrback KF, Roos G. Telomeres a nd telomerase i n normal a nd malig na nt haematopoietic cells. Eur J Cancer 1997: 33: 774-780.

23. Ohyashiki JH, Ohyashiki K, lwama H, Hayashi S, Toyama K, Shay JW. Cli nical implicatio ns o ftelomerase activity levels i n acute leukemia. C/in Cancer Res 1 997: 3: 6 1 9-625.

24. Ohyashiki K, Ohyashiki JH. Telomere dy namics a nd cytoge netic cha nges i n huma n hematologic neoplasias: a worki ng hypothesis. Cancer Genet Cytogenet 1 997: 94: 67-72.

25. Ohyashiki JH, Sashida G, Tauchi T, Ohyashiki K. Telomeres a nd telomerase i n hematologic neoplasia. Oncogene 2002: 2 1 : 680-687.

26. Cou nter CM, Gupta J, Harley CB, Leber B, Sacchetti S. Telomerase activity i n normal leukocytes a nd i n hematologic malig na ncies. Blood 1 995: 85: 23 1 5-2320.

27. Visser 0, Coebergh JWW, va n Dijck JAAM, Siesli ng S. I ncide nce o f ca ncer i n the Netherla nds 1 998. Utrecht: Vere nigi ng va n l ntegrale Ka nkerce ntra, 2002.

28. Cartwright RA, Gurney KA, Moorma n AV. Sex ratios a nd the risks o fhaematological malig na ncies. Br J Haemato/ 2002: 1 1 8: 1 07 1 - 1 077.

29. Gay no n PS, Trigg ME, Heerema NA et a/. Childre n's Ca ncer Group trials i n childhood acute lymphoblastic leukemia: 1 983-1 995. Leukemia 2000: 14: 2223-2233.

30. Harms DO, Ja nka-Schaub GE. Co-operative study group for childhood acute lymphoblastic leukemia (COALL): lo ng-term follow-up o f trials 82, 85, 89 a nd 92. Leukemia 2000: 14: 2234-2239.

3 1 . lmbach P, Fuchs A, Berchtold W et a/. Boys but not girls with T-li neage acute lymphocytic leukemia (ALL) are differe nt from childre n with B-proge nitor ALL. Populatio n-based data results o f i nitial prog nostic factors a nd lo ng-term eve nt- free survival. Swiss Pediatric O ncology Group. J Pediatr Hematol Oncol 1 995 : 17: 346-349.

32. Pui CH, Boyett JM, Rivera GK et a/. Lo ng-term results o f Total Therapy studies 1 1 , 12 a nd 13A for childhood acute lymphoblastic leukemia at St Jude Childre n's Research Hospital. Leukemia 2000: 14: 2286-2294.

33. Silverma n LB, Declerck L, Gelber RD et a/. Results o f Da na-Farber Ca ncer I nstitute Co nsortium protocols for childre n with newly diag nosed acute lymphoblastic leukemia ( 198 1 - 1 995). Leukemia 2000: 14: 2247-2256.

34. Kamps WA, Veerma n AJ, va n Weri ng ER, va n Weerde n JF, Slater R, Does-va n de n Berg A. Lo ng-term follow-up o f Dutch Childhood Leukemia Study Group (DCLSG) protocols for childre n with acute lymphoblastic leukemia, 1 984- 1 99 1 . Leukemia 2000: 14: 2240-2246.

35. Ede n OB, Lilleyma n JS, Richards S. Testicular irradiatio n i n childhood lymphoblastic leukaemia. Medical Research Cou ncil Worki ng Party o n Leukemia i n Childhoods. Br J

31

Chapter 1

Haematol. 1 990: 75: 496-498. 36. Chessells JM, Richards SM, Bailey CC, Lilleyman JS, Eden OB. Gender and treatment

outcome in childhood lymphoblastic leukaemia: report from the MRC UKALL trials. Br J Haematol. 1 995: 89: 364-372.

37. Klemetsdal B, Flaegstad T, Aarbakke J. Is there a gender difference in red blood cell thiopurine methyl transferase activity in healthy children? Med Pediatr Oneal. 1 995: 25: 445-449.

38. Hale JP, Lilleyman JS. Importance of 6-mercaptopurine dose in lymphoblastic leukaemia. Arch Dis Child 1 99 1 : 66: 462-466.

39. Mandelli F, Annino L, Rotoli B. The GIMEMA ALL 0 1 83 trial: analysis of 1 0-year follow-up. GIMEMA Cooperative Group, Italy. Br J Haematol 1 996: 92: 665-672.

40. Fiere D, Lepage E, Sebban C et a/. Adult acute lymphoblastic leukemia: a multicentric randomized trial testing bone marrow transplantation as postremission therapy. The French Group on Therapy for Adult Acute Lymphoblastic Leukemia. J Clin Onco/ 1 993 : 1 1 : 1 990-2001 .

4 1 . Larson RA, Dodge RK, Bums CP et a/. A five-drug remission induction regimen with intensive consolidation for adults with acute lymphoblastic leukemia: cancer and leukemia group B study 88 1 1 . Blood 1 995: 85: 2025-2037.

42. Ribera JM, Ortega JJ, Oriol A et a/. Late intensification chemotherapy has not improved the results of intensive chemotherapy in adult acute lymphoblastic leukemia. Results of a prospective multicenter randomized trial (PETHEMA ALL-89). Spanish Society of Hematology. Haematologica 1998: 83: 222-230.

43. Dekker AW, van't Veer MB, Sizoo W et a!. Intensive postremission chemotherapy without maintenance therapy in adults with acute lymphoblastic leukemia. Dutch Hemato-Oncology Research Group. J Clin Onco/ 1 997: 15: 476-482.

44. Durrant IJ, Prentice HG, Richards SM. Intensification of treatment for adults with acute lymphoblastic leukaemia: results of U.K. Medical Research Council randomized trial UKALL XA. Medical Research Council Working Party on Leukaemia in Adults. Br J Haemato/ 1 997: 99: 84-92.

45. Greaves M. A natural history for pediatric acute leukemia. Blood 1 993: 82: 1 043-1 05 1 . 46. Pui CH. Childhood leukemias. N Eng! J Med 1 995 : 332 : 1 6 1 8- 1 630. 47. Alimena G, Billstrom R, Casalone R, Gallo E, Mitelman F, Pasquali F. Cytogenetic

pattern in leukemic cells of patients with constitutional chromosome anomalies. Cancer Genet Cytogenet 1 985: 16: 207-2 1 8.

48. Rassool FV. DNA double strand breaks (DSB) and non-homologous end joining (NHEJ) pathways in human leukemia. Cancer Lett 2003 : 193: 1 -9.

49. Hoelzer D, Gokbuget N. Recent approaches in acute lymphoblastic leukemia in adults. Crit Rev Oneal Hemato/ 2000: 36: 49-58.

50. Pui CH, Campana D, Evans WE. Childhood acute lymphoblastic leukaemia-current status and future perspectives. Lancet Onco/ 200 1 : 2: 597-607.

5 1 . Simone JV, Verzosa MS, Rudy JA. Initial features and prognosis in 363 children with acute lymphocytic leukemia. Cancer 1 975 : 36: 2099-2 1 08 .

52. Smith M, Arthur D, Camitta B et a!. Uniform approach to risk classification and treatment assignment for children with acute lymphoblastic leukemia. J Clin Oneal 1 996: 14: 1 8-24.

53. Conter V, Arico M, Valsecchi MG et a!. Long-term results of the Italian Association of Pediatric Hematology and Oncology (AIEOP) acute lymphoblastic leukemia studies,

32


1 982- 1995. Leukemia 2000: 14: 2 196-2204.

54. Schrappe M, Reiter A, Zimmermann M et al. Long-term results of four consecutive trials in childhood ALL performed by the ALL-BFM study group from 1981 to 1 995. Berlin-Frankfurt-Munster. Leukemia 2000: 14: 2205-2222.

55. Durrant IJ, Richards SM. Results of Medical Research Council trial UKALL IX in acute lymphoblastic leukaemia in adults: report from the Medical Research Council Working Party on Adult Leukaemia. Br J Haemato/ 1 993: 85: 84-92.

56. Ellison RR, Mick R, Cuttner J et al. The effects of postinduction intensification treatment with cytarabine and daunorubicin in adult acute lymphocytic leukemia: a prospective randomized clinical trial by Cancer and Leukemia Group B. J Clin Oncol 1 99 1 : 9: 2002-20 15 .

57 . Langermann HJ, Henze G, WulfM, Riehm H. Estimation of tumor cell mass in childhood acute lymphoblastic leukemia: prognostic significance and practical application. Klin Padiatr 1 982: 194: 209-2 13 .

58. Schrappe M, Beck J, Brandeis WE et al. Treatment of acute lymphoblastic leukemia in childhood and adolescence: results of the multicenter therapy study ALL-BFM 8 1 . Klin Padiatr 1 987: 199: 133-1 50.

59. Riehm H, Feickert HJ, Schrappe M, Henze G, Schellong G. Therapy results in five ALL-BFM studies since 1 970: implications of risk factors for prognosis. Hamatol Bluttransfus 1 987: 30: 139- 146.

60. Mastrangelo R, Poplack D, Bleyer A, Riccardi R, Sather H, D' Angio G. Report and recommendations of the Rome workshop concerning poor- prognosis acute lymphoblastic leukemia in children: biologic bases for staging, stratification, and treatment. Med Pediatr Onco/ 1 986: 14: 1 9 1 - 1 94.

6 1 . Gokbuget N, Hoelzer D. Meningeosis leukaemica in adult acute lymphoblastic leukaemia. J Neuroonco/ 1 998: 38: 1 67- 1 80.

62. Bleyer WA, Poplack DG. Prophylaxis and treatment of leukemia in the central nervous system and other sanctuaries. Semin Onco/ 1 985: 12: 13 1 - 148.

63. Chessells JM. Central nervous system directed therapy in acute lymphoblastic leukaemia. Baillieres Clin Haemato/ 1 994: 7: 349-363.

64. Pinkel D, Woo S. Prevention and treatment of meningeal leukemia in children. Blood 1 994: 84: 355-366.

65. Gaynon PS, Qu RP, Chappell RJ et a/. Survival after relapse in childhood acute lymphoblastic leukemia: impact of site and time to first relapse--the Children's Cancer Group Experience. Cancer 1 998: 82: 1 387-1 395.

66. Bennett JM, Catovsky D, Daniel MT et a/. The morphological classification of acute lymphoblastic leukaemia: concordance among observers and clinical correlations. Br J

Haemato/ 1 98 1 : 47: 553-561 . 67. Seeker-Walker LM. Chromosomes and genes in acute lymphoblastic leukemia.

Georgetown: Landes Bioscience, 1 997. 68. Miller DR, Leikin S, Albo V, Sather H, Hammond D. Prognostic importance of

morphology (FAB classification) in childhood acute lymphoblastic leukaemia (ALL). Br J Haemato/ 1 98 1 : 48: 1 99-206.

69. Lilleyman JS, Hann IM, Stevens RF et al. Cytomorphology of childhood lymphoblastic leukaemia: a prospective study of 2000 patients. United Kingdom Medical Research Council's Working Party on Childhood Leukaemia. Br J Haematol 1 992: 81: 52-57.

70. Kanerva J, Saarinen-Pihkala UM, Riikonen P, Makipernaa A, Mottonen M, Salmi TT.

33

Chapter 1

Reemphasis on lymphoblast L2 morphology as a poor prognostic factor in childhood acute lymphoblastic leukemia. Med Pediatr Onco/ 1 999: 33: 388-394.

7 1 . Davey FR, Castella A, Lauenstein K, Hubbell C, Oates RP. Prognostic significance of the revised French-American-British classification for acute lymphocytic leukaemia. Clin Lab Haemato/ 1 983: 5: 343-35 1 .

72. Pui CH, Evans WE. Acute lymphoblastic leukemia. N Eng/ J Med 1 998 : 339 : 605-6 1 5 . 73. Patte C, Auperin A, Michon J et a/. The Societe Francaise d'Oncologie Pediatrique

LMB89 protocol : highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 56 1 unselected children with B-cell lymphomas and L3 leukemia. Blood 200 i : 97: 3370-3379.

74. Vilmer E, Suciu S, Ferster A et a/. Long-term results of three randomized trials (5883 1 , 58832, 5888 1 ) in childhood acute lymphoblastic leukemia: a CLCG-EORTC report. Children Leukemia Cooperative Group. Leukemia 2000: 14: 2257-2266.

75. Attal M, Blaise D, Marit G et a/. Consolidation treatment of adult acute lymphoblastic leukemia: a prospective, randomized trial comparing allogeneic versus autologous bone marrow transplantation and testing the impact of recombinant interleukin-2 after autologous bone marrow transplantation. BGMT Group. Blood 1 995 : 86: 1 6 1 9- 1 628.

76. Myeloid-antigen expression in childhood acute lymphoblastic leukemia. N Eng/ J Med

1 99 1 : 325: 1 378- 1 382. 77. Hann IM, Richards SM, Eden OB, Hill FG. Analysis of the immunophenotype of

children treated on the Medical Research Council United Kingdom Acute Lymphoblastic Leukaemia Trial XI (MRC UKALLXI). Medical Research Council Childhood Leukaemia Working Party. Leukemia 1 998: 12: 1 249- 1255.

78. Kurec AS, Belair P, Stefanu C, Barrett DM, Dubowy RL, Davey FR. Significance of aberrant immunophenotypes in childhood acute lymphoid leukemia. Cancer 1 99 1 : 67: 308 1 -3086.

79. Pui CH, Behm FG, Crist WM. Clinical and biologic relevance of immunologic marker studies in childhood acute lymphoblastic leukemia. Blood 1993 : 82: 343-362.

80. Wiersma SR, Ortega J, Sobel E, Weinberg Kl. Clinical importance of myeloid-antigen expression in acute lymphoblastic leukemia of childhood. N Eng/ J Med 1 99 1 : 324: 800-808.

8 1 . Seeker-Walker LM, Chessells JM, Stewart EL, Swansbury GJ, Richards S, Lawler SD. Chromosomes and other prognostic factors in acute lymphoblastic leukaemia: a longterm follow-up. Br J Haemato/ 1 989: 72: 336-342.

82. Whitehead VM, Vuchich MJ, Lauer SJ et a/. Accumulation ofhigh levels of methotrexate polyg1utamates in lymphob1asts from children with hyperdiploid (greater than 50 chromosomes) B-lineage acute lymphoblastic leukemia: a Pediatric Oncology Group study. Blood 1 992 : 80: 1 3 1 6- 1 323.

83. Kaspers GJ, Smets LA, Pieters R, Van Zantwijk CH, van Wering ER, Veerman AJ. Favorable prognosis of hyperdiploid common acute lymphoblastic leukemia may be explained by sensitivity to antimetabolites and other drugs: results of an in vitro study. Blood 1995: 85: 75 1 -756.

84. Seeker-Walker LM, Craig JM, Hawkins JM, Hoffbrand AV. Philadelphia positive acute lymphoblastic leukemia in adults: age distribution, BCR breakpoint and prognostic significance. Leukemia 1 99 1 : 5: 1 96- 1 99.

85. Pui CH, Kane JR, Crist WM. Biology and treatment of infant leukemias. Leukemia 1 995: 9: 762-769.

34


86. Rubnitz JE, Link MP, Shuster JJ et a/. Frequency and prognostic significance of HRX rearrangements in infant acute lymphoblastic leukemia: a Pediatric Oncology Group study. Blood 1 994: 84: 570-573.

87. Barr FG. Translocations, cancer and the puzzle of specificity. Nat Genet 1 998: 19: 1 2 1 - 124.

88. Hongo T, Yajima S, Sakurai M, Horikoshi Y, Hanada R. In vitro drug sensitivity testing can predict induction failure and early relapse of childhood acute lymphoblastic leukemia. Blood 1 997: 89: 2959-2965.

89. Kaspers GJ, Veerman AJ, Pieters R et a/. In vitro cellular drug resistance and prognosis in newly diagnosed childhood acute lymphoblastic leukemia. Blood 1 997: 90: 2723-2729.

90. Pieters R, Huismans DR, Loonen AH et a/. Relation of cellular drug resistance to longterm clinical outcome in childhood acute lymphoblastic leukaemia. Lancet 1 99 1 : 338: 399-403.

9 1 . Pieters R, Kaspers GJ, van Wering ER et a/. Cellular drug resistance profiles that might explain the prognostic value of immunophenotype and age in childhood acute lymphoblastic leukemia. Leukemia 1 993 : 7: 392-397.

92. Pieters R, den Boer ML, Durian M et a/. Relation between age, immunophenotype and in vitro drug resistance in 395 children with acute lymphoblastic leukemia--implications for treatment of infants. Leukemia 1 998: 12: 1 344- 1348.

93. Styczynski J, Wysocki M. In vitro drug resistance profiles of adult acute lymphoblastic leukemia: possible explanation for difference in outcome to similar therapeutic regimens. Leuk Lymphoma 2002: 43: 30 l -307.

94. Maung ZT, Reid MM, Matheson E, Taylor PR, Proctor SJ, Hall AG. Corticosteroid resistance is increased in lymphoblasts from adults compared with children: preliminary results of in vitro drug sensitivity study in adults with acute lymphoblastic leukaemia. Br J Haemato/ 1 995: 91 : 93- 100.

95. Styczynski J, Pieters R, Huismans DR, Schuurhuis GJ, Wysocki M, Veerman AJ. In vitro drug resistance profiles of adult versus childhood acute lymphoblastic leukaemia. Br J Haemato/ 2000: 110: 8 1 3-8 1 8.

96. Gottesman MM, Fojo T, Bates SE. Multidrug resistance in cancer: role of ATPdependent transporters. Nat Rev Cancer 2002: 2 : 48-58.

97. Chauncey TR. Drug resistance mechanisms in acute leukemia. Curr Opin Oneal 200 1 : 13 : 2 1 -26.

98. Marie JP, Legrand 0. MDRl /P-GP expression as a prognostic factor in acute leukemias. Adv Exp Med Bio/ 1 999: 457: 1 -9.

99. Dhooge C, De Moerloose B, Laureys G et a/. Expression of the multidrug transporter P-glycoprotein is highly correlated with clinical outcome in childhood acute lymphoblastic leukemia: results of a long-term prospective study. Leuk Lymphoma 2002: 43 : 309-3 14.

1 00. Tafuri A, Gregmj C, Petrucci MT et a/. MDR1 protein expression is an independent predictor of complete remission in newly diagnosed adult acute lymphoblastic leukemia. Blood 2002: 100: 974-98 1 .

1 0 1 . den Boer ML, Pieters R, Kazemier KM et a/. Relationship between major vault protein/lung resistance protein, multidrug resistance-associated protein, P-glycoprotein expression, and drug resistance in childhood leukemia. Blood 1 998: 91: 2092-2098.

102. Dhooge C, De Moerloose B, Laureys G et a/. P-glycoprotein is an independent

35

Chapter 1

prognostic factor predicting relapse in childhood acute lymphoblastic leukaemia: results of a 6-year prospective study. Br J Haemato/ 1 999: 105: 676-683.

1 03. Goasguen JE, Dossot JM, Fardel 0 et a!. Expression of the multidrug resistanceassociated P-glycoprotein (P- 1 70) in 59 cases of de novo acute lymphoblastic leukemia: prognostic implications. Blood 1 993: 81 : 2394-2398.

1 04. Ivy SP, Olshefski RS, Taylor BJ, Patel KM, Reaman GH. Correlation of P-glycoprotein expression and function in childhood acute leukemia: a children's cancer group study. Blood 1 996: 88: 309-3 1 8.

1 05 . Wattel E, Lepelley P, Merlat A et a!. Expression of the multidrug resistance P glycoprotein in newly diagnosed adult acute lymphoblastic leukemia: absence of correlation with response to treatment. Leukemia 1995 : 9: 1 870-1 874.

1 06. Wuchter C, Leonid K, Ruppert V et a!. Clinical significance ofP-glycoprotein expression and function for response to induction chemotherapy, relapse rate and overall survival in acute leukemia. Haematologica 2000: 85: 7 1 1 -72 1 .

1 07. Plasschaert SLA., Vellenga E , de Bont ESJM et a!. High functional P-glycoprotein activity is more often present in T-cell acute lymphoblastic leukemic cells in adults than in children. Leuk Lymphoma 2003: 44: 85-95 .

1 08. Sauerbrey A, Voigt A, Wittig S, Hafer R, Zintl F. Messenger RNA analysis of the multidrug resistance related protein (MRP 1 ) and the lung resistance protein (LRP) in de novo and relapsed childhood acute lymphoblastic leukemia. Leuk Lymphoma 2002: 43 : 875-879.

1 09. Chen ZS, Lee K, Kruh GD. Transport of cyclic nucleotides and estradiol 1 7-beta-Dglucuronide by multidrug resistance protein 4. Resistance to 6-mercaptopurine and 6-thioguanine. J Bioi Chern 200 1 : 276: 33747-33754.

l l 0. Jedlitschky G, Burchell B, Keppler D. The multi drug resistance protein 5 functions as an ATP-dependent export pump for cyclic nucleotides. J Bioi Chem 2000: 275: 30069-30074.

1 1 1 . Allikmets R, Schriml LM, Hutchinson A, Romano-Spica V, Dean M. A human placentaspecific ATP-binding cassette gene (ABCP) on chromosome 4q22 that is involved in multidrug resistance. Cancer Res 1998: 58: 5337-5339.

1 12. Bunting KD. ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 2002: 20: 1 1 -20.

1 13 . Chaudhary PM, Roninson lB. Expression and activity of P-glycoprotein, a multidrug efflux pump, in human hematopoietic stem cells. Cell l 99 l : 66: 85-94.

1 14. Scharenberg CW, Harkey MA, Torok-Storb B . The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 2002: 99: 507-5 1 2.

1 1 5 . Zhou S, Schuetz JD, Bunting KD et a!. The ABC transporter Bcrp1 /ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 200 l : 7: 1 028- 1034.

1 1 6. Galpin AJ, Schuetz JD, Masson E et a!. Differences in folylpolyglutamate synthetase and dihydrofolate reductase expression in human B-lineage versus T-lineage leukemic lymphoblasts: mechanisms for lineage differences in methotrexate polyglutamylation and cytotoxicity. Mol Pharmaco/ 1 997: 52: 1 55- 163.

1 17 . Goker E, Lin JT, Trippett T et a!. Decreased polyglutamylation of methotrexate in acute lymphoblastic leukemia blasts in adults compared to children with this disease. Leukemia 1 993 : 7: 1 000- 1 004.

36


1 1 8 . Classen CF, Fulda S, Friesen C, Debatin KM. Decreased sensitivity of drug-resistant cells towards T cell cytotoxicity. Leukemia 1 999: 13: 4 1 0-41 8.

1 1 9. Frost P, Ng CP, Belldegrun A, Bona vida B. Immunosensitization of prostate carcinoma cell lines for lymphocytes (CTL, TIL, LAK)-mediated apoptosis via the Fas-Fas-ligand pathway of cytotoxicity. Cell /mmuno/ 1 997: 180: 70-83.

1 20. Strand S, Hofmann WJ, Hug H et a/. Lymphocyte apoptosis induced by CD95 (AP0-1 /Fas) ligand-expressing tumor cells--a mechanism of immune evasion? Nat Med 1 996: 2: 1 36 1 - 1 366.

1 2 1 . Posovszky C, Friesen C, Herr I, Debatin KM . Chemotherapeutic drugs sensitize pre-B ALL cells for CD95- and cytotoxic T-lymphocyte-mediated apoptosis. Leukemia 1 999: 13: 400-409.

1 22. Friesen C, Herr I, Krammer PH, Debatin KM. Involvement of the CD95 (AP0- 1/FAS) receptor/ligand system in drug- induced apoptosis in leukemia cells. Nat Med 1 996: 2 : 574-577.

1 23 . Eischen CM, Kottke TJ, Martins LM et a/. Comparison of apoptosis in wild-type and Pas-resistant cells: chemotherapy-induced apoptosis is not dependent on Fas/Fas ligand interactions. Blood 1 997: 90: 935-943.

124. Villunger A, Egle A, Kos M et a/. Drug-induced apoptosis is associated with enhanced Fas (Apo-l /CD95) ligand expression but occurs independently of Fas (Apol /CD95) signaling in human T-acute lymphatic leukemia cells. Cancer Res 1 997: 57: 333 1 -3334.

1 25 . Bettinger C, Kurz E, Bohler T, Schrappe M, Ludwig WD, Debatin KM. CD95 (AP0-1 /Fas) mutations in childhood T-lineage acute lymphoblastic leukemia. Blood 1 998: 91 : 3943-395 1 .

126. Karawajew L, Wuchter C, Ruppert V et a/. Differential CD95 expression and function in T and B lineage acute lymphoblastic leukemia cells. Leukemia 1 997: 11 : 1 245-1 252.

1 27. Volm M, Zintl F, Sauerbrey A, Koomagi R. Proliferation and apoptosis in newly diagnosed and relapsed childhood acute lymphoblastic leukemia. Anticancer Res 1 999: 19: 4327-433 1 .

128. Bettinger C, Bohler T, Karawajew L, Ludwig WD, Schrappe M, Debatin KM. Mutation analysis of CD95 (AP0-1 /Fas) in childhood B-lineage acute lymphoblastic leukaemia. Br J Haemato/ 1 998: 102: 722-728.

129. Tamiya S, Etoh K, Suzushima H, Takatsuki K, Matsuoka M. Mutation of CD95 (Fas/ Apo- 1 ) gene in adult T-cell leukemia cells. Blood 1 998: 91 : 3935-3942.

1 30. Debatin KM, Krammer PH. Resistance to AP0-1 (CD95) induced apoptosis in T-ALL is determined by a BCL-2 independent anti-apoptotic program. Leukemia 1 995 : 9: 8 1 5-820.

1 3 1 . Campos L, Sabido 0, Sebban C et a/. Expression of BCL-2 proto-oncogene in adult acute lymphoblastic leukemia. Leukemia 1 996: 10: 434-438.

1 32. Coustan-Smith E, KitanakaA, Pui CH et a/. Clinical relevance ofBCL-2 overexpression in childhood acute lymphoblastic leukemia. Blood 1 996: 87: 1 140- 1 146.

1 33. Gala JL, Vermylen C, Cornu G et a/. High expression of bcl-2 is the rule in acute lymphoblastic leukemia, except in Burkitt subtype at presentation, and is not correlated with the prognosis. Ann Hemato/ 1 994: 69: 1 7-24.

1 34. Hogarth LA, Hall AG. Increased BAX expression is associated with an increased risk of relapse in childhood acute lymphocytic leukemia. Blood 1 999: 93: 267 1 -2678.

135 . Prokop A, Wieder T, Sturm I et a/. Relapse in childhood acute lymphoblastic leukemia

37

Chapter 1

is associated with a decrease of the Bax/Bcl-2 ratio and loss of spontaneous caspase-3 processing in vivo. Leukemia 2000: 14: 1 606-1 6 1 3.

1 36. Salomons GS, Smets LA, Verwijs-Janssen M et a/. Bcl-2 family members in childhood acute lymphoblastic leukemia: relationships with features at presentation, in vitro and in vivo drug response and long-term clinical outcome. Leukemia 1 999: 13: 1 574- 1580.

1 37. Uckun FM, Yang Z, Sather H et a/. Cellular expression of antiapoptotic BCL-2 oncoprotein in newly diagnosed childhood acute lymphoblastic leukemia: a Children's Cancer Group Study. Blood 1 997: 89: 3769-3777.

138 .. Del Poeta G, Venditti A, Del Principe MI et a/. Amount of spontaneous apoptosis detected by Bax/Bcl-2 ratio predicts outcome in acute myeloid leukemia (AML). Blood 2003: 101: 2 1 25-2 1 3 1 .

1 39. Hoelzer D . Aggressive chemotherapy of ALL in elderly patients. Hematol Onco/ 1 993: ll(Suppl): 1 2- 14.

140. Copelan EA, McGuire EA. The biology and treatment of acute lymphoblastic leukemia in adults. Blood 1 995: 85: 1 1 5 1- 1 1 68.

1 4 1 . Bassan R, Lerede T, Barbui T. Institutional performance and dose intensity as prognostic factors in adult ALL. Leukemia 1995: 9: 933-934.

142. Chiu EK, Chan LC, Liang R et a!. Poor outcome of intensive chemotherapy for adult acute lymphoblastic leukemia: a possible dose effect. Leukemia 1 994: 8: 1469- 1473.

143. Nachman J, Sather HN, Buckley JD et a/. Young adults 1 6-2 1 years of age at diagnosis entered on Childrens Cancer Group acute lymphoblastic leukemia and acute myeloblastic leukemia protocols. Results of treatment. Cancer 1 993 : 71 : 3377-3385.

144. Rivera GK, Pui CH, Santana VM et a!. Progress in the treatment of adolescents with acute lymphoblastic leukemia. Cancer 1 993 : 71 : 3400-3405 .

145. van Dongen JJ, Macintyre EA, Gabert JA et a!. Standardized RT-PCR analysis of fusion gene transcripts from chromosome aberrations in acute leukemia for detection of minimal residual disease. Report of the BIOMED-1 Concerted Action: investigation of minimal residual disease in acute leukemia. Leukemia 1 999: 13 : 190 1 - 1 928.

146. Brisco MJ, Condon J, Hughes E et a!. Outcome prediction in childhood acute lymphoblastic leukaemia by molecular quantification of residual disease at the end of induction. Lancet 1994: 343: 1 96-200.

147. Cave H, van der Werff ten Bosch, Suciu S et a!. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer--Childhood Leukemia Cooperative Group. N Eng/ J Med 1 998: 339: 59 1 -598.

148. Coustan-Smith E, Behm FG, Sanchez J et a!. Immunological detection of minimal residual disease in children with acute lymphoblastic leukaemia. Lancet 1 998: 351 : 550-554.

149. van Dongen JJ, Seriu T, Panzer-Grumayer ER et a!. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet 1 998: 352: 1 73 1- 1 738.

1 50. Brisco J, Hughes E, Neoh SH et a/. Relationship between minimal residual disease and outcome in adult acute lymphoblastic leukemia. Blood 1 996: 87: 525 1 -5256.

1 5 1 . Foroni L, Hoftbrand AV. Molecular analysis of minimal residual disease in adult acute lymphoblastic leukaemia. Best Pract Res Clin Haemato/ 2002: 15: 7 1 -90.

1 52. Ottmann OG, Hoelzer D. The ABL tyrosine kinase inhibitor STI57 1 (Glivec) m

Philadelphia positive acute lymphoblastic leukemia. Hematol J 2002: 3: 2-6.

38

Chapter 2

High functional P-glycoprotein activity is more often present

in T-cell acute lymphoblastic leukaemic cells in adults than

in children.

Sabine L.A. Plasschaert1, Edo Vellenga2, Eveline S.J.M. de Bont1, Dorina M. van der Kolk2,

Anjo J.P. Veerman4, Wim J. Sluiter1, Simon M. G. Daenen2, Elisabeth G. E. de Vries3, Willem A. Kamps1•4

Departments of1Pediatric Haematology and Oncology, 2Haematology, 3 Medical Oncology, University Hospital Groningen, Groningen,

4Dutch Childhood Leukaemia Study Group, the Hague, The Netherlands.

Leukemia & Lymphoma 2003; 44: 85-95

Chapter 2

ABSTRACT

There is a distinct difference in prognosis between childhood versus adult acute

lymphoblastic leukaemia (ALL). To define whether multidrug resistance (MDR)

genes might contribute to this distinction, the expression and functional activity of

P-glycoprotein (P-gp) and MDR associated proteins (MRP) were determined with

RT-PCR (MDR-1, MRP1, MRP2, MRP3) and flow cytometry (P-gp and MRP).

Patient samples were obtained from 36 children and 35 adults with de novo ALL.

Of these patients, 38 showed a T-lineage and 33 a B-lineage immunophenotype. In

the samples, large variability in P-gp activity (0.8-4.9) and MRP activity (1. 1-13.9)

was observed. Most T-ALL patients with high P-gp activity were adults (89%). The

mRNA expression of MDR-1 correlated weakly with P-gp activity. In contrast, MRP

activity did not correlate with the mRNA expression of MRPl , MRP2 and MRP3.

In T-ALL, a worse overall survival and event-free survival was observed with incre

asing P-gp activity. P-gp activity had no prognostic impact in B-lineage ALL. In

addition, high MRP activity did not influence treatment outcome in either T- or

B-lineage ALL. Multivariate Cox regression analysis, showed P-gp activity to be the

only unfavourable prognostic factor for overall survival in T-ALL. In conclusion,

this study demonstrates the prognostic relevance of P-gp activity in T-ALL. Since

the majority of the patients with high P-gp activity were adults, P-gp might contri

bute to the poor prognosis of adult T-ALL.

40

P-gp and MRPl-2 activity in ALL

INTRODUCTION

Adults with acute lymphoblastic leukaemia (ALL) have a worse prognosis compared

to children. This difference might be explained by various factors, such as age, white

blood cell (WBC) count and cytogenetic abnormalities. The influence of age on cli

nical outcome has been studied in a large cohort of children and adults with ALL,

who were treated according to similar protocols1• Age proved to be an important

prognostic factor, since progressively lower survival rates were found with increa

sing age. Children aged 1-9 years showed overall survival rates at 5 years of 85%

versus 24% of the patients over the age of 40. In addition, a higher incidence of

unfavourable prognostic factors has been observed in adults compared to children2•

Adults more often have high WBC counts and meningeal involvement at diagnosis

than children. Several favourable genetic abnormalities, such as hyperploidy and

TEL-AMLl fusion, occur more often in children, whereas the incidence of the unfa

vourable translocation t(9;22) is higher in adults3.4. Finally, the immunophenotype of

the ALL has an impact on prognosis. Patients with early precursor-B ALL respond

best to chemotherapy and this immunophenotype is seen more often in children than

adults (65% versus 50%). Children with B-lineage ALL have event-free survival

rates (> Syears) of 80-90% and adults of 30-50 % (> 3 years). Furthermore, the T-cell

immunophenotype, which is associated with worse prognosis in children and adults,

is more often expressed in adults with ALL than in children (25 versus 15% )2• Child

ren with T-ALL have event-free survival rates at 5 years between 65 and 75%4,

while adults demonstrate event-free survival rates at 3 years of 45-50 %5• Despite the

improved treatment outcome in adults over the last decade, the prognosis in adults

remains worse than in children.

