granulopoiesis and thrombopoiesis in mice bearing ... · [cancer research 26, 149-159, january...

[CANCER RESEARCH 26, 149-159, January 1966|

Granulopoiesis and Thrombopoiesis in Mice Bearing TransplantedMammary Cancer1

L. DELMONTE, A. G. LIEBELT, AND R. A. LIEBELT

Department of Anatomy, Baylor University College of Medicine, Houston, Texas

Summary

Growth of a transplantable mammary cancer in CE mice (CE1460 MA CA) was paralleled by hyperplasia of the granulocyticelements of the bone marrow and peripheral leukocytosis (up to350,000 leukocytes/cu mm) with a lymphoid-myeloid ratio reversal to 1:4 and a shift to the left. About 5-20% of the granulocytic elements in the blood and bone marrow showed morphologic abnormalities. Hematocrit values decreased 10-15%.Peripheral platelet levels manifested a slight depression in theface of bone marrow megakaryocytopenia and splenic mega-karyocytosis. Extramedullary granulopoiesis, with the presenceof mitoses and all stages of development of the granulocyte seriesin diffuse and discrete foci, was observed in the liver, spleen,lungs, lymph nodes, and adrenals of CE 1460 tumor bearers. Thedevelopment of foci of extramedullary granulopoiesis, despite theabsence of mÃ©tastasesin the bone marrow, suggests that theformer resulted from direct stimulation of both intra- and extra-medullary hemopoietic tissue by a tumor-elaborated, granulo-cyte-specific factor, rather than from a reduction of bone marrowspace available for granulopoiesis.

In BALB/C mice, another transplantable mammary cancer(BALB/C 2301 MA CA) induced periodic fluctuation of plateletsbetween base-line and elevated levels and bone marrow mega-karyocytosis but no change in splenic megakaryoctye levels. Thebone marrow erythroid-myeloid ratio and the peripheral leukocytes' lymphoid-myeloid ratio remained normal. Hematocritvalues decreased 10-15CÂ¿.No foci of extramedullary granulopoiesis developed in the host tissues.

The tumor specificity of the leukemoid and thrombocytic responses was suggested by the host response to transplantation ofthe respective tumors into (BALB/C x CE) FI hybrid mice and bythe regression of the specific CE 1460 tumor-associated andBALB/C 2301 tumor-associated hÃ©matologiechanges followingtumor excision.

Introduction

A recent report from our laboratories (3) described the extraction and partial purification of a granulocytosis-promoting factor

1Supported in part by USPHS Grants SF3CA-17941-01/02,CA-07788-01, CA-04517-03and by American Cancer Society Institutional Grants ACS-IN-27D-Project No. 12 and ACS-IN-27E-Project No. 1.

Received for publication March 16, 1905; revised September 7,1965.

from a transplantable mouse mammary tumor (CE 1460 MA CA)that arose as a spontaneous tumor, associated with a severeleukocytosis and extramedullary granulopoiesis (EMG)2, in a fe

male mouse of the CE strain (22). CE 1460 MA CA has beencarried through more than 100 transplant generations in CE miceand through more than 50 transplant generations in (BALB/C xCE) FI hybrid mice (hereafter referred to as Ft hybrids); it hasshown remarkable stability as regards tumor biology and thetumor-associated hÃ©matologieresponse in both the parent strain(CE) and FI hybrid mice.

During a partial survey of other transplanted mouse tumors inour laboratories, we observed a fluctuating increase in circulatingplatelets but no leukemoid reaction to be associated with anothertransplantable mammary tumor (BALB/C 2301 MA CA) ofspontaneous origin. The specificity of this response was suggested by the finding that none of the other tumors studied in thesurvey to date evoked a comparable platelet response.

Although tumor-associated hÃ©matologie changesâ€”particularly leukemoid reactions and extramedullary hematopoiesisâ€”have been reported frequently, and reviewed extensively byDunn (5) and Ebbenhorst-Tengbergen and MÃ¼hlbock(7), it wasfelt that the extraction of a granulocytosis-promoting factor fromCE 1460 tumor tissue and the consequently implied possibilitythat another humoral (thrombocytosis-promoting) factor iselaborated by the BALB/C 2301 tumor were of sufficient interestto warrant a study of the biology of these tumors. Hence, thepresent report attempts (a) to relate the leukemoid and thrombocytic responses to tumor characteristics (e.g., growth rate, size,and tumor viability), changes in blood morphology, and host tissue histopathology, and (b) to differentiate between the hoststrain-dependent and the tumor-dependent aspects of the hemopoietic responses to the 2 tumors in the FI hybrid system.

Materials and Methods

HÃ©matologieand histopathologic changes were studied in 98control and 155 tumor-bearing mice of the CE, BALB/C, and FIhybrid strains. These included: Group I (control mice), (a) 74 intact young adult (6-8 weeks old) and 6 intact 3-month-old CE,BALB/C, and FI hybrid mice, and (&) 18 intact sham-operatedyoung adult CE, BALB/C, and FI hybrid mice; Group II (micebearing CE 1460 MA CA), 42 young adult, 6 3-month-old, and13 12-month-old CE mice and 45 young adult FI hybrid hosts;Group III (mice bearing BALB/C 2301 MA CA), 31 young adultBALB/C and 18 young adult FI hybrid hosts. Mice were kept ingroups of 3 per cage, separated as to sex, and given Purina LabChow and drinking water ad lib.

JANUARY 1966 149

on April 13, 2017. © 1966 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

L. Delmonte, A. G. Liebelt, and R. A. Liebelt

Each host received a s.c. implant (l cu mm) of fresh donortumor tissue delivered with a trocar into the right flank on Day 1.Of these mice, 105 animals were killed after 2-3 weeks (Group II)or 9-11 weeks (Group III). Another 48 hosts of Groups II andIII were subjected to excision of moderately advanced (16-22

mm in diameter) tumors, while under light ether anesthesia, andwere observed for periods up to 30 days following both successfuland unsuccessful (i.e., local recurrence or mÃ©tastases,or both)tumor excision.

