British Journal ofHaernatology, 1995, 91, 878-884

Combined cytogenetic, FISH and molecular analysis in acute promyelocytic leukaemia at diagnosis and in complete remission

MARCO MANCINI, MAURO NANNI, MICHELE CBDRONE, DANIELA DIVERIO, GIUSEPPE AVVISATI, ROBERTA RICCIONI, MARIA ROSARIA DE CUIA, SUSANNA FENU AND GIULIANA ALIMENA Haematology, Department of Human Bioputhology, University of Rome ‘La Sapienza’, Rome, Italy

Received 23 May 1995; accepted for publication 27 July 1995

Summary. This study reports the results of a simultaneous application of cytogenetic fluorescence in situ hybridization (FISH) and molecular analysis (RT-PCR) in 28 APL cases (23 M3 and five M3v; 26 studied at diagnosis and two at relapse). FISH on metaphases identified the t(15;17) in all cases who were positive for the PML/RAR a transcript by RT-PCR. Conventional cytogenetics revealed the t( 15; 17) in only 68% of cases. However, it enabled the detection of additional chromosome changes in five cases, three of whom were M3v. 11 patients were also investigated during complete remission (CR) by both FISH and RT-PCR, in order to evaluate residual disease; the duration of CR at the time of analysis ranged between 1 and 16 months, with three patients being studied twice. Comparison of RT-PCR and FISH results showed a very good correlation. In fact, of the 10 samples which were RT-PCR positive for residual

disease, all were also recognized by interphase FISH, and eight were positive by metaphase FISH. Of the three samples negative at RT-PCR, all were also negative at the interphase FISH. The results of this study indicate that: (a) the t(15;17) is present in all cases positive for the PML/RAR Q

rearrangement, thus in virtually all true APLs; (b) standard cytogenetics, capable of unravelling the t( 1 5; 17) in only 68% of cases, enables recognition of additional chromosome changes of potential clinical and prognostic significance; (c) FISH on interphase nuclei is a reliable tool for the monitoring of residual disease, with a sensitivity greater than that of FISH on metaphase cells and superimposable to that of RT- PCR.

Keywords: metaphase and interphase FISH, APL, cytogenetics, RT-PCR, residual disease.

Acute promyelocytic leukaemia (APL) is characterized by well-documented biological and clinical features. These include: (1) the rearrangement between the promyelocytic (PML) and retinoic acid receptor o (RAR a ) genes, located on chromosomes 15 and 17, respectively; this leads to the PML/ RAR a fusion gene which encodes for the PML/RAR Q

protein (Grignani et al, 1994), and (2) the high sensitivity of the neoplastic population to differentiation therapy with all- trans retinoic acid (RA) (Huang et al, 1988; Degos et al, 1990; Warrel et al, 1991; Miller et al, 1992; Grignani et al, 1994). Furthermore, the presence of a specific genetic marker enables monitoring of the leukaemic clone during treatment and this has been shown to bear prognostic implications. A persistent signal following induction treat-

Correspondence: Dr Marco Mancini, Haematology, Department of Human Biopathology, University of Rome ’La Sapienza’. Via Benevento 6,00161 Rome, Italy.

8 78

ment has, in fact, been indicated as a predictor of impending relapse, whereas its complete disappearance frequently correlates with long-term survival and, possibly, cure (Lo COCO et al, 1992b; Huang et al, 1993; Miller et al, 1993).

Southern blot and polymerase chain reaction (PCR) analyses, which enable the identification of a PML/RAR a rearrangement in 100% of cases, enable a reliable molecular diagnosis of APL and offer a potential tool for disease monitoring(Biondietal,1991.1992; Chenet al, 1991; Tong et al, 1992; Diverio et al, 1993). However, Southern blotting is a time-consuming and poorly sensitive technique, whereas the use of PCR has so far been restricted to experienced laboratories. Cytogenetic studies recognize the t( 15;17) in only 70-80% of APL cases (Berger et al, 1983; Larson et al, 1984; De Braekeleer & Lin, 1986). The discrepancy with the molecular results may be due to karyotypic analyses carried out on cells which do not belong to the neoplastic clone, or to sub-microscopic rearrangements not detectable by

0 1995 Blackwell Science Ltd

Combined Cgtogenetic, FlSH and Molecular Analysis in APL 879 conventional cytogenetics (Lo COCO et al, 1992a). This technique is, nonetheless, useful for the detection of chromosome changes in addition to the t(15:17), or of variant translocations such as t(11;17) (Licht et ul. 1995), or of not transcribing RAR cr rearrangements.