Biological differences between lymphoblasts of children and adults have also been

demonstrated. With an in vitro chemosensitivity assay, lymphoblasts from adults

show an increased resistance to anticancer drugs, including prednisolone and dau

norubicin6:7. This finding might suggest that the transporter proteins P-glycoprotein

(P-gp), encoded by the MDR-1 gene, and multidrug resistance (MDR) associated

proteins (MRP) play a role in this phenomenon8:9• These proteins act as ATP-depen

dent membrane efflux pumps and actively extrude the anticancer drugs out of the

cell. By lowering the intracellular concentration they confer resistance to anticancer

drugs, such as dexamethasone and anthracyclines.

The protein expression of these drug efflux proteins has been studied in ALL in

children as well as adults, but their clinical significance remains unclear. Several

large studies in children have shown no correlation between P-gp protein expres

sion and known risk factors or clinical outcome10;11• Wattel et a!. has studied the P-gp

41

Chapter 2

protein expression in 50 adult patients and also found no correlation with unfavoura

ble prognosis12. Other studies however, found that P-gp protein positivity resulted in

higher relapse rates and lower overall survival rates in comparison with P-gp nega

tive patients13;14. The functional activity of P-gp has been determined in leukaemic

cells of childhood ALL patients and demonstrated in general no prognostic signifi

cance10;15. These studies have not been performed in adult ALL samples. The MRP

subfamily consists of at least six members; in cancer cell lines, MRP 1-3 and MRP5

have been shown to confer resistance to anticancer drugs'6-20• In childhood ALL, the

MRP1 protein expression had no correlation with unfavourable prognostic factors

and clinical outcome' ' . The functional activity of MRP 1, MRP2 and possibly MRP3

can be determined in a flow cytometric assay21 and has not been assessed in ALL so

far.

Based on the above-described findings, we questioned whether differences in func

tional activity and rnRNA expression of P-gp and MRP might contribute to the dif

ference in clinical outcome between children and adults with ALL. Therefore, P-gp

and MRP activity were studied by flow cytometry in 36 children and 35 adults with

de novo ALL. The mRNA expression of MDR-1, MRP1, MRP2 and MRP3 was

determined by RT-PCR. Our results demonstrate the clinical importance of P-gp and

MRP functional activity and its relation with known clinical risk indicators.

MATERIALS AND METHODS

Patient samples

All childhood samples were obtained from the Dutch Childhood Leukaemia Study

Group (DCLSG). The children with ALL were treated according to the DCLSG

protocols (ALL-6, -7, -8 and -9)22, containing the MDR drugs dexamethasone,

daunorubicin, doxorubicin, methotrexate and vincristine.

Mononuclear cells from bone marrow or peripheral blood were isolated on a

Ficoll-Isopaque (Nycomed, Oslo, Norway) density gradient by centrifugation for 20

minutes at 800g. The cells were cryopreserved in RPMI 1640 medium supplemented

with 10% FCS and 10% dimethyl sulfoxide (Merck, Amsterdam, the Netherlands)

and stored at -196. C.

Upon analysis cells were thawed, centrifuged in new-born calf serum (Life

Technologies, Breda, the Netherlands), treated with DNA-se (Boehringer Mannheim,

Mannheim, Germany) and washed with RPMI 1640 medium. All samples contained

more than 90 % of leukaemic cells. Percentage of blast cells was determined

by May-Griinwald-Giemsa staining and immunophenotyping was performed by

flow cytometry. For immunophenotyping, 0.5 x 106 cells were incubated with

42


5 Jll of FITC- or phycoerythrin-labeled mouse monocloncal antibodies to CD3

(IQ Products, Groningen, the Netherlands), CD7 (IQ Products), COl O (Becton

Dickinson, Woerden, the Netherlands), CD19 (Becton Dickinson), CD33 (Becton

Dickinson), CD34 (Becton Dickinson) or IgG isotype-matched control (Becton

Dickinson) for 30 min at 4°C. Subsequently, cells were washed with RPMI

1 640 medium, and analysed with a FACStar flow cytometer (Becton Dickinson

Medical Systems, Sharon, MA). The immunophenotyping and cytomorphology were

performed by the DCLSG.

Although in general only 20-25% of ALL patients have a T-cell immunophenotype,

this study investigated a comparable number of B- and T-lineage ALL samples, to

make it possible to analyse both groups separately. To determine whether the patients

in our study were a good representation of all patients treated in the DCLSG trial

(n=2 1 8), we analysed the patient characteristics. In our study, more T-ALL patients

were high risk (NCI criteria: WBC�50x 109/L and/or age �10 years) compared to the

patients in the DCLSG trial (n=34). However in the DCLSG trial, high risk patients

had the same survival rates as standard risk22. The B-lineage patients in our study

were comparable with the patients in the DCLSG trial (n=170).

From adult patients, newly diagnosed with ALL in the University Hospital Groningen

during the period of 1989-1999, bone marrow of peripheral blood was collected

for diagnostic evaluation and cryopreserved. The adult patients were included if a

sufficient number of preserved bone marrow or peripheral white blood cells (> 1 Ox1 06

) was available. Patient samples were cryopreserved and immunophenotyped

by the University Hospital Groningen, as described above.

The adults were treated as described by Daenen et aP.3 with or without

pre-induction course of cytarabine and etoposide, of which prednisone,

doxorubicin, vincristine, methotrexate and etoposide are MDR drugs. The

characteristics of the T-ALL patients in our study were comparable with those of the patients in the clinical trial (n=26). The B-lineage patients in our study

had higher WBC counts than the patients in the clinical trial. Nevertheless, in

this trial comparable survival rates were observed between B-lineage patients

with a high and a low WBC count (30x l 09/L)23•

Flow cytometric detection of drug efflux

The ability of leukaemic blasts to extrude the P-gp substrate rhodamine (Rh123)

(Sigma Chemical Co, Bomem, Belgium) and MRP- substrate carboxyfluorescein

diacetate (CFDA) (Sigma Chemical Co) was determined by flow cytometry2 1 ;24• The

cellular retention of the Rh 123 was studied in the absence or presence of the P-gp

inhibitor PSC833 (provided by Novartis, Basel, Switzerland). The cellular retention

43

Chapter 2

of carboxyfluorescein (CF), the intracellular metabolised product of CFDA, was

determined with or without the leukotriene receptor antagonist and MRP inhibitor

MK-57 1 (provided by Dr. Ford-Hutchinson, Merck Sharp, Kirkland, PQ, Canada)25,

and reflects the activity of MRP1, MRP2 and possibly MRP321• Cells (0.5x 106) were

incubated with Rh123 (200 ng/ml) or CFDA (0. 1f..lM) with or without inhibitor (2

f..lg/ml PSC833 or 10 f.!M MK-571) in RPMI 1640 for 20 min at 37°C, 5% C02 •

Subsequently, cells were washed in ice-cold medium and incubated with the inhibi

tors only, for another 60 min at 37°C, 5% C02' Thereafter cells were pelleted and

resuspended in ice-cold medium to stop the efflux. The fluorescence of Rh123 and

CF was determined by using a FACStar flow cytometer. The efflux blocking factors

of the inhibitors were calculated as the ratio of drug fluorescence with modulator

divided by drug fluorescence without modulator after subtraction of the fluorescence

of the control. P-gp activity was defined as the Rh123 efflux blocking factor of

PSC833 and MRP activity as the CF efflux blocking factor of MK-571. Dead cells

were gated out following scatter characteristics. Moreover, the viability of the cells

was also checked using trypan blue.

MDR-1, MRPl, MRP2 and MRP3 mRNA expression measured by

reverse transcription-polymerase chain reaction

Total cellular RNA was extracted from 2-10 x 106 leukaemic cells using 1 ml of

Trizol reagent (Life Technologies, Breda, the Netherlands). RNA was extracted, pre

cipitated and washed according to the manufacturer 's protocol. Subsequently, 2 f.!g

RNA was reverse transcribed in 20 f..ll reverse transcriptase buffer containing 10 mM

DTT, 0.5mM each of dATP, dGTP, dCTP and dTTP, 200 U of Moloney murine leu

kaemia virus reverse transcriptase (M-MLV RT), 5 U RNAse inhibitor and 10 ng/f..ll

pd(N)6 random primers. The PCR reactions were performed under the following

conditions. For MDR-1, 31 cycles (95, 55 and 72°C, 30s) were executed. For MRPl ,

29 cycles (95, 56 and 72°C, 30s), for MRP2, 3 1 cycles (95, 58 and 72°C, 30s) and

for MRP3 32 cycles (95, 56 and 72°C, 30s) were performed. For �-2 microglobulin

(�2m), used as an internal control, 20 cycles (95, 55 and 72, 30s) were executed.

The amplified products contain 161 bp for MDR- 1, 990 bp for MRP1, 1067 bp for

MRP2, 564 bp for MRP3 and 268 bp for �2m. Variations between samples in quan

tity were normalised using the quantity of �2m in each sample. Reaction products

were separated on a 1.5% agarose gel and visualised by ethidium bromide staining.

The absorbance of visualised bands was expressed as absorbance x mm2• The results

were calculated as the ratio of absorbance of the specific mRNA product to the �2m

mRNA product.

44


Statistical analysis

The Mann-Whitney U test was performed to test for differences in ordinal and nume

rical variables in children and adults, and T-ALL and B-lineage ALL. Differences in

frequency distribution in low and high P-gp or MRP activity were studied between

children and adults with the chi-square test. The Spearman rank correlation was used

to determine correlation between P-gp and MRP activities and continuous variables.

To determine the correlation between MRP activity and MRP 1, MRP2 and MRP3

mRNA expression, a multivariate analysis was performed.

To determine whether P-gp or MRP activity had prognostic implications, Cox regres

sion analysis was performed using P-gp and MRP activity as continuous variables.

Then, the optimal cut-off level for P-gp and MRP activity was calculated in such a

way that maximal survival differences occurred between patients with high and low

activity. Overall survival and event-free survival were estimated according to the

method of Kaplan and Meier. Overall survival was defined as the time from diagno

sis until the date of death by any cause, or if no death occurred until the date of last

contact. Event-free survival was defined as the time from diagnosis until the date of

an adverse event (relapse, death or the development of a second malignancy), or if

no such event occurred, until the date of last contact. Patients who did not attain a

complete remission were considered failures at time zero. Furthermore, univariate

and multivariate Cox regression analysis were also used to test for association bet

ween other continuous variables (such as age, WBC count) and survival. P-values

<0.05 were considered significant.

RESULTS

Patient characteristics

In total, samples from 36 children and 35 adults with de novo ALL were studied.

Of these patients, 38 were diagnosed with T-ALL and 33 with B-lineage ALL. The

characteristics of the T-ALL patients are depicted in Table 1. Between children (1-16

years) and adults a significant difference was noticed with regard to the initial WBC

count. Children had a higher initial WBC count than adults. Children also had lower

haemoglobin and platelet count and a higher percentage lymphoblasts in peripheral

blood at diagnosis than adults. In Table 2 the characteristics of the B-lineage ALL

patients are shown. No differences in characteristics between children and adults

with B-lineage ALL were observed.

In our study population, children with T-ALL had a better overall survival (See

figure 1A) and event-free survival than adults with T-ALL, however this difference

was not statistically significant (p=O. I 89 resp. p=O. l 4 1 ). Children with B-lineage

45

Chapter 2

Table 1 Characteristics ofT-ALL patients

Number Male/female (percentage) Aget (range) Induction failure Death during induction (9;22);t( 4; I I ); I I q23 Unfavorable cytogenetics t BMT CR rate (percentage) WBCt (x l 091L) (range) Hbt (mmol/L) (range) Plateletst (x 109/L) (range) %-age blasts BMt (range) %-age blasts PBt (range) P-gp functiont (range) MRP functiont (range)

t median value * p<0.001 ** p<0.05

Children + Adults

38 26/12 (68/32) 13 (3-7 1 )

1 5

I 9

32/33 (97) 70 (2.7-900) 6.4 (2 .7- 1 0.2) 47 (6-263) 89 (72-98) 74 ( 1 0-99) 1 . 1 (0.8-4.9)

2.9 ( 1 . 1 - 1 3.9)

Table 2 Characteristics of B-lineage ALL patients

Number Male/female (percentage) Age (range) Induction failure Death during induction Unfavorable cytogenetics t(9;22);t( 4; 1 1 ); I I q23 BMT CR rate (percentage) WBCt (x i09/L) (range) Hbt (mmol/L) (range) Plateletst (x l 09/L) (range) %-age blasts BMt (range) %-age blasts PBt (range) P-gp functiont (range) MRP functiont (range)

t median value * p<O.OO l

46

Children + Adults

33 1 4/19 (42/58) 1 8 ( 1 -73)

1 3

4 32/33 (97)

32. 1 ( 1 . 1 -446) 5 .4 (2.2-1 0.0)

32 (4-345) 94.7 (27-99) 7 1 (2-98)

1 .0 (0.8-2. 7) 2. 7 ( 1 .2-6.3)

Children

2 1 1417 (67/33) 8* (3- 14)

0 I

0 3

2 1 /2 1 ( 1 00) 1 7 1 .6* (27 .5-900) 5 .6** (2. 7-8.5)

4 1 ** (6-90) 89 (77-98) 82** ( 1 3-99) 1 . 1 ( 1 .0-4.9)

3.9 ( 1 . 1 -1 3.9)

Children

1 5 7/8 (47/53) 7* ( 1 - 16)

0 0 0

0 1 51 15 ( 1 00)

1 2.9 ( 1 . 1 -446) 4.2 (2.2-8.5) 43 (6-239) 95 (86-99) 43 ( 1 5-96) 1 .0 (0.8-2.4) 2.6 ( 1 .2-5.8)

Adults

1 7 1 2/5 (7 1 129) 27* ( 16-7 1 )

1 4

I 6

1 2/1 3 (92.3) 1 7.6* (2.7- 1 5 1 ) 7.6** (4.0- 1 0.2) 64** ( 1 1 -263) 89 (72-97) 5 1 ** ( 10-98) 1 .2 (0.8-4. 1 ) 2.5 ( 1 .2-6.5)

Adults

1 8 711 1 (39/6 1 ) 32.5* ( 14-73)

1 3

4 14/ 1 5 (93)

44.6 (2. 7-332.2) 5.9 (3 .0- 1 0.0)

27 (4-345) 93.3 (27-98.4)

87 (2-98) 1 .0 (0.8-2.7) 3 .2 ( 1 .6-6.3)


A

B

Overall survival T-ALL

children vs. adults 1 0

,8 � ·� � ,6 Q) > �

3 ,4 E ::0

() ,2 N 0 c 21 7

A 17 9 ,0 Lonrank p=O 189

50 100 150 200

Overall survival B-lineage ALL

children vs. adults

c

A

250 300

weeks

1,0 "tT"-1--+-+-+--+-+--++-++-----------

,8 'iii > ·� ::0 6 "' .� � 4 E

::0 () N 0 ,2 c 15 0

A 18 8 Logrank p=O 002

c

A

o.o .�---��-�-��-�-�-�� 0 so 100 150 200 250 300 350 400 4 50

weeks

Figure 1 Overall survival curves of the ALL patients, subdivided in two groups: children (C) and adults (A) for T-ALL patients (1 A) and B-lineage ALL patients (1 B). N = number, 0 = observed deaths.

ALL showed higher overall survival (See figure 1B) and event-free survival rates

than adults (p=0.002 resp. p=0.001). The median follow-up of the patients was 116

weeks ( 1-611 weeks).

Function and expression of P-gp and MRP in ALL patients

The functional activity of P-gp in T-ALL, expressed as efflux-blocking factor, varied

from 1.0 to 4.9 in children (median 1. 1) and in adults from 0.8 to 4.1 (median 1.4). In

adults, more patients showed high P-gp activity (Rh 123 efflux blocking factor > 1.5)

47

Chapter 2

than in children (p=0.002). However, in B-lineage ALL the P-gp activity varied

from 0.8 to 2.7 and no difference in distribution between children and adults was

observed. The P-gp activity in T-ALL was higher than in B-lineage ALL (p=0.030)

(See figure 2A).

The MRP activity in children with T-ALL varied from 1. 1 to 13.9 (median 3.9) and

in adults from 1.2 to 6.5 (median 2.5). Although higher absolute MRP activities were

measured in childhood patients, no difference in distribution of MRP activity was

found between children and adults (p=0.89). In B-lineage ALL, MRP activity ranged

from 1.2 to 5.8 in children (median 2.6) and in adults from 1.6 to 6.3 (median 3.2).

The distribution in MRP activity was similar between children and adults. No dif

ference in MRP activity was observed between T-ALL and B-lineage ALL (p=0.59)

(See figure 2B).

A P-gp activity

� 5 0 1l 4! 0 go • L � � 0

� 2 • 8 0 0 :::: 0 0 0

:c ,., • • 0: 1 if a;$

T·All T-All 8-lin. ALL 8-lin. ALL cnlk! ildull child adult 0-"21 n=H n• 15 n=18

B MRP activity 16

14 0

J 12 0 !!' 10 0

� 8 • S> . � 6 tl 0 o o 0 IE I 0 . � .. 4 • 0 0 l u

t Q9 � 0 <8> T-ALL T-ALL B· 1n ALL B·lln. All chik! adult chHd adult n�<21 n•17 n=15 n=18

Figure 2 (A) Rh 1 23 efflux blocking factors of PSC833 subdivided in four groups: childhood T-ALL, adult T-ALL, childhood B-lineage ALL and adult B-lineage ALL patients. (B) CF efflux blocking factors for MK-57 1 subdivided in the same four groups of patients. Open circles: patients in continuous complete remission; closed circles: induction failures and relapsed patients

48

A

B "

� � 12 ::; 1; � 10 N :r •

i •

, .. • u. u

rtlo-0.35 P-().027 """

0

0

0

0

MDR 1 mRNA expression

""""' " po0260 n-..o o

0

c!' 0


0

�0 g 8 <W o 0 0 o 0 '6 o 0 o0

MRP1 mRNA expression

Figure 3 Correlation between MDRI mRNA expression and Rhl 23 efflux blocking factors of PSC833 (A) and between MRPl mRNA expression and CF efflux blocking factors of MK-57 1 (B) in 4 1 patients. mRNA levels are expressed as absorbance x mm2 of the visualized bands of MDRI or MRP I mRNA divided by absorbance x mm2 of �2m mRNA.

The mRNA expression of MDR-1, MRP l , MRP2 and MRP3 was determined in a

maximum of 18 children and 23 adults with ALL. Because of the limited amount of

patient material available, a few values are missing in the mRNA detection of MRP l ,

MRPl and MRP3. All, but one child, expressed MDR-1 mRNA and the expression

correlated weakly with the Rh123-effiux blocking factor (rho=0.35, p=0.027)(See

figure 3A). MRPl mRNA was expressed by 17 out of 18 children (94%) and 18 out

of 22 adults (82% ). The mRNA expression of MRP 1 did not correlate with the CF

efflux blocking factor (rho=0.18, p=0.260)(See figure 3B). Since CF is effiuxed by

MRP l , MRP2 and possibly MRP3, we also determined MRP2 and MRP3 mRNA

expression. MRP2 expression was found in 89% of the children (16 of 18) and

73% of the adults (16 of 22). The expression of MRP3 was observed in 24% of the

children (4 of 17) and 14% of the adults (3 of 21). No correlation was detected

between MRP2 and MRP3 mRNA expression and the CF efflux blocking factor.

Moreover, multivariate analysis did not show a correlation between a combination

49

Chapter 2

of MRPl , MRP2 and MRP3 mRNA expression and CF effiux blocking factor.

Remarkably, two patients showed MRP-activity in the effiux-assay, but they did not

express detectable levels of MRP 1 , MRP2 and MRP3 mRNA.

Relation between P-gp and MRP activity and clinical data

In T-ALL, P-gp activity did not correlate with MRP activity, age, haemoglobin and

platelet count at diagnosis. A negative correlation was observed between P-gp acti

vity and WBC count (rho=-0.33, p=0.044). MRP activity did not correlate with these

factors.

The same was found in B-lineage ALL. P-gp activity correlated negatively with

WBC count (rho=-0.45, p=0.008) and no other correlations were observed between

MRP activity and P-gp activity and the above described factors.

Since a correlation has been observed between CD34 expression and P-gp activity

A Overall survival T-ALL

B

Ill u..

P-gp function 1.0

,2 N 0 L 29 9 H 9 7

0,0 LOWolnk p=0.002

0 50 1 00

H

ISO

Event-free survival T-ALL P-gp function

1 0

w ,6 � "5 § ,4 '-'

.2 N 0 H � � : '-----+

0 0 logrank p=O 002

0 50 1 00 150

L

200 250 300

weeks

200 250 300

weeks

Figure 4 Survival curves of the T-ALL patients, subdivided in two groups with low (L) and high (H) P-gp activity for overall survival (4A) and event-free survival (4B). N = number, 0 = observed events

50


in acute myeloid leukaemia21 :26, we investigated CD34 expression in our study popu

lation. We found no correlation between CD34 expression in leukaemic blasts and

P-gp activity in T-ALL, nor in B-lineage ALL (p=0. 1 3 resp. p=0.36). Higher P-gp

activity has been found in AML compared to ALV0, and therefore we investigated

a possible correlation between P-gp activity and the presence of the myeloid marker

CD33 on ALL cells. In T-ALL, CD33 expression correlated with P-gp activity

(rho=0.5 1 , p=0.002). CD33 expression did not correlate with P-gp activity in B-line

age ALL (rho=0.26, p=0. 1 88).

The impact of P-gp activity on overall survival and event-free survival was studied

with Cox regression analysis, in which Rh1 23 efflux blocking factor was used as a

continuous variable. In T-ALL, a worse overall and event-free survival was observed

with increasing P-gp activity (RR 1 .78, p=0.005 resp. RR 1 .73, p=0.027). No evi

dence was found for an effect of treatment and/or age on the impact of P-gp on

survival, and therefore the children and adults could be combined. To illustrate the

prognostic impact ofP-gp in survival curves, patients were divided into two groups:

patients with a high Rh123 efflux blocking factor and patients with a low Rh123

efflux blocking factor. The optimal cut-off level was calculated to be 1 .5 to obtain a

maximal survival difference. Patients (both children and adults) with a high Rhl 23

efflux blocking factor (> 1 .5) had lower overall survival rates than patients with a

low Rh1 23 efflux blocking factor (<1 .5) (p=0.002) (See figure 4A). P-gp activity

also affected event-free survival: patients with a high Rh1 23 efflux blocking factor

showed a lower event-free survival than patients with a low Rh l 23 efflux blocking

factor (p=0.002) (See figure 4B). In B-lineage ALL, P-gp activity did not have an

impact on overall survival and event-free survival (p=0.539 resp. p=0.740) (data not

shown).

For the CF efflux blocking factor the optimal cut-off level was determined to be

2.25 and figure 5A shows the relation between MRP function and overall survival

in T-ALL. The MRP activity did not influence overall survival (p=0. 1 82) and event

free survival (p=0.242)(See figure 5B). Also in B-lineage ALL, overall survival and

event-free survival were not influenced by MRP activity (p=0.686 resp. p=0. 1 60)

(data not shown).

The prognostic influence of several other individual factors on overall survival

was determined with univariate Cox regression analysis. Besides P-gp activity, age

proved to be a prognostic factor in T-ALL (p=0.0 16). MRP activity, sex, CD34- and

CD33-expression, WBC count, haemoglobin and platelet count at diagnosis were of

no prognostic value. After multivariate Cox regression analysis using both age and

P-gp, only P-gp activity remained a prognostic factor for overall survival in T-ALL

(RR 4.63 (CI 95% 1 .64- 12 .99), p=0.005).

51

Chapter 2

A Overall survival T-ALL

10 MRP funct1on

N 0 L 13 3 H 25 13

0 logfank P"'O 182 0 50 100

B Event-free survival MRP function

1 0

l N 0 l 13 4 H 25 14 0 \.ograN. p•O 2J2

0 50 tOO

L

150 200 250 300

weeki

L

H

ISO. 200 25C :100

weeks

Figure 5 Survival curves of the T-ALL patients, subdivided in two groups with low (L) and high (H) MRP activity for overall survival (5A) and event-free survival (5B). N = number, 0 = observed events

DISCUSSION

This is the first study, in which the functional activity of P-gp and MRP has been

determined in de novo ALL samples of both children and adults. Multidrug resi

stance caused by the ATP-dependent membrane transporter proteins P-gp and MRP

has been studied in ALL mainly by determining the mRNA and protein expression of

these transporters. Clinically, it may be more relevant to test the functional activity,

since it allows to identify patients who may benefit from the use of inhibitors ofP-gp

and MRP function27• We analysed P-gp function by measuring the modulation of

Rhl 23 fluorescence by PSC833, since it has been demonstrated, that this is the most

sensitive assay to detect P-gp activity28• For assessing MRP activity, we determined

the modulation of CF fluorescence by MK-57121 •

In patients with T-ALL, P-gp activity was mainly observed in adults. Only one child

showed high P-gp activity. This difference in P-gp activity between adults and child

ren is consistent with the hypothesis, that adult leukaemia arises from stem cells

52


and childhood leukaemia from the more mature committed progenitors29• In AML,

the maturation stage plays a role in P-gp activity, given that immature CD34- posi

tive ceiJs showed higher P-gp activity than more mature CD34-negative ce1Js21:26:30•

The connection between maturity and P-gp activity could therefore account for the

higher P-gp activity in adults, observed in the present study. However, our study in

adults and children with ALL, and other studies found no difference between CD34+

and CD34- ALL with respect to P-gp expression3 1 and P-gp activity10; 15 •

P-gp activity in B-lineage ALL was similar in children and adults, while there was a

higher functional activity in T-ALL versus B-lineage ALL. Previous studies have not

observed this correlation with immunophenotypical subgroups, but in these studies

only samples of children were investigated10: 1 5•

The association we observed between the myeloid marker CD33 on ALL ceiJs and

P-gp activity has not been described before. Interestingly, a higher P-gp activity has

been found in AML than in ALL 10; therefore co-expression of the myeloid marker

CD33 on ALL ceiis might be reminiscent of the higher P-gp expression in myeloid

cells.

P-gp activity proved to be a prognostic factor for overaii and event-free survival in

T-ALL, in both children and adults. Patients with high P-gp activity had a lower

overaiJ and event-free survival than patients with low P-gp activity. Previous studies

in ALL have not found this association, but the investigated patients were only child

ren. By comparing the different studies with regard to P-gp activity, a great hetero

geneity in activity is observed. There are several methods to determine P-gp acti

vity, and the assays performed in children with ALL used verapamil as P-gp modu

lator, and not PSC833. PSC833 has proved to be a more potent modulator of P-gp

than verapamiJ32, and therefore the variable sensitivity of the assays might partiaiiy

explain the differences in prognostic implication of P-gp between studies10: 15 •

The children with T-ALL in our study had better survival rates than adults, however

the difference is not statisticaiJy significant, presumably due to the limited number of

samples studied. The selection of children with predominantly high risk T-ALL, i.e.

WBC2:50x I 09/L and/or age 2:10 years, might influence the low P-gp activities found.

In the literature however, there are no indications that P-gp activity and age or WBC

count are inversely related. Moreover, as described above, also in our study WBC

count and age did not have an impact on prognosis, whereas P-gp activity did.

Prednisolone is one of the compounds P-gp is able to transport. Children with

ALL are known to be better responders to prednisolone than adults. This might be

explained by the presence of a higher P-gp activity in leukaemic blasts from adults,

making them less susceptible to prednisolone. Indeed in overaiJ analysis, leukaemic

blasts from adults demonstrated a higher degree of in vitro resistance to prednisolone

53

Chapter 2

than blasts from children, however in the limited number ofT-ALL patients included

in these studies this difference could not be observed6:7•

MRP activity showed a great variability in ALL, with quite high activities in children

in comparison with adults. Data of mRNA expression of MRP 1 -3 are available of a

very limited number of samples, due to lack of sufficient patient material. The MRP

activity did not correlate with MRPl , MRP2, MRP3 mRNA or their combination.

This finding has also been observed in a previous study33 and might be explained

by the presence of dysfunctional MRP homologues or the presence of other trans

porters, which can also extrude CF. The observation of two patients with MRP func

tion and no mRNA expression of MRPl , MRP2 and MRP3 underscores this expla

nation.

The activity of MRP had no prognostic impact on overall survival and event-free

survival. In AML, MRP activity or the simultaneous activity of P-gp and MRP was

found to be predictive of a poor treatment outcome26:33:34• In our study, patients

with simultaneous activity ofP-gp and MRP showed survival rates comparable with

patients with P-gp activity alone. Some explanations can be hypothesised for the

phenomenon that high MRP activity does not seem to influence clinical outcome.

By measuring the CF efflux, we only determined the capacity of MRP to extrude its

substrates. MRPl and MRP2 transport several compounds only in the presence of

glutathione. In our assay, the ability of MRP to transport CF is determined and not

the extent of GSH conjugation, since CF is not conjugated before it is transported by

MRP. By low glutathione levels or impaired conjugation of anticancer drugs, these

drugs might not be transported properly, in spite of high MRP activity16;19• Another

explanation could be, that when a certain level of MRP activity is reached, it does

not matter whether the CF efflux blocking factor varied from 4.0 to 1 6.0, since in

both cases a maximum of transport may already be carried out.

In conclusion, this study shows that P-gp activity is an independent unfavourable

prognostic factor for T-ALL. It might be one of the factors explaining differences

in treatment outcome between children and adults with T-ALL, since it is mainly

present in adults. In B-lineage ALL on the other hand, P-gp activity does not seem

to play a role. MRP activity is present in both T- and B-lineage ALL and can reach

high values in children with T-ALL. The lack of impact of MRP activity needs to

be further studied by incorporating glutathione levels and glutathione-conjugation in

future studies.

54


ACKNOWLEDGEMENTS

This study was supported by the Foundation of Paediatric Oncology Groningen

(SKOG 99-0 1 ). The authors wish to thank all the hospitals participating in the Dutch

Childhood Leukaemia Study Group. Board members of the DCLSG are: H.van den

Berg, J.P.M. Bokkerink, E.S.J.M. de Bont, R.M. Egeler, B . Granzen, W.J.D. Hof

huis, P.M. Hoogerbrugge, W.A. Kamps, R. Pieters, J.A. Rammeloo, T. Revesz and

A.J .P. Veerman.

55

Chapter 2

REFERENCES

l . Chessells, J.M., Hall, E., Prentice, H.G., Durrant, J., Bailey, C.C., and Richards, S.M. The impact of age on outcome in lymphoblastic leukaemia; MRC UKALL X and XA compared: a report from the MRC Paediatric and Adult Working Parties. Leukemia 1 998: 12: 463-473.

2. Perentesis, J.P. Why is age such an important independent prognostic factor in acute lymphoblastic leukemia? Leukemia 1 997: 11 Suppl 4: S4-S7.

3 . Cytogenetic abnormalities i n adult acute lymphoblastic leukemia: correlations with hematologic findings outcome. A Collaborative Study of the Group Francais de Cytogenetique Hematologique. Blood 1996: 87: 3 1 35-3 142.

4. Pui, C.H. and Evans, W.E. Acute lymphoblastic leukemia. N Eng! J Med 1 998: 339: 605-6 1 5.

5 . Hoelzer, D. and Gokbuget, N. Recent approaches in acute lymphoblastic leukemia in adults. Crit Rev Oneal Hemato/ 2000: 36: 49-58.

6. Maung, Z.T., Reid, M.M., Matheson, E., Taylor, P.R., Proctor, S.J., and Hall, A.G. Corticosteroid resistance is increased in 1ymphoblasts from adults compared with children: preliminary results of in vitro drug sensitivity study in adults with acute lymphoblastic leukaemia. Br J Haemato/ 1995: 91: 93- 1 00.

7. Styczynski, J., Pieters, R., Huismans, D.R., Schuurhuis, G.J., Wysocki, M., and Veerman, A.J. In vitro drug resistance profiles of adult versus childhood acute lymphoblastic leukaemia. Br J Haemato/ 2000: 110: 813-8 1 8.

8. Cole, S.P., Bhardwaj, G., Gerlach, J.H., Mackie, J.E., Grant, C.E., Almquist, K.C., Stewart, A.J., Kurz, E.U., Duncan, A.M., and Deeley, R.G. Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science 1 992: 258: 1 650-1 654.

9. Cole, S.P. and Deeley, R.G. Multidrug resistance mediated by the ATP-binding cassette transporter protein MRP. Bioessays 1 998: 20: 93 1-940.

10. Wuchter, C., Leonid, K., Ruppert, V., Schrappe, M., Buchner, T., Schoch, C., Haferlach, T., Harbott, J., Ratei, R., Dorken, B., and Ludwig, W.D. Clinical significance of P-glycoprotein expression and function for response to induction chemotherapy, relapse rate and overall survival in acute leukemia. Haematologica 2000: 85: 71 1 -72 1 .

1 1 . den Boer, M.L., Pieters, R., Kazemier, K.M., Rattier, M.M., Zwaan, C.M., Kaspers, G.J., Janka-Schaub, G., Henze, G., Creutzig, U., Scheper, R.J., and Veerman, A.J. Relationship between major vault protein/lung resistance protein, multidrug resistance-associated protein, P-glycoprotein expression, and drug resistance in childhood leukemia. Blood 1 998: 91: 2092-2098.

12 . Wattel, E., Lepelley, P., Merlat, A., Sartiaux, C., Bauters, F., Jouet, J.P., and Fenaux, P. Expression of the multidrug resistance P glycoprotein in newly diagnosed adult acute lymphoblastic leukemia: absence of correlation with response to treatment. Leukemia 1995: 9: 1 870- 1 874.

1 3 . Goasguen, J.E., Dossot, J.M., Fardel, 0., Le Mee, F., Le Gall, E., Leblay, R., LePrise, P.Y., Chaperon, J., and Fauchet, R. Expression of the multidrug resistance-associated P-glycoprotein (P- 170) in 59 cases of de novo acute lymphoblastic leukemia: prognostic implications. Blood 1 993 : 81: 2394-2398.

14. Dhooge, C., De Moerloose, B., Laureys, G., Kint, J., Ferster, A., De Bacquer, D., Philippe, J., and Benoit, Y. P-glycoprotein is an independent prognostic factor predicting

56

P-gp and MRP1-2 activity in ALL

relapse in childhood acute lymphoblastic leukaemia: results of a 6-year prospective study. Br J Haemato/ 1 999: 105: 676-683 .

1 5 . Ivy, S.P., Olshefski, R.S., Taylor, B .J., Patel, K.M., and Reaman, G.H. Correlation of P-glycoprotein expression and function in childhood acute leukemia: a children's cancer group study. Blood 1 996: 88: 309-3 1 8 .