Morning samples of caudal vein blood were obtained (a) atweekly intervals in intact controls, (6) prior to transplantation,(c) at weekly intervals thereafter in unoperated tumor-bearinghosts, and (d) prior to operation and at 3-day intervals thereafterin sham-operated controls and tumor-excised hosts. A single13-fil blood sample, diluted 1:100 with 1.3 ml of ammoniumoxalate in an Unopette disposable blood-diluting pipette, sufficedfor both total leukocyte and platelet determinations by lightmicroscopy. Heparinized capillary tubes were used for micro-hematocrit determinations. Wright-Giemsa-stained blood filmsand bone marrow imprints were prepared for differential leukocyte counts and studies of blood morphology.

Intact control mice (Group la) were sacrificed at ages comparable to those of the corresponding nonoperated tumor hosts.Sham-operated control mice (Group lÃ²)were sacrificed 21 daysfollowing surgery.

Tissue samples were fixed in 10% neutral formalin, sectionedat 5 n, and stained routinely with hematoxylin and eosin as wellas with (PAS)2 plus alcian blue.

Megakaryocytes were counted in 100 high power fields ( X 450)of 5-/Ã-thickPAS-stained sections of femoral and sternal bonemarrow and of spleenâ€”equivalent to 2.35 X IO5cu n of the fixed

tissue sample.Samples of blood, tumor tissue (with and without central

necrosis), liver, and spleen were taken at autopsy, under asepticconditions, from 12 ~Fihybrid mice bearing CE 1460 MA CA and

cultured both aerobically and anaerobically on blood agar and inthioglycolate broth. Whenever necessary, further diagnostic culture media were used to identify any microorganisms present.

A preliminary study of 6 other transplantable solid tumorscarried in mice of 5 different strains in the Kirschbaum MemorialLaboratory, Baylor University College of Medicine, was undertaken for the purpose of relating possible changes occurring inthe hosts' leukocyte and platelet picture with tumor type, growthrate, and size. These tumors included 1 skin cancer (A+ 9966), 3

mammary cancers [(BALB/C x CE) FI 3206, DBA/2/ff 5199,BALB/C 926)], 1 testicular tumor (CE 8), and 1 lymphoma((BALB/C x CE) F, 1182). Leukocyte and platelet values incaudal vein blood were determined as described above.

Results

CE Host

CE 1460 MA CA, an adenocarcinoma Type B according toDunn's (6) classification, attained a diameter of approximately25 mm within 14-21 days in 6- to 8-week-old and within 21-28

'The following abbreviations are used: PAS, periodic acid-Schiff ; EMG, extramedullary granulopoiesis.

days in 3- and 12-month-old CE mice of both sexes. Macro-scopically, the tumor was a homogeneous white mass that, at adiameter of approximately 15 mm, began to manifest superficialeschar formation and erosion as well as central necrosis and cysticfluid accumulation.

MÃ©tastasesdid not develop in the tissues of young adult CEtumor bearers, but they occasionally occurred in the form of pinpoint to 3-mm foci in the lungs and livers of 3- and 12-month-oldCE hosts during the usual 3-4 week survival period. MÃ©tastaseswere also observed in the lungs of several young adult CE micewhich survived for exceptionally long periods of time (6-12weeks) following tumor implantation.

Tumor growth was paralleled by a marked rise in total circulating leukocytes (frequently to 350,000/cu mm and occasionallyto 750,000/cu mm) as well as by slight depression of both hemato-crit (from 45-54% to 40-46%) and platelet levels (from 1.3-2.2million/cu mm to 1.0-1.7 million /cumm) (Chart 1). Circulatingleukocytes showyeda reversal of the lymphoid-myeloid ratio to1:4 and a shift to the left (5-15% immature granulocytes).

The bone marrow presented severe, predominantly granulo-cytic hyperplasia with a shift to the left (Fig. 16), as well as pronounced megakaryocyte depression (Table 1). Megakaryocytesize remained within normal limits (20-50 fi in diameter).

About 5-20% of the granulocytes in both blood and bone marrow were enlarged up to 1.5-fold and showed other morphologicabnormalities (Fig. lo) commonly seen in tumor-associatedleukemoid reactions (5, 21, 27) as well as in pernicious anemia,drug toxicity, and radiation disease states (36). The cytoplasm ofthe abnormal granulocytes tended towards extreme basophilia inboth immature and mature cells and frequently presented irregular coarse granulation. The nuclei manifested swelling, hyper-segmentation, clumped and accessory chromatin, and occasionally abnormal mitotic figures.

The tumor size-tumor weight curve of the CE 1460 tumor waslinear in transplanted and biphasic in recurrent (postexcision)tumors (Chart 2). In the latter, it increased in a linear fashion upto around the 3.5-gm level, then leveled off as the tumor began tomanifest marked eschar formation, erosion, central necrosis, andcystic fluid. Occasionally, tumors underwent central erosion withsubsequent peripheral regrowth.

Circulating leukocytes averaged 65,000-80,000/cu mm/gm oftumor weight and 8,000-12,000/cu mm/mm of tumor diameter

(Chart 2). Leukocytosis averaging 100,000 leukocytes/cu mmwas associated with a 10-mm tumor; yet macroscopic evidence ofnecrosis was seen only after the tumor had attained a diameter of15 mm or more. Histologie analysis of CE 1460 tumors duringrecent transplant generations in FI hybrid mice showed no consistent relationship between the presence or absence of microscopic foci of necrosis in 5.5- to 6.5-mm tumors and the onset ofthe myeloid reaction. Nor was the degree of microscopicallyevident necrosis in 9.5- to 11.5-mm tumors consistently proportional to the degree of the leukemoid reaction observed.

