relationship of steroid receptor, cell kinetics, and clinical status … · relationship of steroid...

[CANCER RESEARCH 41, 3524-3529, September 1981 ]0008-5472/81 /0041-OOOOS02.00

Relationship of Steroid Receptor, Cell Kinetics, and Clinical Status inPatients with Breast Cancer1

Timothy E. Kute,2 Hyman B. Muss, David Anderson, Kenneth Crumb, Barbara Miller, Debbie Burns, and

Lynn A. Dube

Department of Medicine and the Oncology Research Center, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem. North Caro/ina 27103

ABSTRACT

The fractions of cells in the different phases of the cell cyclewere determined by flow cytometry in 70 human breast tumorsand six human benign breast tissues. This procedure showedthat 44% of the tumors and none of the benign tissues wereaneuploid as determined by mixing experiments using normalperipheral blood as a standard for DNA content per nucleus.The mean percent S-phase fraction (% S) values Â±S.D. for

benign and malignant tissues were 6.9 Â±1.6 and 13.7 Â±6.5,respectively. By our procedure, aneuploid tumors seem to havesignificantly higher mean % S value than do diploid tumors.Breast cancer tissue which contained steroid receptors had amean % S value of 11.3, while those tumors which had neitherthe estrogen nor progesterone receptors had a mean % Svalue of 17.1 (p < 0.01). The estrogen receptor status had abetter inverse relationship to the cell kinetic data than did theprogesterone receptor status. The use of molecular forms ofthe steroid receptor was of some assistance in improving theinverse relationship between cell kinetics and steroid receptorstatus. A trend was observed between lack of steroid receptorsand higher probability of the tumor being aneuploid. From thelimited clinical data, there was no relationship between cellkinetic and aneuploid data with respect to nodal status, meta-

static disease, and menopausal status. The possible use ofthese data is discussed.

INTRODUCTION

Breast cancer has now become the most common cancer inthe United States to afflict women. The incidence continues torise, and it is estimated that there will be 109,000 new casesand 36,000 deaths due to breast cancer in 1980 (5). In spiteof major advances in the chemotherapy and the hormonaltherapy of this disease during the last decade, the death ratehas remained constant, and cure is rare after development ofdistant metastasis.

Recent advances in research have increased our understanding of this illness. The discovery and characterization ofsteroid hormone receptors have enabled the clinician to moreaccurately predict the response to hormonal therapy in patientswith recurrent or metastatic disease (17, 24). In addition, therecent gains in hormonal therapy, including the developmentand successful use of antiestrogen compounds, medical ad-

renalectomy, and improved surgical procedures for endocrine

1Supported by NIH Grant CA26359-01 and the Oncology Research Center-

United Fund intramural grant. Presented at the American Association for CancerResearch. May 1980, San Diego. Calif. (23).

2 To whom requests for reprints should be addressed, at Department ofMedicine, Bowman Gray School of Medicine, 300 Hawthorne Road. Winston-Salem, N. C. 27103.

Received January 19. 1981 ; accepted June 8, 1981.

ablation, have been noteworthy. Several studies have indicatedthat patients with steroid receptor-positive tumors have a more

favorable prognosis and a more indolent course (8, 13, 22);recently, however, other investigators have challenged thisposition (10). Studies have also appeared suggesting that ER3

may be related to response rate and duration in patientsreceiving chemotherapy (2). Initial work indicated that patientswith steroid receptor-containing tumors have a lower response

rate to chemotherapy than do those whose tumors were receptor-negative (2). Studies by other investigators reveal contra

dictory results (12), and recently, retrospective analyses haveindicated that receptor content is probably not related to response rate (10). Other studies in these areas will be neededto determine the role of receptors in predicting clinical course.

Cell kinetic studies have given us a better understanding ofthe biology of breast cancer. Several investigators reportedthat tritiated thymidine uptake was inversely correlated withsteroid receptor activity (20, 27). Meyer and Lee (22) have alsoshown that tritiated thymidine uptake can be related to thedegree of nuclear differentiation.