1 6. Keppler, D., Cui, Y., Konig, J., Leier, 1., and Nies, A . Export pumps for anionic conjugates encoded by MRP genes. Adv Enzyme Regu/ 1 999: 39: 237-246.

1 7. Kool, M., de Haas, M., Scheffer, G.L., Scheper, R.J., van Eijk, M.J., Juijn, J.A., Baas, F., and Borst, P. Analysis of expression of cMOAT (MRP2), MRP3, MRP4, and MRP5, homologues of the multidrug resistance-associated protein gene (MRP 1 ), in human cancer cell lines. Cancer Res 1 997: 57: 3537-3547.

1 8. Koo1, M., van der Linden, M., de Haas, M., Scheffer, G.L., de Vree, J.M., Smith, A.J., Jansen, G., Peters, G.J., Ponne, N., Scheper, R.J., Elferink, R.P., Baas, F., and Borst, P. MRP3, an organic anion transporter able to transport anti-cancer drugs. Proc Nat/ Acad Sci USA 1 999: 96: 69 14-6919 .

1 9. Renes, J . , de Vries, E.G., Nienhuis, E.F., Jansen, P.L., and Muller, M. ATP- and glutathione-dependent transport of chemotherapeutic drugs by the multidrug resistance protein MRPl . Br J Pharmaco/ 1 999: 126: 68 1 -688.

20. Wijnholds, J., Mol, C.A., van Deemter, L., de Haas, M., Scheffer, G.L., Baas, F., Beijnen, J.H., Scheper, R.J., Hatse, S., De Clercq, E., Balzarini, J., and Borst, P. Multidrugresistance protein 5 is a multispecific organic anion transporter able to transport nucleotide analogs. Proc Nat/ Acad Sci USA 2000: 97: 7476-748 1 .

2 1 . van der Kolk, D.M., de Vries, E.G., Koning, J.A., van den Berg, E., Muller,M., and Vellenga, E. Activity and expression of the multi drug resistance proteins MRP 1 and MRP2 in acute myeloid leukemia cells, tumor cell lines, and normal hematopoietic CD34+ peripheral blood cells. Clin Cancer Res 1998: 4: 1 727- 1 736.

22. Kamps, W.A., Veerman, A.J., van Wering, E.R., van Weerden, J.F., Slater, R., and Doesvan den Berg, A. Long-term follow-up of Dutch Childhood Leukemia Study Group (DCLSG) protocols for children with acute lymphoblastic leukemia, 1 984- 1 99 1 . Leukemia 2000: 14: 2240-2246.

23. Daenen, S., van Imhoff, G.W., van den Berg, E., de Kam, P.J., Haaxma-Reiche, H., Vellenga, E., Smit, J.W., and Halie, R.M. Improved outcome of adult acute lymphoblastic leukaemia by moderately intensified chemotherapy which includes a 'pre-induction' course for rapid tumour reduction: preliminary results on 66 patients. Br J Haematol 1 998: 100: 273-282.

24. Leier, 1., Jedlitschky, G., Buchholz, U., Cole, S.P., Deeley, R.G., and Keppler, D. The MRP gene encodes an ATP-dependent export pump for leukotriene C4 and structurally related conjugates. J Bioi Chern 1 994: 269: 27807-278 10.

25. Jones, T.R., Zamboni, R., Belley, M., Champion, E., Charette, L., Ford-Hutchinson, A.W., Frenette, R., Gauthier, J.Y., Leger, S., Masson, P., et al. Pharmacology of L-660,7 1 1 (MK-57 1 ): a novel potent and selective leukotriene D4 receptor antagonist. Can J Physiol Pharmaco/ 1 989: 67: 17-28.

26. Legrand, 0., Simonin, G. , Perrot, J.Y., Zittoun, R. , and Marie, J.P. Pgp and MRP activities using calcein-AM are prognostic factors in adult acute myeloid leukemia patients. Blood 1 998: 91 : 4480-4488.

27. De Moerloose, B., Dhooge, C., and Philippe, J. Discordance of P-glycoprotein expression and function in acute leukemia. Adv Exp Med Bio/ 1 999: 457: 107- 1 1 8.

57

Chapter 2

28. Broxtennan, H.J., Sonneveld, P., Feller, N., Ossenkoppele,G.J., Wahrer, D.C., Eekman, C.A., Schoester, M., Lankelma, J., Pinedo, H.M., Lowenberg, B. , and Schuurhuis, G.J. Quality control of multi drug resistance assays in adult acute leukemia: correlation between assays for ?-glycoprotein expression and activity. Blood 1996: 87: 4809-48 1 6.

29. Jasmin, C., Charpentier, A., Marion, S., and Proctor, S.J. ( 1 99 1 ) Concepts of remission, curability and lineage involvement in relationship to the problems of minimal residual leukaemia. Baillieres Clinical Haematology, 4, 599-607.

30. te Boekhorst, P.A., de Leeuw, K., Schoester, M., Wittebol, S., Nooter, K., Hagemeijer, A., Lowenberg, B. , and Sonneveld, P. Predominance of functional multidrug resistance (MDR- 1 ) phenotype in CD34+ acute myeloid leukemia cells. Blood 1 993 : 82: 3 1 57-3 1 62.

3 1 . Cascavilla, N., Musto, P., D' Arena, G., Ladogana, S., Matera, R., and Carotenuto, M. Adult and childhood acute lymphoblastic leukemia: clinico-biological differences based on CD34 antigen expression. Haematologica 1 997: 82: 3 1 -37.

32. Page, R.L., Hughes, C.S., Huyan, S. , Sagris, J., and Trogdon, M. Modulation of ?-glycoprotein-mediated doxorubicin resistance in canine cell lines. Anticancer Res 2000: 20: 3533-3538.

33. van der Kolk, D.M., de Vries, E.G., van Putten, W.J., Verdonck, L.F., Ossenkoppele, G.J., Verhoef, G.E., and Vellenga, E. ?-glycoprotein and multidrug resistance protein activities in relation to treatment outcome in acute myeloid leukemia. Clin Cancer Res 2000: 6: 3205-3214.

34. Legrand, 0., Simonin, G., Beauchamp-Nicoud, A., Zittoun, R., and Marie, J.P. Simultaneous activity ofMRP I and Pgp is correlated with in vitro resistance to daunorubicin and with in vivo resistance in adult acute myeloid leukemia. Blood 1999: 94: 1046- 1056.

58

Chapter 3

Influence of functional polymorphisms of the MDRJ gene on vincristine pharmacokinetics in childhood acute

lymphoblastic leukemia

Sabine L.A. Plasschaert1, Ellis Groninger, Marike Boezen2, /do Kema3, Elisabeth G.E. de Vriesl, Donald Uges5,

Anjo J.P. Veerman7, Willem A. Kamps1, Edo Vellenga6, Siebold S. de Graaf, Eveline S.J.M. de Bont1

Departments of1Pediatric Hematology and Oncology, 2Epidemiology, 3Pathology and Laboratory Medicine,

4Medical Oncology, 5Pharmacy and Toxicology, 6Hematology, University Hospital Groningen, Groningen,

7Dutch Childhood Oncology Group, the Hague, 8Pediatric Hematology and Oncology, University Medical Centre

St. Radboud, Nijmegen, The Netherlands.

Clinical Pharmacology & Therape11tics 2004; 76: 220-229

Chapter 3

ABSTRACT

Objective: Our objective was to investigate the effect of single nucleotide polymor

phisms (SNPs) in the P-glycoprotein MDRJ gene on vincristine pharmacokinetics

and side-effects in childhood acute lymphoblastic leukemia.

Methods: From 52 of 70 children who participated in a previous study on vin

cristine pharmacokinetics, patient material was available for investigation of the

MDRJ genetic variants. The SNPs C3435T and G2677T were determined by use

of polymerase chain reaction-restriction fragment length polymorphism. Vincristine

side-effects were scored retrospectively from patient records.

Results: No association was observed between C3435T or G2677T and vincristine

pharmacokinetic variables. When haplotypes were assigned, haplotype I l l carriers

(3435C/2677G) showed a longer elimination half-life than noncarriers ( 1 1 56 versus

805 minutes, P = .038). In contrast, haplotype l /2 carriers (3435T/2677G) had a

shorter elimination half-life than noncarriers (805 versus 1 1 80 minutes, P = .044).

However, this significance was lost after Bonferroni correction for multiple testing.

The haplotypes did not affect the other pharmacokinetic parameters, such as clear

ance and area under the concentration-time curve, suggesting that the observed

effect on elimination half-life is of very limited relevance. Moreover, SNPs in the

MDRJ gene did not identify patients with an increased risk for vincristine-induced

constipation.

Conclusion: The genetic variants in the MDRJ gene alone can not explain the large

variability in vincristine pharmacokinetics.

60

MDR-1 genetic variants and vincristine pharmacokinetics

INTRODUCTION

Vincristine plays a key role in the treatment of acute lymphoblastic leukemia (ALL) in children. It exerts a cytotoxic effect by the inhibition of cell cycle progression and the induction of apoptosis1 :2• The antileukemic effect of vincristine in vivo depends on the sensitivity of leukemic cells to the drug and on their exposure to vincristine, determined by its pharmacokinetics. The biliary system is the principal route of vincristine elimination3;4• Recently, we conducted a study of vincristine pharmacokinetics in 70 children newly diagnosed with ALL. This implied that three days before standard induction therapy the patients received a single dose of vincristine as monotherapy, with no other cytotoxic drugs or corticosteroids5• Measurement of vincristine plasma concentrations showed a large variability in vincristine pharmacokinetic parameters such as total body clearance (CL), elimination half-life (t112 6) and apparent volume of distribution at steady state (Vd.J The human MDRJ gene encodes for P-glycoprotein (P-gp ), and acts as an adenosine triphosphate-dependent membrane effi.ux pump. P-gp was originally discovered in cancer cells, causing resistance to natural product antitumor agents6• The effi.ux pump is also present in normal tissues including the gastrointestinal tract, kidney and liver7• P-gp affects the distribution of many drugs by limiting the absorption from the gastro-intestinal tract and secreting drugs into bile and urine7-10• Vincristine is a P-gp substrate, and studies in rats have shown that P-gp could influence the biliary excretion clearance of vincristine• •-13 • This observation suggests that P-gp may influence the pharmacokinetic profile of vincristine. In humans, to our knowledge, no studies have been reported investigating the role of P-gp or P-gp inhibitors in vincristine pharmacokinetics. To date, 29 single nucleotide polymorphisms (SNPs) have been identified in the entire human MDRJ gene14• The SNP C3435T in exon 26 was the first polymorphism for which pharmacologic consequences were shown15• C3435T correlated with intestinal P-gp expression and the uptake of orally administered P-gp substrate digoxin. Individuals homozygous for this polymorphism (TT) had a lower duodenal P-gp expression, lower in vivo P-gp activity, and increased digoxin plasma levels compared to individuals with the CT or CC genotype. Since that report, several studies have shown an association between the SNP C3435T and the pharmacokinetics of drugs that are P-gp substrates, such as cyclosporin A (INN, ciclosporin)16, digoxin17; 18

and fexofenadine19• A linkage disequilibrium exists between C34 3 5T and G2677T I A, and several studies have also shown a connection between G2677T/A and the pharmacokinetics of P-gp substrates16; 1 9-22• Another SNP, C 1 236T, also seems to be

61

Chapter 3

linked to C3435T and G2677T/A, but no effect of this SNP on pharmacokinetics has been described17-1 9:23-25 • These studies suggest that polymorphisms of the MDRI gene that alter protein function or expression may explain differences in the absorption, distribution and elimination of certain drugs8:26•

The dose-limiting side effect of the P-gp substrate vincristine is neurotoxicity. This manifests itself in constipation, loss of deep tendon reflexes, motor weakness, paresthesia and in some cases the syndrome of inappropriate antidiuretic hormone secretion (SIADH)27=28• A direct relationship between vincristine pharmacokinetics and neuropathy has not been demonstrated. However, P-gp is located in the bloodnerve barrier and can limit the vincristine exposure of the nerves29=30• The aim of this study was to investigate the relation among MDRI genetic variants, vincristine pharmacokinetics, and vincristine side effects in childhood ALL. The data from our previous study were exceptionally suitable for this purpose. These patients received vincristine monotherapy during a vincristine pharmacokinetic study, and consequently, no other antileukemic drug could have confounding effects.

METHODS

Patients

Between June 1 997 and January 2000, 70 children newly diagnosed with ALL participated in our study of vincristine pharmacokinetics5. The study protocol was approved by the medical ethics review board of the participating hospitals, and written informed consent was obtained for the pharmacokinetics study, deoxyribonucleic acid (DNA) extraction, and genetic analysis from all patients or parents (as appropriate). From 52 of the 70 children who participated in our previous study, patient material was available for DNA extraction. Serum concentrations of alkaline phosphatase, y-glutamyltransferase, AST, ALT, albumin, total protein, bilirubin, urea nitrogen, and creatinine were recorded at diagnosis.

Study design

The pharmacokinetic study was performed as described by Groninger et al. 5• In brief, at the start of the study, 1 .5 mg/m2 vincristine was administered as an intravenous bolus injection. Blood sampling was performed at 0, 1 0, 30, 1 80 and 1440 minutes after vincristine administration. Plasma was stored at -80°C until analysis. Three days later, standard combination induction therapy was started, consisting of dexamethasone, asparaginase and a second dose of vincristine intravenously, as well as methotrexate, prednisolon and cytosine arabinoside intrathecally, according to

62


Dutch Childhood Leukemia Study Group (DCLSG) protocol ALL-9. Comedication, such as allopurinol, acetoaminophen (INN, paracetamol), and antibiotics were registered during the pharmacokinetic study. Toxicity assessment was graded retrospectively according to the National Cancer Institute common toxicity criteria, by studying the patient records until the end of induction treatment (6 weeks). The incidence of SIADH in childhood ALL patients is generally very low (<5%)27;28;3 1 • Paresthesia is difficult to diagnose in young children, because it is a subjective complaint and hard to express verbally. We focussed on constipation as the main parameter for vincristine-related toxicity, because it is an early and frequently occurring side effect. The highest constipation toxicity score during induction treatment was used for analysis, and severe constipation occurred most frequently during the first three weeks. For analysis, patients with constipation grade 1 and grade 2 were combined, as were patients with grade 3 and grade 4, because these groups are clinically comparable. All patients received the oral laxatives bisacodyl on the day vincristine was administered intravenously (day 1 and day 4), and lactitol the following days.

HPLC analysis of vincristine

Vincristine plasma concentration was measured by HPLC with electrochemical detection32•33• The lower limit of quantitation of the assay was 0.48 )lg/L , and the coefficients of variations were 6.2% (0.48 )lg/L) and 4.2% ( 1 8.40 )lg/L) for withinday precision studies and 10.3% (0.48 )lg/L) and 8.5% ( 1 8.40 )lg/L) for between-day studies. Vincristine concentrations in plasma samples from all participating hospitals were measured in Groningen, The Netherlands.

Pharmacokinetic data analysis

Pharmacokinetic data analysis in our previous study has been described in detaiJS. In short, a two-compartment, first-order pharmacokinetic model was used. Primary pharmacokinetic parameters were estimated using the ADAPT II software package34 (available from the Biomedical Simulations Resource, University of Southern California, Los Angeles, Calif) and a Bayesian algorithm. Secondary pharmacokinetic variables such as clearance (CL), area under the concentration-time curve (exponential) (AUC), apparent volume of distribution at steady state (Vd55), distribution half-life (t112 a), and elimination half-life (t1 12 1l) were calculated from the model.

Extraction of genomic DNA

Bone marrow or peripheral blood was sampled as part of follow-up of patients and

63

Chapter 3

contained no leukemic blasts. Genomic DNA was extracted from the cells via phenolchloroform extraction after proteinase K extraction and ethanol precipitation.

Genotyping of the MDRJ gene

For the MDRI C3435T polymorphism a polymerase chain reaction (PCR)-based restriction fragment length polymorphism (RFLP) assay was performed with the primers MDRI F 5 '-TGCTGGTCCTGAAGTTGATCTGTGAAC-3 ' and MDRJ R

5 '-ACATTAGGCAGTGACTCGATGAAGGCA-3 '35 • PCR was carried out

with an initial denaturation of 1 2 minutes at 95°C followed by 30 cycles of

denaturation 95°C for 1 minute, primer annealing for 1 minute at 60°C, and 1 -minute extension at 72°C. The amplified reaction products were

purified using Qiagen PCR purification kit (Qiagen, Valencia, Calif), followed

by digestion with the restriction enzyme Ndell. When the mutation was

present (3435T), digestion of the 248bp PCR-product produced the 1 72-,

60- and 1 6-base pair fragments. The MDRJ G2677T and MDRJ G2677A

polymorphisms were determined by use of the primers and restriction

enzymes as described by Cascorbi et at.36• The digested PCR products were

analyzed on a 2% agarose gel with ethidium bromide staining. To check

the PCR, the amplified reaction products were analyzed on gel. Two different plasmids were used as a positive and negative control for restriction. The majority of the samples were analysed in duplo. Because it was shown that linkage disequilibrium exists between the SNPs C3435T and G2677T, we determined the association between both SNPs and pharmacokinetics by assigning haplotypes according to Johne et al. (haplotype 1 / l : 3435C, 26770; haplotype 1/2: 3435T, 26770; haplotype 2/1 : 3435C, 2677T and haplotype 2/2: 3435T, 2677T)1 7; 19•


Nonparametric methods were used because in our previous studyS we found that the vincristine pharmacokinetic variables were not normally distributed but were lognormally distributed. Before our study, we performed power calculations on the differences we would be able to detect in the pharmacokinetic variables. The power calculation on the AUC given haplotype carriership was as follows: With an estimated group sample size of 52 subjects a 82% power is achieved to detect a difference of 5.0 mg/Limin. between the null hypothesis that both group mean AUCs are 1 0.0 mg/L/min. and the alternative hypothesis that the mean AUC of haplotype 1 / 1 noncarriers is 1 5 .0 mg/L/min. with an estimated group SD of7.0 mg/Limin. (estimated with the SD estimator) and

64


with an a of 0.05000 by use of a 2-sided Mann-Whitney Test with the assumption that the actual distribution is double exponentiation (because the AUC data were skewed to the right and could be normalized by use of a natural log transformation). The Kruskal-Wallis test was used to calculate differences among the 3 genotype groups for C3455T and G2677T. Observed and expected allele frequencies were compared by means of the Hardy-Weinberg test. For differences between haplotype carriers and non carriers, the Mann-Whitney test was used. In addition, we applied a Bonferroni correction for multiple testing, because we tested for 3 different haplotypes and 2 SNPs across the 4 pharmacokinetic variables. Confounding factors for pharmacokinetic variables were identified by use of the Spearman rank test for continuous variables. For categoric variables, the MannWhitney test was used. The logarithms of the pharmacokinetic variables were used as dependent variables in a regression analysis to investigate the impact of the possible confounding factors on vincristine pharmacokinetics. The association between genotypes or haplotypes and the occurrence of constipation was determined with the chi square test. A P-value of less than .05 was considered to indicate statistical significance.

RESULTS

Patients

The characteristics of the 52 patients, in which MDRJ genetic variants could be related to vincristine pharmacokinetics, are displayed in table 1 . All patients except 1 were white in origin.

Distribution of individual SNPs and haplotypes

Fifty-two patients were genotyped for the C3435T SNP of the MDRI gene. The distribution of the genotype was as follows: TIT n = 1 3 (25%); CIT n = 22 (42%); and CIC n = 1 7 (33%). For the SNP at position 2677, both the G2677T and the G2677 A substitutions were investigated. The 2677 TIT genotype was present om eight patients ( 1 5%), 27 patients (52%) were heterozygous, and the 2677 GIG genotype was present in 17 patients (33%). No patients showed the SNP G2677A. For both SNPs, no significant deviations from Hardy-Weinberg equilibrium were observed (P = .78 for C3435T and P = .56 for G2677T). Because linkage disequilibrium exists between the SNPs C3435T and G2677T, we calculated the linkage in our study group. Of the subjects with 3435 CIC, 84.6% ( 1 1 1 1 3) showed 2677 GIG, and this was higher than the overall 2677 GIG frequency of 32.7% ( 1 7 /52). Moreover, 8 1 .8% ( 1 8122) of the subjects with 3435 TIT showed

65

Chapter 3

Table 1 . Patient characteristics

Age <1 y 1 y 2-5 y 6-9 y 10- 16 y

Sex Male Female

Risk group Not high risk High risk

White blood cell count <10 x 109 /L 10-50 x }09/L >50 x 1Q9 /L

Platelets <20 x }09 /L 20-99 X 109 /L 2: 1 00 X 1 09 /L

lmmunophenotype Unknown Common ALL Pre-B-cell ALL Pro-B-cell ALL T-cell ALL

Cytogenetics Unknown Diploid Hypodiploid Pseudodiploid Hyperdiploid (47-50) Hyperdiploid (>50)

ALL, acute lymphoblastic leukemia

No. ofpatients

0 (0%) 4 (8%)

25 (48%) 9 (1 7%) 14 (27%)

32 (62%) 20 (38%)

37 (7 1 %) 1 5 (29%)

24 (46%) 17 (33%) 1 1 (2 1 %)

1 5 (29%) 23 (44%) 14 (27%)

I (2%) 3 1 (60%) 1 3 (25%)

1 (2%) 6 ( 12%)

1 1 (2 1 %) 1 2 (23%) 4 (8%) 7 ( 1 3%) 7 ( 1 3%) 1 1 (2 1 %)

2677 TIT genotype, which was markedly higher than the overall frequency of 1 5 .4%

(7 /52). In short, the linkage was calculated to be 7 1 .2 % (X2 =3 8 .4, P < .00 I).

The SNPs were combined in haplotypes, as described, and the haplotype distribution, compared with a random white sample36, is shown in table 2.

66


Table 2. Comparison of genotype frequencies with a Caucasian random sample36•

Genotype Study sample* Random sample Haplotype pair (No.) (n=687) (%) I n

0/0 1 1 (2 1 %) 1 7.9 I l l I l l 0/1 4 (8%) 10.9 1 / 1 1/2 0/2 2 (4%) 2.3 1 12 1 12 110 2 (4%) 1 .9 I l l 2/1 I l l 1 8 (35%) 38.6 I l l 2/2 112 7 ( 13%) 1 0.5 1 /2 2/2 2/0 0 (0%) 0 2/1 211 2/ 1 0 (0%) 1 .2 2/ 1 2/2 2/2 8 ( 1 5%) 1 6.7 2/2 2/2

* One patient was not from Caucasian origin. Genotype and haplotype pairs were deduced from MDRJ single nucleotide polymorphisms (SNPs) 2677G > T and 3435C > T according to Johne et al. 1 7 • Genotype coding is as follows: 0, homozygous for nucleotides identical with the reference sequence for the position at both chromosomes; I , heterozygous; 2, homozygous for nucleotides different from the reference sequence. Haplotype coding as follows: 1 , identical with the reference sequence (2677 G; 3435C); 2 different from the reference sequence (2677T; 3435T).

Table 3. Impact of MDRJ SNPs C3435T and G2677T on vincristine pharmacokinetics

CL AUC Vd .. tl/2 11 tl/2 � (rnL·min 1 ·m·2) (mg·L·1 ·min·1 ) (Um2) (min) (min)

SNP C3435T CC (n = I 3) 223(63-720) 7.3 (2. 1 -24.3) 330 (59- 1 278) 7.3 (4.3-1 1 . 1 ) 1 1 55 (545-5030) CT (n = 22) 205 (44- 1 009) 7.3 ( 1 .4-34.5) 254 (49-2088) 6.5 (3.6-1 0. I) 1 1 94 (641 -3231 ) TT (n = 1 7) 256 (34- 1 1 38) 5.8 ( 1 .4-44. 7) 2 1 8 (75- 1408) . 7.0 (5.6- 1 0.0) 805 (434-3623) ? value .89 1 .89 1 705 . .060 094

SNP G2677T GG (n = 1 7) 235 (63- 1 1 38) 6.4 ( 1 .4-24.3) 330 (59- 1 278) 7.5 (3.6- 1 1 . 1 ) 1 1 55 (545-5030) GT (n = 27) 26 1 (44- 1 04 1 ) 5.8 ( 1 .4-34.5) 294 (49-2088) 6.5 (4.6- 1 0.0) 898 (583-323 1 ) TT (n = 8) 1 90 (34-552) 7.8 (2.8-44.7) 1 87 (75-986) 7.0 ( 6.0-9 .3) 1 200 (434-3623)

? value .430 .422 .350 .080 .749

The G2677 A SNP was not present in our study population. Values are given as medians and range. CL, Total body clearance; AUC, area under the concentration-time curve; Vd

ss.

apparent volume of distribution at steady state; t112 a' distribution half-life; t1 '2 P' elimination half-life. • Kruskal-Wallis test for differences among the 3 groups.

67

Chapter 3

Influence of individual SNPs and haplotypes on vincristine pharmacokinetics

The possible differences in vincristine pharmacokinetics among the 3 C3435T and 02677T genotypes were calculated. No differences were found in the vincristine CL, AUC, Vdss' t112

a and t112 � among 3435 C/C, 3435 CIT and 3435 TIT genotypes (table 3). The SNP 02677T also did not affect vincristine pharmacokinetic variables (table 3). Because of the linkage disequilibrium between C3435T and 02677T, haplotypes were assigned as described. The effect ofhaplotypes on vincristine pharmacokinetics is displayed in table 4. Carriers of haplotype 1/1 (26770 3435C) showed a longer t112 P

than noncarriers ( 1 1 56 versus 805 minutes, P = .038) (Figure 1 ) . In contrast, the haplotype 1 /2 (26770 3435T) carriers had a shorter t112 P than noncarriers (805 versus 1 1 80 minutes, P = .044). However, when a Bonferroni correction for multiple testing was applied, the impact of haplotypes I l l and 1/2 on elimination half-life lost its significance. Haplotype 1 / 1 and 1/2 did not influence vincristine CL, AUC, Vd •• and t112 a· Haplotype 2/2 did not show any effect on the pharmacokinetic variables.

Table 4. Impact of MDRJ haplotypes 1 / 1 , 1 12 and 2/2 on vincristine pharmacokinetics

CL AUC Vd,. tll2a tl/2 p (mL·min·1·m·Z) (mg·L·1 ·min·1) (Um2) (min) (min)

Haplotype Ill Carrier 223 (44-1 009) 7.3 ( 1 .4-34.5) 294 (49-2088) 6.6 (3.6- 1 1 . 1 ) 1 1 56 (545-5030)

(n = 35) Noncarrier 256 (34- 1 1 38) 5.8 ( 1 .4-44.7) 2 1 8 (75- 1408) 7.0 (5.6-1 0.0) 805 (434-3623)

(n = 1 7) P va1ue .633 .646 .489 .297 .038' Haplotype 112 Carrier 264 (74- 1 1 38) 5.6 ( 1 .4-20.5) 320 (75-1408) 7.0 (3.6- 1 0. 1 ) 805 (625-2 1 43)

{n = 1 3)

Noncarrier 202 (34-1 009) 7.5 ( 1 .4-44.7) 261 (49-2088) 6.8 (4.3- 1 1 . 1 ) 1 1 80 ( 434-5030) (n = 39) P va1ue . 1 36 . 1 25 .775 .841 .044" Haplotype 212 Carrier 203 (34- 1 04 1 ) 7.4 ( 1 .4-44. 7) 246 (49-2088) 6.6 (4.6-1 0.0) 969 (434-3623) (n = 33) Noncarrier 235 (63- 1 1 38) 6.4 ( 1 .4-24.3) 330 (59-1278) 7.3 (3.6- 1 1 . 1) 1 140 (545-5030)

(n = 1 9)

P va1ue .

.725 .752 .500 .206 . 9 1 7

Haplotype 2 1 1 carriers were not present in our study population. Values are given as medians and range. · Mann-Whitney test for differences among the 2 groups. "Not significant when corrected with Bonferroni for multiple testing.

68


Elimination half-life

6000

5000 • •

4000 Cl) • • Cl) - • • :I 3000 c .E • •

2000 •• •• • • I

1 000

•• :: . •• ••

� ••• •• ...:as. •••••• -:n:- "1!r ·•·i•·· •:. • •

• •

0 haplotype 1/1 haplotype 1/2

carrier non·carrier earner non·camer

Figure 1 . Effect of haplotypes on elimination half-life. Scatterplot showing the elimination half-life of haplotype 1 / 1 carriers versus noncarriers (P == .038), and haplotype 1 /2 carriers and noncarriers (P == .044) in minutes.

Correlations of vincristine pharmacokinetics

When the impact of haplotypes on vincristine CL is being studied, other factors might be confounding by influencing vincristine pharmacokinetics. Therefore the effect of possible confounding factors, such as sex, cytogenetics, and immunophenotype, was studied on vincristine CL. Vincristine CL correlated negatively with y-glutamyltransferase at diagnosis (n = 25; r = -0.66; P < .00 1 ),

but not with other biochemical measurements in addition to age at diagnosis, weight, height, body surface area and sex. Vincristine CL was also influenced by immunophenotype, with T-cell ALL patients showing a lower vincristine CL than pre-B-cell ALL patients (n = 5 1 ; Kruskal-Wallis, P = .0 1 6) (table 5). Moreover, the cytogenetics of leukemic blasts correlated with vincristine CL as well (n = 40;

Kruskal-Wallis, P = .0 1 6). Vincristine CL was faster in patients with hyperdiploid blasts (>50 chromosomes) than in patients with diploid or hyperdiploid blasts (46-50

chromosomes) (table 5). In multiple linear regression analysis, the 3 factors were analyzed in combination with the effect ofhaplotypes on vincristine CL to check for possible confounding. y-Glutamyltransferase, immunophenotype, and cytogenetics did not show any influence on the effect of haplotypes on vincristine CL.

During the vincristine pharmacokinetic study, no other cytotoxic drugs or steroids were given. However, individual patients received allopurinol (n = 24),

69

Chapter 3

Table 5. Factors affecting vincristine clearance

Immunophenotype (P =.0 1 6*) Common ALL (n = 3 1) Pre-B-cell ALL (n = 13) Pro-B-cell ALL (n = 1 ) T-cell ALL (n = 6)

Cytogenetics (P = .01 6*) Diploid (n = 12) Hypodiploid (n = 4) Pseudodiploid (n = 7) Hyperdiploid (46-50) (n = 6) Hyperdiploid (>50) (n = 1 1 )

Values are given as medians and range.

Vincristine clearance (mL·min·Lm·2)

223.2 (43 .7 - 1 1 38.0) 329.7 (72.7 - 1 041 .0)

8 14.5 99.3 (34. 1 - 260.7)

1 1 1 .7 (34. 1 - 617 . 1 ) 1 85.0 ( 1 15 .4 - 225.7) 256. 1 (69.5 - 814.5) 1 20.2 (72.7 - 260.7)

329.7 (46.0 - 1 1 38.0)

* Kruskal-Wallis test for differences among more than two groups

cephalosporins (n = 10), acetoaminophen (n = 7), or furosemide (n = 6). Of these drugs, cephalosporins are known P-gp substrates and could, therefore, influence vincristine pharmacokinetics by competing for P-gp-mediated transport or inducing cytochrome P450 (CYP) 3A4, a vincristine-metabolizing enzyme. Therefore the effect of the drugs on the vincristine pharmacokinetic variables was determined. Patients who received allopurinol or furosemide had a shorter distribution half-life (t112 ex) than patients who did not receive the drugs (P = .046 and P = .039, respectively). However, these drugs did not show any effect on the vincristine CL,

AUC, V dss and t112 P" To investigate whether allopurinol or furosemide influenced the effect of haplotypes on t112 a' multiple linear regression analysis was performed and no confounding effect of the drugs on t112a was observed.

Influence of individual SNPs and haplotypes on vincristine side effects

The correlation between vincristine pharmacokinetic variables and severity of constipation was investigated. All patients received bisacodyl on the day vincristine was administered and lactitol the following days. Patients with constipation grade 3 or 4 did not have a lower CL, larger AUC, or longer t112 a and t112 P than patients with constipation grade 1 or 2. Thus we found no connection between vincristine exposure and severity of constipation. Furthermore, we determined whether genetic variants of the MDR-1 gene contribute to the occurrence of severe constipation, a frequently occurring side effect of vincristine (table 6). We found no difference in the occurrence of severe constipation

70


Table 6. SNPs, haplotypes, and occurrence of severe constipation

SNPs Constipation Constipation X2 Test grade 1 -2 grade 3-4 P value

C3435T P = .82 cc 8/1 3 (62%) 5/13 (38%) CT 13/ 1 8 (72%) 5/1 8 (28%) TT 8/1 2 (67%) 4/1 2 (33%)

G2677T P = . 1 2 GG 9/1 7 (53%) 8/1 7 (47%) GT 1 8/22 (82%) 4/22 ( 1 8%) TT 2/4 (50%) 2/4 (50%)

Haplotype 1 1 1 P = .95 Carrier 2 1/3 1 (68%) 10/3 1 (32%) Noncarrier 8/1 2 (67%) 4/1 2 (32%)

Haplotype 1 12 P = .95 Carrier 8/1 2 (67%) 4/1 2 (32%) Noncarrier 2113 1 (67%) 10/3 1 (32%)

Haplotype 2/2 P = .24 Carrier 1 8/24 (75%) 6/24 (25%) Noncarrier 1 1/ 19 (58%) 8/1 9 (42%)

(grade 3 or 4) among the 3 C3435T and G2677T genotypes (P = .82 and P = . 1 2, respectively). Moreover, haplotype Ill carriers, with a longer vincristine t112 11, did not have severe constipation more often than noncarriers (P = .95). There were also no differences in grade of constipation between haplotype 1 /2 carriers and noncarriers or between haplotype 2/2 carriers and noncarriers, respectively (P = .95 and P = .24, respectively).