Fluorescence in situ hybridization (FISH) has greatly increased the potential of standard cytogenetics towards identifying chromosomal or gene rearrangements both in metaphase or interphase cells. The feasibility and usefulness of this technique in APL has been previously recognized (Speicher et al, 1993).

With the aim of better assessing the sensitivity and role of FISH in detecting the t( 15: 17) and/or the PML/RAR a rearrangement as compared to both classic cytogenetics and PCR, we have applied these three techniques simultaneously to 28 cases, classified as APL according to morphological criteria and molecular grounds. Samples were collected at diagnosis and/or at relapse, and in a proportion of cases also after the achievement of complete remission (CR), in order to evaluate minimal residual disease (IvlRD).

MATERIALS AND METHODS

Patients. Bone marrow (BM) specimens were collected from 28 patients (16 females, 12 males; median age 34 years, range 2-77) with morpho-cytochemical diagnosis of FAB M3 (23 patients) or M3v (five patients), observed between June 1991 and July 1994 at the Haematology Department of the University ‘La Sapienza’ of Rome. Of the 28 patients, 26 were studied at diagnosis, and two at relapse. 11 patients were also studied in CR after induction therapy: three of them (nos. 6, 8 and 10) were analysed twice at different time intervals.

Cytochemistry and immunology. Morphological and cyto- chemical analyses were performed by Wright-Giemsa, myeloperoxidase, chloroacetate esterase, alpha naphtyl- acetate and butyrate esterase in all patients.

The immunophenotypic characterization was performed on Ficoll-Hypaque-isolated mononuclear cells using a panel of monoclonal antibodies, as previously reported (De Rossi et al, 1990).

Cytogenetics. GTG-banded chromosomes of BM cells were studied according to standard methods and classified according to the ISCN (1 99 1).

In situ hybridization. (A) A library of chromosome 17 labelled with biotin (Oncor, Gaithesburg, Md.) was used as probe for chromosomal in situ suppression (CISS) hybridiza- tion on metaphase cells. Hybridization and detection of chromosome 1 7 sequences were carried out according to the protocol of Wiegant et al(l991). Samples were visualized on a photomicroscope (Zeiss Axiophot) equipped with a double- band pass filter.

(B) A mixture of digoxigenin-labelled cosmid DNA probes specific for the 17q2 1 chromosome breakpoint and a biotin- labelled cosmid DNA probe specific for the 15q22 chromo- some breakpoint (Oncor, Gaithesburg, Md.) were used on interphase cells. Dual-colour hybridization was performed as previously described by Arnoldus et a l ( l990) and according to the manufacturer’s instructions. Samples were evaluated

with a Zeiss microscope Axiophot equipped with a triple- band pass filter.

Hybridization signals were evaluated independently by three microscopists in 1000 nuclei.

Molecular biology. (A) RNA extraction. Mononuclear cells were separated by Ficoll-Hypaque centrifugation and total RNA extracted according to the method of Chomczynski & Sacchi (1987). The quality of RNA was assessed in an ethidium bromide-stained 1.5% agarose gel containing 2.2 mol/l formaldehyde.

(B) RT-PCR amplification. In vitro reverse transcription of 1 pg of APL total RNA to cDNA was performed using a commercial kit (Gene Amp RNA kit, Perkin Elmer-Cetus). The RT-PCR conditions and the oligonucleotides used to amplify the PML,/RAR a cDNA have been reported elsewhere (Diverio et al, 1993). Amplification of the normal RAR a cDNA was performed to further verify RNA integrity and the efficiency of the reverse transcription reaction. To assess the sensitivity of the RT-PCR assay, total RNA isolated from one APL patient was serially diluted by mixing with the myeloid cell line GF-D8 RNA, as previously described (Diverio et al, 1993).

RESULTS

Clinical data The main clinical and haematological data of the 28 cases are summarized in Table I. 23 patients were diagnosed morphologically as having the classic M3 form and five were classified as M3v. CR was achieved in 24/28 APL patients after induction therapy.

Cytogenetics The cytogenetic results are reported in Table 11. Evaluable data were obtained in 20126 patients studied at diagnosis and in 21 2 patients observed at relapse. A mean number of 15 metaphases (range 12-30) could be analyzed for each sample. Of the 26 patients studied at diagnosis, 18 (69%)) revealed the standard t(15;17)(q22;q21). In 14 cases (13 M3 and one M3v). this was present as an isolated abnormality, whereas in four, two of which with M3v (1 3 and 25) and one (24) with a percentage of microgranular blasts > 25% it was associated to additional chromosome changes: these consisted of a t(6;17)(q23;q21), a del(l7)(p12) and a + 8 (two cases), respectively. Two cases had an apparently normal karyotype, whereas six (five M3 and one M3v) were not evaluable because of the poor quality of the metaphases.