In cases where necrosis advanced to the stage of tumor erosion,the leukemoid reaction became markedly depressed, and the depression appeared to be reversed only upon renewed growthalong the tumor periphery. Cystic-necrotic material (devoid ofviable cells) from large CE 1460 tumors, when injected i.p., failedto evoke significant leukocytosis in young adult CE mice (L.Delmonte and R. A. Liebet, unpublished data). The peripheral

150 CANCER RESEARCH VOL. 26



Granulopoiesis and Thrombopoiesis in Mammary Cancer

leukocyte, or more specifically the peripheral granulocyte levelapparently was directly proportional to the volume of viabletumor tissue present.

Mitotically active foci of EMG developed consistently in theliver, spleen and lungs and occasionally in the lymph nodes andadrenals of CE mice bearing CE 1460 MA CA (Figs. 1 c-e; 2f-k). EMG was more pronounced in 6- to 8-week-old than in 3-and 12-month-old hosts. EMG foci were not usually associatedwith metastatic proliferation. Granulocytes in EMG foci encompassed both normal and morphologically aberrant forms.Only rarely did erythropoietic and thrombopoietic proliferationoccur at sites of EMG.

Hepatic EMG foci occurred mainly within the walls of thelarger venous vessels and occasionally among the liver cord cells

Â¡esoIs20Â§~

100â€¢

CEHOST^,jf^.^r-""^.Â»â€¢â€¢""''

ii ii

z

o<J^

OCA(DO*5

(A

O

60

40I

20

0

300

250

200

150

IOO

SO

7 10 14 17 21DAYS POST-TRANSPLANTATION

i -M.IS.E.# 99.7% CONTROL CONFIDENCE LIMITS

CHART1. Peripheral blood changes accompanying growth ofCE 1460 MA CA in CE mice.

TABLE 1TUMOR SPECIFICITYOF HEMOPOIETICTISSUE RESPONSEIN 3STRAINSOF MICE BEARINGCE 1460 MA CA ORBALB/C 2301

MA CA

TISSUECE

1460 MACACE(BALB/C

x CE)KI hybridBALB/C

2301 MACA(BALE/

C x CE) F,hybridBALB/C

LiverSpleen

LungsAdrenalsLymph

nodes+

++++++Â±+ ++++++Â±-â€”-â€”

Megakaryocytes6

Bone marrowSpleen38%278%38% 330%123% 165%146%126%

0 Degree of extramedullary granulopoiesis.* Percentage of control (100%) levels.

themselves (Fig. 2f, <?).Occasionally megakaryocytes were foundlodged in hepatic sinusoids.

The spleen, which doubled or tripled its size within 2-3 weeksof CE 1460 tumor implantation, showed evidence of diffuse EMGthat disrupted the splenic architecture and occasionally evendisplaced the lymphoid elements in germinal centers (Fig. le).Megakaryocytes, easily distinguished from other giant cells bytheir strong affinity for the PAS stain, increased about 3-fold(Table 1); they occurred mainly in discrete nests of both immature and mature cells and occasionally .singly within the EMGfoci.

The lungs presented diffuse granulocyte permeation of thealveolar walls as well as a few discrete foci abutting bronchiolesand small blood vessels and showing a high degree of mitoticgranulocytic activity (Fig. 2h, i).

The lymph nodes occasionally presented diffuse granulocytepermeation or discrete round areas consisting entirely of immature and mature granulocytic elements (Fig. Id, e). These areasof EMG stained a distinctive light purple with hematoxylin andeosin and thus were easily distinguishable from the more deeplystaining lymphoid tissue.

The adrenals developed EMG mainly at the corticomedullaryjunction (Fig. 2j, k). Granulocytic elements appeared to concentrate particularly around clumps of cells whose granules showeda strong affinity for PAS.

BALB/C Host

BALB/C 2301 MA CA, an adenoacanthoma according toDunn's (6) classification, had the macroscopic aspect of a homo

geneous white mass throughout the course of its development,although it occasionally showed eschar formation towards theterminal stage (30-38 mm in diameter).

In young adult hosts of both sexes, the tumor developed to adiameter of approximately 25 mm within 9-11 weeks and at-

JANUARY 1960 151



-> JLli*<#Â°

Â®fflS^&&**f -Â»* â€¢s

felÂ» &Ã•152 CANCER RESEARCH VOL. 26




50

LuK 40LU

<~30

KO

'20

IO

TRANSPLANTRECURRENCE

I 234567TUMOR WEIGHT (GRAMS )

50

LUl- 40LU5<~30

IO

o

1

200 400 600WBC(THOUSANDS/CU MM)

800

o 'â€” CD

< 4

O 200 400 600 800WBC(THOUSANDS/CU MM)

CHART 2. Correlation of leukocytosis, tumor size, and tumorweight in mice bearing transplanted and recurrent CE 1460 MACA. Upper graph: Ratio of tumor size to tumor weight. Transplant: Y = Ã•.95X+ 15.28, r = 0.782, P < 0.001. Recurrence:YÂ¡= 2.63X + 16.7Â«, ri = 0.646, PÂ¡< 0.010. Recurrence: Y2 =-2.44X + 38.90, r, = 0.812, I\ < 0.025. Interclass: r = 0.739,P < 0.001. Center graph: Ratio of tumor size to leukocytosis.Transplant: Y = 0.03X + 10.93, r = 0.581, P < 0.001. Recurrence: Y = 0.04X + 8.42, r = 0.624, P < 0.001. Interclass: r =0.979, P < 0.001. Lower graph: Ratio of tumor weight to leukocytosis. Transplant: F = 0.007X + 1.55, r = 0.509, P < 0.001.Recurrence: Y = 0.005X + 1.77, r = 0.329, P < 0.200. Interclass:r = 0.613, P < 0.100.

tained a maximal diameter of 30-38 mm within 13-15 weeks,although some hosts survived up to 24 weeks.