DISCUSSION

We observed no effect of C3435T and/or G2677T on vincristine pharmacokinetics. The incidence of C3435T varies according to racial background. The 3435TT genotype is observed in approximately 25 % of white persons, with a much higher prevalence in black persons19;35;37• The distribution ofC3435T genotypes in our study sample was comparable to data from a large cohort of white subjects (table 2)17• In addition, the allelic distribution of G2677T is similar to that found in a large cohort (table 2). Recently, it was suggested that the use of haplotypes is superior to single SNP analysis to predict pharmacokinetics16;17• After all, a genetic variant of C3435T may cause higher P-gp activity, whereas another genetic variant of G2677T can have

71

Chapter 3

the opposite effect. Therefore we assigned haplotypes and analyzed their impact on vincristine pharmacokinetics. Significant differences in elimination half-life were found between haplotype 1 1 1 carriers (26770, 3435C) and noncarriers and also between haplotype 1/2 carriers (2677G, 3435T) and noncarriers. However, these differences lost their significance when we applied a Bonferroni correction for multiple testing. Moreover, total body clearance was not significantly different in these groups of carriers and noncarriers, which means that differences in half-life must be (partly) attributed to differences in distribution volume. Total body clearance is inversely related to AUC, a reflection of systemic exposure. The absence of a significant difference in vincristine clearance, therefore, suggests that haplotype carrier status does not affect in vivo exposure of leukemic cells to vincristine. In brief, we were unable to show a significant impact of MDRJ genetic variants on vincristine exposure. Until now, only a few studies have investigated the clinical consequences of MDRJ polymorphisms and pharmacokinetics of anticancer drugs. The drugs studied, irinotecan, docetaxel, and etoposide38-4°, revealed a relationship between MDRJ polymorphisms and pharmacokinetics only for etoposide. Several factors may play a role for the limited observed effect of MDRJ genotypes on pharmacokinetics. First, the SNPs C3435T and G2677T/A are also linked to the synonymous SNP C l236T. We did not determine the influence of Cl 236T on vincristine pharmacokinetics, because no correlation between this SNP and pharmacokinetics was found so far. However, we cannot exclude that other MDRJ polymorphisms, including C 1 236T, may contribute to the large interindividual variability of vincristine pharmacokinetics. Another important factor is the route of administration of the drugs. By studying intravenously administered drugs, the possible influence of MDRJ genetic variants on the gastrointestinal uptake is excluded. The impact of genetic variants on pharmacokinetics is, therefore, diminished. Moreover, not only is vincristine a substrate for P-gp but it can also be transported by other adenosine triphosphate-dependent drug efflux pumps, such as the members of the multidrug resistance-associated protein (MRP) family MRPl and MRP241;43• In the liver, MRPs could compensate for low P-gp functional activity by upregulating their own expression and thus mask the effect of P-gp haplotypes on biliary excretion. Besides drug transporters, drug metabolizing enzymes (eg CYPs) are also major important biologic factors determining drug pharmacokinetics8• P-gp and CYP3A4 have a significant overlap in substrates, and vincristine is also a substrate for CYP3A4-metabolizing enzymes44;45• CYP3A4 activity could be influenced by P-gp

72


functional activity in our study, because interactions between both proteins have been described previously"6• Finally, environmental factors affecting P-gp functional activity, including food constituents and comedication, may play an important role47• The use of drugs that induce CYP3A4 metabolic activity or compete with vincristine for P-gp mediated transport can affect vincristine pharmacokinetics. We corrected for the coadministration of these drugs in our analysis and found no effect on vincristine pharmacokinetics. Therefore it can be concluded that the co-administration of these drugs had no confounding effect on the impact of haplotypes on vincristine pharmacokinetics. In addition to comedication, diet may affect P-gp activity. All patients received a normal hospital-diet. Possible dietary differences such as herbs or grapefruit juice were not taken into account, but these food constituents are not generally used by children. We observed a lower vincristine clearance in T-cell ALL patients than pre-B-cell ALL patients. It is unlikely that the vincristine clearance is directly influenced by the leukemic blasts. Patients with T-cell ALL have associated characteristics that may influence the vincristine clearance, such as high white blood cell count, which could affect liver or renal function. The impact of cytogenetics on vincristine clearance can also probably be explained by associated characteristics and not features of the hyperdiploid leukemic blasts themselves. One study found an association between MDRJ genetic variants and the occurrence of side effects of the antidepressant amitriptyline48• We were interested in the role of MDRJ genetic variants in the occurrence of vincristine side effects and focussed on constipation. Of the additional drugs our patients received, only ondansetron can cause constipation, but this drug was administered to only 1 patient. No effect of individual SNPs or haplotypes on the severity of constipation could be observed. Although our study groups were relatively small, it seems unlikely that an existing association between haplotypes and grade of constipation was missed, because there was not even a remote suggestion for such a relation in the data, as depicted in table 6. In conclusion, the genetic variants in the MDRJ gene alone can not explain the large variability in vincristine pharmacokinetics in childhood ALL. Polymorphisms in the MDRJ gene alone do not identify patients with an increased risk for the occurrence of vincristine-induced constipation. Many factors are involved in determining vincristine pharmacokinetics, and proteomics may help to identify protein activity and interactions of interest, to better predict vincristine pharmacokinetics and to identify a group at risk for vincristine-induced toxicity.

73

Chapter 3

ACKNOWLEDGEMENTS

We thank the medical, nursing and laboratory staff of the following hospitals for enrolling patients in the study: University Hospital, Groningen; University Medical Centre St. Radboud, Nijmegen; Bosch Medicentrum, 's Hertogenbosch; Catharina Hospital, Eindhoven; Clara Hospital, Rotterdam; University Medical Centre, Leiden; University Medical Centre, Utrecht; Rijnstate Hospital, Arnhem; Tjongerschans Hospital, Heerenveen; Vrije Universiteit Medical Centre, Amsterdam.

74


REFERENCES

1 . Huschtscha LI, Bartier WA, Ross CE, Tattersall MH. Characteristics of cancer cell death after exposure to cytotoxic drugs in vitro. Br J Cancer 1 996: 73: 54-60.

2. Jordan MA, Thrower D, Wilson L. Mechanism of inhibition of cell proliferation by Vinca alkaloids. Cancer Res 1 99 1 : 51 : 2212-2222.

3 . Bender RA, Castle MC, Margileth DA, Oliverio VT. The pharmacokinetics of [3H]vincristine in man. C/in Pharmacal Ther 1 977: 22: 430-435.

4. Jackson DV, Jr., Castle MC, Bender RA. Biliary excretion of vincristine. C/in Pharmacal Ther 1 978: 24: 1 0 1 - 1 07.

5. Groninger E, Meeuwsen-De Boer T, Koopmans P, Uges D, Sluiter W, Veerman A et al. Pharmacokinetics of vincristine monotherapy in childhood acute lymphoblastic leukemia. Pediatr Res 2002: 52: l l3 - l l 8.

6. Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, Gottesman MM. Biochemical, cellular, and pharmacological aspects of the multi drug transporter. Annu Rev Pharmacal Toxico/ 1 999: 39: 361-398.

7. Fojo AT, Ueda K, Slamon DJ, Poplack DG, Gottesman MM, Pastan I. Expression of a multidrug-resistance gene in human tumors and tissues. Proc Nat/ Acad Sci US A 1 987: 84: 265-269.

8. Sakaeda T, Nakamura T, Okumura K. Pharmacogenetics of MDR1 and its impact on the pharmacokinetics and pharmacodynamics of drugs. Pharmacogenomics 2003 : 4 : 397-4 10.

9. Schinkel AH. Pharmacological insights from P-glycoprotein knockout mice. Int J Clin Pharmacal Ther 1 998: 36: 9-1 3 .

1 0. Evans WE, McLeod HL. Pharmacogenomics-drug disposition, drug targets, and side effects. N Eng/ J Med 2003 : 348: 538-549.

1 1 . Song S, Suzuki H, Kawai R, Sugiyama Y. Effect ofPSC 833, a P-glycoprotein modulator, on the disposition of vincristine and digoxin in rats. Drug Metab Dispos 1 999: 27: 689-694.

1 2. Watanabe T, Miyauchi S, Sawada Y, Iga T, Hanano M, Inaba M et al. Kinetic analysis of hepatobiliary transport of vincristine in perfused rat liver. Possible roles of P-glycoprotein in biliary excretion of vincristine. J Hepato/ 1 992: 16: 77-88.

13 . Watanabe T, Suzuki H, Sawada Y, Naito M, Tsuruo T, Inaba M et al. Induction of hepatic P-glycoprotein enhances biliary excretion of vincristine in rats. J Hepato/ 1 995: 23: 440-448.

14. Marzolini C, Paus E, Buclin T, Kim RB. Polymorphisms in human MDRl (P-glycoprotein): recent advances and clinical relevance. C/in Pharmacal Ther 2004: 75: 1 3-33.

15 . Hoffmeyer S, Burk 0, von Richter 0, Arnold HP, Brockmoller J, Johne A et al . Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P- glycoprotein expression and activity in vivo. Proc Nat/ Acad Sci U S A 2000: 97: 3473-3478.

1 6. Chowbay B, Cumaraswamy S, Cheung YB, Zhou Q, Lee EJ. Genetic polymorphisms in MDRl and CYP3A4 genes in Asians and the influence ofMDRl haplotypes on cyclosporin disposition in heart transplant recipients. Pharmacogenetics 2003 : 13: 89-95.

17 . Johne A, Kopke K, GerloffT, Mai I, Rietbrock S, Meisel C et al. Modulation of steadystate kinetics of digoxin by haplotypes of the P- glycoprotein MDRl gene. C/in Pharmacal Ther 2002: 72: 584-594.

75

Chapter 3

1 8. Sakaeda T, Nakamura T, Horinouchi M, Kakumoto M, Ohmoto N, Sakai T et al. MDRl genotype-related pharmacokinetics of digoxin after single oral administration in healthy Japanese subjects. Pharm Res 200 1 : 18: 1400-1404.

1 9. Kim RB, Leake BF, Choo EF, Dresser GK, Kubba SV, Schwarz UI et al. Identification of functionally variant MDRl alleles among European Americans and African Americans. Clin Pharmacal Ther 200 1 : 70: 1 89- 1 99.

20. Siegmund W, Ludwig K, Giessmann T, Dazert P, Schroeder E, Sperker B et al. The effects of the human MDRl genotype on the expression of duodenal P-glycoprotein and disposition of the probe drug talinolol. Clin Pharmacal Ther 2002: 72: 572-583.

2 1 . Verstuyft C, Schwab M, Schaeffeler E, Kerb R, Brinkmann U, Jaillon P et al. Digoxin pharmacokinetics and MDRl genetic polymorphisms. Eur J Clin Pharmacol 2003: 58: 809-8 1 2.

22. Kurata Y, Ieiri I, Kimura M, Morita T, Irie S, Urae A et al. Role of human MDRl gene polymorphism in bioavailability and interaction of digoxin, a substrate of P-glycoprotein. Clin Pharmacal Ther 2002: 72: 209-2 1 9.

23. Anglicheau D, Verstuyft C, Laurent-Puig P, Becquemont L, Schlageter MH, Cassinat B et al. Association of the multidrug resistance-! gene single-nucleotide polymorphisms with the tacrolimus dose requirements in renal transplant recipients. JAm Soc Nephrol 2003 : 14: 1 889-1 896.

24. Goto M, Masuda S, Saito H, Uemoto S, Kiuchi T, Tanaka K et al. C3435T polymorphism in the MDRl gene affects the enterocyte expression level of CYP3A4 rather than Pgp in recipients ofliving-donor liver transplantation. Pharmacogenetics 2002: 12: 45 1 -457.

25. Illmer T, Schuler US, Thiede C, Schwarz UI, Kim RB, Gotthard S et al. MDRl gene polymorphisms affect therapy outcome in acute myeloid leukemia patients. Cancer Res 2002: 62: 4955-4962.

26. Brinkmann U, Eichelbaum M. Polymorphisms in the ABC drug transporter gene MDR l . Pharmacagenamics J 200 1 : 1 : 59-64.

27. Gidding CE, Kellie SJ, Kamps WA, de Graaf SS. Vincristine revisited. Crit Rev Oneal

Hematal l999: 29: 267-287. 28. Hussain M, Wozniak AJ, Edelstein MB. Neurotoxicity of antineoplastic agents. Crit Rev

Oneal Hematol l993 : 14: 6 1 -75. 29. Saito T, Zhang ZJ, Shibamori Y, Ohtsubo T, Noda I, Yamamoto T et al. P-glycoprotein

expression in capillary endothelial cells of the 7th and 8th nerves of guinea pig in relation to blood-nerve barrier sites. New·asci Lett 1 997: 232: 4 1 -44.

30. Saito T, Zhang ZJ, Ohtsubo T, Noda I, Shibamori Y, Yamamoto T et al. Homozygous disruption of the mdrla P-glycoprotein gene affects blood-nerve barrier function in mice administered with neurotoxic drugs. Acta Otalaryngal 200 1 : 121 : 735-742.

3 1 . Hammond IW, Ferguson JA, Kwong K, Muniz E, Delisle F. Hyponatremia and syndrome of inappropriate anti-diuretic hormone reported with the use of Vincristine: an over-representation of Asians? Pharmacaepidemial Drug Saf2002: 11 : 229-234.

32. Bloemhof H, Van Dijk KN, de Graaf SS, Vendrig DE, U ges DR. Sensitive method for the determination of vincristine in human serum by high-performance liquid chromatography after on-line column-extraction. J Chramatogr 1 99 1 : 572: 1 7 1 - 1 79.

33. Koopmans P, Gidding CE, de GraafSS, Uges DR. An automated method for the bioanalysis of vincristine suitable for therapeutic drug monitoring and pharmacokinetic studies in young children. Ther Drug Manit 200 1 : 23 : 406-409.

76


34. de Graaf SS, Bloemhof H, Vendrig DE, Uges DR. Vincristine disposition in children with acute lymphoblastic leukemia. Med Pediatr Onco/ 1 995: 24: 235-240.

35. Ameyaw MM, Regateiro F, Li T, Liu X, Tariq M, Mobarek A et al. MDR l pharmacogenetics: frequency of the C3435T mutation in exon 26 is significantly influenced by ethnicity. Pharmacogenetics 200 1 : 1 1 : 2 1 7-22 1 .

36. Cascorbi I , Gerloff T, Johne A, Meisel C, Hoffmeyer S , Schwab M et al. Frequency of single nucleotide polymorphisms in the P-glycoprotein drug transporter MDR1 gene in white subjects. Clin Pharmacal Ther 200 1 : 69: 1 69- 1 74.

37. Schaeffeler E, Eichelbaum M, Brinkmann U, Penger A, Asante-Poku S, Zanger UM et al. Frequency ofC3435T polymorphism ofMDRl gene in African people. Lancet 200 1 : 358: 383-384.

38. Goh BC, Lee SC, Wang LZ, Fan L, Guo JY, Lamba J et al. Explaining interindividual variability of docetaxel pharmacokinetics and pharmacodynamics in Asians through phenotyping and genotyping strategies. J Clin Onco/ 2002: 20: 3683-3690.

39. Sparreboom A, Marsh M, Mathijssen RHVJ, McLeod H. Irinotecan (CPT- 1 1 ) disposition in relation to single-nucleotide polymorphisms (SNPs) of ABC transporters and drug-metabolizing enzymes. Proc Am Soc Clin Onco/ 2002: 21 : 83a (abstract) . .

40. Kishi S, Yang W, Boureau B, Morand S, Das S, Chen P et al. Effects of prednisone and genetic polymorphisms on etoposide disposition in children with acute lymphoblastic leukemia. Blood 2004: 103: 67-72.

4 1 . Cole SP, Sparks KE, Fraser K, Loe DW, Grant CE, Wilson GM et al. Pharmacological characterization ofmultidrug resistant MRP-transfected human tumor cells. Cancer Res 1 994: 54: 5902-5910.

42. Cui Y, Konig J, Buchholz JK, Spring H, Leier I, Keppler D. Drug resistance and ATPdependent conjugate transport mediated by the apical multidrug resistance protein, MRP2, permanently expressed in human and canine cells. Mol Pharmacal 1 999: 55: 929-937.

43. Zaman GJ, Flens MJ, van Leusden MR, de Haas M, Mulder HS, Lankelma J et al. The human multidrug resistance-associated protein MRP is a plasma membrane drug-efflux pump. Proc Nat! A cad Sci U S A 1 994: 91: 8822-8826.

44. Villikka K, Kivisto KT, Maenpaa H, Joensuu H, Neuvonen PJ. Cytochrome P450-inducing antiepileptics increase the clearance of vincristine in patients with brain tumors. Clin Pharmacal Ther 1999: 66: 589-593.

45. Yao D, Ding S, Burchell B, Wolf CR, Friedberg T. Detoxication of vinca alkaloids by human P450 CYP3A4-mediated metabolism: implications for the development of drug resistance. J Pharmacal Exp Ther 2000: 294: 387-395.

46. Lan LB, Dalton JT, Schuetz EG. Mdr1 limits CYP3A metabolism in vivo. Mol Pharmacol 2000: 58: 863-869.

47. Tian R, Koyabu N, Takanaga H, Matsuo H, Ohtani H, Sawada Y. Effects of grapefruit juice and orange juice on the intestinal efflux of P-glycoprotein substrates. Pharm Res

2002: 19: 802-809. 48. Roberts RL, Joyce PR, Mulder RT, Begg EJ, Kennedy MA. A common P-glycoprotein

polymorphism is associated with nortriptyline-induced postural hypotension in patients treated for major depression. Pharmacogenomics J 2002: 2: 1 9 1 - 1 96.

77

Chapter 4

Breast Cancer Resistance Protein (BCRP) in acute leukemia

Sabine L.A. Plasschaert1, Dorina M. van der Kolk1, Eveline S.J.M. de Bont1, Edo Vellenga2, Willem A. Kamps1,

Elisabeth G.B. de Vries3

Departments of Paediatric Oncology and Haematology1, Haematology and Medical Oncology, University Hospital Groningen,

Groningen, The Netherlands

Leukemia & Lymphoma 2004; 45: 649-654

Chapter 4

ABSTRACT

Multidrug resistance, cross-resistance to structurally and functionally unrelated drugs, is an important cause of treatment failure in acute leukemia. Multidrug resistance can result from the overexpression of ATP-dependent efflux pumps, such as P-glycoprotein and members of the multidrug resistance associated protein (MRP) family. Recently a novel transporter has been identified, which is called breast cancer resistance protein (BCRP), ABCG2 or mitoxantrone resistance protein. BCRP confers resistance to chemotherapeutic agents, such as mitoxantrone, doxorubicin and daunorubicin. This review describes BCRP detection techniques and the normal physiology of BCRP. The role of BCRP in the physiology of hematopoietic stem cells is addressed as well as the involvement of BCRP in multidrug resistance in acute leukemia. In AML and ALL, several studies showed that BCRP is expressed and functionally active at low, but variable levels. However, further studies are warranted to investigate its effect on clinical outcome, and explore whether patients could benefit from the combination of BCRP inhibitors and chemotherapy.

80

BCRP in acute leukemia

INTRODUCTION

Development of resistance to multiple structurally and functionally unrelated chemotherapeutic drugs, the so-called multidrug resistance, is an important cause of treatment failure in acute leukemia. One mechanism of multidrug resistance is the overexpression of members of the ABC superfamily of membrane transporters1 • This family of transporters consists of an increasing number of proteins, of which P-glycoprotein and multidrug resistance associated proteins have been extensively studied in leukemia2;3 • Another ABC transporter has been identified in a multidrug resistant human breast cancer cell line MCF-7/AdrVp. This cell line shows an ATP-dependent reduction in the intracellular accumulation of anthracycline anticancer drugs, in the absence of the overexpression of known multidrug resistance transporters such as P-gp and MRP4-6• The transporter, first described in 1 998, received several names, including breast cancer resistance protein (BCRP), ABCG2, placental transporter or mitoxantrone resistance protein7;8• The BCRP protein is a 655 amino acid protein and its gene is mapped on chromosome 4q227• The BCRP gene spans over 66kb and consists of 1 6 exons and 1 5 introns9• The protein is a half-transporter with 1 intracellular ATP-binding site, and 6 putative transmembrane domains, and may form homodimers or heterodimers bridged by disulfide bonds in order to function as an active transporter (Figure 1 ) 10• Enforced expression of the full length BCRP eDNA in MCF-7 breast cancer cells confers resistance to mitoxantrone, doxorubicin and daunorubicin, bisantrene and topo-isomerase II

inhibitors5;1 1 ; 12• A summary of all the chemotherapeutic drugs, which were found to be BCRP substrates, is given in Table 1 . This review gives an overview of the current knowledge concerning BCRP and will specifically address the role of BCRP in acute leukemia.

I ntracellular

Figure 1 Membrane topology model for BCRP. The half-transporter contains six transmembrane regions and the ATP binding cassette (ABC).

81

Chapter 4

Table 1. Chemotherapy substrates of BCRP

Anthracyc1ines

Topo-isomerase !-inhibitors

Topo-isomerase II-inhibitors

Cell-cycle inhibitors

Mitoxantrone Doxorubicin Daunorubicin Epirubicin Irinotecan ( CPT- 1 1 ) SN-38 9-aminocamptothecin Topotecan Bisantrene Etoposide (VP-1 6) Teniposide F1avopirido1

BCRP expression in normal tissues

References

5; 1 1 ; 1 2;38;5 1 -53 5;8; 1 1 ; 12;5 1 5;8 ; 1 1 ;5 1 52 12;32;52;54 32;52;53 32;52 1 1 ; 12;32;5 1 -54 1 1 ; 12 12;5 1 ;54 12 53

The physiological function of BCRP can be explained by its localization in normal tissues13• In drug-resistant cell lines BCRP was found to be localized at the plasma membrane14: 1 5 • It was also called placental transporter, because of its high expression in the placental syncytiotrophoblasts. This abundant expression suggests an important role for BCRP protecting the fetus against xenobiotics. BCRP is also expressed in the epithelium of the small intestine and colon, where it regulates the uptake of certain orally administered substrates from the gastro-intestinal tract, by transporting substrates back into the gut lumen. In the liver, BCRP is localized in the canicular membrane, where it is responsible for the excretion of substances in bile. BCRP is also present in ducts and lobules of the breast, where its function is still subject of investigations. Furthermore, in almost all tissues, the venous and capillary endothelium express BCRP, in contrast to the arterial endothelium13• In the blood-brain barrier BCRP, just like P-gp, is mainly located at the luminal surface of microvessel endothelium, suggesting a role in protection of the brain against drugs16• Both the central nervous system and the testis are sanctuary sites for chemotherapy. These sanctuary sites are maintained by the blood-brain and -testis barrier, which hinder the delivery of cytotoxic drugs. For prevention of CNS or testicular relapses of leukemia, the ability to deliver more chemotherapy over the barrier would be helpful and this could potentially be achieved by inhibiting transporters in this barrier1 7• BCRP was recently found to play an important role in the physiology of normal pluripotent hematopoietic stem cells. Hematopoietic stem cells can be identified by a 'side population' phenotype, based on the efflux of fluorescent dyes such

82


as rhodamine 123 and Hoechst 3334218;19• P-gp as well as BCRP is capable of transporting Hoechst, but the presence of P-gp is not required for the 'side population' phenotype (as was shown in MDRl knock-outs20). Interestingly, BCRP knock-out mice have a normal number of hematopoietic stem cells in the total bone marrow, but an almost complete loss of 'side population' cells. This suggests that BCRP expression is required for hematopoietic stem cells to display the 'side population' , but with the loss of BCRP the hematopoietic stem cells are still maintained and appear outside the 'side population' region. Furthermore, hematopoietic stem cells from BCRP knock-out mice were significantly more sensitive to mitoxantrone after transplantation in drug-treated mice, implying a physiological role of BCRP in providing protection from cytotoxic substrates (2 1 ) . BCRP is expressed at high levels in primitive hematopoietic stem cells, and is sharply downregulated with differentiation20;2 1 • Zhou et a/. showed that enforced expression of BCRP in hematopoietic cells caused a defect in hematopoietic reconstitution, suggesting an important role in stem cell biology22• BCRP might regulate maturation by expelling substrates that are capable of inducing differentiation23 • This is particularly interesting for a disease like acute leukemia, in which a maturation arrest takes place. However, in BCRP knock-out mice no defect in hematopoiesis could be observed20, but alternative transporters such as P-gp might well have compensated for the lack of BCRP function. Mice lacking BCRP developed necrotic ear laesions, when exposed to standard fluorescent light and fed with a particular batch of food. Further investigations showed, that the phototoxicity was caused by the chlorophyl-catabolite pheophorbide A, a substrate in various plant-derived food and food supplements24• BCRP transports pheophorbide A and can limit the intestinal uptake from pheophorbide A-containing food. These mice also displayed protoporphyria, characterized by increased levels of protoporphyrin IX, which is structurally related to pheophorbide A. In summary, BCRP knock-out mice showed the role of BCRP in the 'side population' of hematopoietic stem cells and their protection from cytotoxic substrates, and its role in processes involving the handling of porphyrins. The function of BCRP in differentiation of stem cells remains elusive.

Detection of BCRP

To detect the BCRP protein expression 2 monoclonal antibodies have been developed: BXP-34 and BXP-2 1 13; 15 . These antibodies can be used for immunoblots, immunohistochemistry and flow cytometry13; 15;25-27• Recently polyclonal antibodies against peptides corresponding to 3 different epitopes on the BCRP protein were developed, which can also be used for immunoblot analysis and

83

Chapter 4

immunohistochemistry28 • The functional activity of BCRP can be studied by measuring the intracellular accumulation of BCRP substrates with or without inhibition of BCRP specific modulators. Since BCRP has differing substrate specificities, depending on mutations in the BCRP gene (see BCRP polymorphisms)29, the BCRP substrate used in the functional assay should not be influenced by these mutations. In the functional assays for BCRP detection, mitoxantrone and prazosin (a-adrenergic antagonist) turned out to be suitable substrates1 1 :30 QUOTE '"' . Mitoxantrone however, can also be transported by P-gp and MRP l . To distinguish BCRP-mediated from P-gp and MRP l mediated mitoxantrone transport, a BCRP specific modulator is required. BCRP modulators include the acridone carboxamide derivative GF1 209 1 83 1 ;32 and fungal FTC12;Jo;33• FTC specifically modulates BCRP, whereas GF1 209 1 8 also modulates P-gp mediated transport34• Nevertheless, FTC itself is not suitable for clinical application because of its neurotoxicity, and therefore other BCRP modulators are being developed and evaluated35• Recently a new tetracyclic analogue of FTC, Ko143, was detected as highly potent and applicable in vivo (in mice), making it a good candidate for clinical use36• In conclusion, BCRP functional activity can be determined by flow cytometry using mitoxantrone or prazosin as substrates and FTC or Ko143 as modulators.

BCRP polymorphisms

Several studies reported a disparity concerning the ability of BCRP to transport rhodamine 123. Since it was known that mutations in a transporter can be responsible for differing substrate specificities, the BCRP mRNA was screened for mutations in 1 1 different cell lines, of which 2 were capable of effluxing rhodamine29• Differences at the position of amino acid 482 were found to predict the ability of BCRP to transport rhodamine. Cell lines overexpressing wild-type BCRP with an arginine at position 482 could efflux mitoxantrone, but not rhodamine or doxorubicin. Whereas in cell lines with threonine or glycine at position 482, BCRP was capable of transporting rhodamine, doxorubicin and methotrexate as we1F9;37;38• Imai et a!.

investigated 1 1 tumour cell lines and found 2 polymorphisms and one splicing variant39• In the C42 1 A polymorphism lysine is substituted for glutamine at amino acid position 1 4 1 . Transfection studies in murine fibroblast PA3 1 7 cells showed lower protein expression by Western blotting and a higher topotecan accumulation. The 14 1st amino acid of BCRP is located in the ATP-binding region, but the increased drug sensitivity is a consequence of decreased protein expression and not altered protein structure. This decreased protein expression is a result of protein instability, since the mRNA levels were not lower in the transfected cell line.

84


Another polymorphism was the G34A, which causes the substitution of valine for methionine at amino acid position 1 2 . No effect of this polymorphism on protein level and drug accumulation was observed. In a healthy Japanese population, the polymorphism C376T was observed in genomic DNA, which substitutes a stop codon for glutamine. The implication of this allele is currently under investigation, but it is likely that the BCRP protein is not transcribed from this allele. In short, the polymorphism at amino acid position 482 alters substrate specificity, while the C42 1 A polymorphism is associated with lower BCRP protein expression and increased drug sensitivity. Screening for C42 1 A polymorphism in patients treated with BCRP substrates could be useful for the prevention of (toxic) side effects of these substrates.

BCRP in acute myeloid leukemia

Patients with acute myeloid leukemia (AML) are commonly treated with BCRP substrates such as mitoxantrone, daunorubicin and etoposide. The first report that appeared40, described a relatively high BCRP mRNA expression in 30% of 20 tested patient samples. An additional study on BCRP mRNA expression in 20 paired AML samples at diagnosis and at relapse of the disease, reported a significant upregulation of BCRP mRNA, but not P-gp or MRPl , in relapsed/refractory AML as compared to diagnosis41 • An upregulation of BCRP mRNA at relapse was also observed by another group, who studied 59 samples of children with newly diagnosed AML, of which 9 were also analyzed in relapse. Furthermore, the median BCRP mRNA expression was more than 10 times higher in patients who did not achieve remission after the first chemotherapy course versus patients who did achieve remission42• BCRP protein expression in AML was studied by several groups15·27;43;44• One study did not observe detectable levels of BCRP with the monoclonal antibody BXP-34 in AML samples15, whereas another study, using the same technique, found 27% out of 22 samples with > 1 0% cells staining positively. These BCRP positive samples also appeared to be less sensitive to daunorubicin than BCRP negative samples44• Our group reported a study in which BCRP protein expression was measured by flow cytometry. We found low but variable BCRP protein levels in 20 paired samples of de novo and relapsed/refractory AML. In contrast to the above described mRNA data41 ;42, no BCRP upregulation at the protein level was observed at relapse/refractory disease. BCRP protein expression was correlated with the immature, CD34 positive, immunophenotype in the AML samples. Moreover, relatively high BCRP protein expression correlated with a high functional capacity to export mitoxantrone27. However, only a small effect of the inhibitor FTC was observed on mitoxantrone accumulation. The low BCRP protein expression measured by flow cytometry was

85

Chapter 4

confirmed by the studies of Abbott et a/. 43and Van der Pol et a/. 45• The latter study only observed BCRP activity in 3 out of 26 AML cases, using prazosin as BCRP substrate, and did not find changes in BCRP function at follow-up of the disease in 1 3 out of 1 4 studied cases. No correlation with CD34 expression was observed in 9 studied AML samples. In summary, BCRP mRNA and protein data in these small groups of studied AML patients appear to be somewhat conflicting. BCRP protein expression and activity should be studied in larger patient groups to determine the precise role of BCRP in clinical resistance in AML.

BCRP in acute lymphoblastic leukemia

BCRP may also play a role in treatment failure in acute lymphoblastic leukemia (ALL), since ALL patients are treated with BCRP substrates such as methotrexate, daunorubicin, doxorubicin and mitoxantrone. Few studies have investigated the role of BCRP in ALL46;47• BCRP mRNA, as determined by real-time polymerase chain reaction, was found to be expressed at a low level, when compared to the MCF7 cell line. In T-lineage ALL, a lower level could be observed than in B-lineage ALL. No relation was found between BCRP expression and clinical parameters, such as white blood cell count at diagnosis and bone marrow response. Blasts of infants with ALL showed a lower level of BCRP mRNA expression in comparison to these of older children47• We determined the functional activity of BCRP in ALL, by measuring the mitoxantrone accumulation in combination with the BCRP inhibitor FTC, and observed a higher functional activity in B-lineage than in T-lineage ALL. Since BCRP has been associated with an immature phenotype in normal haematopoietic cells, we determined the immunophenotype and found no correlation between BCRP and an immature immunophenotype48• In conclusion, BCRP is expressed and functionally active in B-lineage ALL, and at lower levels in T-lineage ALL. This finding indicates that BCRP plays a small role in drug resistance in B-lineage ALL. Larger studies are necessary to investigate the prognostic impact ofBCRP functional activity and to elucidate whether patients with ALL could benefit from treatment adaptations based on this knowledge.

Concluding remarks

Multidrug resistance (MDR) is a cause of therapeutic failure in acute leukemia. The role of ABC transporters in MDR has been studied mainly for P-gp and MRP1 and led to the development of P-gp blockers for clinical use. Recently, the identification of BCRP initiated growing interest in BCRP inhibitors. To predict whether BCRP

86


inhibitors can effectively reverse chemotherapy resistance, the functional activity of BCRP in acute leukemia should give a good indication. B-lineage ALL patient samples, compared to T-lineage ALL and AML patient samples, showed higher BCRP activity, since a larger effect of FTC on mitoxantrone accumulation could

be observed27;48• This indicates that namely B-lineage ALL patients would

benefit from BCRP inhibitors. However, no comparative studies have been

performed investigating BCRP in both AML and ALL. In addition, even

when BCRP is present in only a small subpopulation of leukemic blasts,

it can be of major importance in treatment failure because it can form a

small clone of drug resistant cells. Therefore, it may be useful to develop

BCRP inhibitors. In trials using P-gp inhibitors, one of the problems was the

short duration of action and the lack of specificity. In leukemic patients in

whom several pumps are playing a role, using 1 blocker capable of inhibiting

multiple pumps might be very useful. However, if only 1 transporter is active,

inhibiting other transporters may only portend potential side effects. The

mainly disappointing results of the clinical trials with P-gp inhibitors may

also be explained by the presence of the other drug transporters. The multiple

transporters expressed on leukaemic cells act in concert, and inhibition

of 1 transporter may lead to compensation by other transporter proteins.

Moreover, in the trials the impact of P-gp inhibitors on pharmacokinetics of

chemotherapeutic drugs was observed. P-gp inhibitors caused an increase in

toxicity, making it necessary to reduce the chemotherapy49;so. Recently, the

BCRP inhibitor Ko143 was developed, which proved to be highly potent

and applicable in vivo36• In contrast to FTC, Ko143 is not neurotoxic. In the

clinical use of BCRP inhibitors, the possible effect on normal hematopoietic

stem cells should be considered, because they show a relatively high BCRP

expression. The combination with chemotherapy may render the leukemic

stem cells sensitive to chemotherapy, but simultaneously affect the normal

hematopoietic stem cells leading to an increased risk on bone marrow

toxicity.

In conclusion, the role of BCRP in acute leukemia needs to be studied in a

larger group of patients to investigate its effect on clinical outcome, in order

to investigate whether patients could benefit from the combination of BCRP

inhibitors and chemotherapy.

87

Chapter 4

REFERENCES

1 . Gottesman, M.M., Fojo, T. and Bates, S.E. Multidrug resistance in cancer: role of ATPdependent transporters. Nat Rev Cancer 2002: 2: 48-58.