Of the two cases analysed at relapse, one with a M3v( 7) showed an add( 7)(pl5) chromosome associated to the t(15;17) and the other displayed an apparently normal karyotype.

In situ hybridization on metaphase cells An average of 29 metaphases (range 15-98) could be analysed for each case at first examination (Table II). In all 24 cases examined at diagnosis and the two cases analysed at relapse, CISS-hybridization technique clearly demon- strated the t(15:17), and also in poor quality metaphases not analysable by standard cytogenetics (Table 11). In all 26 cases the t( 15: 17) coexisted with normal cells at a rate ranging between 30% and 93% of examined cells.

0 1995 Blackwell Science Ltd, British Journal of Haematology 91: 878-884

880 Marc0 Mancini et a1 Table I. Biological and clinical data of 28 patients with a morphological diagnosis of APL.

BM-blasts

Immunology

Case Agelsex Morphology status DIC x109/1 % DR CD34 CD33 CD13 CD9 0 2 Therapy Outcome Disease PB-WBC

- 1 2/F 2 14/F 3 26/F 4 30/M 5 32/F 6 28/M 7 2O/M 8 29/M 9 38/F

10 53/F 11 63/M 12 19/F 13 56/M 14 64/F 15 77/F 16 51/F 17 50/M 18 lS/F 19 28/M 20 51/F 21 14/M 22 65/M 23 36/F 24 64/P 25 9/F 26 37/M 27 40/M 28 31/F

M3 Dia M3 Dia M3 Re1 M3v Dia M3 Dia M3 Dia M3v Re1 M3 Dia M3 Dia M3 Dia M3 Dia M3 Dia M3v Dia M3 Dia M3 Dia M3 Dia M3v Dia M3 Dia M3 Dia M3 Dia M3 Dia M3 Dia M3 Dia M3 Dia M3v Dia M3 Dia M3 Dia M3 Dia

Yes - Yes 2.7 NO - Yes 238 Yes 1.6 Yes 15.7 Yes 19.7 Yes 3.2 Yes 0.4 Yes 1.3 Yes 1.7 Yes 2.0 Yes 2.5 Yes 4.0 Yes 1.2 Yes 1.5 Yes 18.2 Yes 1.4 Yes 0.6 Yes 0.8 Yes 2.3 NO 1.7 Yes 2.3 Yes 18-9 Yes 3.1 Yes 1.1 Yes 7.9 Yes 5.5

85 0 0 90 2 2

98 7 25 60 - - 90 2 2 78 - - 80 0 0 92 1 1 81 1 0 93 3 0 80 0 0 93 8 16 79 3 1 83 0 0 80 0 0 90 1 0 85 18 1 83 0 0 79 2 0 9s 4 9 74 3 4 9 s 1s 19 94 15 30 80 2 1 83 8 4 75 4 11 76 5 0

- - _ 65 78

84

79

81 78 80 85 69 72 61 91 90 90 90 96 53 80 93 88 72 76 58 60 34

-

-

-

77 70 ND 77 85 72

79 96 ND

87 ND 2

26 ND 3 35 ND 2 90 85 1 75 86 0 49 80 2 60 90 2 77 93 2 66 93 ND 72 87 2 4 s 88 1 89 90 ND 12 95 ND 46 ND ND 60 67 ND 57 91 1 83 90 21 61 91 28 69 ND 3 20 ND 1 57 5 8 78 ND 5

- - -

- - -

- - -

Idarubicin + cytarabine CR Idarubicin + cytarabine CR ATRA CR Died before treatment Dead Idarubicin CR Idarubicin+ cytarabine CR ATRA CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin Dead ATRA Dead ATRA + idarubicin CR ATRA + idarubicin Dead Idarubicin CR Idarubicin + cytarabine CR Idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR ATRA + idarubicin CR

Dia. diagnosis: Rel, relapse: ND, not done: ATRA. all-trans retinoic acid: CR, complete remission.