Neither mÃ©tastasesnor foci of EMG developed in the tissues ofBALB/C hosts bearing BALB/C 2301 MA CA.

During the course of BALB/C 2301 tumor growth, total anddifferential leukocyte numbers remained normal, while hemato-

t-3UliÂ»I'Â»>

0a6040

I-20o300_2503200m

z150Â§

100K5O2.55

2.OS|

='â€¢â€¢Â¿B

* LÂ°05:-c-P--P?t-BALB

/C HOST..*

*-a"f

f i i i i i i i iiâ€¢-*.,-.,.*

^ jt ,LJlrâ€¢â€¢â€¢%â€¢â€¢â€¢â€¢*â€¢-â€¢â€¢*.....:..â€¢.â€¢..::.ti

i i i i i i i i ii,

^ .^-M1**^ v//,~ â€¢ *J'&Â¿jÂ¿Â¿Â¿&&Â¿Â¿6%%Â¿Â¿&LWrtfy//

I 10 II3456789WEEKS POST-TRANSPLANTATION

I â€¢M.Â±S.E.

â€” â€¢99 7% CONTROL CONFIDENCE LIMITS

CHART 3. Peripheral blood changes accompanying growth ofBALB/C 2301 MA CA in BALB/C mice.

crit levels were depressed slightly (from 47-60% to 40-48%).The striking characteristic of this tumor was that plateletsshowed periodic fluctuation between baseline (0.8-1.8 million/cu mm) and elevated levels (to 3.0 million/cu mm), albeit withoutany discernably constant rhythm or correlation with tumor sizeand weight (Chart 3). Thrombocytosis was evident as early as2-3 weeks following transplantation of BALB/C 2301 MA CA,when the tumor measured approximately 3-5 mm in diameter;the phenomenon of recurrent thrombocytosis became well established during the ensuing 9-11 weeks (Chart 3) and persisted

thereafter. Individual platelet fluctuations were far more irregular in the circulation of tumor hosts than in that of control mice

FIG. 1. Leukemoid blood picture and granulopoiesis in tissues of CE 1460 MA CA hosts: a, peripheral blood with aberrant granulo-cytic forms (X 1000); 6, bone marrow (X 150); c, spleen (X 050); d, lymph node (X 300); e, lymph node (X 750).

JANUARY 1960 153



Granulopoicsis and Thrombopoiesis in Mammary Cancer

bJÂ¡IHaSi3020

10604O

20-F|

HYBRIDHOST; /*..._.** **"""*'â€”'*^aT^ Til I|fr

*" 3bfc.-"^-rw:-g-.-.-.-:-:.-Tr- -ar- â€” "*Â»v â€” --.III!

_ 250 -

roo O

"S I50h

3*<

mO

â€¢Â» â€¢. Â«â€¢Â¡a a Â¿Â¿-__ _ T'" T â€¢> ^

I

Ã®3-

Â¿iO

OI

V.T

"w^

I 10 II3456789WEEKS POST-TRANSPLANTATION

!â€¢ M.Â±SE. CE 1460 MACA

Â«p. 99.7% CONTROL CONFIDENCE LIMITS BALB/C 2301 MACA

CHART4. Peripheral blood changes accompanying growth ofCE 1460MA CA and BALB/C 2301 MA CA in (BALB/C x CE)Fi hybrid mice.

and periodically attained peak levels in the former that were upto 50% higher than any ever observed in the latter. At 9-11weeks, tissue megakaryocytc concentrations manifested an increase of about 50% in the bone marrow and 2590 in the spleen(Table 1). The tumor dependence of these changes in the mega-karyocyte-platelet axis has been studied extensively and will bereported elsewhere.3

The bone marrow of BALB/C 2301 MA CA hosts presented anunchanged erythroid-myeloid ratio. There were no aberrant

3L. Delmonte and R. A. Liebelt, Granulopoiesis and Thrombopoiesis in Mice Bearing Transplanted Mammary Cancer. II.Megakaryocyte-Platelet Axis, in manuscript.

50

40

30

20

10

CETRANSPLANTRECURRENCE

ceo

50

4O

30

20

IO

(BALB/C X CE) F,

O 7 14 21 28DAYS POST-TRANSPLANTATION

CHART5. Growth rate of transplanted and recurrent CK 1400MA CA in CE and (BALB/C x CE) F, hybrid mice. Upper graph:CE Host, Transplant: Y = 1.24A"- 0.57, r = 0.501, P < 0.001.

Recurrence: Y = 2.13X - 0.12, r = 0.727, P < 0.001. Interclass:r = 0.494,P < 0.500.Lower graph: (BALB/C x CE) Fi hybrid host,Transplant: Y = 0.54X + 4.%, r = 0.624, P < 0.001. Recurrence:Y = 1.98X - 0.10, r = 0.942, P < 0.001. Interclass: r = 0.143,P < 0.500.

morphologic forms and no macro- or microscopically discerniblemetastatic foci.

(BALB/C x CE) I'\ Hybrid Host

In Fi hybrid mice, both CE 1460 MA CA and BALB/C 2301MA CA retained their biologic and specific hÃ©matologiecharacteristics (Chart 4; Table 1).

CE 1460 tumor growth rate in young adult F[ hybrids lagged 1week behind that observed in young adult CE hosts (Chart 5).This lag, which was significant within 99.9% confidence limits,was reproduced in the development of both the leukemoid response and the platelet depression (Charts 1 and 4), althoughleukocyte and platelet values as related to tumor size were statistically identical in the parent strain and the FI hybrid. It isof interest that the CE 1460 tumor growth rate in young adultFi hybrids was comparable to that observed in 12-month-oklCE hosts.