The development and use of flow cytometry to determine cellcycle kinetics has been a major advance in tumor biology (21 ).Previous methods of cell kinetic analysis using tritiated thymidine labeling methods have been cumbersome, expensive, andtime consuming. Flow cytometry enables the investigator toassay cell cycle parameters precisely and rapidly, since smallamounts of tissue (50 to 60 mg) with as few as 105 cells can

give an accurate DNA histogram. Flow cytometry also has theadvantage of producing precise data analysis when comparedto the tritiated thymidine uptake studies, which require subjective assessment in the actual counting procedure. In addition,flow cytometry offers the investigator the ability to sort cellsaccording to DNA content. An excellent review describing themethodology and application of flow cytometry has recentlybeen published (19). The disadvantages of flow cytometrymeasurements of cell cycle kinetics are lack of histologicalidentification of individual cells analyzed and sampling selectivity of cell population from tumor tissue. The present study wasdesigned to determine the relationship of steroid receptorcontent to cell kinetics as defined by flow cytometry studies. Italso includes limited clinical data which were compared withcell kinetics and steroid receptor activity.

MATERIALS AND METHODS

Steroid Receptor Assay. Primary and metastatic breast cancertissues were received between September 1978 and February 1980from the North Carolina Baptist Hospital, Gaston Memorial Hospital(Gastonia, N. C.), and Cleveland Memorial Hospital (Shelby, N. C.).

3 The abbreviations used are: ER, estrogen receptor; % S, percentage of S-

phase fraction; PI, proliferative index; PR, progesterone receptor.

3524 CANCER RESEARCH VOL. 41

Research. on February 11, 2021. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Tissue samples were frozen immediately after surgery and were storedat -85Â°. Approximately 0.5 to 1.0 g of tissue was homogenized, and

the cytosol was prepared as described previously (15). The receptorbinding activities were determined either by sucrose density gradientassay (85% of patients) or the dextran-coated charcoal assay (15% of

patients). The methods used for these assays have been reportedelsewhere and utilize a computer program (15).

Preparation and Staining of Nuclei. The preparation of the nuclearfraction of the tumor tissue for use in flow cytometry was performed asfollows. The frozen tissue was removed, cut into small pieces, andadded to Buffer A (0.5 M sucrose: 1.5 mM MgCI2:40 rriM Tris, pH 7.4).The ratio of tissue to Buffer A was approximately 1:20. This tissue wasthen homogenized with a 4- to 5-sec burst of the Polytron (Brinkmann

Instruments, Inc., Westbury, N. Y.) at very slow speeds and wascontinued until most of the solid tissue was completely eliminated. Thissuspension was then successively passed through 40-, and 100-, and200-wire mesh strainers to remove connective and nonhomogenizedtissue. The nuclear suspension solution was then placed in an HS-4

rotor (Ivan Sorvall, Inc., Newtown, Conn.) and centrifuged for 10 minat 1000 rpm (192 x g), and then the pellet was concentrated into onetest tube and washed once by centrifugation. After the nuclear pelletwas suspended in 1 to 2 ml of Buffer A, the number of nuclei per mlwas determined in an electronic counter (Model ZB Coulter Counter;Hialeah, Fla.). A nuclear suspension of 1 x 106 nuclei/ml was made

with a propidium iodide staining solution as described by Krishan (14).This staining solution contains RNase (3 units/ml) for removal of theRNA-propidium iodide fluorescent interaction and 0.1% Nonidet P-40

to remove the cytoplasmic debris from the prepared nuclei (14). After10 min of incubation at 0Â°,the stained nuclei were passed through a

70-/xm-pore-size mesh for removal of any further debris. The stained

nuclei were examined prior to flow cytometry to check for adequatedispersal of nuclei. This stained solution was then analyzed by theCoulter TPS-1 fluorescent cell sorter. In several patients, the DMA

histograms were obtained from fresh and frozen specimens of thesame tumor. No differences were noted.

Analyses of DNA Histograms. The cells were excited at 488 nmusing an American Standard laser, and the fluorescence was analyzedwith the green photomultiplier which included the various necessaryfilters for removing laser excitation and nonfluorescent light. The analysis of the DNA histogram was continuously formed until one channelreached a maximum of 10,000 cells. The total amount of cells analyzedvaried depending on the DNA histogram profile.

A representative sample of a DNA histogram done by the aboveprocedure on various tumors and on normal tissue is shown in Chart 1.Chart ^A represents the DNA histogram obtained from normal peripheral blood leukocytes. One sees a predominant peak at the 2n region(diploid) which corresponds to the cells in the Gt (or G0) phase of thecell cycle. There are very few nuclei which contain greater fluorescence(i.e., DNA content) than this peak, indicating that there are few to noDNA-synthesizing cells (S phase) or cells ready for division (G2). The

nuclei in the G2 and/or mitotic (M) phases would have a normaldistribution of fluorescence twice that of the G, nuclei, but one cannotanalyze the M-phase cells in the DNA histogram by this method. S-

phase nuclei would have fluorescence between these 2 peaks.A DNA histogram of a benign fibrocystic tumor is shown in Chart 1ÃŸ.