In 5/8 cases in CR who were studied only once (Table 111: nos. 1 , 2 , 5, 7,9, 16,20 and 22) CISS hybridization analysis revealed the t(15;17) in 1-12% of the metaphases (an average of 100 metaphases was analysed for each case, range 19-283), whereas three cases were negative (Table III; nos. 1, 7 and 22). The other three cases (Table III; nos. 6, 8 and lo), who were studied twice during CR. were positive at the first determination and negative at the second.

In situ hybridization on interphase cells The 15q22 labelled breakpoint resulted in a green signal and the 1 7q2 1 in a red signal. Normal cells showed four separate signals (two green and two red), whereas all translocations involving the fusion of PML and RAR Q regions resulted in a close red-green signal, plus two separated red and green signals. Only overlapping or touching signals were classified as positive for PML/RAR a rearrangement.

(A) Controls. The mean percentage of interphase cells with an apparent PML/RAR a fusion signal, evaluated in 11 selected specimens (normal, BM donors, and patients in CR after a BM transplantation for chronic myelogenous

leukaemia) was 3.3% (range 1*8-3.9%) with a standard deviation of 0*4% These apparent fusion signals were interpreted as due to coincidental overlapping of the PML and RAR a domains. A PML/RAR a fusion in the test specimens was therefore diagnosed if > 4.5% (mean f 3 SD of the controls) (Fig 1).

(B) APL patients. 11 cases were analysed in CR following induction treatment. Cases 6, 8 and 10 were studied twice. In 11/14 specimens fusion signals were detected in a rate greater than the cut-off level of 4.5% (range 5*7-10-7%) established in the controls, thus indicating persistence of residual disease. Three cases (nos. 7 ,8 [2nd control] and 22) showed a percentage of fusion signals in the normal range (Fig 1, Table III).

Molecular biology The 28 cases were studied by RT-PCR for the presence of PML/RAR a fusion gene transcripts (26 at diagnosis and two at first relapse). A positive RT-PCR was observed in all 28: 15 cases had a breakpoint in bcr 1 or 2, and 13 in bcr 3 (Table 11).

0 1995 Blackwell Science Ltd. British Journal of Haemutobgy 91: 878-884

Combined Cytogenetic, FISH and Molecular Analysis in APL 88 1 Table 11. Cytogenetic, FISH and RT-PCR data of 28 patients.

Disease CISS RT-PCR Patient status Morphology Karyotype t(15;17) (type of breakpoint)

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Dia Dia Re1 Dia Dia Dia Re1 Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia Dia

M3 M3 M3 M3v M3 M3 M3v M3 M3 M3 M3 M3 M3v M3 M3 M3 M3v M3 M3 M3 M3 M3 M3 M3 M3v M3 M3 M3

NE ND 46,XX ND 46,XX pos

POS 46,XX,t(l5:17) NE POS

POS 46,XY,t(l5:17) 46,XX,add(7),t(15;17) POS

POS 46,XY,t(15;17) 46,XX POS

POS 46,XX.t(15;17) Pos 46,XY,t(15:17) POS 46,XX,t(l5;17)

46,XY,t(6;17),t(15:17) Pos POS 46,XY,t(l5:17)

NE POS NE POS NE Pos

Pos 46,XX.t(15;17) POS 46,XY,t(l5:17) POS 46,XX,t(15;17) POS 46,XY,t(15;17) POS 46,XY,t(15:17) POS 46,XX,t( 15;17)

46,XX,del(l7p),t(15;17) pos 47,XX,+8~(15;17) pos 47,XY,+8,t(l5;17) pos NE POS

POS 46 ,XX,t( 1 5; 1 7)

pos (bcr 1) pos (bcr 1-2) pos (b 3) pos (bcr 3) pos (bcr 1-2) pos (bcr 1) pos (bcr 3) pos (bcr 1) ~ o s (bcr 2) pos (bcr 1) Pos ( b a 1) pos (bcr 1) pos (bcr 3) pos (bcr 3) pos (bcr 1) pos (bcr 1) pos (bcr 1)

pos (bcr 1) pos (bcr 1) pos (bcr 3) pos (bcr 3) pos (bcr 3) pos (h 3) P (bcr 3) pos (bcr 3) pos (bcr 3) pos (bcr 1)

~ o s ( b a 3)

Dia. diagnosis: Rel, relapse: NE. not evaluable: ND, not done.

Table 111. Results of FISH and RT-PCR in the BM cells of patients studied at CR.