The incidence of lung and liver mÃ©tastasesin CE 1460 tumorbearers appeared to be dependent on time rather than on tumorsize or strain. In 3- to 12-week-old CE and in young adult FIhybrid hosts bearing advanced tumors (20-25 mm in diameter),

FIG. 2. Leukemoid blood picture and granulopoiesis in tissues of CE 1460MA CA hosts:/, liver (X 150); g, liver (X 475); h, lung(X 200); i, lung (X 600); j, adrenal (X 60); k, adrenal (X 475).

JANUARY 1966 155




the incidence of mÃ©tastaseswas negligible at 2 weeks, slight at 3weeks, and increasingly marked at 4 or more weeks after tumortransplantation.

BALB/C 2301 tumor growth rate was identical in young adultFi hybrid and BALB/C hosts (Charts 3 and 4). No leukemoidreaction occurred in either host strain. The fluctuating plateletelevation was slightly less pronounced in Fi hybrids than inBALB/C hosts (Charts 3 and 4). Unlike CE 1460 MA CA,BALB/C 2301 MA CA was already associated with a rise inplatelet levels by the time a 5-mm tumor diameter had been attained in the FI hybrid host.

Hematocrit changes were slight (10-15% maximum decrease)

and were comparable for the 2 tumors and all 3 strains of mice(Charts 1, 3, 4).

The bone marrow and EMG and megakaryocyte picture in FIhybrids bearing CE 1460 MA CA or BALB/C 2301 MA CA wascomparable to that of hosts of the respective strain of tumor

origin.Splenomegaly in FI hybrids, as well as in the strain of tumor

origin, depended on tumor size. Splenic megakaryocyte levels

were elevated more than 3-fold in CE 1460 tumor bearers butonly about Â§in BALB/C 2301 tumor bearers (Table 1).

Tumor Excision

Moderate transitory leukocytosis (25,000-45,000 leukocytes/cu mm)â€”probablydue toa transi tor y increase in leukopoiesis compensating for blood loss at surgeryâ€”followed not only CE 1460and BALB/C 2301 tumor excision in hosts of the respectiveparent strain (CE and BALB/C) and in FI hybrid hosts, but alsosham operation in all 3 strains of mice (Table 2).

Renormalization of total and differential leukocyte values anddisappearance of aberrant granulocytic forms occurred in CE andFi hybrid hosts within 3-9 days of excision of CE 1460 MA CA(Table 2). Excision of a 16- to 20-mm tumor was usually followed within 5-21 days by tumor recurrence along the line of

excision. Recurrence of the leukemoid response was observed asearly as 24-48 hr prior to macroscopically evident tumor re-growth. The growth rate of CE 1460 tumors recurring at the siteof excision was accelerated by comparison to that of the transplanted tumors (Chart 5).

TABLE 2PERIPHERALBLOODPICTUREFOLLOWINGSHAMOPERATIONANDTUMOREXCISIONIN CE, BALB/C AND(ALB/C x CE) FI HYBRID

MICE (Mean Â±S.E.)

MOUSESTRAINCE(BALB/C

x CE)F,BALB/CCE(BALB/C

x CE)F,BALB/CTUMORNoneCE

1460 MACANoneCE

1460 MACABALB/C

2301 MACANoneBALB/C

2301 MACANoneCE

1460 MACANoneCE

1460 MACABALB/C

2301 MACANoneBALB/C

2301 MA CATYPE

EXCISIONShamTumorShamTumorTumorShamTumorShamTumorShamTumorTumorShamTumorDAYS

POSTSURGERY036912is21\VBC

(thousands/cumm)10.6Â±2.9185.3Â±7.113.3Â±0.7175.1Â±19.612.0Â±1.59.6Â±1.812.6Â±4.014.8Â±1.835.9Â±2.310.3Â±0.944.0Â±3.924.4Â±2.110.3Â±1.124.8Â±2.014.0Â±1.432.4Â±8.913.0Â±1.125.7Â±3.421.3Â±1.08.9Â±1.330.5Â±2.022.9Â±2.221.5Â±5.319.6Â±1.919.5Â±1.525.0Â±1.518.4Â±2.732.4Â±3.222.8Â±1.420.4Â±1.222.4Â±1.921.3Â±1.524.9Â±2.218.2Â±0.523.0Â±2.916.0Â±1.723.5Â±3.522.4Â±1.518.6Â±0.720.1Â±3.610.8Â±3.241.8Â±6.211.9Â±5.120.2Â±1.811.9Â±1.011.3Â±0.511.8Â±0.611.8Â±2.114.1Â±2.4Platelets

(millions/eumm)1.863Â±0.1301.239Â±0.0811.548Â±0.1241.465Â±0.2131.783Â±0.0381.468Â±0.1211.948Â±0.2171.803Â±0.1323.093Â±0.3051.675Â±0.0452.897Â±0.1982.824Â±0.1461.660Â±0.1112.773Â±0.3022.723Â±0.3504.454Â±0.5802.116Â±0.0694.334Â±0.2312.649Â±0.2841.719Â±0.0944.474Â±0.4732.630Â±0.4152.643Â±0.7292.163Â±0.2682.461Â±0.1782.468Â±0.4281.758Â±0.1672.934Â±0.1762.191Â±0.1721.827Â±0.0221.819Â±0.0902.832Â±0.7842.783Â±0.1742.003Â±0.1961.602Â±0.0551.902Â±0.0972.493Â±0.0081.889Â±0.1712.122Â±0.2862.196Â±0.1261.676Â±0.0593.061Â±0.5911.984Â±0.1492.277Â±0.1021.800Â±0.0771.710Â±0.1962.364Â±0.2131.678Â±0.0961.884Â±0.080





With CE 1460 MA CA recurrence after unsuccessful excision,the tumor size-tumor weight relationship was initially (up to the3.5-gm level) similar to that observed before excision; then, withthe development of a disproportionate weight increase due tomassive accumulation of cystic fluid, the ratio began to show atrend towards a linear decrease rather than towards an increase(Chart 2). However, the pre- and postexcision ratios of tumor sizeto leukocyte number were statistically identical (interclassr = 0.979) (Chart 2). Following recurrence, the relationship oftumor weight to leukocyte number Ix-came random (r = 0.329),although it manifested a trend towards the same ratio as that ofthe pre-excision group (interclass r = 0.613) (Chart 2).