One observes that the predominant type of cell in this tissue is also aresting cell in the G, phase of the cell cycle. There are small populationsof cells in the S and G2 phases. In Chart 1C, one observes a DNAhistogram obtained from breast cancer tissue. There are 2 dominantpeaks which correspond to the Gi-phase of diploid cells and the G,

phase of tumor cells that contain an increased DNA content (aneuploid).In order to determine which peaks were aneuploid or diploid, a mixingexperiment as described by Barlogie ef al. (4) was performed and isshown in Chart 1D. With this method, normal DNA-containing stained

nuclei from peripheral blood cells (Chart 1/4) were mixed with thestained tumor nuclei. The normal nuclei enhanced the diploid peak anddiminished the aneuploid population. A large concentration of cells in

Cell Kinetics to Steroid Receptor and Clinical Status

lorA c ,1

I m$

râ€”12N 4N 2N 4N

DNA Content (Relative)

Chart 1. DNA histograms of nuclei from several sources. A, normal peripheralblood; B, benign fibrocystic breast tumor; C, malignant breast tumor with aneu-ploidy; D. mixture of nuclei obtained from normal peripheral blood and malignantbreast tumor of C. The 2n and 4n peaks occurred at Channels 30 and 60,respectively. The GÃ¬aneuploid peak (/) and G2 aneuploid peak (//) occurred atChannels 51 and 101, respectively.

the S and G2 phases for the aneuploid cells is also seen in Chart 1C,which is different from Chart 1, A and B.

DNA histograms were recorded on a Texas Instruments Silent 700(Texas Instruments, Houston, Texas), and the data were stored onmagnetic tapes. The analysis of the DNA histogram data was performedusing a program developed by Fried ef al. (7), which determines thepercentage of cells in each phase after curve-fitting the raw data. In all

the tumors analyzed, the fit of the raw data corresponded with the fit ofthe curves defined by the Fried program. PI in this study is defined bythe equation

PI =S + G,

x 100G, + S + G2

\n the case of aneuploid tumors or histograms showing 2 separate G,peaks, the cell kinetics data were determined only for the aneuploidsubpopulations of the tumor. Only DNA histograms on tumor tissuewith clear separation of diploid and aneuploid peaks were used in thisstudy.

Pathological Evaluation. All tissue samples were histologically reviewed to confirm the presence of malignant tissue. Portions of thesample used for steroid receptor and cell kinetic analysis were submitted for further histological confirmation. Histological and cytologicalevaluation used to group tumors by variables relating to differentiation,proliferative activity, and multifactorial correlation with these assayswill be the subject of a separate paper.

Computed Analysis of Generated Data. The data obtained from thetissue for steroid receptor results (molecular form and capacity), cellkinetic analyses (PI, % G,, % S, % G2), presence of anueploid cells,and clinical data (age, lymph node status, and site of tumor) wereobtained and stored on the General Electric DMS II data managementsystem.

RESULTS

An analysis of clinical and biochemical data for breast cancer

SEPTEMBER 1981 3525



T. E. Kufe ef al.

tissues was performed with objective computer programs andstandardized definitions (see "Materials and Methods") which

were designated prior to the analysis of the data. A correlationcoefficient of 0.84 was obtained when comparing the % Sactivity with the PI activity (Chart 2/4). A correlation coefficientof only 0.60 was obtained for the comparison of PI with thepercent G2 fraction activity (Chart 2ÃŸ).Because of the relativelyhigh positive correlation between PI and % S, it was decided touse % S activity in most of the analysis as it is probably a moreimportant biological parameter relating to cell kinetic activity,and since it would allow us to compare our data with similarstudies using tritiated thymidine in breast cancer (20, 26, 27).At the present time, we have not used G2 activity alone asdescribed by Irvin and Bagwell (11) for cell kinetic analysis.