FISH RT-PCR Clinical status at

Case FISH and PCR date CISS t(15;17) PML/RAR a PML/RAR 01

1 CRx 1 month 2 CR x 2 months 5 CRx 1 5 months 6 CRx 1 month

CR x 1 6 months 7 CR x 1 month 8 CR x 1 month

CR x 15 months 9 CRxl month POS POS POS

10 CRx 1.5 months POS POS POS CR x 2 months neg POS Pos

16 CRx 1 month POS POS PoS

20 CR x 2 months POS POS POS 22 CRx 12 months neg neg neg

CR, complete remission: ND, not done.

0 1995 Blackwell Science Ltd. British Journal of HuernatoZogy 91: 878-884

Marc0 Mancini et a1 882

12 -

10 -

8 -

6 -

4 -

2 -

% O f NUCLEI

Fig 1. Percentage of nuclei with PML/RAR LY fusion signals in 11 controls and 11 API, patients in CR. The cut-off level was obtained from the statistical evaluation of the controls: the mean f3 SD of false-positive nuclei is

PML/RAR 01 fusion. Number in oarentheses is 11 1 11 the duration of CR (in months) i t FISH date.

0 - 1

Controls APL patienta *Second control.

Eleven patients were investigated also following the achievement of CR (Table 111): two of them (nos. 8 and 10) were studied twice. Nine cases studied within 1 or 2 months after CR showed positive RT-PCR. Three cases (nos. 7 ,8 [2nd control] and 22) investigated at 1, 1 5 and 12 months from CR, respectively, displayed negative RT-PCR. Case 5 had persistent positive RT-PCR 15 months after CR (Table III).

DISCUSSION

The risk of a potentially fatal coagulopathy, as well as the unique clinical response to RA in APL, requires an accurate and prompt diagnosis of this disease (Gralnik & Sultan, 1975; Rodeghiero et al, 1990; Tallman & Kwaan, 1992).

Though the identification of the hypergranular APL form is usually easy, a misclassification of some AML cases with signs of maturation as APL may occur: also in cases of M3v the distinction from acute myelomonocytic leukaemia may represent a diagnostic problem (Castoldi et al, 1994). However, in both the classic and variant forms of APL the PML/RAR a! hybrid gene is a diagnostic hallmark. It is therefore mandatory that this is searched for when a diagnosis of APL is suspected (Grignani et al. 1994). The use of molecular biology (Southern, PCR) enables the identification of RAR n rearrangements or PMT,/RAR a transcripts in 100% of cases, whereas cytogenetics recog- nizes the t(15;17) in only 70430% of APL cases (De Braekeleer & Lin, 1986; Warrel et d, 1991; Lo COCO et al, 1992a; Grignani et al, 1994).

In the present study cytogenetic analysis of 28 cases with a diagnosis of APL confirmed at molecular level identified the t( 15;17) in 19 of them (68%) (Table 11). This finding is in line with previously reported cytogenetic results for APL (Berger et ul, 1983; De Braekeleer & Lin, 1986). A low number of evaluable metaphases (a mean of 15 per case) and a poor

chromosome morphology are the main limitations of the cytogenetic investigation.

Recently the ability of FISH to identify the t( 15;17) in cases of APL with an apparently normal karyotype has been reported, thus suggesting that this technique represents an important tool for the recognition of the aberration (Speicher et al, 1993). We have applied painting of chromosome 17 in 26/28 cases, 24 at diagnosis and two at first relapse, with the aim of: (a) confirming the power of FISH in detecting the t(15:17), and (b) evaluating the real percentage of t(15:17)- positive cells in a large number of PCR-positive patients. Using this approach, clear-cut results were obtained even in cases with chromosomes of poor morphology; moreover, a higher number of metaphases (a median of 29 per case) was analysable over a short period of time, compared to standard cytogenetics. The t(l5;17) was detected in all cases positive for the PML/RAR N fusion transcript by PCR, including those not evaluable or negative by conventional G-banding method (Table 11). These results further support the suggestion that virtually all true APLs carry the t(15;17) and indicate that CISS hybridization is an easy method to unravel the anomaly, with a sensitivity which appears superimposable to that of molecular analyses.