Following CE 1460 tumor excision, pinpoint to 3-mm mÃ©tastases were found frequently in the lungs and liver and occasionally in the lymph nodes, kidneys, and adrenals, but not infemoral or sternal bone marrow of the mice that presented norecurrence at the tumor implantation site. Microscopically evident foci of lung mÃ©tastasesbut no foci or EMG were found uponhistologie examination of all CE and FI hybrid hosts withmacroscopic-ally and hematologically diagnosed successful CE

1460 tumor excision.In BALB/C and F! hybrid mice bearing BALB/C 2301 MA

CA, tumor excision was usually successful, recurrence being rarewithin the 21-day postsurgical observation period of our experiments. Neither of 2 mice that manifested incipient (palpable)tumor recurrence presented mÃ©tastases.

A transitory 2- to 3-fold increase in platelet levels with a peakat 6 days followed both tumor excision and sham operation in all3 strains of mice, persisted for 9-12 days, and manifested a trendtowards renormalization by 21 days (Table 2). The immediatereaction, following sham operation or tumor excision, appearedto be dependent on (i.e., compensatory for) blood loss or trauma,or both, at surgery rather than on tumor ablation itself, but it wasless pronounced after sham operation than after tumor excision.In all 3 strains of mice, the tumor-dependent platelet changeswere totally or partially reversed within 21 days of excision ofeither CE 1460 MA CA or BALB/C 2301 MA CA.

Microbiology of CE 1460 MA CA

Although all the mice from which these tissue samples had beentaken presented a marked leukemoid reaction, only some of thesamples of tumor (67%) and spleen (50%), a single sample ofblood (8%), and no sample of liver (0%) from mice bearing advanced (20- to 25-mm diameter) CE 1460 tumors manifestedevidence of microorganisms in thioglycolate broth or on bloodagar. Among the positive cultures, 4 different Gram-negativeorganismsâ€”Pseudomonas aeruginosa, Aerobacter aerogenes, Proteus vulgaris, and Staphylococcus aureusâ€”were identified; a

Bacillus (species not identified) also was isolated. The presenceor absence of bacterial contamination had no effect on either theoccurrence or the magnitude of the CE 1460 tumor-associatedleukemoid reaction.

HÃ©matologieSurvey of Other Transplantable Mouse Tumors

The hÃ©matologiestudy of hosts of 5 different strains of micebearing additional tumors of various types of tissues revealed thatneither the tumor tissue type and growth rate nor the host strainand survival time following tumor transplantation appeared to

affect the type of hÃ©matologieresponse associated with thetumor: A+ 9966 skin cancer, (BALB/C x CE) FI 3206 mam

mary cancer, CE 8 testicular tumor, and DBA/2/ff 5199 mammary cancer hosts survived approximately 4-5, 5-6, 6-7, and19-20 weeks, respectively; yet all showed normal platelet values.A+ 9966 and (BALB/C x CE) F! 3206 tumor hosts also mani

fested a leukemoid reaction, whereas CE 8 and DBA/2/ff 5199hosts showed a normal leukocyte picture. BALB/C 962 mammary cancer and (BALB/C x CE) Ft 1882 lymphoma killed theirrespective hosts within approximately 4-5 weeks; hosts bearingboth these tumors manifested a trend towards platelet depressionsimilar to that seen in CE 1460 tumor hosts, whose survival was3-4 weeks, but only BALB/C 962 hosts also manifested Icuko-cytosis. None of the tumors studied evoked a recurrent throm-bocytosis like that observed in BALB/C 2301 tumor hosts.

Discussion

The tumor specificity of the leukemoid and thrombocytic responses associated with CE 1460 MA CA and BALB/C 2301 MACA, respectively, was demonstrated by the response of genetically isologous hosts (Fi hybrid mice) to the transplantation ofthese genetically unrelated tumors and by the regression of thespecific tumor-associated hÃ©matologieresponses following tumorexcision.

Over the past 50 years, leukocytosis associated with extra-medullary hematopoiesisâ€”both leukopoiesis and erythropoiesisâ€”has been reported repeatedly in tumor-bearing mice (1, 2, 5, 7,13, 14, 19-21, 26-28), rats (21, 23), hamsters (15), rabbits (8),dogs (8), monkeys (8), and man (4, 9, 10-12, 16-18, 24, 25,29-35). Thrombocytosis associated with leukemoid reactions hasbeen reported both in man and in tumor-bearing mice manifesting extramedullaiy hematopoiesis (2). A direct correlation between splenic megakaryocytosis and peripheral leukocytosis hasbeen reported by Parsons et al. (28) in mice bearing transplantedsarcomas. Most of the tumor-associated leukemoid reactionsdescribed in the literature have been associated with moderateto extreme concurrent effects on erythropoiesis or thrombopoiesis,or both, and have resulted in extramedullaiy hemopoiesis involving all cell lines rather than a single one, as is the case for theCE 1460 and BALB/C 2301 tumor systems.