In the 70 breast tumors analyzed for cell kinetics using flowcytometry, a wide distribution of % S activity was noted (Chart3). There is a steady decrease in the distribution of patients as% S activity increases. The mean value of % S activity was13.7 Â± 6.5 (S.D.). In a series of 6 fibroadenomas obtainedfrom Dr. Kenneth McCarty, Jr. (Duke University) and analyzedin our laboratory, the % S activity was 6.9 Â±1.8. None of thesetumorous tissues, along with several normal breast tissues,showed any aneuploid activity. In the 70 breast cancers studied, aneuploidy was noted in 44%. The distribution of PI and% S values for diploid and aneuploid tumors is shown in Chart4. Values for diploid tumors were 10.4 Â±5.5 for % S and 19.0Â±7.1 for PI. Values for aneuploid tumors were 17.9 Â±5.1 for% S and 27.1 Â±6.2 for PI. Using t tests, the mean values forPI and % S are significantly different between the aneuploidand diploid groups to a p < 0.01 value. In 103 DNA histogramsdone on breast tumors in our laboratory, only one tumor hasbeen proven to be hypodiploid as determined by mixing experiments. The average coefficient of variation for the GÃ¬and G2peaks in the DNA histogram was between 3 and 7% as determined from the Fried program. The DNA histogram for peripheral blood always gave a lower coefficient of variation for the

30

25

t 20

^ 15

tf.

10

20

t> 15Ho< 10

10 20 30

PROLIFERATIVE INDEX

40

1is?NUMBEROFi3191815II4

0-5 6-10 11-15 16-20 21-25 26-30

% S

Chart 3. Distribution of % S activity in the 70 tumors analyzed in this study.Range, 2 to 27; mean % S activity, 13.7 Â±6.5.

4032XLUQ?

24LUHOCLUEL0

16Â£0-8DIPLOID

( 56%) ANEUPLOID(44%)PI

%S PI%S9

â€¢â€¢MÂ«

â€¢$â€¢*

- â€¢;Â«**Â»

.0 *** i ' 0'â€¢â€¢

OOâ€¢â€¢* 0

â€¢â€¢ O ;50

OOâ€¢â€¢â€¢Â«; 0

â€¢â€¢â€¢O â€¢â€¢OOâ€¢â€¢â€¢oooo

â€¢â€¢ o â€¢ooâ€” Â«o*â€¢*â€¢000 :0*Ã¬*ÃŒioiiocPooÂ»

0OOO Â» : Oâ€¢ i: OOO i iOOioooooiÂ«â€¢

i oo;ooo:.

\^Â°\oi00i0o,,.....,,

L :. :,-. limaliâ€”â€”4032m24

5CO'S168

Chart 2. Relationship of PI to % S and % G2 activity obtained from DNAhistogram analysis. A, PI compared with % S activity; B, PI compared with % G2activity.

Chart 4. Relationship between % S and PI activity of diploid and aneuploidtumors.

G, phase nuclei than did the DNA histograms for the tumortissue.

Using a threshold level of 10 fmol/mg of protein as the valuefor a steroid receptor-positive tissue, overall analyses of thedistribution of ER- and PR-positive tumors were 54 and 45%,respectively. Further distribution into ER-PR categories isshown in Table 1. The ER-positive-PR-positive group is ap-proximatey equal in number to the ER-negative-PR-negativegroup, while the ER-positive-PR-negative and ER-negative-PR-

positive are significantly smaller in number. The distribution isin good agreement with previous investigations (17, 18) andwould indicate that the 70 tumors analyzed are a good samplingof breast cancer tissue as related to steroid receptor analysis.

A comparison of the steroid receptor status with cell kineticsand % aneuploidy was performed with the data in Table 1 usingan analysis of variance. There is evidence of a significantdifference between the mean values of PI and % S betweenER-positive-PR-positive and ER-negative-PR-negative tumors(p < 0.01). The mean values of % S for the ER-positive-PR-





Table 1

Comparison ol steroid receptors to cell kinetics and to aneuploidy

ReceptorstatusER-positive-PR-positive

ER-positive-PR-negativeER-negative-PR-positiveER-negative-PR-negativeNo.

ofpatients24

(34)"

14 (20)7(10)

25 (35)Proliterative

index19.6Â±6.66

21.1 Â±7.423.3 Â±9.727.1 Â±7.7%S

activity11.3

Â±5.410.9 Â±5.415.6 Â±7.617.1 Â±6.2%

aneu-ploid38.0

36.043.052

Numbers in parentheses, percentage.0 Mean Â±S.D.