2. Chauncey, T.R. Drug resistance mechanisms in acute leukemia. Curr Opin Onco/ 200 1 : 13: 2 1 -26.

3 . Marie, J.P. and Legrand, 0. MDRl!P-GP expression as a prognostic factor in acute leukemias. Adv Exp Med Bio/ 1 999: 457: 1 -9 .

4. Chen, Y.N., Mickley, L.A., Schwartz, A.M., Acton, E.M., Hwang, J.L. and Fojo, A.T. Characterization of adriamycin-resistant human breast cancer cells which display overexpression of a novel resistance-related membrane protein. J Bioi Chem 1 990: 265: 1 0073-1 0080.

5 . Doyle, L.A., Yang, W., Abruzzo, L.V., Krogmann, T., Gao, Y., Rishi, A.K. and Ross, D.D. A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc Nat/Acad Sci U S A 1998: 95: 1 5665-1 5670.

6. Lee, J.S., Scala, S., Matsumoto, Y., Dickstein, B., Robey, R., Zhan, Z., Altenberg, G. and Bates, S.E. Reduced drug accumulation and multidrug resistance in human breast cancer cells without associated P-glycoprotein or MRP overexpression. J Cell Biochem 1997: 65: 5 1 3-526.

7. Allikmets, R., Schriml, L.M., Hutchinson, A., Romano-Spica, V. and Dean, M. A human placenta-specific ATP-binding cassette gene (ABCP) on chromosome 4q22 that is involved in multidrug resistance. Cancer Res 1 998: 58: 5337-5339.

8. Miyake, K., Mickley, L., Litman, T., Zhan, Z., Robey, R., Cristensen, B., Brangi, M., Greenberger, L., Dean, M., Fojo,T. and Bates, S.E. Molecular cloning of cDNAs which are highly overexpressed in mitoxantrone-resistant cells: demonstration of homology to ABC transport genes. Cancer Res 1 999: 59: 8- 13 .

9 . Bailey-Dell, K.J., Hassel, B. , Doyle, L.A. and Ross, D.D. Promoter characterization and genomic organization of the human breast cancer resistance protein (ATP-binding cassette transporter G2) gene. Biochim Biophys Acta 200 1 : 1520: 234-24 1 .

1 0. Kage, K., Tsukahara, S. , Sugiyama, T., Asada, S. , Ishikawa, E., Tsuruo, T. and Sugimoto, Y. Dominant-negative inhibition of breast cancer resistance protein as drug efflux pump through the inhibition of S-S dependent homodimerization. Int J Cancer 2002: 97: 626-630.

I I . Litman, T., Brangi, M., Hudson, E., Fetsch, P., Abati, A., Ross, D.D., Miyake, K., Resau, J.H. and Bates, S.E. The multidrug-resistant phenotype associated with overexpression of the new ABC half-transporter, MXR (ABCG2). J Cell Sci 2000: 113: 201 1 -202 1 .

12 . Rabindran, S.K., He, H., Singh, M., Brown, E., Collins, K.I., Annable, T. and Greenberger, L.M. Reversal of a novel multidrug resistance mechanism in human colon carcinoma cells by fumitremorgin C. Cancer Res 1998: 58: 5850-5858.

13. Maliepaard, M., Scheffer, G.L., Faneyte, I.F., van Gastelen, M.A., Pijnenborg, A.C., Schinkel, A.H., van de Vijver, M.J., Scheper,R.J., and Schellens, J.H. Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer Res 200 1 : 61 : 3458-3464.

14. Rocchi, E., Khodjakov, A. Yolk, E.L., Yang, C.H., Litman, T., Bates, S.E. and Schneider, E. The product of the ABC half-transporter gene ABCG2 (BCRPIMXR/ABCP) is expressed in the plasma membrane. Biochem Biophys Res Commun 2000: 271 : 42-46.

88


1 5. Scheffer, G.L., Maliepaard, M., Pijnenborg, A.C., van Gastelen, M.A., de Jong, M.C., Schroeijers, A.B., van der Kolk, D.M., Allen, J.D., Ross, D.D., van der Valk, P., Dalton, W.S., Schellens, J.H. and Scheper, R.J. Breast cancer resistance protein is localized at the plasma membrane in mitoxantrone- and topotecan-resistant cell lines. Cancer Res 2000: 60: 2589-2593.

1 6. Cooray, H.C., Blackmore, C.G., Maskell, L. and Barrand, M.A. Localisation of breast cancer resistance protein in microvessel endothelium of human brain. Neuroreport 2002: 13: 2059-2063.

1 7. Bart, J., Groen, H.J., van der Graaf, W.T., Hollema, H., Hendrikse, N.H., Vaalburg, W., Sleijfer, D.T. and de Vries, E.G. An oncological view on the blood-testis barrier. Lancet Onco/ 2002: 3: 357-363 .

1 8. Goodell, M.A., Brose, K., Paradis, G., Conner, A.S. and Mulligan, R.C. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 1 996: 183: 1 797-1 806.

19 . Spangrude, G.J. and Johnson, G.R. Resting and activated subsets of mouse multipotent hematopoietic stem cells. Proc Nat/ A cad Sci U S A 1 990: 87: 7433-7437.

20. Zhou, S., Morris, J.J., Barnes, Y., Lan, L., Schuetz, J.D. and Sorrentino, B.P. Bcrpl gene expression is required for normal numbers of side population stem cells in mice, and confers relative protection to mitoxantrone in hematopoietic cells in vivo. Proc Nat/ Acad Sci U S A 2002: 99: 1 2339-12344.

2 1 . Scharenberg, C.W., Harkey, M.A. and Torok-Storb, B. The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 2002: 99: 507-5 12 .

22. Zhou, S., Schuetz, J.D., Bunting, K.D., Colapietro, A.M., Sampath, J . , Morris, J.J., Lagutina, 1., Grosveld, G.C., Osawa, M., Nakauchi, H. and Sorrentino, B.P. The ABC transporter Bcrpl !ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 200 1 : 7: I 028-1034.

23 . Good, J.R. and Kuspa, A. Evidence that a cell-type-specific efflux pump regulates cell differentiation in Dictyostelium. Dev Bio/ 2000: 220: 53-6 1 .

24. Jonker, J.W., Buitelaar, M., Wagenaar, E., van der Valk, M.A., Scheffer, G.L., Scheper, R.J., Plosch, T., Kuipers, F., Elferink, R.P., Rosing, H., Beijnen, J.H. and Schinkel, A.H. The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. Proc Nat/ Acad Sci U S A 2002: 99: 1 5649- 15654.

25. Diestra, J.E., Scheffer, G.L., Catala, 1., Maliepaard, M., Schellens, J.H., Scheper, R.J., Germa-Lluch, J.R. and Izquierdo, M.A. Frequent expression of the multi-drug resistance-associated protein BCRP/MXR/ABCP/ABCG2 in human tumours detected by the BXP-2 1 monoclonal antibody in paraffin-embedded material. J Patho/ 2002: 198: 2 1 3-2 1 9.

26. Minderman, H., Suvannasankha, A., O'Loughlin, K.L., Scheffer, G.L., Scheper, R.J., Robey, R.W. and Baer, M.R. Flow cytometric analysis of breast cancer resistance protein expression and function. Cytometry 2002: 48: 59-65.

27. van der Kolk, D.M., Vellenga, E., Scheffer, G.L., MUller, M., Bates, S.E., Scheper, R.J. and de Vries, E.G. Expression and activity of breast cancer resistance protein (BCRP) in de novo and relapsed acute myeloid leukemia. Blood 2002: 99: 3763-3770.

28. Litman, T., Jensen, U ., Hansen, A., Covitz, K.M., Zhan, Z., Fetsch, P., Abati, A., Hansen, P.R., Hom, T., Skovsgaard, T. and Bates, S.E. Use of peptide antibodies to probe for the mitoxantrone resistance- associated protein MXRIBCRP/ABCP/ABCG2. Biochim

89

Chapter 4

Biophys Acta 2002: 1565: 6- 1 6. 29. Honjo, Y., Hrycyna, C.A., Yan, Q.W., Medina-Perez, W.Y., Robey, R.W., van de Laar,

A., Litman, T., Dean, M. and Bates, S.E. Acquired mutations in the MXRIBCRP/ABCP gene alter substrate specificity in MXR/BCRP/ABCP-overexpressing cells. Cancer Res 200 1 : 61: 6635-6639.

30. Robey, R.W., Honjo, Y., van de Laar, A. , Miyake, K., Regis, J.T., Litman, T. and Bates, S.E. A functional assay for detection of the mitoxantrone resistance protein, MXR (ABCG2). Biochim Biophys Acta 200 I : 1512: 1 7 1 - 1 82.

3 1 . de Bruin, M., Miyake, K., Litman, T., Robey, R. and Bates, S.E. Reversal of resistance by GF 1209 1 8 in cell lines expressing the ABC half-transporter, MXR. Cancer Lett 1 999: 146: 1 1 7- 1 26.

32. Maliepaard, M., van Gastelen, M.A., Tohgo, A., Hausheer, F.H., van Waardenburg, R.C., de Jong, L.A., Pluim, D., Beijnen, J.H. and Schellens, J.H. Circumvention of breast cancer resistance protein (BCRP)-mediated resistance to camptothecins in vitro using non-substrate drugs or the BCRP inhibitor GFI 209 1 8. C/in Cancer Res 200 1 : 7: 935-941 .

33. Rabindran, S.K., Ross, D.D., Doyle, L.A., Yang, W. and Greenberger, L.M. Fumitremorgin C reverses multidrug resistance in cells transfected with the breast cancer resistance protein. Cancer Res 2000: 60: 47-50.

34. Hyafil, F., Vergely, C., Du, V.P. and Grand-Perret, T. ( 1993) "In vitro and in vivo reversal of multi drug resistance by GF 1 209 1 8, an acridonecarboxamide derivative" Cancer Research 53, 4595-4602.

35. van Loevezijn, A., Allen, J.D., Schinkel, A.H. and Koomen, G.J. Inhibition of BCRPmediated drug effiux by fumitremorgin-type indolyl diketopiperazines. Bioorg Med Chern Lett 200 1 : 11 : 29-32.

36. Allen, J.D., van Loevezijn, A., Lakhai, J.M., van der Valk, M., van Tellingen, 0., Reid, G., Schellens, J.H., Koomen, G.J. and Schinkel, A.H. Potent and specific inhibition of the breast cancer resistance protein multi drug transporter in vitro and in mouse intestine by a novel analogue of fumitremorgin C. Mol Cancer Ther 2002: 1 : 4 1 7-425.

37. Allen, J.D., Jackson, S.C. and Schinkel, A.H. A mutation hot spot in the Bcrp l (Abcg2) multidrug transporter in mouse cell lines selected for doxorubicin resistance. Cancer Res 2002: 62: 2294-2299.

38. Yolk, E.L., Farley, K.M., Wu, Y., Li, F., Robey, R.W. and Schneider, E. Overexpression of wild-type breast cancer resistance protein mediates methotrexate resistance. Cancer Res 2002: 62: 5035-5040.

39. Imai, Y., Nakane, M., Kage, K., Tsukahara, S., Ishikawa, E., Tsuruo, T., Miki, Y. and Sugimoto, Y. C42 1A polymorphism in the human breast cancer resistance protein gene is associated with low expression of Q141K protein and low-level drug resistance. Mol Cancer Ther 2002: 1 : 6 1 1 -6 1 6.

40. Ross, D.D., Karp, J.E., Chen, T.T. and Doyle, L.A. Expression of breast cancer resistance protein in blast cells from patients with acute leukemia. Blood 2000: 96: 365-368.

4 1 . Heuvel-Eibrink, M.M., Wiemer, E.A., Prins, A., Meijerink, J.P., Vossebeld, P.J., van der Holt, B., Pieters, R. and Sonneveld, P. Increased expression of the breast cancer resistance protein (BCRP) in relapsed or refractory acute myeloid leukemia (AML). Leukemia 2002: 16: 833-839.

42. Steinbach, D., Sell, W., Voigt, A., Hermann, J., Zintl, F. and Sauerbrey, A. BCRP gene

90


expression is associated with a poor response to remission induction therapy in childhood acute myeloid leukemia. Leukemia 2002: 16: 1443-1447.

43. Abbott, B.L., Colapietro, A.M., Barnes, Y., Marini, F., Andreeff, M. and Sorrentino, B .P. Low levels of ABCG2 expression in adult AML blast samples. Blood 2002: 100: 4594-4601 .

44. Sargent, J.M., Williamson, C.J., Maliepaard, M., Elgie, A.W., Scheper, R.J. and Taylor, C.G. Breast cancer resistance protein expression and resistance to daunorubicin in blast cells from patients with acute myeloid leukaemia. Br J Haemato/ 200 1 : 115: 257-262.

45. van der Pol, M.A., Broxterman, H.J., Pater, J.M., Feller, N., van der Maas M., Weijers, G.W., Scheffer, G.L., Allen, J.D., Scheper, R.J., van Loevezijn, A., Ossenkoppele, G.J. and Schuurhuis, G.J. Function of the ABC transporters, P-glycoprotein, multidrug resistance protein and breast cancer resistance protein, in minimal residual disease in acute myeloid leukemia Haematologica 2003 : 88: 1 34-147.

46. Sauerbrey, A., Sell, W., Steinbach, D., Voigt, A. and Zintl, F. Expression of the BCRP gene (ABCG21MXR/ ABCP) in childhood acute lymphoblastic leukaemia. Br J Haemato/ 2002: 118: 147- 1 50.

47. Starn, R.W., van den Heuvel-Eibrink, M.M., den Boer, M.L., Ebus, M.E.G., JankaSchaub, G.E., Allen, J.D. and Pieters, R. Multidrug resistance genes in infant acute lymphoblastic leukemia; Ara-C is not a substrate for BCRP. Leukemia 2003 : 17: 678 Abstract P33.

48. Plasschaert, S.L.A., de Bont, E.S.J.M., Kamps, W.A., van der Kolk, D.M., Morisaki, K., Bates, S.E., Scheffer, G.L., Scheper, R.J., Vellenga, E. and de Vries, E.G.E. Functional activity of breast cancer resistance protein (BCRP) in acute lymphoblastic leukemia. Leukemia 2003 : 17: 662 Abstract 0 1 6.

49. Baer, M.R., George, S.L., Dodge, R.K., O'Loughlin, K.L., Minderman, H., Caligiuri, M.A., Anastasi, J., Powell, B.L., Kolitz, J.E., Schiffer, C.A., Bloomfield, C.D. and Larson, R.A. Phase 3 study of the multidrug resistance modulator PSC-833 in previously untreated patients 60 years of age and older with acute myeloid leukemia: Cancer and Leukemia Group B Study 9720. Blood 2002: 100: 1 224- 1232.

50. Goldman,B. Multidrug resistance: can new drugs help chemotherapy score against cancer? J Nat! Cancer !nsf 2003 : 95: 255-257.

5 1 . Allen, J.D., Brinkhuis, R.F., Wijnholds, J. and Schinkel, A.H. The mouse Bcrp l /Mxr/ Abcp gene: amplification and overexpression in cell lines selected for resistance to topotecan, mitoxantrone, or doxorubicin. Cancer Res 1999: 59: 4237-424 1 .

52. Brangi, M., Litman, T. , Ciotti, M . , Nishiyama, K., Kohlhagen, G., Takimoto, C., Robey, R., Pommier, Y., Fojo, T. and Bates, S.E. Camptothecin resistance: role of the ATPbinding cassette (ABC), mitoxantrone-resistance half-transporter (MXR), and potential for glucuronidation in MXR-expressing cells. Cancer Res 1 999: 59: 5938-5946.

53. Robey, R.W., Medina-Perez, W.Y., Nishiyama, K., Lahusen, T., Miyake, K., Litman, T., Senderowicz, A.M., Ross, D.O. and Bates, S.E. Overexpression of the ATP-binding cassette half-transporter, ABCG2 (Mxr/BCrp/ABCP 1 ), in flavopiridol-resistant human breast cancer cells. Clin Cancer Res 200 1 : 7: 145-1 52.

54. Komatani, H., Kotani, H., Hara, Y., Nakagawa, R., Matsumoto, M., Arakawa, H. and Nishimura, S. Identification of breast cancer resistant protein/mitoxantrone resistance/ placenta-specific, ATP-binding cassette transporter as a transporter of NB-506 and J-1 07088, topoisomerase I inhibitors with an indolocarbazole structure. Cancer Res 200 1 : 61: 2827-2832.

91

Chapter 5

The role of Breast Cancer Resistance Protein

in acute lymphoblastic leukemia

Sabine L.A. Plasschaert1, Dorina M. van der Kolk2, Eveline S.J.M.de Bont1, Willem A. Kamps1, Kuniaki Morisaki5,

Susan E. Bates5, George L. Scheffer", Rik J. Scheper", Edo Vellenga2, Elisabeth G.E. de Vries3

Departments of Pediatric Oncology and Hematology1, Hematology and Medical Oncology3, University Hospital Groningen, Groningen;

Division of Pathology, Free University Medical Center, Amsterdam; The Netherlands; National Cancer Institute, Medicine Branch,

National Institutes of Health5, Bethesda, USA

Clinical Cancer Research 2003; 9: 5171-7

Chapter 5

ABSTRACT

Purpose: Overexpression of the transporter ABCG2, also known as breast cancer resistance protein and mitoxantrone resistance protein, can confer resistance to a variety of cytostatic drugs, such as mitoxantrone, topotecan, doxorubicin, and daunorubicin. This study analyzes the ABCG2 expression and activity in 46 human de novo acute lymphoblastic leukemia B- and T-lineage (ALL) samples. Experimental design: ABCG2 expression was measured flow cytometrically with the BXP-34 monoclonal antibody. ABCG2 functional activity was determined flow cytometrically by measuring mitoxantrone accumulation in combination with the ABCG2 inhibitor fumitremorgin C (FTC). To determine a possible effect of the transporters P-glycoprotein and multidrug resistance-associated protein (MRP1 and MRP2) on mitoxantrone accumulation, the accumulation was investigated in the presence of the P-glycoprotein inhibitor PSC 833 and MRP inhibitor MK-571 . The ABCG2 gene was sequenced to investigate the amino acid at position 482. Results: In B-lineage ALL (n == 23), the median BXP-34:IgG 1 ratio was higher, namely 2.4 (range, 1 .7-3 .7), than in T-lineage ALL (n == 23; 1 .9, range 1 .2-6.6; P ==

0.003). The addition of FTC to mitoxantrone treatment caused a median increase in mitoxantrone accumulation of 2 1 % (range, 0- 140%) in B-lineage ALL. In T-lineage ALL, this FTC effect was less pronounced (5%; range, 0-256%; P = 0.0 1 3). The influence of FTC on mitoxantrone accumulation correlated with ABCG2 protein expression (r = 0.52; p < 0.00 1 ; n = 43). The increase in mitoxantrone accumulation, when FTC was added to cells treated with both PSC 833 and MK-57 1 , correlated with the ABCG2 expression in B-lineage ALL, but not in T-lineage ALL. Sequencing the ABCG2 gene revealed no ABCG2 mutation at position 482 in patients who accumulated more rhodamine following FTC. Conclusions: This study shows that ABCG2 is expressed higher and functionally more active in B-lineage than in T-lineage ALL.

94

BCRP in ALL

INTRODUCTION

The prognosis of acute lymphoblastic leukemia (ALL) has improved over the last decades, in children even more than in adults. Still, many ALL patients do not achieve a complete remission or develop a relapse1 ;2• Resistance to chemotherapeutic agents plays an important role. Multidrug resistance (MDR) is defined as resistance to multiple, structurally and functionally unrelated natural product drugs. One mechanism of MDR is the overexpression of members of the ABC superfamily of membrane transporters. This family of transporters consists of a number of proteins, including P-glycoprotein (P-gp) and the multi drug resistance-associated proteins (MRPI and MRP2). Contradictory results have been observed with respect to the prognostic impact of P-gp expression and functional activityJ.S. For MRP I protein and mRNA expression, neither a difference has been found between initial and relapse samples, nor a correlation with complete response or survivaP. We have studied previoously the functional activity of MRP I and MRP2 in children and adults with ALL, and observed no correlation between functional activity and clinical outcome9• Another ATP-binding cassette transporter has been identified in a MDR human breast cancer cell line MCF-7/AdrVp10; 1 1 • This cell line shows an ATP-dependent reduction in the intracellular accumulation of anthracycline anticancer drugs, in the absence of overexpression of known MDR transporters such as P-gp or MRP. This transporter protein was named ABCG2, and is also known as breast cancer resistance protein, placental ABC transporter or mitoxantrone resistance protein. Several human cancer cell lines selected for resistance to mitoxantrone or topotecan have shown a marked ABCG2 overexpression. In addition, enforced expression of the full-length ABCG2 eDNA in MCF-7 breast cancer cells confers resistance to mitoxantrone, bisantrene, topotecan, and doxorubicin or daunorubicin in the presence of verapamil1 1 • 14 • The physiological function of ABCG2 may be similar to P-gp, since it is also located in the blood-brain barrier, the placenta, the liver, and intestinal tract15 • The spectrum of ABCG2 substrates shows an overlap with that of P-gp and MRP. All three of the transporters can extrude the anthracyclines, such as daunorubicin, doxorubicin and epirubicin. P-gp16, ABCG2 and probably MRP 1 17; 18

can also transport mitoxantrone. There are indications that the spectrum of drugs transported by ABCG2 might differ in cell lines because of a mutation in the ABCG2

gene at amino acid position 48219• When a mutation is present in the ABCG2 gene with threonine or glycine situated at position 482 instead of arginine, ABCG2 is able to transport rhodamine and doxorubicin additionallyl9 but not methotrexate20• Mitoxantrone, methotrexate, daunorubicin and doxorubicin are used frequently

95

Chapter 5

during the induction or consolidation phase in ALL, indicating that ABCG2 might have clinical impact in this patient group. The present study demonstrates that the ABCG2 protein is expressed in ALL. The functional activity was measured by mitoxantrone accumulation in the absence or presence of the ABCG2 inhibitor fumitremorgin C (FTC)21 , and FTC in combination with the inhibitors of the other transporters P-gp and MRP1 , which might obscure the ABCG2 function12;22•

The ABCG2 gene was sequenced in patients who accumulated more rhodamine following FTC, to investigate the amino acid at position 482. Since ABCG2 expression is correlated with an immature phenotype in normal human hematopoietic cells15;23;24 andAML blasts2 1 , we also determined the immunophenotype of the patient samples.


Cell lines

The drug-sensitive human breast cancer cell line MCF-7 and its mitoxantrone resistant counterpart MCF-7 MR, which overexpresses ABCG2 but not P-gp or MRP- 1 , were cultured in RPMI 1 640 (Bio Whitaker, Brussels, Belgium) supplemented with 1 0% fetal calf serum (FCS) (Hyclone, Logan, UT). The MCF-7 MR cell line was cultured in the presence of 80 nM mitoxantrone. The human small cell lung cancer cell line GLC4, the in vitro doxorubicin-selected MRP 1 -overexpressing sub line GLC4/ ADR and the MDR-1 transfected cell line GLC4/P-gp were also cultured in RPMI 1 640 with 1 0% FCS. The doxorubicin-selected cell line was cultured in the presence of doxorubicin (2 ).1M; Pharmacia and Upjohn, Woerden, the Netherlands) and the MDR-1 transfected cell 1ine in the presence of 50 nM vincristine (Teva Pharma BY, Mijdrecht, the Netherlands). Cells were cultured in drug-free medium at least 7 days before the analyses. The cell lines served as controls for the ABCG2 expression and functional assay.

Patient samples

From all of the patients newly diagnosed with ALL at the University Hospital Groningen during the period of 1 989-200 1 , bone marrow and peripheral blood samples were collected for diagnostic evaluation and cryopreserved. Patients were included in the study if a sufficient number of preserved bone marrow or peripheral white blood cells (> 1 Ox 1 06) was available. The Dutch Childhood Leukemia Study Group performed the immunophenotyping and cytomorphology of pediatric samples. Adult patient samples were immunophenotyped by the University Hospital Groningen. The children were treated according to the Dutch Childhood

96

BCRP in ALL

Leukemia Study Group protocols ALL-8 or -925, containing the MDR drugs dexamethasone, daunorubicin, doxorubicin, methotrexate and vincristine. The adult patients received a regimen including the MDR drugs prednisone, doxorubicin, vincristine, methotrexate and etoposide26. Mononuclear cells from bone marrow or peripheral blood were isolated on a Ficolllsopaque (Nycomed, Oslo, Norway) density gradient by centrifugation for 20 min at 800 x g. The cells were cryopreserved in RPMI 1 640 supplemented with 1 0 % FCS and 1 0 % dimethyl sulfoxide (Merck, Amsterdam, the Netherlands) and stored at - 1 96 ' C. Upon analysis cells were thawed, centrifuged in newborn calf serum (Life Technologies, Breda, the Netherlands), treated with DNase (Boehringer Mannheim, Mannheim, Germany) and washed with RPMI 1 640. Immunophenotyping showed that all samples contained more than 90 % ofleukemic cells. For immunophenotyping, 0.5 x 1 06 cells were incubated with 5 J.LL of

FITC- or phycoerythrin-labeled mouse monoclonal antibodies to CD117,

CD3, CD7 (all IQ Products, Groningen, the Netherlands), CD2, CD10,

CD19, CD20, CD33, CD34 or IgG isotype-matched controls (all Becton

Dickinson, Woerden, the Netherlands) for 30 min at 4°C. Subsequently, cells

were washed with RPMI 1640, and analyzed with a FACStar flow cytometer (Becton Dickinson Medical Systems, Sharon, MA). The immature markers

CD34 and CD 117, expressed only on progenitor cells, were analyzed in both

B-lineage and T-lineage ALL. For B-lineage ALL samples, the B-lineage

markers CD10, CD19 and CD20 were used. The T-lineage markers CD3

and CD7 were investigated in T-lineage ALL samples. Because the myeloid

marker CD33 correlated with P-gp functional activity in an earlier study9, CD33 expression was also measured.

Flow cytometric detection of ABCG2 protein expression

ABCG2 protein expression was measured with the monoclonal antibody BXP-3427, which recognizes an internal epitope of the ABCG2 protein. Cells (0.5 x 1 06) were incubated with FACS lysing solution (Becton Dickinson) to permeabilize the cell membranes for 1 0 min at room temperature. Then the cells were incubated for 1 5 min at room temperature, with 1 % goat serum i n phosphate buffered saline (PBS: 1 40 mM NaCI, 9.0 mM Na2HP042Hp, 1 .3 mM NaHl042Hp, pH 7 .4) containing 2% bovine serum albumine (BSA). Subsequently, the cells were incubated for 60 min at room temperature with 4 JlL ( 1 .0 Jlg) of BXP-34 mouse antibody or with 20 JlL ( 1 .0 Jlg) IgG 1 isotype control. Cells were washed with PBS and incubated for 20 min with PE-conjugated rabbit-anti-mouse F(ab)2 fragments. PE fluorescence was measured on a FACScalibur flow cytometer and expressed as median fluorescence

97

Chapter 5

intensity (MFI). ABCG2 protein expression was expressed as adjusted for IgG 1 control, i.e., the ratio of BXP-34: IgG 1 antibody MFI. Control cell lines used to standardize the ABCG2 protein expression assay, included the MCF-7 cell line, expressing low levels of ABCG2, and the mitoxantrone resistant subline MCF-7 MR, expressing high levels of ABCG2.

Flow cytometric detection of mitoxantrone accumulation

The capacity of ALL blasts to extrude mitoxantrone in the absence or presence of the ABCG2 inhibitor FTC12, the P-gp inhibitor PSC 833 (provided by Novartis Pharma Inc., Basel, Switzerland), the MRP inhibitor MK-571 (Merck Sharp, Kirkland, PG, Canada)28, or combinations of these inhibitors, was measured in a flow cytometric assay21 • The cell lines MCF-7 and MCF-7 MR served as controls. To exclude a possible inhibitory effect of FTC on P-gp or MRP, mitoxantrone accumulation in combination with FTC was studied in the P-gp and MRP overexpressing cell lines GLC4/ ADR and GLC4/P-gp. Cells (0.5 x I 06) were preincubated with the inhibitors for 20 min at 37°C, 5% C0

2, in the following combinations: RPMI 1 640 alone (0.5

mL), RPMI 1 640 plus FTC (10 J.lM), the maximum non-toxic doses of PSC 833 (2 J.lg/mL) and MK-571 ( 1 0 J.!M), FTC plus PSC 833, FTC plus MK-57 1 , PSC 833 plus MK-57 1 , or FTC plus PSC 833 plus MK-57 1 . Thereafter, mitoxantrone ( 10 J.lM) was added and the cells were incubated for 60 min at 37°C, 5% C0

2• Cells were washed with ice-cold RPMI 1 640. Mitoxantrone fluorescence was analyzed with a FACScalibur flow cytometer equipped with an argon laser. The blast population was gated by forward and side scatter characteristics. The mitoxantrone fluorescence of 5000 gated events was logarithmically measured at a laser excitation wavelength of 635 nm through a 670-nm band-pass filter. Before assessing the mitoxantrone accumulation, the flow cytometer was calibrated for the specific wavelength using sphero particles (Spherotech Inc, Libertyville, IL). Mitoxantrone accumulation was expressed as MFI. The effects of the inhibitors were calculated as the ratio of the shift ofMFI of the mitoxantrone accumulation divided by the mitoxantrone accumulation without the inhibitors, multiplied by 100%. When an inhibitor caused a decrease in mitoxantrone accumulation, the effect of the inhibitor was equated to zero. The mitoxantrone accumulation was determined in normal B- and T-lymphocytes from healthy donors in combination with CD2-FITC and CD 19-PE monoclonal antibodies.

Flow cytometric detection of P-gp and MRP functional activity

Functional activity of the P-gp and MRP transporter proteins was demonstrated as described previously29• Shortly, P-gp activity was analyzed with rhodamine 123

98

BCRP in ALL

(Rh 1 23) (Sigma Chemical, Bornem, Belgium) in combination with PSC 833. Since a mutation in the ABCG2 gene can render the protein capable of transporting Rh 1 23, the effect of FTC on Rh 1 23 accumulation was also studied19• To determine MRP (MRPl and homologue MRP2) activity the compound carboxyfluorescein (CF) (Sigma Chemical Co.) was used, in combination with MK-57 1 . Rh 123 and CF fluorescence of 5000 events was measured with a FACScalibur flow cytometer. The effects of the inhibitors were expressed as a ratio of the shift of MFI divided by the MFI in unblocked cells, multiplied by 1 00%.

RNA isolation, eDNA and genomic DNA sequencing.

The ABCG2 gene was sequenced to investigate the amino acid at position 482 (arginine, threonine, or glycine)19• Total cellular RNA was extracted from leukemic cells using 1 mL ofTrizol reagent (Life technologies, Breda, the Netherlands). RNA was extracted, precipitated and washed according to the manufacturer's protocol. Subsequently, RNA was reverse transcribed in 20 J.!L reverse transcriptase buffer containing 10 mM DTT, 0.5 mM each of dATP, dGTP, dCTP and dTTP, 200 units of Moloney murine leukemia virus reverse transcriptase (M-MLV RT), 5 units of RNAse inhibitor and 10 ng/J.lL pd(N)6 random primers. Sequencing was performed using the ABI PRISM Rhodamine Terminator Cycle Sequencing Ready Reaction kit with an ABI PRISM 377 DNA sequencer (Applied Biosystems, Foster City, CA).


The Mann-Whitney U and Wilcoxon non-parametric tests were performed to test for differences in variables between children and adults, B-lineage ALL and T-lineage ALL, and effects of combination of inhibitors on mitoxantrone accumulation. The Spearman rank correlation was used to determine correlation among ABCG2, P-gp and MRP activities and continuous variables. Ps < 0.05 were considered significant.

RESULTS


Samples from 46 patients with de novo ALL were studied, including 23 patients with B-lineage and 23 patients with T-lineage ALL. The patient characteristics are summarized in Table 1 .

ABCG2 protein expression

To standardize the ABCG2 protein expression assay, the ABCG2 expression was studied in the ABCG2 low-expressing cell line MCF-7 and the ABCG2

99

Chapter 5

Table 1 Characteristics of ALL patients

Number Gender Age* (male/female) (years)

B-lineage ALL Total 23 1 1 I 1 2 14 ( 1 -73) Children 1 2 6 / 6 9 ( 1 - 14) Adults 1 1 5 / 6 33 ( 19-73)

T-lineage ALL Total 23 1 2 I 1 1 22 (3-75) Children 1 0 7 / 3 7 (3- 1 5) Adults 1 3 5 / 8 28 ( 1 7-75)

*The results represent median values with range tWBC: white blood cell count at diagnosis

WBC*t Unfavorable (x l09/ L) cytogenetics

49.6 ( 1 . 1 -446.0) 4 (8 unknown) 49.0 ( 1 . 1 -446.0) 1 (5 unknown) 52.0 (8.3-332.2) 3 (3 unknown)

7 1 .5 (3.5-590.0) 0 (3 unknown) 1 56.0 (34. 1 -590.0) 0 (1 unknown) 1 7.6 (3 .5-206.0) 0 (2 unknown)

Unfavorable cytogenetics include the translocations t(9;22), t(4; 1 1 ) and l lq23 rearrangements.

overexpressing cell line MCF-7 MR. The monoclonal antibody BXP-34 was used in combination with the IgG 1 isotype control in a flow cytometric assay. A median BXP-34:IgG1 ratio of 8.3 (range, 5.4- 1 1 . 1 ) was observed in the MCF-7 cell line versus a high median expression of 86.4 (range, 59.8- 1 1 2.7) in the MCF-7 MR cell line. Subsequently, the ABCG2 protein expression was investigated in ALL blasts (table 2). In B-lineage ALL samples, the median BXP-34:IgG 1 ratio was 2.4 (range, 1 .7-3 .7). The T-lineage ALL samples showed a median expression of 1 .9 (range, 1 .2-6.6). The BXP-34:IgG 1 ratios in all ALL samples were lower than the observed ratios in the MCF-7 parent cell line. In B-lineage ALL, ABCG2 expression was higher compared to T-lineage ALL (P = 0.003). The ABCG2 expression in B-lineage ALL showed also a trend toward a higher expression in children than adults (P =

0.07 1) . In T-lineage ALL, no difference was observed between the two age groups (P = 0.574).