The tumor-host relationship has been shown to be affected bya variety of factors. Whereas the growth rate of CE 1460 MA CAwas influenced by factors such as host age and genetics, the development of granulocytosis and EMG foci appeared to be dependent upon the volume of viable tumor tissue present and onthe hemopoietic mitotic and maturation potential of intra- andextramedullary host tissue in response to the tumor-elaboratedstimulus. The occurrence of EMG foci, despite an actual quantitative increase rather than a reduction of granulocytic tissue inthe bone marrow cavities by metastatic or other factors, furthersupports the concept that the host's granulocytic response is

tumor dependent.Although several investigators including Bateman (2) found

that tumor age, necrosis, ulcÃ©ration,and eschar formation had agreater influence on tumor-associated leukemoid reactions inmice than did tumor tissue viability or volume, our data, likethose of Ebbenhorst-Tengbergen and MÃ¼hlbock(7), suggested

JANUARY 1906 157



L. Delmontc, A. G. Liebelt, and R. A. Liebelt

that the amount of viable tumor tissue present rather than thepresence of necrotic tissue was the prime factor in the development of the tumor-associated leukemoid reaction. In fact, in theCE 1460 tumor system, whenever the ratio of tumor weight totumor size increased disproportionately owing to the appearanceof cystic fluid and necrosis at the core of the tumor, the peripheral leukocytosis tended to level off despite a continuous increase in absolute tumor weight. Furthermore, association of erosion with necrosis, resulting in an even greater loss of viabletumor tissue, was actually accompanied by regression of theleukocytosis, which recurred only upon renewal of growth at thetumor periphery.

The possibility that tumor-associated microorganisms inducethe leukemoid reaction has been proposed by Barnes and Sisman(1). Bateman (2) found leukemoid reactions to occur in tumor-bearing mice without demonstrable bacterial infection. In thepresent study, we found the presence or absence of bacterial contamination of CE 1460 tumor or host tissue to have no effect onthe induction or the magnitude of the leukemoid reaction, suggesting that there probably is no bacterial pathogenesis of thetumor-associated leukocytosis and EMG. However, the possibility of viral pathogenesis cannot be excluded, since recent preliminary electron microscope studies in our laboratories (L.Delmonte, T. Smith, and R. A. Liebelt, unpublished data) haverevealed the presence of virus-like particles in CE 1460 tumortissue.

The disparity between the ratios of tumor size to leukocytosisin mice with transplanted and those with recurrent CE 1460tumors raises an interesting point: Is the disproportionate decrease in this ratio due to reactivation of residual EMG focithat have become dormant following tumor ablation, superimposed on the laying down of new EMG foci during recurrenttumor growth? Or has the host become "sensitized" to hemo-

poietic stimuli by the original tumor transplant? The phenomenon only serves to emphasize the delicate balance of tumorand host factors in tumor-host relationships.

Similarly to the CE 1460 tumor-associated leukemoid andEMG reaction, the BALB/C 2301 tumor-associated thrombo-cytosis and bone marrow megakaryocytosis showed direct dependence on the presence of viable tumor tissue. However, itappeared that even a small volume of BALB/C 2301 tumor tissuesufficed to initiate a platelet releaseâ€”as manifested by the appearance of the 1st platelet peak 2-3 weeks following tumor implantationâ€”but that the bone marrow megakaryocyte build-up(hyperplasia) was a more gradual process paralleling tumorvolume increase.

The specificity of BALB/C 2301 MA CA for evoking a fluctuating thrombocytosis is further supported by the lack of consistentrelationship found by both ourselves and Bateman (2) betweentumor tissue type, size, and growth rate and host strain and survival time, following tumor transplantation, on the one hand, andthe development of tumor-associated thrombocytic (and/orleukocytic) reactions, on the other hand. Further investigationis necessary to determine whether a thrombocytosis-promotingfactor can be extracted from BALB/C 2301 tumor, just as agranulocytosis-promoting factor was extracted from CE 1460mammary tissue (3). The BALB/C 2301 tumor system appearsto present us with a biologic model from which we can postulatethe probable presence of a tumor-dependent factor influencingcirculating platelet levels.

Acknowledgments

The authors wish to express their appreciation to Miss Kath-aryn Kemp and Mr. William T. Miller for their respective rolesin the survey of the Kirschbaum Memorial Laboratory's transplanted mouse tumors for tumor-associated leukemoid and thrombocytic reactions.

References

1. Barnes, W. A., and Sisman, I. E. Myeloid Leukemia and Non-malignant Extramedullary Myelopoiesis in Mice. Am. J. Cancer, 37:1-35, 1939.

2. Bateman, J. C. Leukemoid Reactions to Transplanted MouseTumors. J. Nati. Cancer Inst., //.-671-87, 1951.

3. Delmonte, L., and Liebelt, R. A. Granulocytosis-promotingExtract of Mouse Tumor Tissue: Partial Purification. Science,148:521-23, 191.5.

4. Downey, II., Major, S. C., and Noble, J. F. Leukemoid BloodPictures of the Myeloid Type. Folia Haematol., 4/.-493-511,

1930.5. Dunn, T. B. Normal and Pathologic Anatomy of the Reticu

lar Tissue in Laboratory Mice, with a Classification and Discussion of Neoplasms. J. Nati. Cancer Inst., 14:1281-1433,1954.

6. . Morphology of Mammary Tumors in Mice. In: F.Homburger (ed.), The Physiopathology of Cancer, pp. 38-84,New York: Paul B. Hoeber, Inc., 1959.

7. Ebbenhorst-Tengbergen, W. J. P. It. van, and MÃ¼hlbock,O.Disorders of the Blood of Mice Bearing Spontaneous or Transplanted Tumors. Brit. J. Cancer, ^2:81-97, 1958.

8. Engelbreth-Holm, J. Spontaneous and Experimental Leukaemia in Animals. Edinburgh and London: Oliver and Boyd,Ltd., 1942.