negative group and ER-negative-PR-negative group are also

significantly different (p < 0.01). However, there is no significant difference at the p < 0.05 level between the ER-negative-PR-positive and the ER-negative-PR-negative group with respect to PI or % S. The percentages aneuploid of ER-positive-PR-positive and ER-positive-PR-negative were not significantlydifferent in occurrence as determined by x2- However, the

trend is for % S, PI, and % aneuploidy to increase as thetumors go from ER-positive-PR-positive to ER-negative-PR-positive to ER-negative-PR-negative. The range of % S for

each subgroup of steroid receptor status is large. In this study,the ranges of % S values for ER-positive and ER-negative

tumors are 2 to 23 and 4 to 27, respectively. The ranges of% S values for PR-positive and PR-negative tumors are 2 to 23

and 4 to 27, respectively.Tumors were categorized as to ER positive, ER negative, PR

positive, and PR negative and related to % S and % aneuploidy(Table 2). Again using t tests, a statistically significant difference was noted in the % S mean value of ER-positive tumors(11.1 Â± 5.2) versus ER-negative tumors (16.8 Â± 6.4; p <

0.01). There was, however, no significant difference betweenthe % S mean values of the PR-positive and PR-negative

tumors.Since 85% of the tumors were analyzed by sucrose density

gradients, the molecular forms of the steroid receptor could beobtained and compared with the cell kinetic data. The resultsof this comparison are summarized in Table 3. The mean % Svalue is lower in tumors which contain both 4S and 8S estrogenreceptors than in tumors which have only the 4S estrogenreceptor. If the tumor is negative for both 4S and 8S molecularforms, as defined by having less than 10 fmol of ER per mg ofprotein, the mean % S value is significantly higher. The mean% S value is lowest in tumors which have only the 8S molecularform of the PR. There is an increase in the mean % S value oftumors which have both forms of PR, but the highest % S isseen in patients without receptor. The tumors which have onlythe 4S PR have a mean % S value not significantly differentfrom those tumors which contain neither the 8S nor 4S molecular forms of this steroid. The importance of steroid receptorsand their molecular forms will be discussed later.

The relationship of cell kinetics and % aneuploidy was correlated with the clinical status of the patient (Table 4). Onlypatients who had 5 or more nodes examined at the time ofprimary treatment were included in this evaluation. Analysis ofvariance revealed no significant difference in kinetic parameters between the groups with respect to positive nodal statusor metastatic disease. The number of positive nodes and met-astatic disease was also compared with receptor status, andno statistically significant differences were observed. This is inagreement with other reported studies (24).

Finally, the cell kinetic data and % aneuploidy were compared between 20 premenopausal (<50 years old) and 50postmenopausal (>50 years old) patients. The % S values were15.1 Â±6.7 and 13.2 Â±6.4 for premenopausal and postmenopausal patients, respectively. Aneuploidy was 50% in thepremenopausal patients and only 42% in the postmenopausalpatients. Neither of these differences is statistically significant.

DISCUSSION

This study was designed to measure several biochemicaland clinical parameters of breast cancer and compare themwith cell kinetic activity. Our laboratory has also compared cellkinetic data from mouse mammary tumors obtained by flowcytometry and autoradiography (6). Flow cytometry on mousemammary tumors is faster, is easier to perform, and is achemical analysis, while autoradiography is time consuming, istedious, and requires a biologically active system. Flow cytometry has the advantage of accurate and reproducible quan-

titation of the total cell cycle but does not supply the histologicalcharacterization of tumor cells analyzed in the tissue slice.Preliminary results on our mouse mammary tumors comparingDNA histograms and autoradiography profiles indicated thatflow cytometry consistently gave a higher % S value (6). Furtherstudies are being planned to compare these 2 methods of cellcycle analysis for human breast tissue.

In this present study, our major concern is the sampling oftumor nuclei for analysis. It is quite probable that normal nucleiwere analyzed with the tumor nuclei. This problem exists whenever one performs biochemical assays on a tissue sample.

Table 2

Comparison of ER and PR status to % S and aneuploidy

ReceptorstatusER-positive

ER-negativePR-positivePR-negativeNo.

ofpatients38

323139%S11.1

Â±5.2a

16.8 Â±6.412.2 Â±6.314.8 Â±6.7%

aneuploid3750

3949

" Mean Â±S.D.

Table 3

Comparison of molecular forms of steroid receptors with % S

Molecularforms84S

alone8Salone4S

+8SNeitherformNo.

ofpatients712033ER%S12.6

Â±6.26610.1

Â±5.116.3Â±6.6No.

ofpatients731442PR%S13.6

Â±5.79.7Â±4.010.9Â±6.714.7Â±6.7

a Determined by sucrose density gradient assay where S is equal to Svedberg

coefficient.b Mean Â±S.D.