Cytogenetics is, however, of fundamental importance in detecting additional karyotypic changes which may occur in 10-30% of cases. Berger et aZ(1991), in a series of 73 APLs examined at diagnosis, observed additional chromosomal changes in 20 cases with no significant differences between the M3 and M3v forms. In our series of 28 patients, 22 of whom were cytogenetically evaluable, we found chromo- some anomalies in addition to the t(15;17) in five cases (Table 11: nos. 7, 13, 24, 25 and 26). three of which were M3v and one M3 with a high number of microgranular blasts. The anomalies consisted of a t(6;17)(q23;q21). an add( 7)(pl 5 ) , a del (1 7)(p12) (one case each) and trisomy 8

Q 1995 Blackwell Science Ltd, British Journal of Huemutology 91: 878-884

Combined Cytogenetic, FISH and Molecular Analysis in APL 883 (two cases). The APL variant form, although bearing the PML/RAR a protein and responding to RA treatment, exhibits clinical, cytological and, possibly, molecular dis- tinctive features, suggesting that it may represent an evolutive form of APL. When compared to classic APL, the M3v shows the presence of hyperleucocytosis at onset, a more severe coagulopathy, the expression of the CD2 antigen, and a higher incidence of BCR3 breakpoint in the PML gene. This type of breakpoint has been recently related to poor prognosis (Huang et al, 1993; Grignani eta] , 1994; Vahdat et al, 1994). The presence of karyotypic anomalies in addition to the t( 15; 17) in the M3v may therefore be related to an increased chromosome instability due to tumour progression, in a fashion similar to that observed when chronic myelogenous leukaemia enters acute phase (Heim & Mitelman, 1987). The evaluation of the prognostic sig- nificance of additional karyotypic changes, as well as their nonrandom association with M3v, requires further inves- tigations.

Monitoring of residual disease through the presence of the genetic marker is another very important issue in APL the incidence of CR is, in fact, very high, and survival after therapy is probably long-term in this disease (Grignani et al, 1994). In a recent study Zhao et aZ(1995) applied the CISS to detect residual leukaemic cells in 10 APL patients in CR; although less sensitive than PCR, owing to its confinement to dividing cells only, FISH appeared to be more efficient than conventional cytogenetics. In our study, 11 APL patients in CR were investigated with FISH both on interphase and on metaphase nuclei, by using probes for the PML/RAR a: gene or specific for the entire chromosome 17. The results were compared with those based on RT-PCR analysis. Of these 11 patients, two were re-analysed both by FISH and PCR, and one only by FISH (Table III). FISH applied to interphase cells was positive in all cases which were positive for the PML/ RAR a transcript by RT-PCR. The cases negative at the molecular analysis showed a percentage of rearranged nuclei lower than the cut-off level. CISS hybridization, applied to metaphases, revealed the presence of the t( 15;17) in only eight of the 10 RT-PCR-positive samples, though it enabled the evaluation of an extremely large number of metaphases per patient (median 100, range 19-283).

Based on the results of this study, it emerges that FISH on interphase nuclei using probes exploring the PML/RAR Q

junction represents an effective tool for detecting MRD in APL patients. Its high sensitivity appears greater than that of CISS hybridization and superimposable to that of RT-PCR. It represents an easy procedure, which enables an accurate quantitative assessment of the PML/RAR cy rearranged cell fraction. The combined use of the FISH and PCR techniques in prospective studies may be of relevance towards a better definition of the reproducibility of these two methods in the evaluation of MRD and of its significance with respect to the outcome of the disease.

ACKNOWLEDGMENTS

We thank Professor Robert Foa for critically reading the manuscript. M. Cedrone is an AIRC fellow.

REFERENCES

Arnoldus, E.P.J.. Wiegant, J., Noordmeer. LA.. Wessels. J.W., Beverstock. G.C.. Grosveld, J.C., van der Ploeg, M. & Raap. A.K. (1990) Detection of Philadelphia chromosome in interphase nuclei. Cytogenetics and Cell Genetics, 54. 108-111.

Berger, R., Bernheim, A,, Daniel, M.T., Valensi, F. & Flandrin. 1. (1983) Cytological types of mitosis and chromosomal abnormal- ities in acute leukemia. Leukemia Research, 7, 221-236.

Berger, R.. Le Coniat, M., Derre, J.. Vecchione, D. & Jonveaux. P. (1991) Cytogenetic studies in acute promyelocytic leukemia: a survey of secondary chromosomal abnormalities. Genes, Chromo- somes and Cancer, 3, 332-337.

Biondi, A.. Rambaldi, A., Alcalay. M.. Pandolfi, P.P., Lo Coco. F.. Diverio. D., Rossi, V., Mencarelli. A., Longo, L., Zangrilli, D.. Masera. G.. Barbui, T., Mandelli, F.. Grignani, F. & Pelicci. P.G. (1991) RAR-rw gene rearrangements as a genetic marker for diagnosis and monitoring in acute promyelocytic leukemia. Blood.