Mitoxantrone accumulation

The functional activity of ABCG2 was examined by measuring the intracellular mitoxantrone accumulation, expressed as MFI, in the absence or presence of the ABCG2 inhibitor FTC. In the MCF-7 MR cell line, the mean mitoxantrone accumulation was lower than in the MCF-7 cell line (465 versus 1 004), which was expected as the overexpression of ABCG2 causes a decrease in mitoxantrone accumulation. Next, the mitoxantrone accumulation was determined in ALL blasts. In the B-lineage ALL samples, the median mitoxantrone accumulation was 148 (range, 77-555 MFI).

100

BCRP in ALL

The median mitoxantrone accumulation in T-lineage ALL was 108 (range, 65-28 1 MFI) (table 2). To investigate whether the efflux of mitoxantrone was ABCG2-mediated, the ABCG2 inhibitor FTC was added. In the MCF-7 cell line, FTC caused an increase in mitoxantrone accumulation of 1 8% (range, 1 5-38%). In the ABCG2 overexpressing cell line MCF-7 MR an increase of 1 68% was observed (range, 1 4 1 -227%). FTC did not affect mitoxantrone accumulation in P-gp and MRP-overexpressing cell lines (data not shown). The addition of the ABCG2 inhibitor FTC caused a higher median increase in mitoxantrone accumulation in B-lineage ALL, namely 2 1 % (range, 0- 1 40% ), than in T-lineage ALL (5%; range, 0-256%; P = 0.0 1 3). From the ALL samples, 27 (59%) patient samples showed a smaller effect of FTC on mitoxantrone accumulation than the MCF-7 cell line. Eighteen (39%) patient samples showed a FTC effect between the MCF-7 and MCF-7 MR cell lines, and 1 (2%) patient sample showed a larger increase in mitoxantrone accumulation by FTC than the MCF-7 MR cell line. These results indicate that ABCG2 is functionally active on a low, but significant level in B-lineage ALL.

In normal B- and T-lymphocytes from healthy donors, the effect of FTC on mitoxantrone accumulation was also studied. No effect of FTC could be observed in both B and T lymphocytes. The FTC effect on mitoxantrone accumulation correlated positively with ABCG2 expression (r = 0.52; P < 0.00 1 ; n = 43 ; fig. 1) . Even when the outlier is removed, the

Table 2 ABCG2 protein expression and mitoxantrone accumulation in ALL patients

B-lineage ALL T-lineage ALL

ABCG2 protein expression 2.4 ( 1 .7-3.7) 1 .9 ( 1 .2-6.6) Mito accumulation (MFI*) 148 (77-555) 1 08 (65-28 1 )

FTC shift 2 1 (0- 140) 5 (0-256) PSC 833 shift 22 (6-70) 24 (0-8 1 ) MK-571 shift 1 6 (0- 1 73) 1 8 (0- 14 1 )

Rh 123 accumulation PSC 833 shift 0 (0-1 97) 14 (0-362)

CF accumulation MK-57 1 shift 1 92 (23-639) 239 (68-500)

ABCG2 protein expression calculated as the BXP-34:IgG 1 ratio. Shifts calculated as percentage of accumulation without inhibitors. The results represent median values with range. *MFI: mean fluorescence intensity

101

Chapter 5

300 c: r=0.52 .Q ]j p<0.001 0

::::l n=43 E a 200

(.) ro Q)

0 c: e 0 c 100 ro X 0

·"E 0 c: 0 (.) ti: tl & -100 w

2 3 4 5 II 7

ABCG2 protein expression (BXP-34ngG1 ) Figure 1. Correlation between ABCG2 protein expression and effect of FTC on mitoxantrone accumulation. ABCG2 expression was measured with the BXP-34 monoclonal antibody, the effect of FTC was measured after 60 min of incubation with mitoxantrone with and without FTC.

correlation remains present (r = 0.49; P = 0.00 1 ; n = 42). No difference was observed between children and adults in mitoxantrone accumulation or in the effect of FTC (B-lineage P = 0.268 and P = 0.566; T-lineage ALL P = 0.352 and P = 0.226).

P-gp functional activity

Mitoxantrone can also be transported by P-gp and MRP. To get an impression of the impact of ABCG2 on mitoxantrone transport, P-gp and MRP activity, and the effect of their blockers on mitoxantrone accumulation were studied as well. In B-lineage ALL, there was no median effect of PSC 833 on Rh l 23 accumulation (range, 0- 1 97%; table 2), whereas PSC 833 gave a median increase of 14% in T-lineage ALL (range, 0-362%; P = 0.00 1 ; table 2). In brief, P-gp activity was observed only in T-lineage ALL. To study the specific transport of mitoxantrone by P-gp, the effect of PSC 833 on mitoxantrone was studied. Surprisingly, PSC 833 increased the mitoxantrone accumulation in all of the B-lineage and most T -lineage ALL patients (table 2).

MRP functional activity

The functional activity of MRP was determined by assessing the effect of MK -571 on CF accumulation. In B-lineage ALL, MK-57 1 increased the CF accumulation

102

BCRP in ALL

with 1 92% (range, 23-639%) and in T-lineage ALL with 239% (range, 68-500%), indicating that MRP is functionally active in both B- and T-lineage ALL (table 2). To investigate the role of MRP in mitoxantrone transport in ALL, mitoxantrone accumulation was studied in the presence of MK-5 7 1 . MK-57 1 increased the mitoxantrone accumulation in 20 out of 2 1 B-lineage samples and 1 8 out of 22 T -lineage samples (table 2).

Mutation analysis of the ABCG2 gene and substrate spectrum

The ability of ABCG2 to transport Rh 1 23 was studied by measuring the effect of FTC on Rh 123 accumulation. In three of the 46 samples ( 1 B-lineage, 2 T-lineage ALL) the addition of FTC showed an increase in Rh 1 23 accumulation. These three samples, as well as five control samples, were sequenced for mutation at amino acid 482 in the ABCG2 gene. All 8 of the samples showed the wild-type variant.

Combination of inhibitors on mitoxantrone accumulation

Finally, the inhibitors were combined to analyze whether their effects are the result of the activity of one transporter or a combined action of transporters. First, the addition of FTC to PSC 833 on mitoxantrone accumulation was studied, to explain the effect of PSC 833 in samples without P-gp activity, as assessed by the Rh 1 23 efflux assay. Our results demonstrated an increased effect of the combination of FTC and PSC 833 on mitoxantrone accumulation in comparison with just PSC 833 or FTC, in both B-lineage and T-lineage ALL samples (tables 3 and 4). By adding FTC to MK-57 1 , the role of ABCG2 in mitoxantrone transport in patients

Table 3 Effect of the combination of inhibitors on mitoxantrone accumulation in ALL patients

Mito accumulation (MFI*) FTC shift PSC 833 shift MK-571 shift FTC+PSC 833 shift FTC+MK-57 1 shift PSC 833+MK-57 1 shift FTC+PSC 833+MK-57 1 shift

B-lineage ALL

148 (77-555) 2 1 (0- 140) 22 (6-70) 1 6 (0- 1 73) 27 (0- 1 56) 39 (0-226) 32 (8- 1 66) 44 (6-2 1 7)

T-lineage ALL

1 08 (65-28 1 ) 5 (0-256) 24 (0-8 1 ) 1 8 (0- 141 ) 43 (0-308) 22 (0-28 1 ) 5 0 (0-225) 46 (0-340)

Shifts calculated as percentage of mitoxantrone accumulation without inhibitors. The results represent median values with range. *MFI: mean fluorescence intensity

103

Chapter 5

Table 4 Wilcoxon rank test calculating the effect of the addition of one inhibitor to a combination of inhibitors on mitoxantrone accumulation in ALL patients

Combination of inhibitors Shift p

B-lineage ALL FTC+PSC833 vs. PSC833 27 vs. 22 0.002 FTC+MK57 1 vs. MK571 39 vs. 16 0.002 FTC+PSC833+MK57 1 vs. PSC833+MK571 44 vs. 32 0.35 1

T-lineage ALL FTC+PSC833 vs. PSC833 43 vs. 24 0.01 6 FTC+MK57 1 vs. MK571 22 vs. 1 8 0. 1 70 FTC+PSC833 +MK57 1 vs. PSC833+MK571 46 vs. 50 0.408

Shifts calculated as percentage of mitoxantrone accumulation without inhibitors. The results represent median values.

with MRP activity was studied. The addition of FTC to MK-57 1 caused an increase in mitoxantrone accumulation in only B-lineage ALL (table 3, table 4). The addition of FTC to the combination of PSC 833 and MK-571 gives a reflection of the ABCG2 activity not biased by P-gp and MRP functional activity. In the Band T-lineage ALL samples, this did not cause an overall increase in mitoxantrone accumulation (B-lineage P = 0.35 1 respectively. T-lineage ALL P = 0.408; table 3 and 4). However, in B-lineage ALL samples in which the addition of FTC caused an increase in mitoxantrone accumulation, this increase correlated with the ABCG2 protein expression (r = 0.57; P = 0.009; n = 20).

Immunophenotype

ABCG2 expression has been associated with an immature phenotype in normal hematopoietic cells15;23;24 and AML blasts2 1 • Therefore, the immunophenotype of the patient blasts was determined and correlated with the maturation stage. In B-lineage ALL, CD 1 0 ("common" ALL antigen) expression correlated with ABCG2 protein expression and ABCG2 functional activity (r = 0.5 1 , P = 0.02 1 ; n =

20 resp. r = 0.52, P = 0.0 1 8 ; n = 20) . The CD 1 9 expression was also associated with ABCG2 activity (r = 0.48, P = 0.039; n = 1 9). No correlation was observed between the markers CD20, CD33, the immature markers CD34 or CD 1 1 7 (c-kit), and the ABCG2 protein expression and ABCG2 functional activity (i. e. , the effect of FTC on mitoxantrone accumulation). In T-lineage ALL, the expression of the CD markers CD34, CD1 17, CD3, CD7 and CD33 did not correlate with ABCG2 expression and functional activity. Of all investigated markers, CD33 and CD1 1 7 were scarcely expressed on ALL blasts.

104

BCRP in ALL

In summary, a number of correlations were found between more mature markers

in B-lineage ALL and ABCG2. No relation was observed between the protein

expression and functional activity of ABCG2, and an immature phenotype in our

patient samples.

DISCUSSION

This is the first study, which addresses the ABCG2 expression and its functional

activity in ALL in children and adults. ABCG2 expression and functional activity

was low but higher in B-lineage ALL than in T-lineage ALL. The functional ABCG2

activity correlated well with the ABCG2 expression. Sauerbrey et a/. have measured

ABCG2 expression in ALL only at the mRNA leveP0, and also observed higher

mRNA expression in B-lineage ALL than in T-lineage ALL. In contrast, the present

study showed that in normal B- and T-lymphocytes no ABCG2 functional activity

could be determined, and consequently no difference between B- and T-lymphocytes

was observed.

In normal hematopoietic cells, it was shown that ABCG2 is expressed in pluripotent

stem cells and sharply down-regulated at the point of commitment to lineage

specific development15;23;24• Also in AML, an immature phenotype, as determined by

the expression of CD34, correlated with ABCG2 expression21 • In ALL, we did not

observe an association between the immature markers CD 1 1 7 or CD34 and ABCG2.

The expression of other transporters, such as P-gp and MRP, is also associated

with an immature phenotype in normal hematopoietic cells31;32 and AML blasts33·36,

but not in ALL blasts6;8;37• Apparently, the connection between transporters and

maturation stage is differently regulated in ALL than normal hematopoietic cells and

AML blasts.

A mutation in the ABCG2 gene can alter substrate specificity in drug-resistant

cell lines19• Expression of the mutated form, with threonine or glycine, can confer

resistance to a broader spectrum of drugs, including mitoxantrone, Rh123 and

doxorubicin. The three patient samples, in which ABCG2 was suggested to transport

Rh123, did not show the mutation. This finding underscores the limited role of this

mutation in the ABCG2 gene in clinical samples.

Although no gold standard exists for the detection of ABCG2 functional activity,

Minderman et a/. 38 showed that flow cytometry using mitoxantrone in combination

with ABCG2 inhibitor FTC can be used to detect functional activity. Evidently, other

inhibitors could have been used, such as the FTC-analogue Ko143, but we have no

reason to expect them to be more suitable than FTC. Moreover, FTC is used most

often in functional studies and the application of a standardized detection method is

105

Chapter 5

recommended39• Mitoxantrone is a substrate for several other transporters, including P-gp and MRP. Therefore, the functional activity of these transporters was investigated and their role

in mitoxantrone transport. The contribution to mitoxantrone transport of the

individual transporters is reflected in table 3 . ABCG2-mediated mitoxantrone

transport in B-lineage ALL is comparable with that of P-gp and MRP. In

T-lineage ALL, the mitoxantrone transport by ABCG2 is negligible compared

to P-gp- and MRP-mediated mitoxantrone transport.

Remarkable was the effect of PSC 833 on mitoxantrone accumulation, in

patients who did not show P-gp activity, as assessed with the Rh 1 23 efflux

assay. This could possibly be explained by PSC 833 interfering with another

transporter than P-gp, which is also capable of transporting mitoxantrone. In

general, the lack of specificity of P-gp and ABCG2 inhibitors could explain

the inconclusive data, because new transporters are still being identified that

may be inhibited by P-gp and/or ABCG2 inhibitors . Also the occurrence of

polymorphisms in the ABCG2 or P-gp encoding genes, altering the substrate

specificity, may be of influence.

In patients with functionally active P-gp, FTC showed an increase in

mitoxantrone accumulation, when P-gp was inhibited. This suggests that even

in patients in which mitoxantrone is transported by P-gp, ABCG2 contributes

to mitoxantrone resistance.

The ability of MRP to transport mitoxantrone is not very clear. We observed

an effect of MK-57 1 in mitoxantrone transport in most of the ALL cases.

The MRP overexpressing cell line GLC4/ADR also showed an increase in

mitoxantrone accumulation when MK-57 1 was added. However, there are

studies suggesting that mitoxantrone is not a substrate of MRP18;40• The

conflicting reports could also be the result of a functional polymorphism that

has not been discovered yet.

When P-gp and MRP were inhibited, FTC did not cause an overall increase

in mitoxantrone accumulation in B- and T-lineage ALL. In B-lineage ALL,

an effect of FTC could be observed in half of the patients, and the increase

in mitoxantrone accumulation correlated with the ABCG2 protein expression.

In patient samples in which PSC 833 and MK-57 1 showed a significant

effect on mitoxantrone accumulation, the addition of FTC did not cause an

extra increase. Consequently, in these samples the effect of FTC is most

likely masked by the effect of PSC 833 and MK-57 1 . The effect of inhibitors

was expressed as a percentage and consequently, if two inhibitors have a large effect, the effect of the addition of a third inhibitor is likely to be

106

BCRP in ALL

underestimated.

The distribution of the functional activity of transporters within patients was

determined by calculating the correlation between the effect of FTC and P-gp

and/or MRP functional activity. No correlation was found between the three

transporters in B- and T-lineage ALL. Surprisingly, the effect of FTC on

mitoxantrone accumulation correlated with the effect of only PSC 833 or

MK -57 1 on mitoxantrone accumulation. Cross-resistance of inhibitors could

account for this phenomenon, although this has not been described in studies

using cell lines overexpressing only one of the transporters 1 2;2 1 • In cell lines

overexpressing P-gp or MRP, we and others showed that FTC did not reverse

resistance12;2 1 • These observations in cell lines imply that FTC acts as a specific

ABCG2 inhibitor. The specificity of PSC833 was described in the same

study2 1 , in which PSC833 showed little effect on the ABCG2-overexpressing

cell line. However, in contrast to cell lines overexpressing specific transporters,

clinical samples could express other (unknown) transporters, which can also

be inhibited by FTC or PSC833 . The discrepancies in our data could be

explained by the presence of these transporters, which are capable of effluxing

mitoxantrone as well.

In summary, this study shows that ABCG2 is expressed and functional in

B-lineage ALL and at a lower level in T-lineage ALL. The next step will be

to determine the clinical impact of this transporter in a larger prospective

trial. Because ABCG2 is mainly functionally active in B-lineage ALL, the

trial will only be useful in this ALL subtype. Hopefully, it will elucidate

whether these patients with ALL could benefit from treatment adaptations based on this knowledge.

ACKNOWLEDGMENTS

This study was supported by the Foundation of Pediatric Oncology Groningen (SKOG 99-0 1 ) .

107

Chapter 5

REFERENCES

1 . Chessells, J. M., Hall, E., Prentice, H . G., Durrant, J., Bailey, C . C. , and Richards, S. M. The impact of age on outcome in lymphoblastic leukaemia; MRC UKALL X and XA compared: a report from the MRC Paediatric and Adult Working Parties. Leukemia 1 998: 12: 463-473.

2. Mauer, A. M. Adult and childhood acute lymphocytic leukemia: are they different diseases? Am J Hemato/ 1 993 : 42: 1 27- 1 3 1 .

3 . den Boer, M. L., Pieters, R., Kazemier, K . M. , Rattier, M. M. , Zwaan, C. M. , Kaspers, G. J., Janka-Schaub, G., Henze, G., Creutzig, U., Scheper, R. J., and Veerman, A. J. Relationship between major vault protein/lung resistance protein, multidrug resistanceassociated protein, P-glycoprotein expression, and drug resistance in childhood leukemia. Blood 1 998: 91 : 2092-2098.

4. Dhooge, C., De Moerloose, B., Laureys, G., Kint, J., Ferster, A. , De Bacquer, D. , Philippe, J., and Benoit, Y. P-glycoprotein is an independent prognostic factor predicting relapse in childhood acute lymphoblastic leukaemia: results of a 6-year prospective study. Br J Haemato/ 1 999: 105: 676-683.

5 . Goasguen, J . E. , Dossot, J . M., Fardel, 0., Le Mee, F. Le Gall, E., Leblay, R. , LePrise, P. Y., Chaperon, J., and Fauchet, R. Expression of the multidrug resistance-associated P-glycoprotein (P- 170) in 59 cases of de novo acute lymphoblastic leukemia: prognostic implications. Blood 1 993 : 81 : 2394-2398.

6. Ivy, S. P., Olshefski, R. S., Taylor, B. J., Patel, K. M., and Reaman, G. H. Correlation of P-glycoprotein expression and function in childhood acute leukemia: a children's cancer group study. Blood 1996: 88: 309-3 1 8.

7. Wattel, E., Lepelley, P., Merlat, A., Sartiaux, C., Bauters, F., Jouet, J. P., and Fenaux, P. Expression of the multidrug resistance P glycoprotein in newly diagnosed adult acute lymphoblastic leukemia: absence of correlation with response to treatment. Leukemia

1 995 : 9: 1 870- 1 874. 8 . Wuchter, C . , Leonid, K., Ruppert, V., Schrappe, M. , Buchner, T., Schoch, C . , Hafer

lach, T., Harbott, J., Ratei, R., Darken, B., and Ludwig, W. D. Clinical significance of P-glycoprotein expression and function for response to induction chemotherapy, relapse rate and overall survival in acute leukemia. Haematologica 2000: 85: 71 1 -72 1 .

9. Plasschaert, S. L. A., Vellenga, E., de Bont, E. S. J. M., van der Kolk, D. M., Veerman, A. J. , Sluiter W. J., Daenen S. M. G., de Vries, E. G. E., and Kamps, W. A. High functional P-glycoprotein activity is more often present in T-cell acute lymphoblastic leukemic cells in adults than in children. Leuk Lymphoma 2002: 44: 85-95 .

10. Allikmets, R., Schriml, L. M., Hutchinson, A., Romano-Spica, V., and Dean, M. A human placenta-specific ATP-binding cassette gene (ABCP) on chromosome 4q22 that is involved in multidrug resistance. Cancer Res 1 998: 58: 5337-5339.

1 1 . Doyle, L. A., Yang, W., Abruzzo, L. V., Krogmann, T., Gao, Y., Rishi, A. K., and Ross, D. D. A multidrug resistance transporter from human MCF -7 breast cancer cells. Proc Nat! Acad Sci US A 1 998: 95: 1 5665-1 5670.

12. Rabindran, S. K., He, H., Singh, M., Brown, E., Collins, K. I., Annable, T., and Greenberger, L. M. Reversal of a novel multi drug resistance mechanism in human colon carcinoma cells by fumitremorgin C. Cancer Res 1 998: 58: 5850-5858.

1 3 . Maliepaard, M., van Gastelen, M. A., Tohgo, A., Hausheer, F. H., van Waardenburg, R. C., de Jong, L. A., Pluim, D., Beijnen, J. H., and Schellens, J. H. Circumvention of

108

BCRP in ALL

breast cancer resistance protein (BCRP)-mediated resistance to camptothecins in vitro using non-substrate drugs or the BCRP inhibitor GF 1 209 1 8. Clin Cancer Res 200 1 : 7: 935-94 1 .

14. Litman, T., Brangi, M., Hudson, E., Fetsch, P., Abati, A., Ross, D . D., Miyake, K., Resau, J. H., and Bates, S. E. The multidrug-resistant phenotype associated with overexpression of the new ABC half-transporter, MXR (ABCG2). J Cell Sci 2000: 113: 201 1 -202 1 .

1 5 . Maliepaard, M., Scheffer, G. L., Faneyte, I . F., van Gastelen, M. A., Pijnenborg, A. C., Schinkel, A. H., van De Vijver, M. J., Scheper, R. J., and Schellens, J. H. Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer Res 200 1 : 61: 3458-3464.

1 6. Consoli, U., Van, N. T., Neamati, N., Mahadevia, R., Beran, M., Zhao, S., and Andreeff, M. Cellular pharmacology of mitoxantrone in p-glycoprotein-positive and - negative human myeloid leukemic cell lines. Leukemia 1 997: 11 : 2066-2074.

1 7. Schneider, E., Horton, J. K., Yang, C. H., Nakagawa, M., and Cowan, K. H. Multidrug resistance-associated protein gene overexpression and reduced drug sensitivity oftopoisomerase II in a human breast carcinoma MCF7 cell line selected for etoposide resistance. Cancer Res 1 994: 54: 1 52- 1 58.

1 8. Diah, S. K., Smitherman, P. K., Aldridge, J., Yolk, E. L., Schneider, E., Townsend, A. J., and Morrow, C. S. Resistance to mitoxantrone in multidrug-resistant MCF7 breast cancer cells: evaluation of mitoxantrone transport and the role of multidrug resistance protein family proteins. Cancer Res., 61 : 5461 -5467, 200 1 .

1 9. Honjo, Y., Hrycyna, C. A., Yan, Q. W. , Medina-Perez, W. Y., Robey, R. W. , van de, L. A., Litman, T., Dean, M., and Bates, S. E. Acquired mutations in the MXRIBCRP/ABCP gene alter substrate specificity in MXRIBCRP/ABCP-overexpressing cells. Cancer Res 200 1 : 61: 6635-6639.

20. Yolk, E. L., Farley, K. M., Wu, Y., Li, F., Robey, R. W., and Schneider, E. Overexpression of wild-type breast cancer resistance protein mediates methotrexate resistance. Cancer Res 2002: 62: 5035-5040.

2 1 . van der Kolk, D. M., Vellenga, E., Scheffer, G. L., Muller, M., Bates, S. E., Scheper, R. J., and de Vries, E. G. Expression and activity of breast cancer resistance protein (BCRP) in de novo and relapsed acute myeloid leukemia. Blood 2002: 99: 3763-3770.

22. Robey, R. W., Honjo, Y., van de Laar, A., Miyake, K., Regis, J. T., Litman, T., and Bates, S. E. A functional assay for detection of the mitoxantrone resistance protein, MXR (ABCG2). Biochim Biophys Acta 200 1 : 1512: 1 7 1 - 1 82.

23. Scharenberg, C. W., Harkey, M. A., and Torok-Storb, B. The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 2002: 99: 507-5 12 .

24. Zhou, S., Schuetz, J. D., Bunting, K. D. , Colapietro, A. M., Sampath, J . , Morris, J. J., Lagutina, 1., Grosveld, G. C., Osawa, M., Nakauchi, H., and Sorrentino, B. P. The ABC transporter Bcrpl /ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 200 1 : 7: 1 028-1 034.

25. Kamps, W. A., Veerman, A. J., van Wering, E. R., van Weerden, J. F., Slater, R., and Does-van den Berg, A. Long-term follow-up of Dutch Childhood Leukemia Study Group (DCLSG) protocols for children with acute lymphoblastic leukemia, 1 984- 1 99 1 . Leukemia 2000: 14: 2240-2246.

26. Daenen, S., van Imhoff, G. W., van den Berg, E., de Kam, P. J. , Haaxma-Reiche, H., Vel-

109

Chapter 5

lenga, E., Smit, J. W., and Halie, R. M. Improved outcome of adult acute lymphoblastic leukaemia by moderately intensified chemotherapy which includes a 'pre-induction' course for rapid tumour reduction: preliminary results on 66 patients. Br J Haematol 1 998: 100: 273-282.

27. Scheffer, G. L., Maliepaard, M., Pijnenborg, A. C., van Gastelen, M. A., de Jong, M.C., Schroeijers, A. B., van der Kolk, D. M., Allen, J . D., Ross, D. D., van der Valk, P., Dalton, W. S., Schellens, J. H., and Scheper, R. J. Breast cancer resistance protein is localized at the plasma membrane in mitoxantrone- and topotecan-resistant cell lines. Cancer Res 2000: 60: 2589-2593 .

28. Jones, T. R., Zamboni, R., Belley, M., Champion, E., Charette, L., Ford-Hutchinson, A. W., Frenette, R., Gauthier, J. Y., Leger, S., Masson, P., et al.. Pharmacology of L-660, 7 1 1 (MK-57 1 ): a novel potent and selective leukotriene D4 receptor antagonist. Can J Physiol Pharmaco/ 1 989: 67: 1 7-28.

29. van der Kolk, D. M., de Vries, E. G., Koning, J . A., van den Berg, E., Muller, M., and Vellenga, E. Activity and expression of the multidrug resistance proteins MRP 1 and MRP2 in acute myeloid leukemia cells, tumor cell lines, and normal hematopoietic CD34+ peripheral blood cells. Clin Cancer Res 1998: 4: 1 727-1 736.

30. Sauerbrey, A., Sell, W., Steinbach, D., Voigt, A., and Zintl, F. Expression of the BCRP gene (ABCG2/MXR/ ABCP) in childhood acute lymphoblastic leukaemia. Br J Haemato/ 2002: 118: 147- 1 50.

3 1 . Chaudhary, P. M. and Roninson, I. B. Expression and activity of P-glycoprotein, a multidrug efflux pump, in human hematopoietic stem cells. Cel/ 1 99 1 : 66: 85-94.

32. Bunting, K. D. ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 2002: 20: 1 1 -20.

33. Laupeze, B., Amiot, L., Drenou, B., Bernard, M., Branger, B., Grosset, J. M., Lamy, T., Fauchet, R., and Fardel, 0. High multidrug resistance protein activity in acute myeloid leukaemias is associated with poor response to chemotherapy and reduced patient survival. Br J Haemato/ 2002: 116: 834-838.

34. Legrand, 0., Simonin, G., Perrot, J. Y., Zittoun, R., and Marie, J. P. Pgp and MRP activities using calcein-AM are prognostic factors in adult acute myeloid leukemia patients. Blood 1 998: 91: 4480-4488.

35. te Boekhorst, P. A., de Leeuw, K., Schoester, M., Wittebol, S., Nooter, K., Hagemeijer, A., Lowenberg, B., and Sonneveld, P. Predominance of functional multi drug resistance (MDR- 1) phenotype in CD34+ acute myeloid leukemia cells. Blood, 1 993: 82: 3 1 57-3 1 62.

36. van der Kolk, D. M., de Vries, E. G., Noordhoek, L., van den Berg, E., van der Pol, M. A., Muller, M., and Vellenga, E. Activity and expression of the multidrug resistance proteins P- glycoprotein, MRP 1 , MRP2, MRP3 and MRP5 in de novo and relapsed acute myeloid leukemia. Leukemia 200 1 : 15: 1 544-1 553.

37. Cascavilla, N., Musto, P., D'Arena, G., Ladogana, S . . Matera, R., and Carotenuto, M. Adult and childhood acute lymphoblastic leukemia: clinico-biological differences based on CD34 antigen expression. Haematologica 1 997: 82: 3 1 -37.

38. Minderman, H., Suvannasankha, A., O'Loughlin, K. L., Scheffer, G. L., Scheper, R. J., Robey, R.W. and Baer, M. R. Flow cytometric analysis of breast cancer resistance protein expression and function. Cytomet1y 2002: 48: 59-65.

39. Broxterman, H. J., Sonneveld, P., Feller, N., Ossenkoppele, G. J., Wahrer, D. C., Eekman, C. A., Schoester, M., Lankelma, J . , Pinedo, H. M., Lowenberg, B. and Schuurhuis, G. J.

110

BCRP in ALL

Quality control of multi drug resistance assays in adult acute leukemia: correlation between assays for ?-glycoprotein expression and activity. Blood 1 996: 87: 4809-48 1 6.

40. Cole, S. P., Sparks, K. E., Fraser, K., Loe, D. W., Grant, C. E., Wilson, G. M., and Deeley, R. G. Pharmacological characterization ofmultidrug resistant MRP-transfected human tumor cells. Cancer Res 1 994: 54: 5902-59 1 0.

111

Chapter 6

Expression of multidrug resistance-associated proteins

predicts prognosis in acute lymphoblastic leukemia in

children and adults

Sabine L.A. Plasschaert1, Eveline S.J.M. de Bont1, Marike Boezen2, Dorina M. van der Kolk3,

Simon M. J. G. Daenen3, Klaas Nico Fabert, Willem A. Kamps1, Elisabeth G.E. de Vries5, Edo Vellenga1

Departments of1Pediatric Oncology and Hematology, 2Epidemiology, 3Hematology, 4Gastroenterology and Hepatology, 5Medical Oncology,

University Hospital Groningen, Groningen, The Netherlands.

Submitted

Chapter 6

ABSTRACT

Patients with acute lymphoblastic leukemia (ALL) are treated with a variety of

chemotherapeutic drugs, which can be transported by six multidrug resistance

associated proteins (MRPs). These MRPs have strongly overlapping functional

activities and therefore studying only one or two of these transporters can potentially

underestimate the clinical relevance of these proteins. The aim of this study was to

investigate the expression levels of MRP l -6 and study their impact on prognosis.

The mRNA expression level of MRP 1-6 was analyzed by quantitative real-time

PCR in leukemic blasts of 105 de novo ALL patients (adults, n=49; children, n=56)

including 70% B-lineage and 30% T-lineage ALL patients. Adults showed a higher

expression of MRPl (P=.008), MRP2 (P=.026) and MRP3 (P=.039) than children.

Interestingly, this difference disappeared when patients were categorized based on

clinical outcome. Relapsed patients showed a higher expression of all MRP genes,

except MRP4. For the total group of ALL patients, the expression of MRP l -3 and

MRP5-6 predicted relapse. Moreover, high expression of all MRP genes, except

MRP4, was associated with a reduced relapse-free survival in children and adults

(MRPl P=.005, MRP2 P=.008, MRP3 P=.OO l , MRP5 P=.0 16, MRP6 P=.037).

In summary, the present study demonstrates that a subset of ALL patients with high

MRP expression has an unfavorable prognosis independently of age.

114

MRPl-6 mRNA expression in ALL

INTRODUCTION

The prognosis of acute lymphoblastic leukemia (ALL) has improved considerably

during the last years. However, children persistently have a better survival than

adults1;2• This is linked to a higher tolerance of children for chemotherapy, with the

result that they can receive more intensive treatment. Additionally, intrinsic cellular

differences might exist between childhood versus adult ALL cells. This is supported

by studies suggesting that the malignant transformation originates at different

levels in the lymphoid stem cell compartment3;4• These differences might be

translated in an altered drug sensitivity profile5• Chemosensitivity is mainly

regulated by the ATP-binding cassette superfamily of membrane transporters. Of this

family, especially P-glycoprotein (P-gp), Breast Cancer Resistance Protein (BCRP),

multi drug resistance-associated protein 1 (MRP 1 ) and MRP2 have been studied in

ALL. So far, high P-gp activity has been associated with poor clinical outcome

mainly in adults6·9• For MRP1 and MRP2 no correlation with prognosis has been

observed in childhood ALU;s;IO; I I , while the impact of BCRP on prognosis is yet

unknown12• Currently, eight MRP genes have been identified, of which the MRP

transporters (MRP1 -6) are known to be involved in extruding substrates that are

generally used in the treatment of ALL, including doxorubicin, vincristine, etoposide,

6-mercaptopurine and methotrexate (summarized in table 1 )13•22• In view of the

strong overlapping activities of the different MRPs, studying only one or two

of these transporters can potentially underestimate the clinical relevance of these

proteins. The aim of this study was to investigate the mRNA expression of all six

relevant MRP genes (MRP 1 -6) in childhood and adult ALL and study their impact

on patient prognosis. Our results demonstrate that a subset of patients with high

MRP expression has an unfavorable prognosis independently of age.


Patients

From all 1 89 patients newly diagnosed with ALL at the University Hospital

Groningen during the period of 1989-2002, bone marrow and peripheral blood

samples were collected and cryopreserved after informed consent. Patients were

included in the study, if a sufficient number of bone marrow or peripheral

mononuclear cells was available. The clinical data from all patients were obtained,

including age at diagnosis, gender, white blood cell (WBC) count at diagnosis,

lactate dehydrogenase (LDH), immunophenotype, cytogenetics and clinical outcome.

Complete remission was defined as less than 5% leukemic blasts in bone marrow,

115

Chapter 6

Table 1 Resistance by the MRP subfamily to drugs used in ALL treatment

MRP l

MRP2

MRP3

MRP4

MRP5

MRP6

Ref. 13-22.

Chemotherapeutic drug

Daunorubicin Doxorubicin Mitoxantrone Vincristine Etoposide Methotrexate Doxorubicin Vincristine Etoposide Methotrexate Doxorubicin Vincristine Etoposide Methotrexate Methotrexate 6-Mercaptopurine Methotrexate 6-Mercaptopurine Daunorubicin Doxorubicin Etoposide

absence of blasts m peripheral blood, absence of leukemic blasts in spinal

fluid or other extramedullary sites. Sufficient patient material was available

from 1 05 patients. The Dutch Childhood Leukemia Study Group performed

the immunophenotyping and cytomorphology of pediatric samples. Adult patient

samples were immunophenotyped by the central laboratory of the University

Hospital Groningen.