Table 4

Comparison of clinical status to cell kinetics and aneuploidy

Nodalstatus01-3>4Metastatic

diseasePI22.4

Â±8.2a20.3

Â±9.326.4Â±8.422.8

Â±7.0%S13

Â±6.810.8Â±6.515.9Â±6.713.5

Â±6.5%

aneuploid43254745No.

ofpatients2181211a

Mean Â±S.D.

SEPTEMBER 1981 3527



T. E. Kufe ef al.

Other investigators have shown a relationship among moredetailed cellular and nuclear characteristics, receptor activity,and cell kinetic data (3, 16). We are presently trying to confirmthese data.

The distribution of % S activity for the 70 tumors (Chart 3) isvery wide and does not seem to be in a lognormal distributionas described by other investigators (22). The % S activity forbenign breast disease is significantly different when comparedwith breast cancer tissue. In the limited number of benign andnormal tissues, none of the DMA histograms obtained showedaneuploid characteristics. However, the 70 breast cancer tissues studied here showed a 44% aneuploidy. This value issignificantly lower than the 75 to 78% aneuploidy obtained byBarlogie's group (4, 25). The reason for this discrepancy is

unknown at this time but could be due to methodology andequipment used, or sampling variability. A total of 103 tumorswere analyzed in this study. Thirty-three of these tumors were

insufficient for complete analysis of the parameters used in thisstudy. Of these 33 tumors, 28 were aneuploid, giving a totalpercentage of aneuploids in this study of 57%. Of these 28tumors, 7 had overlap of the diploid and aneuploid peaks,precluding histogram analysis. The remaining 70 patients formthe basis of this report.

The relationship between PI and % S activity obtained fromdiploid and aneuploid tumors is significantly different (Chart 4).This would agree with the general hypothesis that more aneuploid tumors are usually more poorly diffentiated histologicallyand usually faster growing. It is possible that a systematic errorin the calculations could have been obtained from overlappingof the diploid histogram with the aneuploid histogram. At present, there are no computer programs available to correct forthis possible error. The tumors analyzed which did not give agood separation between the diploid cells and the aneuploidcells were removed from any further anaysis. It is also possiblethat the diploid histogram has a larger percentage of noncan-

cerous cells which do not divide as rapidly. Finally, we plan touse vital DNA stains to sort out the different cell populations byDNA content using the capability of the fluorescence-activated

cell sorter. This might answer our sampling problem and giveus the potential to sort out monoclonal cells with aneuploidDNA content.

Our data support a firm relationship of steroid receptor statusand cell kinetic parameters as defined by flow cytometry (Tables 1 and 2) and are in agreement with reports from otherinvestigators (20, 25, 27). A greater PI, % S activity, and morefrequent aneuploidy is associated with low steroid receptoractivity and supports the hypothesis that steroid receptor activity is indirectly related to the proliferative capacity of breastcancer. Although the correlation between receptor activity andcell kinetic parameters is strong, considerable overlap hasbeen noted. Further studies, will, hopefully, define whether PIor % S activity is related to response rates and duration inpatients undergoing therapy. Protocols which use these datain a prospective manner are currently being developed at ourinstitution.

Our data also indicate that the ER status is more predictiveof lower % S activity than the PR status (Table 2). This iscontrary to the general view that, since PR is a product of ERaction, it should be more predictive of estrogen-sensitive cells.However, the present data by Knight ef al. (13) and Meyer andLee (22) indicate that ER-positive tumors are slow growing and

that the patient usually has a better prognosis compared to thepatients with the faster growing ER-negative tumors who have

a poorer prognosis. Allegra eÃal. ( 1) have shown no relationshipbetween prognosis of the patient and the presence of PR. Thisagrees with our data.

Patients whose tumors contain the 8S molecular form of thereceptor have been shown to have a higher probability ofresponse to hormone therapy (28). The relationship betweenthe molecular forms of the steroid receptor and cell kinetics(Table 3) indicates that tumors which contain the 8S molecularspecies have a lower mean % S activity than do those whichhave only the 4S molecular form or neither molecular form. Itshould be noted, however, that the number of tumors in themolecular species categories is low and that there is widedistribution of % S activity in each category. Therefore, arelationship of cell kinetic activity and response to hormonetherapy might exist through the use of molecular forms, butmore analyses are needed.