Biondi, A.. Rambaldi, A., Pandolfi, P.P., Rossi, V., Giudici. G.. Alcalay. M., Lo Coco, F., Diverio. D., Pogliani, E.M., Lanxi, E.M., Mandelli, F., Masera, G., Barbui, T. & Pelicci, P.G. (1992) Molecular monitoring of the My1 retinoic/acid receptor-cu fusion gene in acute promyelocytic leukemia by polymerase chain reaction. Blood. 80, 492-497.

Castoldi, G.L., Liso. V.. Specchia, G . & Tomasi. P. (1994) Acute promyelocytic leukemia: morphological aspects. Leukemia. 8.

Chen. Z., Chen, S.J., Tong, J.H., Dong, S., Huang, W., Chen. Y.. Xiang. W.M., Zhang, L., Song, X., Qian, X.S., Wang, Z.Y., Chen. Z., Larsen. C.J. & Berger, R. (1991) The retinoic acid o receptor gene is frequently disrupted in its 5’ part in Chinese patients with acute promyelocytic leukemia. Leukemia, 5, 288-292.

Chomczynsky, P. & Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyauate-phenol-chloroform extraction. Annals of Biochemistry, 162.156-160.

De Braekeleer. M. & Lin, C.C. (1986) The occurrence of the 15:17 translocation in acute promyelocytic leukemia. Cancer Gmetics and Cytogenetics, 19, 311-319.

Degos, L., Chomienne, C.. Daniel, M.T.. Berger, R., Dombret. H.. Fenaux. P. & Castaigne, S. (1990) Treatment of first relapse in acute promyelocytic leukaemia with all-trans retinoic acid. Luncct,

De Rossi, G., Avvisati, G., Coluzzi. S., Fernu. S., Lo Coco, P.. Lopex. M., Nanni, F., Masqualetti. D. & Mandelli, F. (1990) Immunolo- gical definition of acute promyelocytic leukaemia (FAR M3): a study of 39 cases. European Journal of Haematology, 45. 168- 171.

Diverio, D., Pandolfi, P.P.. Biondi, A.. Avvisati. G., Petti. M.C., Mandelli. F., Pelicci, P.G. &Lo Coco. F. (1993) Absence of RT-PCR detectable residual disease in patients with acute promyelocytic leukemia in long-term remission. Blood, 82, 3556-3559.

Gralnik, H.R. & Sultan, C. (1975) Acute promyelocytic leukaemia: haemorrhagic manifestation and morphologic criteria. British Journal of Haematolgy, 29, 373-376.

Grignani, F., Fagioli. M., Alcalay. M.. Longo, L., Pandolfi, P.P., Ilonti. E.. Biondi, A., Lo Coco. F., Grignani, F. & Pelicci, P.G. ( I YY4) Acute promyelocytic leukemia: from genetics to treatment. Blood,

Heim, S. & Mitelman. F. (1987) Cancer Cytogenetics. Alan R. Liss. New York.

Huang. W., Sun, G.L., Li, X.S., Cao, Q.. Lu, Y., Jang, GS.. Zhang. P.Q.. Chai. J.R., Wang, Z.Y., Waxman, S. & Chen, S.J. (1993) Acute promyelocytic leukemia: clinical relevance of two major PML- RAR A isoforms and detection of minimal residual disease by

77, 1418-1422.

1441-1446.

336, 1440-1441.

83.10-25.

0 1995 Blackwell Science Ltd, British Journal of Huematology 91: 878-884

884 Marc0 Mancini et a1 retrotranscriptase/polymerase chain reaction to predict relapse.

Huang, M., Yu-Chen. Y., Shu-Rong, C.. Chai, J., Lin. Z., Long, J. & Wang, Z. (1988) Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood, 72, 567-572.

ISCN (1991) Guidelines for Cancer Cytogenetics, Supplement to an International System for Human Cutogenetic Nomenclature (ed. by F. Mitelman). Karger, Basel.

Larson, R.A., Kondo. K., Vardiman, J.W., Butler, A.E., Golomb. H.M. & Rowley, J.D. (1984) Evidence for a 15;17 translocation in every patient with acute promyelocytic leukemia. American lournal of Medicine, 76, 827-831.

Licht, J.D., Chomienne C., Goy, A.. Chen. A., Scott, A.A., Head, D.R., Michaux, J.L., Wu, Y., DeBlasio, A., Miller, W.H., Jr, Zelenetz, A.D., Willman. C.L.. Chen, Z., Chen, S.J., Zelent. A,, MacIntyre. E., Veil, A,, Cortes, J., Kantarjian, H. & Waxman. S. (1995) Clinical and molecular characterization of a rare syndrome of acute promyelocytk leukemia associated with translocation (1 1:17).