Children were treated according to the Dutch Childhood Leukemia Study Group

(DCLSG) protocol ALL-9, except three patients who were treated according to

protocol DCLSG ALL-823• Childhood patients received vincristine, dexamethasone

and asparaginase as induction therapy. Maintenance therapy consisted of

6-mercaptopurine and methotrexate, alternating with vincristine and dexamethasone.

High-risk patients (WBC count more than 50 x 1 09/L, mediastinal mass, CNS

involvement, unfavorable cytogenetics or T-cell immunophenotype) additionally

received daunorubicin during induction therapy and intensification treatment with

one cycle comparable with induction treatment and a second cycle consisting

of cytosine arabinoside and cyclophosphamide. The adult patients received a

116


regimen including prednisone, vincnstme, adriamycin, cytosine arabinoside and

asparaginase, as induction and consolidation therapy. Maintenance therapy consisted

of 6-mercaptopurine and methotrexate in combination with cytosine arabinoside and

etoposide or cyclophosphamide and mitoxantrone. Initially, high-risk patients (WBC

count more than 30 x 1 09/L or T-cell immunophenotype with mediastinal mass)

received extra therapy with cytosine arabinoside and etoposide24• From 1996, all

adult patients received extra therapy with cytosine arabinoside and etoposide, and

the high-risk patients were treated additionally with methotrexate.

Mononuclear cells from bone marrow or peripheral blood were isolated on Ficoll

Isopaque (Nycomed, Oslo, Norway) density gradient by centrifugation. The cells

were cryopreserved in RPMI 1 640 medium supplemented with 1 0% fetal calf serum

and 1 0% dimethyl sulfoxide (Merck, Amsterdam, The Netherlands) and stored at

-1 96 °C. The percentage of blasts in patient material was 86 ± 1 5% (mean ± SD)

RNA isolation

Total cellular RNA was isolated from ALL blasts using RNeasy Mini Kit including

DNase digestion (Qiagen, Hilden, Germany). From some samples total cellular

RNA was extracted using 1 mL of Trizol reagent (Life Technologies, Breda, The

Netherlands). The amount of RNA was measured by photometry. Subsequently,

1 flg RNA was reverse transcribed in 20 flL reverse transcriptase buffer containing

1 0 mM DTT, 0.5 mM each of dATP, dGTP, dCTP, dTTP, 200 U of Moloney

murine leukemia virus reverse transcriptase (M-MLV RT), 5 U RNAse inhibitor and

10 ng/flL pd(N)6 random primers (MBI Fermentas, St.Leon-Rot, Germany).

Quantitative PCR was performed using the ABI Prism 7700 Sequence Detector

(Applied Biosystems, Foster City, Ca). Primers and TaqMan probes are described

in Table 2. We used 4 J.l.L of a 22.5-fold diluted eDNA in each PCR reaction

in a final volume of 20 J.l.L, containing 900 nM of sense and antisense primers,

200 nM of the TaqMan probe, 5 mM MgCl2, KCl, Tris HCl, 0.2 mM dATP,

dCTP, dGTP, dTTP, and dUTP, and 0.5 U of AmpliTaq DNA polymerase (qPCR

Core Kit; Eurogentech, Seraing, Belgium). TaqMan probes were labeled by a 5'

FAM (6-carboxy-fluorescein) reporter and a 3 ' TAMRA (6-carboxyl-tetramethyl

rhodamine) quencher. The PCR program was 95 oc for 1 0 minutes, followed by

40 cycles of 1 5 seconds at 95 °C and 1 minute at 60 °C.

The expression of the MRP-genes was standardized for expression of glyceraldehyde-

3-phosphate dehydrogenase (GAPDH). Serial eDNA dilutions of a mixture of all

patient samples were used to generate standard curves. The expression of each gene

in each sample was analyzed in duplicate. The regression coefficients of the standard

curves were 2::0.99.

117

Chapter 6

Table 2. Sequences of the PCR primers and probes used for real-time detection PCR analysis

CDNA Primers

Gapdh Forward 5 '-GGTGGTCTCCTCTGACTTCAACA-3 ' Reverse 5' -GTGGTCGTTGAGGGCAATG-3' Probe 5 ' -ACACCCACTCCTCCACCTTTGACGC-3 '

MRP l Forward 5 ' -CTTCTGGAGGAATTGGTTGTATAGAAG-3' Reverse 5 ' -GGTAGACCCAGACAAGGATGTTAGA-3 ' Probe 5 '-TCTTTGAGATGCTTCTGGCTCCCATCAC-3'

MRP2 Forward 5 '-TGCAGCCTCCATAACCATGAG-3 ' Reverse 5 '-CTTCGTCTTCCTTCAGGCTATTCA-3 ' Probe 5 '-CAGCTTTCGTCGAACACTTAGCCGCA-3 '

MRP3 Forward 5 '-GCCATCGACCTGGAGACTGA-3 ' Reverse 5 '-GACCCTGGTGTAGTCCATGATAGTG-3 ' Probe 5 ' -CATCCGCACCCAGTTTGATACCTGCAC-3 '

MRP4 Forward 5 ' -AAGTGAACAACCTCCAGTTCCAG-3 ' Reverse 5 ' -GGCTCTCCAGAGCACCATCT-3 ' Probe 5 '-CAAACCGAAGACTCTGAGAAGGTACGATTCCT-3 '

MRP5 Forward 5 '-TGAAAGCCATTCGAGGAGTTG-3 ' Reverse 5 ' -CGGAAAAGCTCGTCATGCA-3 ' Probe 5 ' -CTCGCAGCGTGCCCTTGACAAAG-3 '

MRP6 Forward 5 ' -AGACACGGTTGACGTGGACAT-3 ' Reverse 5 '-GCTGACCTCCAGGAGTCCAA-3 ' Probe 5 '-CCAGACAAACTCCGGTCCCTGCTGAT-3 '

Statistical methods

Because the levels ofMRP expression were not normally distributed, non-parametric

methods were used. The Mann Whitney U test was performed to compare MRP

expression between two groups. For more than two groups, the Kruskal Wallis test

was employed. With the Spearman rank test the correlation between MRP expression

and other continuous variables was determined. The ability to discriminate between

patients who will remain in complete remission and patients who will relapse

was investigated by receiver operating characteristic (ROC) curves, and the area

under the curve (AUC) was calculated. Kaplan-Meier statistics and log-rank tests

were calculated to estimate the significance of differences between survival curves.

Median values were used as cutoffs for high versus low MRP expression. Relapse

free survival was defined as the time interval between the date that complete

remission was established until the first evidence of disease relapse. Second

malignancy, early death and toxic death in remission were censored at the time of

occurrence. Relapse-free survival was censored at the date of last contact for patients

118


with no report of relapse.

All ?-values are given for 2-sided tests and ?-values ::; .05 were considered

significant. Analyses were performed using SPSS 1 1 .0 for Windows software (SPSS,

Chicago, IL).

RESULTS


A total of 1 05 patients with de novo ALL were studied, with an age ranging between

2 and 60 years. The main characteristics of the studied patient population are

depicted in Table 3 . In total, 56 children were treated according to the DCLSG

protocol. Forty-five adults received treatment according to the adult protocol, and

four children under the age of 1 7 received this treatment as well. To exclude the

possibility of a selection bias, the patient characteristics were compared between

the studied patients (n= 1 05) and the patients that did not participate in our study

(n=84) due to lack of sufficient patient material. No differences were observed

in age, gender, WBC, LDH, immunophenotype and unfavorable chromosomal

Table 3. Patient characteristics at diagnosis

Patient characteristics Children (n=56) Adults (n=49)

Age, y (mean ± SD) 5.9 ± 4. 1 36.6 ± 1 8.7 Sex, male/female 34/22 32/1 7 Median WBC, 1 09/L (range) 1 8.3 (0.8-998.9) 1 7.8 ( 1 . 1 -332.2) Median LDH, U/L (range) 762.5 (263-1 1 789) 1 0 1 0 (5 1 -6375) lmmunophenotype

T-ALL 14 (25%) 1 8 (37%) Common ALL 23 (41%) 1 8 (37%) Pre-B ALL 1 7 (30%) 4 (8%) Pro-B ALL 2 (4%) 9 ( 1 8%)

Cytogenetics Translocation t(9;22) 2 % (n=5 1 ) 1 2 % (n=43) Translocation t( 4; 1 1 ) 4 % (n=5 1 ) 7 % (n=43) Translocation t( 1 2;2 1 ) 4 % (n=50) 0 % (n=43)

Hyperdiploid (more than 50 chromosomes) 22 % (n=49) NA Relapse/continuous complete remission 1 0/45 14/24* Death during induction 6 Induction failure 0 2

WBC indicates white blood cell count; LDH lactate dehydrogenase; and NA, not available * From 2 adults no clinical data were available. One patient died from treatment-related

toxicity in complete remission.

119

Chapter 6

aberrations (t(9;22), t(4; 1 1) or l lq23 rearrangements) between the two groups (data

not shown).

The overall survival was 82% in children, with a median follow-up of 3.6 years for

the total group of childhood patients. For the children who remained in complete

remission the median follow-up was 3 .8 years. The overall survival for adults was

48%, with a median follow-up of 2.0 years for the total group. The group of adult

patient� who remained in complete remission had a median follow-up of 3.9 years.

Expression of MRP in ALL

The expression of the MRP genes m leukemic blasts of children and adults

were studied and showed a large variability in expression level (Table 4). MRPl

expression could be detected in 48 children (86%) and 48 adults (98%). Similarly,

MRP2 expression was present in 49 children (87%) and 46 adults (94%). The lowest

expression was found for MRP3, which could only be demonstrated in 29 childhood

(52%) and 40 adult (82%) patients. MRP4, MRP5 and MRP6 were observed in

almost all patients (98%, 99% and 98% respectively). When comparing the MRP

expression between children and adults, the expression levels of MRP l , MRP2

and MRP 3 were higher in adults than in children (P=.008, P=.026 and P=.039

respectively). Correlations were observed between the expression of the six MRP

genes in ALL blasts, but with low Spearman's correlation coefficients ranging from

0.23-0.56. This suggests that high expression of one MRP in individual patients was

correlated with increased expression of multiple MRPs.

MRP expression and survival

The median MRPl expression was higher in patients who relapsed than in patients

who remained in complete remission (P=.003). The same pattern was observed for

MRP2 (P=.002), MRP3 (P=.OO l ), MRP5 (P=.004) and MRP6 (P=.009, Table 5).

Table 4 MRP expression in ALL

MRP expression

MRPI expression MRP2 expression MRP3 expression MRP4 expression MRP5 expression MRP6 expression

Children (n=56)

0.69 (0.37-1 .45) 0.53 (0. 1 9- 1 .40)

0.07 (0-0.8 1) 0.65 (0.33-2. 1 5) 0.95 (0.63-2.02) 0.54 (0.3 1 - 1 .96)

Adults (n=49) Mann Whitney U p

1 .08 (0.63-1 .90) .008 0.96 (0.60- 1 .96) .026 0.3 1 (0.09- 1 .04) .039 0.78 (0.41 - 1 .69) .445 1 .04 (0.60-1 . 75) .928 0.65 (0.39- 1 .59) .693

The values are given as median (25th - 75th percentile)

120


Table 5 MRP expression and clinical outcome children and adults

MRP expression Continuous Complete Relapset Remission*

MRPl expression Children 0.56 (0.33-0.93) 1 .60 (0.62-7 . 12) Adults 0.83 (0.59-1 .93) 1 .40 (0.82-2.34) Children & adults 0.66 (0.42- 1 .27) 1 .44 (0.80-2.59) MRP2 expression Children 0.50 (0. 1 7- 1 . 10) 1 .73 (0.22-1 3 .35) Adults 0.87 (0.49-1 .25) 1 . 76 (0.91-2.66) Children & adults 0.69 (0.24- 1 . 1 8) 1 . 76 (0.68-3. 1 7) MRP3 expression Children 0 (0-0.33) 1 .70 (0.06-59.72) Adults 0.23 (0.06-0.5 1 ) 0.69 (0.23- 1 . 73) Children & adults 0.08 (0-0.44) 0.82 (0.20-3.27) MRP4 expression Children 0.64 (0.34-2. 1 4) 0.92 (0.25-23.42) Adults 0. 70 (0.40- 1 .02) 0.85 (0.5 1 - 1 .67) Children & adults 0.65 (0.37-1 .53) 0.85 (0.48- 1 . 80) MRP5 expression Children 0.93 (0.58- 1 .76) 2.20 (0.84-12.57) Adults 0.93 (0.45-1 .49) 1 .33 (0.93-2.34) Children & adults 0.93 (0.57- 1 .6 1 ) 1 .54 (0.88-3 .01) MRP6 expression Children 0.53 (0.30-1 .55) 1 .33 (0.44-43 . 1 5) Adults 0.46 (0.33-1 .03) 1 .09 (0.58-1 .95) Children & adults 0.49 (0.3 1 - 1 .26) 1 .09 (0.54-2.59)

The values are given as median (25th-75th percentile) * Continuous complete remission: children n=45; adults n=24 t Relapse: children n=l O; adults n=14

Mann Whitney U p

.028

. 1 44

.002

. 1 37

.01 2

.002

.0 1 3

.076

.00 1

.492

.287 .238

.01 6

.082

.004

. 127

.0 1 5

.009

When the children were analyzed separately, a higher expression level was observed

in relapsed patients for MRPl (P=.028), MRP3 (P=.Ol 3) and MRP5 (P=.O l 6). The

remaining MRPs, (MRP2, MRP6) showed a trend towards a higher expression in

relapsed patients (MRP2 P=. l 37 and MRP6 P=. l 27). In adults who relapsed, MRP2

and MRP6 expression were higher compared to adults who did not relapse (P=.O l 2,

P=.Ol 5 respectively). The expression of MRP l , MRP3 and MRP5 was also higher

in relapsed adults, but this was not statistically significant. MRP4 expression was

similar in all patients who remained in complete remission and all patients who

relapsed.

In Table 6 the median MRP expression is compared between children and adults

who were in continuous complete remission and between children and adults who

121

Chapter 6

Table 6 MRP expression and clinical outcome children versus adults

MRP expression Children* Adultst

MRP1 expression CCR 0.56 (0.33-0.93) 0.83 (0.59-1 .93) Relapse 1 .6 1 (0.62-7 . 12) 1 .40 (0.82-2.34) MRP2 expression CCR 0.50 (0. 1 7- 1 . 1 0) 0.87 (0.49-1 .25) Relapse 1 .73 (0.22-1 3 .35) 1 .76 (0.91 -2.66) MRP3 expression CCR 0 (0-0.33) 0.23 (0.06-0.5 1 ) Relapse 1 .70 (0.06-59.72) 0.69 (0.23-1 .73) MRP4 expression CCR 0.64 (0.34-2. 14) 0.70 (0.40-1 .02) Relapse 0.92 (0.25-23 .42) 0.85 (0.5 1 - 1 .67) MRP5 expression CCR 0.93 (0.58-1 .76) 0.93 (0.45-1 .49) Relapse 2.20 (0.84-1 2.57) 1 .33 (0.93-2.34) MRP6 expression CCR 0.53 (0.30- 1 .55) 0.46 (0.33-1 .03) Relapse 1 .33 (0.44-43 . 1 5) 1 .09 (0.58-1 .95)

The values are given as median (25th-75th percentile) CCR indicates continuous complete remission. * Children: CCR n=45; relapse n=1 0 t Adults: CCR n=24; relapse n=14

Mann Whitney U p

0.0 14 0.84 1

0.099 0.886

0.056 0.508

0.801 1 .000

0.623 0. 1 72

0.682 0.841

relapsed. Children who remained in complete remission showed lower MRPl

and MRP3 levels than adults (P=.0 14 and P=.056 respectively). The other MRPs

did not differ between children and adults in complete remission. Interestingly,

the expression of MRPs was comparable in children and adults who eventually

relapsed.

By ROC curves, the ability of MRPs to discriminate between patients who will

remain in complete remission versus patients that will relapse was analyzed. The

ROC curves and AUCs for both children and adults are shown in Figure 1 and

Table 7. All individual MRP genes, except MRP4, predicted whether patients would

relapse. The AUCs were comparable for the five MRPs, which indicates that they

can all predict prognosis with comparable power. A high expression of all five MRPs

did not lead to a stronger prediction of relapse than the MRPs individually. When

the children were analyzed separately, the same MRPs could predict which patients

would relapse, with statistical significance for MRP l , MRP3 and MRP5 (Table 8A).

122


,8

f ·� .s Jl

-- MRP1 -- MRP2 • • • • • • • MRP3

,3 ,5 ,8 1 ,0

1 - Specificity

1 ,0

,8

� > E ,5 :!! ., <n

• • • • • • • MRP5 -- MRP6 -- MRP1/213/516

,3 ,5 ,8 1 ,0

1- Specificity

Figure 1 . ROC curves predicting relapse for the different MRPs in children and adults. (A) ROC curves for MRP l , MRP2 and MRP3. (B) ROC curves for MRP5, MRP6 and the cluster of MRPl , MRP2, MRP3, MRP5 and MRP6.

Table 7. MRP expression and area under the ROC curve for children and adults

AUC 95% CI p

MRP l 0.72 0.60-0.83 .002 MRP2 0.71 0.58-0.85 .002 MRP3 0.72 0.6 1 -0.84 .001 MRP4 0.58 0.45-0.72 .238 MRP5 0.70 0.59-0.8 1 .004 MRP6 0.68 0.56-0.80 .009 MRP l /2/3/5/6 Cluster 0.70 0.60-0.8 1 .003

AUC indicates area under the curve; CI indicates confidence interval

123

Chapter 6

Table SA. MRP expression and area under the ROC curve for children

AUC 95% CI p

MRP l 0.72 0.53-0.92 .028 MRP2 0.65 0.42-0.88 . 1 38 MRP3 0.74 0.56-0.92 .01 8 MRP4 0.57 0.34-0.80 .492 MRP5 0.74 0.58-0.9 1 .0 1 6 MRP6 0.66 0.47-0.84 . 1 27 M RP l/3/5 Cluster 0.77 0.63-0.9 1 .009

Table 8B. MRP expression and area under the ROC curve for adults

AUC 95% CI p

MRPI 0.65 0.47-0.82 . 1 38 MRP2 0.74 0.57-0.92 .014 MRP3 0.67 0.50-0.85 .077 MRP4 0.6 1 0.42-0.79 .276 MRP5 0.67 0.50-0.84 .079 MRP6 0.74 0.58-0.90 .0 1 5 MRP2/6 Cluster 0.75 0.59-0.92 .0 1 0

AUC indicates area under the curve; CI indicates confidence interval

Of the traditional predictors, initial WBC count also discriminated for outcome

(AUC 0.79, P=.007). ROC curves for adults showed the prognostic impact of the

same MRPs, of which MRP2 and MRP6 were statistically significant (Table 8B). For

the adults, initial WBC count, age, LDH or immunophenotype were not predictive

for outcome. Kaplan-Meier curves and log-rank tests analyzed the impact of MRPs

on relapse-free survival. The median MRP expression was used as cutoff point for

high and low MRP expression. Table 9 provides the data for relapse-free survival

in all patients with low versus high MRP expression. It shows that patients with

high expression of all MRPs, except MRP4, had a shorter relapse-free survival than

patients with low MRP expression (MRPl P=.005, MRP2 P=.008, MRP3 P=.00 1 ,

MRP5 P=.01 6, MRP6 P=.037). In figure 2 the relapse-free survival curves for the

various MRPs are depicted. When childhood patients were analyzed alone, high

MRP3 expression levels lead to shorter relapse-free survival (5.3 years vs. 4.9 years,

P=.033). For adults, MRP2 and MRP6 influenced relapse-free survival (MRP2 1 0.7

years vs. 5.8 years, P=.049; MRP6 1 1 .2 years vs. 3 .3 years, P=.008).

In summary, ALL patients who will eventually relapse can be recognized at diagnosis

124


Table 9. Impact of high MRP expression on relapse-free survival in children and adults

Variable Relapse-free survival, Relapse- Log-rank test, y (95% CI) free survival p

MRPl Low expression 1 1 .7 ( 1 0.4-1 2.9) 87.2 % High expression 8.2 (6.3- 1 0. 1 ) 60.0 % .0047

MRP2 Low expression 1 1 .0 (9.8-12.2) 86.4 % High expression 8.2 (6.2- 1 0.2) 62.5 % .0084

MRP3 Low expression 1 2.4 ( 1 1 .2- 13 .6) 89.4 % High expression 7.6 (5.6-9.5) 57.8 % .0006

MRP4 Low expression 1 0.6 (9. 1 - 12. 1 ) 78.7 % High expression 8.9 (6.8- 1 1 . 1 ) 68.9 % .2077

MRP5 Low expression 1 1 .8 ( 10.4-1 3.2) 84.8 % High expression 7.6 (5 .8-9.5) 63.0 % .01 58

MRP6 Low expression 1 1 .5 ( 1 0. 1 - 1 3 .0) 83.3 % High expression 7.8 (5.9-9.7) 63.6 % .0374

The median value was used as cutoff for high and low expression.

by high expression levels of MRPl , MRP2, MRP3, MRPS and MRP6 in leukemic

cells. This expression profile is comparable in children and adults, but is more often

present in adults.

MRP expression and clinical features

To investigate whether high MRP expression is an independent prognostic factor,

the association with clinical features, including age, gender, WBC count, LDH,

immunophenotype and unfavorable chromosomal aberrations (t(9;22), t(4; 1 1 ) or

l l q23 rearrangements) was determined. MRPI expression was higher in T-lineage

ALL compared to B-lineage ALL (including pro-B, pre-B and common ALL)

(median MRPI expression 1 .00 vs. 0.66 (P=.03 1 ). In contrast, MRP4 was expressed

at a lower level in T-lineage ALL in comparison with B-lineage ALL (median

MRP4 expression 0.48 vs. 0.83 (P=.007). For MRP2, MRP3, MRPS and MRP6 no

association between immunophenotype and MRP expression was observed.

No effect of gender was found on the expression of MRP genes in children and

adults. Age correlated positively with MRPI expression. Higher initial WBC counts

were associated with a lower expression of MRP3, MRP4 and MRP6. High LDH

levels were correlated to low expression of MRP3 and MRP6. The Spearman

correlation coefficients for age, initial WBC counts and LDH were very small

(r=0.23-0.27).

There were few ALL patients with chromosomal aberrations. We analyzed whether

these patients differed in their MRP expression pattern. Five adults and one child

125

Chapter 6

A 1 . B 1 .

MRP1 l MRP2 l

1 1 i!: i!: :I o . iii 0. ..

t MRP1 H I MRP2 H

0.4 t 0.4 i -II a: N 0 a: N 0 0.2 L 47 e 0.2 l 44 6

H 45 18 H 48 18

o.o. nk P...005 .. Q o.o •

rank P=.008 I) " "

Yenalor dllgnooil Yon alor dllgnooil

D 1 .

MR l

1 1 i!: i!: :I :I 0 . .. ..

t MR I 0.4 t 0.4

� -II N 0 a: N 0 0.2 l 47 5 0.2 l 47 10

H 45 1 9 H 45 14 nmk P=.001 l rank P=.208

0 . • I) Q " o.o, I) " "

Yon ator dllgnooil Yon atordllgnooil

F 1 .

1 1 i!: i!: :I :I 0 . .. ..

i MRPS H

t 0.4 0.4

� -II N 0 a: N 0 0.2 l 48 7 0.2 l 48 8

H 48 1 7 H 44 1 8 l rankP=.016 rank P...037

0.0, I) " " o.o, I) " "

vo .. ator dagn .. ll Yon alor dllgnooil

Figure 2. Relapse-free survival in children and adults from the various MRPs, subdivided in two groups with low (L) and high (H) MRP expression. N, number; 0, observed events. (A) MRPl . (B) MRP2. (C) MRP3. (D) MRP4. (E) MRP5. (F) MRP6.

126


had a BCR/ ABL translocation. No difference in MRP expression was observed

in these patients compared to patients without the translocation. The translocation

t( 4; 1 1 ) was present in three adults and two children, and these patients showed a

higher MRP2 and MRP4 expression (P=.027 and P=.033 respectively) than patients

without the translocation. Hyperdiploidy (more than 50 chromosomes) of leukemic

blasts, which has an incidence of 1 5-20 % in childhood ALL, was present in 1 1

children and was not associated with a high or low expression ofMRP genes.

DISCUSSION

This is the first study describing the association between MRP 1 -6 expression and

clinical outcome in childhood and adult ALL. In children, high expression of MRP 1 ,

MRP3 and MRP5 i s associated with unfavorable outcome, whereas MRP2 and

MRP6 can predict relapse in adults. In addition, adults showed a higher expression

of MRP 1 , MRP2 and MRP3 compared to children. Interestingly, this difference

disappeared when children and adults were categorized on the basis of clinical

outcome. Relapsed patients showed a higher expression of all MRP genes, except

MRP4, than patients who remained in complete remission. These results suggest

that the prognosis of childhood ALL is comparable with adult ALL, if childhood

leukemic cells have intrinsic cellular properties similar to adult leukemic cells.

This supports the studies of Greaves et al. , who suggested that the transforming

events in adult and childhood ALL occur in general at a different maturation level

in the hematopoietic stem cell compartrnent3;4. In adult ALL the transformation is

considered to occur in the multipotent stem cell, whereas childhood ALL originates

from the more committed lymphoid progenitor cell. The level of the transforming

event is not strictly connected to a certain age group, because there is an overlapping

spectrum of the course of the disease between children and adults with ALL.

The association between maturation stage and ABC-transporters has already been

observed in normal hematopoiesis. The expression and functional activity of P-gp

and BCRP was found to be higher in normal hematopoietic stem cells with an

immature phenotype25-27•

So far, only a few studies have addressed the role of a number of MRPs in ALL.

No association between expression levels and clinical outcome was observed for

MRP1 and MRP2 in childhood ALL in these studies7;s; IO; I I . For MRP3, children with

high expression showed lower overall survival rates than patients with low MRP

expression, which is in agreement with our study1 1 • The discrepancy with our study for MRP 1 and MRP2 might be caused by the limited number of studied childhood

ALL patients with an unfavorable outcome in their study. In view of the fact that

127

Chapter 6

we also studied adult patients, we included more patients with an unfavorable MRP

profile, which makes it easier to detect a possible influence of MRP expression on

prognosis. This is supported by the finding that the significance of MRP expression

in our study was even stronger when the combined group of children and adults was

analyzed.

A remarkable finding was that five of the six studied MRPs have a higher expression

in relapsed ALL patients. These results indicate that it will be of limited value

to counteract the effects of a single MRP with regard to drug resistance. A more

general approach is required, especially because the different MRPs have strongly

overlapping substrate specificities and are capable of transporting many different

chemotherapeutic drugs. Recently the results of micro-array analysis were reported

in patients with childhood ALL in relation to drug sensitivity and clinical outcome28•

The gene expression patterns were investigated in relation to in vitro cellular drug

resistance to prednisone, vincristine, asparaginase and daunorubicin. In this analysis,

no differences were found in the expression of genes known to play a role in

drug resistance, including MDR-1 and MRP l . The discrepancy with our study

might be related to the fact that the selection of genes was based on the drug

sensitivity assay, which is another approach than clinical endpoints, such as relapse

free survival as used in the present study. Moreover, MRPs are not involved in the

transport of prednisone or asparaginase and will subsequently not be recognized as

an unfavorable prognostic factor.

In summary, the present study demonstrates that a subset of ALL patients with

high MRP expression has an unfavorable prognosis independently of age. Early

recognition of a profile with high MRP expression could identify patients with an

increased risk for relapse that could benefit from treatment adaptations based on this

knowledge.

128

MRP1-6 mRNA expression in ALL

REFERENCES

l . Gokbuget N, Hoelzer D . Recent approaches in acute lymphoblastic leukemia in adults. Rev Clin Exp Hemato/ 2002: 6: 1 14- 14 1 .

2. Schrappe M, Camitta B, Pui CH, et a! . Long-term results of large prospective trials in childhood acute lymphoblastic leukemia. Leukemia 2000: 14: 2 193-2 194.

3 . Greaves M. Molecular genetics, natural history and the demise of childhood leukaemia. Eur J Cancer 1 999: 35: 1 94 1 - 1953.

4. Greaves MF. Stem cell origins of leukaemia and curability. Br J Cancer 1 993 : 67: 41 3-423.

5. Styczynski J, Pieters R, Huismans DR, Schuurhuis GJ, Wysocki M, Veerman AJ. In vitro drug resistance profiles of adult versus childhood acute lymphoblastic leukaemia. Br J Haemato/ 2000: 110: 8 1 3-8 18 .

6. Del Principe MI, Del Poeta G, Maurillo L , et a!. P-glycoprotein and BCL-2 levels predict outcome in adult acute lymphoblastic leukaemia. Br J Haematol 2003 : 121: 730-738.

7. Plasschaert SL, Vellenga E, de Bont ES, et a!. High functional P-glycoprotein activity is more often present in T-cell acute lymphoblastic leukaemic cells in adults than in children. Leuk Lymphoma 2003: 44: 85-95.

8. Tafuri A, Gregorj C, Petrucci MT, et a!. MDRl protein expression is an independent predictor of complete remission in newly diagnosed adult acute lymphoblastic leukemia. Blood 2002: 100: 974-98 1 .

9 . Wuchter C , Leonid K, Ruppert V, et al. Clinical significance of P-glycoprotein expression and function for response to induction chemotherapy, relapse rate and overall survival in acute leukemia. Haematologica 2000: 85: 7 1 1 -72 1 .

10. Sauerbrey A, Voigt A, Wittig S, Hafer R, Zintl F. Messenger RNA analysis of the multidrug resistance related protein (MRP l ) and the lung resistance protein (LRP) in de novo and relapsed childhood acute lymphoblastic leukemia. Leuk Lymphoma 2002: 43: 875-879.

1 1 . Steinbach D, Wittig S, Cario G, et a!. The multidrug resistance-associated protein 3 (MRP3) is associated with a poor outcome in childhood ALL and may account for the worse prognosis in male patients and T-cell immunophenotype. Blood 2003 : 102: 4493-4498.

12 . Plasschaert SL, van der Kolk DM, de Bont ES, et al . The role of breast cancer resistance protein in acute lymphoblastic leukemia. Clin Cancer Res 2003 : 9: 5 1 7 1 -5 1 77.

13. Belinsky MG, Chen ZS, Shchaveleva I, Zeng H, Kruh GD. Characterization of the drug resistance and transport properties of multidrug resistance protein 6 (MRP6, ABCC6). Cancer Res 2002: 62: 6 1 72-61 77.

14. Borst P, Evers R, Kool M, Wijnholds J. A family of drug transporters: the multidrug resistance-associated proteins. J Nat! Cancer Ins! 2000: 92: 1 295- 1 302.

1 5. Jedlitschky G, Burchell B, Keppler D. The multidrug resistance protein 5 functions as an ATP-dependent export pump for cyclic nucleotides. J Bioi Chern 2000: 275: 30069-30074.

16 . Kool M, van der Linden M, de Haas M, Scheffer GL, de Vree JM, Smith AJ et a!. MRP3, an organic anion transporter able to transport anti-cancer drugs. Proc Nat! Acad Sci U S A 1 999: 96: 69 14-691 9.

1 7. Kool M, van der Linden M, de Haas M, Baas F, Borst P. Expression of human MRP6, a

129

Chapter 6

homologue of the multidrug resistance protein gene MRP1 , in tissues and cancer cells. Cancer Res 1999: 59: 1 75-1 82.

18. Reid G, Wielinga P, Zelcer N, et al . Characterization of the transport of nucleoside analog drugs by the human multidrug resistance proteins MRP4 and MRP5. Mol Pharmaco/ 2003: 63: 1094- 1 1 03.

19 . Wielinga PR, Reid G, Challa EE, et al. Thiopurine metabolism and identification of the thiopurine metabolites transported by MRP4 and MRP5 overexpressed in human embryonic kidney cells. Mol Pharmaco/ 2002: 62: 1321- 133 1 .

20. Wijnholds J, Mol CA, Van Deemter L, et al. Multidrug-resistance protein 5 i s a multispecific organic anion transporter able to transport nucleotide analogs. Proc Nat/ Acad Sci US A 2000: 97: 7476-7481 .

2 1 . Yoshida M , Suzuki T, Komiya T, et al. Induction of MRP5 and SMRP mRNA by adriamycin exposure and its overexpression in human lung cancer cells resistant to adriamycin. Int J Cancer 200 1 : 94: 432-437.

22. Zeng H, Bain LJ, Belinsky MG, Kruh GD. Expression ofmultidrug resistance protein-3 (multispecific organic anion transporter-D) in human embryonic kidney 293 cells confers resistance to anticancer agents. Cancer Res 1 999: 59: 5964-5967.

23. Kamps WA, Bokkerink JP, Hakvoort-Cammel FG, et al. BFM-oriented treatment for children with acute lymphoblastic leukemia without cranial irradiation and treatment reduction for standard risk patients: results of DCLSG protocol ALL-8 ( 1 99 1 - 1996). Leukemia 2002: 16: 1099- 1 1 1 1 .

24. Daenen S, van Imhoff GW, van den Berg E , et al. Improved outcome of adult acute lymphoblastic leukaemia by moderately intensified chemotherapy which includes a 'pre-induction' course for rapid tumour reduction: preliminary results on 66 patients. Br J Haemato/ 1998: 100: 273-282.

25. Bunting KD. ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 2002: 20: 1 1 -20.

26. Scharenberg CW, Harkey MA, Torok-Storb B. The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 2002: 99: 507-5 12 .

27 . Zhou S, Schuetz JD, Bunting KD, et al . The ABC transporter Bcrp l /ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 200 1 : 7: 1028-1 034.

28. Holleman A, Cheok MH, den Boer ML, et al. Gene-expression patterns in drug-resistant acute lymphoblastic leukemia cells and response to treatment. N Eng/ J Med2004: 351 : 533-542.

130

Chapter 7

Summary, general discussion and future perspectives

Chapter 7

Summary

Acute lymphoblastic leukemia is a disease characterized by an uncontrolled

proliferation and maturation arrest of lymphoid progenitor cells in the bone marrow,

resulting in an excess of malignant cells. The disease has a peak incidence between

the age of 2-5 years, and a low and steady rise from the age of 40. The prognosis

in children with ALL is one of the most favorable of all disseminated cancers: more

than 95% achieve complete remission and disease-free survival rates are 63-83%.