A direct relationship of the actual capacity of steroid receptors (fmol/mg of protein) and the % S value was not observed.This has also been studied by Silvestrini ef al. with similarresults (27). Raber ef al, (25) have found an inverse relationshipbetween ER content and proliferative activity. There are severalreasons why a direct relationship of this type may not exist, (a)An exact relationship between molecules of receptors per celland biological activity has never been shown. If a threshold ofreceptors per cell is reached, it should be active as an estrogen-sensitive cell. Any more receptor molecules per cell would

not enhance this biological activity. Therefore, the steroidreceptor concentration would not be the limiting factor forsteroid sensitivity or biological activity, (b) The assay of fmol/mg of protein is generally an inaccurate estimation of moleculesof receptor per cell because of methodological considerations.These include dissociations of radioactive steroids from thereceptor, nonsaturation of receptor, and use of mg of proteinas a standardization procedure. It has been estimated that themg of "cytosol" protein observed in a tumor extract can be

contaminated with as much as 45% albumin obtained fromblood and interstitial fluids (9). It is possible that cell kineticdata can be related to receptor content, but new methodologiesmust be developed.

The lack of correlation of cell kinetic data with age, nodalstatus, and metastatic disease is noteworthy. Our sample sizewas not large enough to detect small but possibly significantdifferences. In addition, the fact that this was a retrospectivestudy may have biased our results in these analyses. Prospective studies which contain prognostic variables related to tumorsize and more detailed histological characteristics will be necessary to adequately determine any relationship that mightexist.

Flow cytometry as used in this study may prove useful as atool in both the basic and clinical areas. It will be important todetermine the role of cell kinetics in treatment planning and inthe natural history of the disease and to determine whetherkinetic data supply important independent clues that cannot beobtained by standard methods of evaluation.

ACKNOWLEDGMENTS

We acknowledge the members of the surgery and oncology staffs at BowmanGray School of Medicine, especially Dr. Robert Cooper who initiated the cell





kinetic analysis work, and Dr. Leonard Rhyne who helped transfer the Friedprogram to our computer. We would also like to thank Dr. Ken McCarty, Jr., forsupplying the benign breast tissues. Finally, we would like to thank Dr. RÃ¼sselHilf and Dr. Richard Marshall, who gave constructive comments on this manuscript.

REFERENCES

1. Allegra. C. J., Lippman, M. E., Simon, R., Thompson, E. B., Barlock, A.,Green. L., Huff. K. K., Hoan, M. T. D., Aithen, S. C., and Warren, R.Association between steroid hormone receptors status and disease-freeinterval in breast cancer. Cancer Treat. Rep., 63. 1271-1277, 1979.

2. Allegra, J. C., Lippman, M. E., Thompson, E. B., and Simon, R. An association between steroid hormone receptors and response to cytotoxic chemotherapy in patients with metastatic breast cancer. Cancer Res., 38. 4299-4304, 1978.

3. Antoniades, K., and Spector. H. Correlation of estrogen receptor levels withhistology and cytomorphology in human mammary cancer. Am. J. Clin.Pathol., 71: 497-503. 1979.

4. Barlogie. B., Gohde, W., Johnston, D. A., Smallwood. L., Schumann, J.,Drewinko. B.. and Freireich, E. J Determination of ploidy and proliferativecharacteristics and human solid tumors by pulse cytophotometry. CancerRes.. 38 3333-3339. 1978.

5. Cancer Facts and Figures 1980, p. 8. New York: American Cancer Society.1979.

6. Crumb, K.. Kute. T. E., Johnson, C.. and Witcofski, K. Cell kinetic analysisof mouse mammary tumors. Med. Pediat. Oncol., 9. 95. 1981.

7. Fried. J., Perez, A., and Clarkson, B. D. Rapid hypotonie method of flowcytofluorometry of monolayer cell cultures. J. Histochem. Cytochem.. 26921-933, 1978.

8. HÃ¤hnel, R.. Woodings. T., and Vivian, A. B. Prognostic value of estrogenreceptors in primary breast cancer. Cancer (Phila.), 44: 671-675. 1979.

9. Hess, P., Feldhoff, R. C., Kute, T. E., and Wittliff, J. L. Albumin content ofhuman mammary tumor cytosols: possible influence of estrogen receptorassays. In: Proceedings of Southeastern Cancer Research Association.Sixth Annual Meeting. Kiawah Island, S. C.: Southeastern Cancer ResearchAssociation, 1978.

10. Hilf, R., Feldstein, M. L., Gibson, S. L., and Sovlov, E. D. The relativeimportance of estrogen receptor analysis as a prognostic factor for recurrence or response to chemotherapy in women with breast cancer. Cancer(Phila.), 45: 1993-2000, 1980.