Lo Coco, F.. Diverio, D., D’Adamo, F.. Awisati, G.. Alimena, G.. Nanni, M., Alcalay, M., Pandolfi, P.P. & Pelicci, P.G. (1992a) PML RAK-a rearrangement in acute promyelocytic leukaemias appar- ently lacking the t( 15;17) translocation. European Journal of Haematology, 48, 173-176.

Lo Coco, F., Diverio, D.. PandoK, P.P., Biondi, A.. Rossi, V., Avvisati, G.. Rambaldi. A., Arcese, W., Petti. M.C., Meloni. G., Mandelli. P., Grignani. F.. Masera. G.. Barbui, T. & Pelicci, P.G. (1992b) Molecular evaluation of residual disease as a predictor of relapse in acute promyelocytic leukaemia. Lancet. 340, 1437-1438.

Miller, W.H.. Jr, Kakizuka, A., Frankel. S., Warrel, R.P., Jr, De Blasio, A.. Levine, K., Evans, R.M. & Dmitrowsky, E. (1992) Reverse transcription polymerase chain reaction for the rearranged retinoic acid receptor a clariEes diagnosis and detects minimal residual disease in acute promyelocytic leukemia. Proceedings of the National Academy of Sciences of the United States of America. 89,

Miller. W.H., Jr. Levine. K., De Blasio, A.. Frankel. S.. Dmitrowsky, E. & Warrel, R.P.. Jr (1993) Detection of minimal residual disease in

Blood, 82,1264-1269.

Blood, 85,1083-1094.

2694-2698.

acute promyelocytic leukemia by reverse transcription polymer- ase chain reaction assay for the PML/RARA fusion mRNA. Blood.

Rodeghiero, P., Avvisati, G., Castman, G., Barbui. T. & Mandelli, F. (1990) Early deaths and anti-hemorragic treatments in acute promyelocytic leukemia: a GIMEMA retrospective study in 268 consecutive patients. Blood, 7 5 , 2112-2117.

Speicher, M.R., Janch, A.. Parr. A. & Becher, R. (1993) Delineation of translocation t( 15;17) in acute promyelocytic leukemia by chromosomal in situ suppression hybridization. Lt.ukem.a Research, 17, 359-364.

Tallman, M.S. & Kwaan, H.C. (1992) Reassessing the hemostatic disorder associated with acute promyelocytic leukemia. Blood, 79, 543-553.

Tong, J.H., Dong, S.. Geng, J.P., Huang, W., Wang, Z.Y., Sun, G.L., Chen, S.J., Chen, Z., Larsen, CJ. & Berger, R. (1992) Molecular rearrangements of the My1 gene in acute promyelocytic leukemia (APL M3) d e h e a breakpoint cluster region as well as some molecular variants. Oncogene, 7, 311-316.

Vahdat, L., Maslak, P., Miller, W.H.. Eardley. A., Heller, G., Scheinberg, D.A. & Warrell, R.P., Jr (1994) Early mortality and retinoic acid syndrome in acute promyelocytic leukemia: impact of leukocytosis, low-dose chemotherapy, PML/RAR N isoform. and CD13 expression in patients treated with all trans retinoic acid. Blood, 84, 3843-3847.

Warrel, R.P. Jr.,Frankel. S.R., Miller, W.H.. Jr, Scheinberg. D.A.. Itri. L.M., Hittelman, W.N., Vyas, R., Andreeff. M., Tafuri. A., Jakubowski. A., Gabrilove, J., Gordon. M.S. & Dmitrowsky. E. (199 1) Differentiation therapy for acute promyelocytic leukaemia with tretionin (all-trans-retinoic acid). New England Journal of Medicine, 324,13 8 5 - 1 3 9 3.

Wiegant, J., Ried, T.. Nederlof, P.M., van der Ploeg, M., Tanke, H.J. & Raap, A.K. (1991) In situ hybridization with fluoresceinated DNA. Nucleic Acids Research, 19, 3237-3241.

Zhao. L., Chang. K.S., Estey, E.H.. Hayes, K., Deisseroth, A.B. & Liang. J.C. (1995) Detection of residual leukemic cells in patients with acute promyelocytic leukemia by the fluorescence in situ hybridm- tion method: potential for predicting relapse. Blood, 85,49 5-499.

82,1689-1694.

0 1995 Blackwell Science Ltd, British Journal ofHaematology 91: 878-884