Adults with ALL have a worse prognosis, with a complete remission rate of 75-89%

and disease-free survival rates of 28-39%. From the age-dependent incidence of ALL

and differences in characteristics and prognosis, it could be concluded that ALL in

childhood and adults are distinct diseases. Recent evidence has underscored that the

difference in characteristics and biology of adult versus childhood ALL might be the

result of a different origin. According to the two-hit paradigm of Knudson to develop

cancer, two genetic events are necessary. It has been suggested by Greaves et al.,

that in childhood ALL the first genetic event happens in the more mature lymphoid

committed progenitor cells, whereas in adult ALL the first hit occurs in multi potent

stem cells. In light of this hypothesis, it is assumed that the hematopoietic cells in

which the transformation takes place, share some characteristics with their normal

hematopoietic counterpart. The committed progenitors have a limited selfrenewal

capacity and are more sensitive to apoptosis induced by chemotherapy in contrast to

the multipotent stem cells. Chapter 1 compares various patient characteristics, the

extent of the disease, leukemic cell characteristics and treatment between childhood

and adult ALL. This is discussed in relation to the hypothesis that the maturation

stage of the cells, from which the leukemia arises, is responsible for the differential

behavior of adult and childhood ALL.

Differences in intrinsic cellular differences, e.g. the expression of ABC-transporters,

may be a result of this difference in origin. Chapter 2 describes the role of the ABC

transporters P-gp and MRPl -2 in ALL in children and adults. The functional activity

of P-gp in leukemic blasts is determined in a flow cytometric assay, measuring

the accumulation of rhodamine 123 with or without the P-gp inhibitor PSC833.

For MRP activity, the carboxyfluorescein accumulation with or without the MRP

inhibitor MK-57 1 is determined. We have observed that most T-ALL patients with

high P-gp activity are adults (89% ). P-gp activity in B-lineage ALL is similar

in children and adults. When determining the prognostic impact of P-gp activity,

lower overall and event-free survival rates are found in patients with high P-gp

activity. Therefore, high functional P-gp activity in leukemic blasts may be one of

the factors explaining differences in treatment outcome between children and adults

132


with T-ALL. MRPI -2 activity shows a great variability in ALL cells, with quite

high activities in children in comparison with adults. The activity of MRP has no

prognostic impact on overall survival and event-free survival. However, in this small

study population only few patients suffered from a relapse.

The role of P-gp in ALL cells is described in chapter 2. However, P-gp is also present

in normal tissues, including the liver and kidney, and affects the distribution of many

drugs. In chapter 3 we investigate the role of germline MDR- 1 genetic variants in

vincristine pharmacokinetics and side-effects in childhood ALL. From 52 out of 70

children, who participated in our previous study on vincristine pharmacokinetics,

patient material was available for investigation of the MDR-1 genetic variants.

The SNPs C3435T and G2677T are determined using restriction fragment length

polymorphism polymerase chain reaction. We observed no association between

C3435T or G2677T and vincristine pharmacokinetics. Since it was suggested that

the use of haplotypes is superior to single SNP analysis to predict pharmacokinetics,

we have assigned haplotypes according to Johne et al. Haplotype 1/1 carriers

(3435C/2677G) show a longer elimination half-life than non-carriers. In contrast,

haplotype 1/2-carriers (3435T/2677G) have a shorter elimination half-life than non

carriers. However, the significance is lost after Bonferroni correction for multiple

testing. The haplotypes do not affect the additional pharmacokinetic parameters,

such as clearance and area under the concentration-time curve, suggesting that the

observed effect on elimination half-life is of very limited relevance. Moreover, SNPs

in the MDR-1 gene do not identify patients with an increased risk for vincristine

induced constipation.

Overexpression of the Breast Cancer Resistance Protein, first identified in a

multidrug resistant human breast cancer cell line, was shown to confer resistance

to mitoxantrone, doxorubicin and daunorubicin, which are frequently used in the

treatment of ALL. In chapter 4 an overview of the current knowledge of BCRP is

given and specifically the role of BCRP in acute leukemia. The role of BCRP in

hematopoietic stem cells is demonstrated by a "side population" phenotype in the

hematopoietic stem cell compartment. This side population is based on the efflux of

fluorescent dyes such as rhodamine 1 23 and Hoechst 33342. BCRP knock-out mice

have an almost complete loss of "side population" cells. The knock-out mice also

showed an increased sensitivity to mitoxantrone, implying a physiological role of

BCRP in providing protection from cytotoxic substrates.

The functional activity of BCRP can be determined by measuring the effect of BCRP

inhibitor fumitremorgin C on the accumulation of the BCRP substrate mitoxantrone.

Studies so far showed that B-lineage ALL patient samples have a higher BCRP

activity compared to T-lineage ALL and AML patient samples. The impact of

133

Chapter 7

BCRP functional activity on clinical outcome needs to be studied in larger groups

of patients in order to investigate whether these patients could benefit from the

combination of BCRP inhibitors and chemotherapy.

In chapter 5, we investige BCRP in de novo ALL samples of children and adults.

BCRP expression is measured flow cytometrically with the BXP-34 monoclonal

antibody. BCRP functional activity is determined flow cytometrically by measuring

mitoxantrone accumulation (expressed as Median Fluorescence Intensity (MFI)) in

combination with the BCRP inhibitor fumitremorgin C (FTC). We show a higher

BCRP expression in B-lineage ALL compared to T-lineage ALL. BCRP functional

activity can be detected in B-lineage ALL, whereas in T-lineage ALL this activity

is less pronounced. The influence of FTC on mitoxantrone accumulation correlates

with BCRP protein expression. In summary, we show that BCRP is expressed higher

and functionally more active in B-lineage, than in T-lineage ALL.

The study described in chapter 6 focuses on the role of transporters of the MRP

family in ALL. The mRNA expression of MRPI -6 is determined by real-time

quantitative PCR. Since numerous transporters can be functionally active in ALL,

the connection between the expression of the six MRP genes is also addressed. In

adult ALL a significantly higher expression of MRP I , MRP2 and MRP3 is observed

than in childhood ALL. Interestingly, the expression of the MRPs is comparable

in children and adults who eventually relapsed. High expression of one MRP in

individual patients is correlated with increased expression of multiple MRPs. The

ability to discriminate between patients who will remain in continuous complete

remission and patients who will relapse was determined by receiver operating

characteristic (ROC) curves. For children and adults, the area under the ROC curves

show that MRPI , MRP2, MRP3, MRP5 and MRP6 predict relapse. Using median

values as cutoffs for high versus low expression, high expression of all MRP

genes, except MRP4, is significantly associated with reduced event-free survival in

children and adults. In summary, we show that a subset of patients with high mRNA

expression of multiple MRPs has an unfavorable prognosis independently of age.

Early recognition of a profile with high MRP expression could identify patients with

an increased risk for relapse that could benefit from treatment adaptations based on

this knowledge.

General discussion and future perspectives

This thesis describes the role of ABC-transporters in ALL in children and adults. The

worse survival rates of adult ALL patients compared to children and the differences

in incidence of ALL depending on age has led to the hypothesis of Greaves 1 '2• He

134


suggests that the transformation of normal hematopoietic cells to leukemia occurs

at a different level of lymphoid differentiation between children and adults. In adult

ALL the transformation is considered to occur in the multipotent stem cells, whereas

childhood ALL orginates from the more mature committed lymphoid progenitor

cells. This difference in cells of origin is not limited to a certain age group, but to

certain subgroups of ALL, most likely with specific chromosomal translocations.

Limited studies have investigated these assumptions3-6. The studies so far were

focused on the presence of several chromosomal translocations in different ALL

subpopulations, based on their immunophenotype. For example, the translocation

TEL-AML1 , which is mainly present in childhood ALL, was found in the more

committed progenitor cells (CD34+/CD 1 9+ ), and not in the stem cell compartment

(CD34+/CD1 9-)3:4. In contrast, in Philadelphia-positive ALL (t(9;22)), which is

more frequently diagnosed in adults, the t(9;22) translocation was also present in

B-cells (CD 19+ ), T-cells (CD3+) and myeloid cells (CD 1 3+ ), suggesting a stem cell

origin3;5;6•

To further support the hypothesis of Greaves, leukemic blasts from childhood and

adult patients with t(9;22) and t( 1 2;2 1 ) could be sorted into different stages of

lymphoid commitment. By FISH-technique the presence of the translocation could

be detected in the different subpopulations.

So far, we studied the expression of ABC-transporters in the total ALL population.

However, future investigations have to be focused on the above-described

subpopulations based on maturation stage. This can be especially interesting, because

ALL patients who relapse were found to have a higher MRP expression at diagnosis

in comparison with patients in continuous complete remission. This could possibly

be related to the maturation stage of the originating leukemic cell. After all, the

expression of several transporters, including BCRP and P-gp, is associated with an

immature phenotype in normal hematopoietic stem cells7- 10•

Also micro-array analysis might be of great value to further unravel the differences

between childhood and adult ALL. By analyzing expression patterns in the

subpopulations with different maturation stages, more insights can be obtained in the

similarities and differences between childhood and adult ALL.

Moreover, micro-arrays can also be useful to predict response to chemotherapy.

Recently, Holleman et al. used micro-arrays in childhood ALL, and correlated the

expression profile with in vitro drug sensitivity". In this study genes were identified,

that were differentially expressed in ALL samples that were sensitive or resistant

to to prednisolone, vincristine, asparaginase and daunorubicin. No differences were

found in the expression of MDR-1 and MRPl between sensitive and resistant ALL

samples. The discrepancy with our study might be related to the fact that the

135

Chapter 7

selection of genes was based on the drug sensitivity assay, which is another approach

than clinical endpoints, such as relapse-free survival as used in our study.

When we started this research years ago, less than 1 0 ABC-transporters were known.

To date, 49 ABC-transporter genes have been identified in the human genome (for

an update see nutrigene.4t.comlhumanabc.htm). In ALL most of these transporters

still need to be investigated. With the use of micro-arrays, the mRNA expression can

easily be studied in ALL samples. However, it is essential to investigate whether the

functional activity of these ABC-transporters also correlates with clinical outcome.

A specific a group of patients might be identified, that could benefit from drug

efflux pump inhibitors. So far, few clinical trials have investigated the use of P-gp

inhibitors in AMU2-14• However, the results have been disappointing. Unacceptable

toxicity was observed when the P-gp inhibitor as used in combination with standard

chemotherapy. Moreover, other ABC-transporters could compensate the decreased

functional activity of P-gp.

In conclusion, this thesis focuses on the differences between childhood and adult ALL

in relation to the hypothesis of Greaves, who suggested that transformation of normal

hematopoietic cells to leukemia occurs at a different level of lymphoid differentiation

between children and adults. Moreover, the expression and/or functional activity of

ABC-transporters (P-gp, MRPl -6 and BCRP) in ALL is described. In view of the

strong overlapping activities of the ABC-transporters, and the correlation between

the expression of the different transporters, it could be suggested that a network of

ABC-transporters, instead of a single transporter, is responsible for an unfavorable

outcome in a subset if ALL patients.

136


REFERENCES

1 . Greaves M. Molecular genetics, natural history and the demise of childhood leukaemia. Eur J Cancer 1 999: 35: 1 941 - 1 953.

2 . Greaves MF. Stem cell origins of leukaemia and curability. Br J Cancer 1 993: 67:

4 1 3-423. 3. Castor A, Nilsson L, Astrand-Grundstrom I, Bekassy A, Garwicz S, Johansson B et

al. The hematopoietic stem cell pool is normal in size and function in children with acute lymphoblastic lekuemia and 12 ;21 translocation, but not in Philadelphia positive patients. Blood 2002: 100: 1 52a Abstract 5 7 1 .

4. Hotfilder M, Rottgers S, Rosemann A, Jurgens H, Harbott J, Vormoor J. Immature CD34+. Blood 2002: 100: 640-646.

5 . Kasprzyk A , Harrison CJ, Seeker-Walker LM. Investigation of clonal involvement of myeloid cells in Philadelphia- positive and high hyperdiploid acute lymphoblastic leukemia. Leukemia 1 999: 13: 2000-2006.

6. Schenk TM, Keyhani A, Bottcher S, Kliche KO, Goodacre A, Guo JQ et al. Multilineage involvement of Philadelphia chromosome positive acute lymphoblastic leukemia. Leukemia 1 998: 12: 666-674.

7. Bunting KD. ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 2002: 20: 1 1 -20.

8 . Chaudhary PM, Roninson lB. Expression and activity of P-glycoprotein, a multidrug efflux pump, in human hematopoietic stem cells. Cell 199 1 : 66: 85-94.

9. Scharenberg CW, Harkey MA, Torok-Storb B. The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 2002: 99: 507-5 12 .

1 0. Zhou S, Schuetz JD, Bunting KD, Colapietro AM , Sampath J, Morris JJ et al. The ABC transporter Bcrp 11 ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 200 1 : 7: 1 028-1034.

1 1 . Holleman A, Cheok MH, den Boer ML, Yang W, Veerman AJ, Kazemier KM et al . Gene-expression patterns in drug-resistant acute lymphoblastic leukemia cells and response to treatment. N Eng[ J Med 2004: 351 : 533-542.

12 . Leonard GO, Fojo T, Bates SE. The role of ABC transporters in clinical practice. Oncologist 2003: 8: 41 1 -424.

13 . Baer MR, George SL, Dodge RK, O'Loughlin KL, Minderman H, Caligiuri MA et al. Phase 3 study of the multidrug resistance modulator PSC-833 in previously untreated patients 60 years of age and older with acute myeloid leukemia: Cancer and Leukemia Group 8 Study 9720. Blood 2002: 100: 1 224- 1232.

1 4. Greenberg PL, Lee SJ, Advani R, Tallman MS, Sikic BI, Letendre L et al. Mitoxantrone, etoposide, and cytarabine with or without valspodar in patients with relapsed or refractory acute myeloid leukemia and high-risk myelodysplastic syndrome: a phase III trial (E2995). J Clin Oncol 2004: 22: 1 078- 1 086.

137

Nederlandse samenvatting

Nederlandse samenvatting voor niet-ingewijden

Leukemie is een vorm van bloedkanker. In het bloed bevinden zich verschillende

soorten cellen. De rode bloedcellen zorgen voor zuurstof transport naar de weefsels.

De witte bloedcellen, ook wel leukocyten genoemd, zijn verantwoordelijk voor

onze afweer tegen ziekteverwekkers, zoals bacterien en virussen. De leukocyten

kunnen, atbankelijk van hun functie, worden onderverdeeld in lymfocyten (de T- en

B-cellen), monocyten en granulocyten. Tot slot zijn er de bloedplaatjes, die zorgen

voor de bloedstolling bij wondjes.

De vorming van bloedcellen vindt voomamelijk plaats in het beenmerg, het

binnenste deel van onze botten. In het beenmerg bevinden zich multipotente

stamcellen in specifieke niches waaruit, na een proces van celdeling en uitrijping,

de verschillende bloedcellen ontstaan. Een schematische weergave van de uitrijping

van de bloedcellen is afgebeeld in onderstaand figuur.

Voorloper � B-l.ymfoeyt

LymfoTde o h o StaiTI(;el,. ----1,..�

0 '-.

Voorloper

/ '4 o ----�--�T·o� • Myeloblast Granuloeyt

Multlpotente 0 ----1111-� 0 Stamcel J/1 / . Monoblast Monoeyt Mvelolde

0 Stamce� ----1111-� o 0 Megakaryoblast Bloedplaatje

'\:.52., :�9-o�-- o

Bij leukemie ontstaat een stoomis in het uitrijpingsproces van de bloedcellen. Hierbij

kan onderscheid worden gemaakt tussen acute en chronische leukernie. Bij acute

leukemie treden binnen enkele weken klachten op, als de bloedcellen niet uitrijpen

140


en zich ophopen in het beenmerg. Bij chronische Ieukemie verloopt het proces veel

trager en kunnen de cellen nog redelijk goed uitrijpen. Leukemie kan verder worden

onderverdeeld in Iymfatische en myeloi"de leukemie, afhankelijk van het celtype

dat kwaadaardig ontaard is. Bij acute lymfatische leukemie (ALL) vindt er een

woekering plaats van de lymfoi"de voorlopercellen, die normaal gesproken zouden

uitrijpen tot B- of T-cellen. ALL komt op alle leeftijden voor, maar heeft een piek

incidentie op de leeftijd van twee tot vijf jaar. Elk jaar wordt er in Nederland

bij ongeveer 1 20 kinderen en 70 volwassenen ALL vastgesteld (gebaseerd op de

gegevens van de Nederlandse Kanker Registratie 1 999-2000). De behandeling van

ALL bestaat uit een combinatie van verschillende celdelingremmende medicijnen

(cytostatica) en een preventieve behandeling van het centraal zenuwstelsel, soms

gevolgd door een stamceltransplantatie. Na deze behandeling zijn bij de meeste

patienten met de gebruikelijke technieken geen Ieukemiecellen meer aantoonbaar:

zij hebben dan complete remissie bereikt. Toch blijft er een kans dat de ziekte

terugkomt. Men spreekt dan van een recidief. De prognose van patienten met ALL is

de afgelopen decennia aanzienlijk verbeterd. De vijf-jaars overleving van kinderen

met ALL is 63-83% en bij volwassenen 28-39%.

Het leeftijdsafhankelijke verschil in incidentie en de verschillen in leukemie

karakteristieken (zoals immuunfenotype en chromosoomafwijkingen) en prognose,

doen vermoeden dat de ALL cell en op de kinderen volwassen leeftijd zich gedragen

als verschillende ziekten, mogelijk als gevolg van een verschil in oorsprong. Recente

studies hebben aangetoond, dat het verschil in karakteristieken en biologie van ALL

op de kinderen volwassen leeftijd verklaard zou kunnen worden door een verschil

in origine van de leukemie. Volgens het 'two-hit model' van Knudson, zijn er twee

genetische mutaties nodig om kanker te verkrijgen. Greaves suggereerde dat bij ALL

op de kinderleeftijd de eerste mutatie ontstaat in de rijpere Iymfoi"de voorlopercel,

terwijl bij volwassenen met ALL de eerste mutatie plaatsvindt in de multipotente

stamcel (zie figuur).

In hoofdstuk 1 worden de verschillende klinische kenmerken, uitgebreidheid

van de ziekte, eigenschappen van de leukemiecellen en behandeling vergeleken

tussen ALL op de kinderen op de volwassen Ieeftijd. Dit wordt besproken

vanuit de hypothese dat het uitrijpingsstadium van de cellen, waaruit de leukemie

ontstaat, verantwoordelijk is voor het verschillende gedrag van ALL bij kinderen

en volwassenen. In het Iicht van deze hypothese kan worden aangenomen dat de

voorlopercellen, van waaruit de leukemie ontstaat, dezelfde eigenschappen hebben

als hun normale fysiologische tegenhangers. De rijpere lymfoi"de voorlopercellen

hebben een beperkt vermogen tot celdeling en zijn gevoeliger voor geprogrammeerde

celdood door chemotherapie dan de multipotente stamcellen. Bovenstaande zou dan

141

ook een verklaring kunnen zijn voor het feit dat ALL op de kinderleeftijd gevoeliger

is voor chemotherapie en daardoor beter genezen kan worden.

De verschillen in behandelingsresultaten tussen kinderen en volwassenen met

ALL kunnen dus mogelijk verklaard worden door een variatie in gevoeligheid

van de leukemiecellen voor chemotherapie als gevolg van intrinsieke cellulaire

verschillen. Resistentie tegen verschillende soorten cytostatica wordt ook wel

multidrugresistentie genoemd. Een van de mechanismen van multidrugresistentie

is de aanwezigheid van energie-afhankelijke transporteiwitten, ook wel drug efflux

pompen genoemd.

In hoofdstuk 2 wordt de rol beschreven van de energie-afhankeljike transporteiwitten

P-glycoprote"ine (P-gp) en de multidrugresistentie prote"ines (MRPs) 1 en 2 in

leukemiecellen van kinderen en volwassenen. Om de functionele activiteit van P-gp

te meten, hebben we gebruik gemaakt van een fluorescerende stof Rhodamine 123,

die door P-gp de eel uitgepompt wordt, in combinatie met PSC833, een stof die de

transportfunctie van P-gp blokkeert. Bij een hoge P-gp functionele activiteit wordt

veel Rhodamine1 23 de eel uitgepompt en is er een lage fluorescentie te meten. Als

de P-gp blokker wordt toegevoegd, verwacht je een hogere fluorescentie, omdat

P-gp niet in staat is om aile Rhodamine 123 de eel uit te pompen. De verhouding

tussen de Rhodamine 123 fluorescentie zonder en met de P-gp blokker is een maat

voor de functionele activiteit van P-gp. De functionele activiteit van MRP werd

op een vergelijkbare manier gemeten met carboxyfluoresce"ine, een stof die MRP

kan transporteren, en MK-57 1 als MRP-blokker. De functionele activiteit van P-gp

en MRP werd gemeten in leukemiecellen van 36 kinderen en 35 volwassenen.

We vonden dat de meeste T-cel leukemie patienten met hoge P-gp activiteit

volwassenen waren. Bovendien hadden patienten met hoge P-gp activiteit slechtere

overlevingskansen, dan patienten met lage P-gp activiteit. Verder zagen we een

grote variabiliteit in de functionele activiteit van MRP in leukemiecellen van zowel

kinderen als volwassenen. MRP activiteit had echter geen invloed op de prognose.

De rol van P-gp in leukemiecellen wordt beschreven in hoofdstuk 2. Echter, P-gp is

ook aanwezig in normale weefsels, zoals de lever, de nieren en de hersenen, waar het

de distributie van vele medicijnen be"invloedt. Zo wordt vincristine, een cytostaticum

gebruikt in de behandeling van ALL, door P-gp uitgescheiden in de gal en verlaat

dan het lichaam via de ontlasting. P-gp belnvloedt dus mogelijk de distributie

van vincristine in het lichaam, hetgeen ook kan worden teruggezien in de

vincristine bloedspiegels. In de bevolking komen genetische variaties voor, waarvan

single-nucleotide polymorfismen (SNP's) een voorbeeld zijn. Sommige genetische

varianten leiden ertoe dat een bepaald eiwit, waarvoor het gen codeert, meer wordt

aangemaakt en/of een sterkere werkzaamheid heeft. Zulke genetische varianten zijn

142


ook beschreven voor het MDR- 1 gen. Als iemand een bepaalde genetische variant

van het MDR-1 gen heeft, dat leidt tot een hogere P-gp functionele activiteit, zou

deze variant ook resulteren in lagere bloedspiegels, of andere farmacokinetische

variabelen die de distributie van vincristine in het lichaam weergeven.

In hoofdstuk 3 hebben we onderzocht of er inderdaad een verband bestaat tussen

bepaalde genetische varianten van het MDR- 1 gen en de vincristine farmacokinetiek.

We vonden geen verband tussen de twee SNPs, C3435T en G2677T, en vincristine

farmacokinetiek. Aangezien blootstelling aan vincristine ook kan leiden tot specifieke

bijwerkingen, zoals obstipatie, hebben we tevens onderzocht of er een samenhang

is tussen specifieke MDR- 1 genetische varianten en obstipatie. We vonden geen

verhoogde kans op obstipatie bij specifieke genetische varaties van het MDR-1 gen.

Naast de bekende drug efflux pompen P-gp en MRP is een aantal jaren geleden

terug het borstkanker resistentie eiwit (BCRP) beschreven. Dit transporteiwit is

tevens in staat verschillende cytostatica te transporteren, die gebruikt worden in de

behandeling van ALL. In hoofdstuk 4 wordt een overzicht gegeven van de huidige

kennis van BCRP en de rol in acute leukemie. Van BCRP is bekend dat het een rol

speelt in de fysiologie van stamcellen. Stamcellen kunnen herkend worden doordat

ze fluorescerende stoffen, zoals Rhodamine 1 23 en Hoechst 33342 uitpompen en

dus een lagere fluorescentie hebben. Bij muizen waarbij het gen voor BCRP niet

aanwezig is, zijn de stamcellen dan ook niet te herkennen en hebben ze allemaal een

'hogere' fluorescentie. Ook zijn deze muizen extra gevoelig voor mitoxantrone, een

van de cytostatica die BCRP kan uitpompen.

De functionele activiteit van BCRP kan worden gemeten met een methode

vergelijkbaar met die voor P-gp activiteit. Mitoxantrone wordt gebruikt als

fluorescerende stof in combinatie met een stof die de BCRP-activiteit blokkeert,

namelijk Fumitremorgin C. Tot nu toe hebben studies aangetoond, dat leukemiecellen

van B-ALL patienten een hogere BCRP-activiteit hebben in vergelijking tot T-ALL

en AML patienten. De invloed van BCRP functionele activiteit op de prognose dient

nog bestudeerd te worden in grotere groepen patienten, om vast te stellen of deze

patienten baat zouden kunnen hebben van de toevoeging van BCRP-blokkers aan de

chemotherapie.

In hoofdstuk 5 beschrijven we onze studie waarin BCRP is onderzocht in

leukemiecellen van kinderen en volwassenen met ALL. Het eiwitgehalte van

de leukemiecellen werd gemeten met behulp van het BXP-34 antilichaam. De

functionele activiteit werd gemeten zoals hierboven beschreven met mitoxantrone in

combinatie met Fumitremorgin C. We vonden een hager BCRP eiwitgehalte in B-cel

ALL patienten vergeleken met T-cel ALL patienten. Ook konden we de functionele

activiteit meten in B-ALL en in mindere mate in T-ALL. Er was een correlatie tussen

143

bet eiwitgebalte en de functionele activiteit van BCRP.

Het onderzoek beschreven in hoofdstuk 6 betreft de rol van de farnilie van

multidrugresistentie protei'nes (MRPs) in ALL. Toen met bet onderzoek werd

begonnen waren er slechts enkele drug efflux pompen bekend. Inmiddels zijn er

meer dan 50 verscbillende ABC-transporters gei'dentificeerd. Van de MRP subfamilie

is van MRP 1 -6 bekend dat ze in staat zijn cytostatica te pompen, die gebruikt worden

in de behandeling van ALL. Aangezien we nu weten dat meerdere transporteiwitten

actief kunnen zijn in ALL, hebben we de samenhang tussen de zes verscbillende

MRP genen bestudeerd. We vonden een bogere mRNA expressie van MRP 1 , MRP2

en MRP3 in volwassenen dan in kinderen. Het was opvallend, dat de expressie van

de MRPs vergelijkbaar was in de leukemiecellen van kinderen en volwassenen

die uiteindelijk een recidiefkregen. Een hoge mRNA expressiee van een enkele MRP

ging gepaard met een hogere expressie van meerdere andere MRPs. Patienten die

een recidief ontwikkelden badden bij diagnose een boger gebalte van MRP l , MRP2,

MRP3, MRP5 en MRP6. Bovendien badden patienten met een boger gebalte van

dezelfde MRPs een kortere ziektevrije overleving. Kortom, we vonden dat een

subpopulatie van patienten met een boger gehalte van meerder MRPs een ongunstige

prognose beeft, onafhankelijk van de leeftijd.

Samengevat beschrijft dit proefschrift de verschillen tussen ALL bij kinderen en

volwassenen in relatie tot de hypothese van Greaves, die suggereerde dat het

ontstaan van leukemie uit de lyrnfoi'de voorlopercellen plaatsvindt in een verschillend

uitrijpingsstadium bij kinderen en volwassenen. Bovendien, wordt bet gehalte en

de functionele activiteit van energie-afhankelijke transporteiwitten (P-gp, MRP en

BCRP) beschreven in ALL. Vanwege de sterk overlappende activiteiten van de

verscbillende transport-eiwitten, zou gesuggereerd kunnen worden dat een netwerk

van verschillende transporters, in plaats van een enkele transporter, verantwoordelijk

is voor de ongunstige prognose in een subgroep van ALL-patienten.

144

DANKWOORD

Nu mijn proefschrift bijna gereed is, is het moment aangebroken om iedereen te

bedanken die, op welke manier dan ook, hieraan heeft bijgedragen. Graag wil ik

enkele mensen in het bijzonder noemen.

Het feit dat patienten met leukemie, kinderen en volwassenen, bereid zijn bij te

dragen aan wetenschappelijk onderzoek is heel waardevol en daarvoor ben ik hen

zeer erkentelijk. Ik hoop dan ook dat mijn onderzoek een beetje bijdraagt aan een

betere toekomst voor hun allemaal. Verder wil ik de SKJON, voorheen SNWLK,

bedanken voor het patientenmateriaal en de patientengegevens die ik van hen

ontvangen heb.

De Stichting Kinderoncologie Groningen bedank ik voor het financieel mogelijk

maken van mijn onderzoeksperiode, en aile bijbehorende zaken zoals posters en

congres-bezoeken.

Mijn (co-)promotoren, die aile vier op eigen wijze veel tijd en energie hebben

gestoken in mijn promotie wil ik enorm bedanken.

Allereerst mijn eerste promotor Prof. dr. W.A. Kamps, beste Willem, ik leerde je

tijdens mijn geneeskunde-doctoraal kennen, toen ik voor mijn wetenschappelijke

stage op het gebied van de kinderoncologie naar het National Cancer Institute,

Bethesda, ging. Daarna adviseerde je me en hielp me te bereiken wat ik graag wilde:

een AGIKO-plek in de kindergeneeskunde. Gedurende mijn onderzoeks-periode kon

ik in moeilijke tijden altijd op je steun rekenen. Bedankt voor het vertrouwen dat je

in me had en voor je sturende kracht achter mijn onderzoek.

Prof. dr. E.G.E. de Vries, beste Liesbeth, bedankt voor je kritische inhoudelijke en

taalkundige blik en de snelheid waarmee je altijd mijn artikelen nakeek. Het feit dat

jouw deur, ondanks je voile agenda, altijd voor me openstond en ik laagdrempelig

met je kon overleggen heb ik enorm gewaardeerd.

Prof. dr. E. Vellenga, beste Edo, het project was er voornamelijk dankzij jou.

Je gedrevenheid en jouw brede wetenschappelijke visie, waarmee je vanuit

verschillende invalshoeken onderzoeksdata kan interpreteren, bewonder ik zeer.

Dr. E.S.J.M. de Bont, beste Eveline, mijn co-promotor en directe begeleider, in het

begin was het wat zoeken, maar vanaf het moment dat je intensiever betrokken

raakte, hielp je me geweldig met plannen, structureren en het oog hebben voor

de grote lijnen. Je probeerde altijd het beste in mij naar hoven te halen. Jouw

enthousiasme, je eeuwige optimisme, ook wanneer mijn experimenten weer eens

mislukten, en de wijze waarop je kliniek met wetenschap combineert, zijn een ware

inspiratie voor mij .

Dr. Dorina van der Kolk, mijn paranimf: als analist, zelf bezig met een promotie-

145

onderzoek, kwam je op mijn project terecht. Je leerde mij nieuwe lab-technieken,

hielp me wanneer experimenten niet liepen, en dacht mee over de interpretatie van

mijn data. Naast dat alles kon ik ook altijd bij je terecht om stoom af te blazen of

gewoon voor de gezelligheid. Zander jou was het me nooit gelukt en ik ben dan ook

heel dankbaar voor alles wat je voor me gedaan hebt.

De leden van de beoordelingscommissie, Prof. Dr. P.M. Kluin, Prof. Dr. R. Pieters

en Prof. Dr. R.J. Scheper wil ik bedanken voor hun beoordeling van het manuscript.

Dr. S.M.G. Daenen, beste Simon, als hematoloog intensief betrokken bij de

behandeling van volwassenen met ALL, heel hartelijk dank voor de vlotte aanlevering

van patienten-gegevens en de bereidheid om kritisch naar mijn manuscripten te

kijken.

Zander statistiek ben je nergens, en ik ben blij dat Dr. H. Marike Boezen me hierbij

geholpen heeft. Door te achterhalen wat nou mijn concrete vraagstelling was, en niet

zo maar verschillende testen op mijn data los te Iaten, heb ik veel van je geleerd.

Dan wil ik nog de FACS-operators Geert Mesander en Henk Moes bedanken voor

hun geduld en hulp bij de vele functionele assays op de flow cytometer. Klaas-Nico

Faber en Hans Blokzijl wil ik bedanken voor hun expertise en hulp bij het gebruik

van het Taq-Man real-time PCR-apparaat.

Mijn artikel over vincristine-farmacokinetiek was er nooit gekomen zonder de hulp

van Dr. Ellis Groninger, Dr. Siebold de Graaf en Prof. Dr. D. Uges. Bedankt voor

het bijbrengen van de ingewikkelde beginselen van farmacokinetiek en de hulp bij

de interpretatie en analyse van de vincristine-data. Ook wil ik graag Dr. Ido Kema en

Astrid Brugman bedanken voor de analyse van de MDR-1 polymorfismen.

Het grootste gedeelte van mijn onderzoeksperiode heb ik met veel plezier

doorgebracht op het lab. Pipetteren is een stuk leuker met gezellige mensen om

je been, en wat dat betreft heb ik enorm geboft. De mensen van het lab van de

kinderoncologie: Susan, Tiny, Frank, Ellen, Ellis, Claudi, Gineke, Rony en Esther;

en van het hematologie research lab de oud-labgenoten (Arjen-Kars, Dolf, Gerlinde,

Henny, Kim en Mariet) en Bart-Jan, Bettie, Hein, Irene, Kyrjon en Lyndsay, bedankt

voor jullie gezelligheid. En natuurlijk Gwenny: met jou was het op het lab, tijdens

de ASH-congressen, en later ook daarbuiten, nog leuker!

Behalve pipetteren, heb ik veel tijd doorgebracht achter mijn computer op de

'kindergeneeskunde-AIO' kamer. Margriethe, Shalini, Lisethe, Terry, Danielle, Greet

en Elisabeth, dank voor jullie luisterend oor als ik weer eens mijn ei kwijt moest en

voor jullie gezelligheid. En Saskia, mijn andere paranimf, begonnen als kamergenoot,

maar nu zo veel meer, als steun en toeverlaat is het voor mij vanzelfsprekend dat je

vandaag als paranimfnaast me staat !

146

De secretaresses Roos, Wilma en later Maaike (kinderoncologie) en Sylvia, Gretha

en Bianca (interne geneeskunde) dank ik voor hun secretariele ondersteuning.

Tot slot, mijn lieve ouders, Alain en Veronique, bedankt voor jullie vertrouwen en

onvoorwaardelijke steun. En als laatste mijn 'andere helft', lieve Nicole, hopelijk

kun je er vandaag wei bij zijn, met of zonder dikke buik; want op een dag als

vandaag ben je voor mij echt onrnisbaar!

Iedereen heel erg bedankt,

Sabine

147

university of groningen the role of abc-transporters in ... · university of groningen the role of...

Documents