11. Irvin, G. L., and Bagwell, C. B. Identification of histologically undetectablehyperplasia by flow cytometry. Am. J. Surg., Â»38.567-571, 1979.

12. Kiang, D. T., Frenning, D. H., Goldman, A. I., Ascensao. V. G., and Kennedy,B. J. Estrogen receptors and responses to chemotherapy and hormonaltherapy in advanced breast cancer. N. Engl. J. Med., 299. 1330-1334,

1978.13. Knight. W. A., Livingston. R. B., Gregory, E. J., and McGuire, W. L Estrogen

receptor: an independent prognostic factor for early recurrence in breastcancer. Cancer Res.. 37 4669-4671, 1977.

14. Krishan. A. Rapid flow cytofluorometric analysis of mammalian cell cycle bypropidium iodide staining. J. Cell Biol., 66. 188-193, 1975.

15. Kute, T. E., Huske, M. S., Shore, A., and Rhyne, A. L. Improvements insteroid receptor assays including rapid computer analysis of data. Anal.Biochem.. Â»03.272-279. 1980.

16. Maynard, P. V.. Davies. C. J., BlÃ¢mer,R. W.. Elston, C. W., Johnson. J., andGriffith. K. Relationship between estrogen receptor content and histologicalgrade in human primary breast tumors. Br. J. Cancer. 38. 745-748. 1978.

17. McGuire. W. L., Carbone, P. P.. and Vollmer, E. P. (eds.). Estrogen Receptors in Human Breast Cancer. New York: Raven Press. 1975.

18. McGuire, W. L., Raynard. J. P., and Baulieu, E. E. (eds.). ProgesteroneReceptors in Normal and Neoplastic Tissues. New York: Raven Press, 1977.

19. Melamed. M. R.. Mullaney. P. F., and Mendelsohn, M. L. Flow Cytometryand Sorting. New York: John Wiley & Sons, Inc., 1979.

20. Meyer, J. S., Bauer, W. C., and Rao, B. R. Subpopulations of breastcarcinoma defined by S-phase fraction, morphology, and estrogen recetporcontent. Lab Invest., 39. 225-235, 1978.

21. Meyer, J. S., and Conner, R. E. In vitro labeling of solid tissues with tritiatedthymidine for autoradiographic detection of S-phase nuclei. Stain Technol..52. 185-195, 1977.

22. Meyer, J. S., and Lee, J. Y. Relationships of S-phase fraction of breastcarcinoma in relapse to duration of remission, estrogen receptor content,therapeutic responsiveness, and duration of survival. Cancer Res.. 40:1890-1896, 1980.

23. Muss, H. B., Kute, T. R.. Cooper, M. R., and Marshall, R. C. Correlation ofestrogen and progesterone receptors with DMA histograms in patients withadvanced breast cancer. Proc. Am. Assoc. Cancer Res.. 21: 172, 1980.

24. National Institutes of Health Consensus Development Conference. SteroidReceptors in Breast Cancer. Washington. D.C.: National Institutes of Health.1979.

25. Raber. M., Barlogie, B., Latreille, J., Fu, C. T.. and Pritsche. H. Proliferativeactivity and ploidy abnormality in human breast cancer. In: Proceedings ofthe Cell Kinetic Society, Seventh Annual Meeting, p. 35. Lexington, Ky.: CellKinetic Society, 1980.

26. Schiffer, L. M., Brawnschweiger, P. G.. Stragand, J. J.. and Poulakos. L.The cell kinetics of human mammary cancers. Cancer (Phila.), 43: 1707-

1719, 1979.27. Silvestrini. R., Diadone, M. G., and DiFronzo, G. Relationship between

proliferative activity and estrogen receptors in breast cancer. Cancer(Phila.), 44. 665-670, 1979.

28. Wittliff, J. L.. Beatty, B. W., Baker. D. T., Jr., Savlov, E. P., and Cooper, R.A., Jr. Clinical significance of molecular forms of estrogen receptor in humanbreast cancer. Res. Steroids. 7. 393-403, 1977.

SEPTEMBER 1981 3529



1981;41:3524-3529. Cancer Res Timothy E. Kute, Hyman B. Muss, David Anderson, et al. Status in Patients with Breast CancerRelationship of Steroid Receptor, Cell Kinetics, and Clinical

Updated version

http://cancerres.aacrjournals.org/content/41/9_Part_1/3524

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

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/41/9_Part_1/3524To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



relationship of steroid receptor, cell kinetics, and clinical status … · relationship of steroid...

Documents