9. Fahey, R. J. LTnusual Leukocyte Response in Primary Carcinoma of the Lung. Cancer, 4/930-35, 1951.

10. Hensler, L.: Hohe Leukoytose durch Karinom. Schweiz.Med. Wochschr., 83:1032-34, 1953.

11. Hill, J. M., and Duncan, C. N. Leukemoid Reactions. Am. J.Med. Sci., 20/.-847-57, 1941.

12. Hinshaw, D. B., and Hoxie, H. J. Leukemoid Reactions inCarcinoma. Calif. Med., 71:300-1, 1949.

13. Hueper, W. C. Leukemoid and Leukemic Conditions in WhiteMice with Spontaneous Mammary Carcinoma. Folia Haematol., 52:167-78, 1934.

14. â€” â€”. Megakaryocytosis in White Mice with SpontaneousMammary Carcinoma. Am. J. Med. Sci., Ã•SS.-41-48,1934.

15. Kelsall, M. A. Hematopoiesis in the Spleen of Tumor-bearingHamsters. J. Nati. Cancer Inst., /0.-625-36, 1949.

16. Krumbhar, E. B. Leukemoid Blood Pictures in Various Clinical Conditions. Am. J. Med. Sci., Ã•72.-519-33,1320.

17. Kugelmeier, L. M. LeukÃ¤moide Reaktionen bei Karzinom.Folia Haematol., 53:370-81, 1935.

18. Lang, F. J. Myeloid Metaplasia. In: H. Downey (ed.), Handbook of Hematology, pp. 2105-45. New York: Paul B. Hoeber,Inc., 1938.

19. Lappat, E. J., and Cawein, M. A Study of the Leukemoid Response to Transplantable A-280 Tumor in Mice. Cancer Res.24:302-11, 1964.

20. Lewis, M. R. Myeloid Hyperplasia Brought About in Mice bythe Growth of Dibenzanthracene Tumors and Its Relation tothe Transplantability of the Tumors into Mice of AlienStrains. Am. J. Cancer, 29:510-16, 1937.

21. â€¢ . Myeloid Infiltrations Occurring in the Adrenals ofAnimals Bearing Certain Tumors. Ibid., 30:95-101, 1937.

22. Liebelt, R. A., Lay, M., and Davis, B. Extramedullary He-





matopoiesis Associated with a Transplantable Mouse Mammary Cancer. Proc. Am. Assoc. Cancer Res., 3:130, 1960 (abstract).

23. McKucn, C. S., and Selye, H. Histologie Changes in the Adrenals of Tumor-bearing Rats. Am. J. Mcd. Sci., /S9.-423-24,

1935.24. McFadzcan, A. J. S., Todd, D., and Tsang, K. C. Polycythe-

mia in Primary Carcinoma of the Liver. Blood, J3:427-35,

1958.25. Meyer, L. M., and Rotter, S. D.: Leukemoid Reaction (Hy-

perleukocytosis) in Malignancy. Am. J. Clin. Pathol., IS:218-22, 1942.

20. Parsons, L. D. Leukemia Coincident with and Transmissibleby a Spindle-celled Sarcoma in the Mouse. J. Pathol. Bac-teriol, 40:45-54, 1935.

27. â€” â€”. Blood Changes in Mice Bearing Experimental Sarcomas. Ibid., 43:1-22, 1930.

28. Parsons, L. D., and Warren, F. L. Cellular Changes in theSpleen and Lymph (Â¡lands of Mice Used for Carcinogenic andRelated Experiments, with Special Reference to Giant Cellsof the Spleen. Ibid., 52:305-21, 1941.

29. Paunz, T. Ueber die Rundzellhcrde der Nebenniere. (Ein

Beitrug zur histopathologischen Bedeutung des makrophagen(retikulo-endothelialen) Systems.) Yirchow. Arch. Pathol.Anat. Physiol. Klin. Med., Â£42:138-94,1923.

30. Plonskier, M. Ueber tumorformige (extramedullaere, hetero-tope) subpleurale Knochenmarksherde. Ibid., 277:804-9, 1930.

31. Saleeby, E. R.: Heterotopia of Bone Marrow without Appar-rent Cause. Am. J. Pathol., Ã•.-69-76,1925.

32. Sonnenfeld, A. Leukaemische Reaktionen bei Karzinom. Z.Klin. Med., Ã•Ã•/.-108-1G,1929.

33. Starr, G. F., Stroebel, C. F., Jr., and Kearnes, T. P. Poly-cythemia with Papilledema and Infratentorial Vascular Tumors. Ann. Internal Med., 4Â«:978-8(>,1958.

34. Steigbigcl, N. H., Oppenheim, J. J., Fishman, L. M., and Carbone, P. P. Metastatic Embryonal Carcinoma of the Test isAssociated with Elevated Plasma TSH-like Activity and Hy-perthyroidism. New Engl. J. Med., 27/:345-49, 1904.

35. Waldmann, T. A., and Bradley, J. E. Polycythemia Secondaryto a Pheoehromocytoma with Production of an ErythropoiesisStimulating Factor by the Tumor. Proc. Soc. Exptl. Biol,Med., / 08:425-27, 1901.

36. Wintrobe, M. M. Clinical Hematology, 1180 pp. Philadelphia:Lea & Febiger, 1901.

JANUARY 1900 159



1966;26:149-159. Cancer Res L. Delmonte, A. G. Liebelt and R. A. Liebelt Transplanted Mammary CancerGranulopoiesis and Thrombopoiesis in Mice Bearing

Updated version

http://cancerres.aacrjournals.org/content/26/1/149

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications


http://cancerres.aacrjournals.org/content/26/1/149

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

mailto:[email protected]


granulopoiesis and thrombopoiesis in mice bearing ... · [cancer research 26, 149-159, january...

Documents