Influence of chemotherapy on hormone receptor concentration in a xenotransplanted endometrial cancer

European Journal of Obstetrics & Gynecology and Reproductive Biology, 45 (1992) 131-138 0 1992 Elsevier Science Publishers B.V. All rights reserved 0028-2243/92/$05.00

EUROBS

131

01361

Influence of chemotherapy on hormone receptor concentration in a xenotransplanted endometrial cancer A. Vering

a, R.Th. Michel

a, M. Mitze b, M. Stegmiiller

a and H.G. Bender

a

aDepartment of Gynecology and Oncology, Main Vniuersi&, Medical Center, Frankfurt and ’ Department of Obstetrics and Gynecology, University Medical Center, Mainz, FRG Accepted

for publication

10 January

1992

Summary There is growing evidence that chemotherapy may influence the hormone receptor capacity in human carcinomas. The consideration of this assumption may be of importance for the therapeutic management of tumors with metastatic spread which underwent previous adjuvant chemotherapy. Therefore we investigated the influence of six different kinds of chemotherapy on the hormone receptor concentration and the percentage of receptorpositive cells in xenotransplanted endometrial cancer. Our results can be summarized as follows: (1) We find neither a significant decrease in hormone receptor capacity after chemotherapeutic treatment (biochemical determination), nor do we see a decrease in the percentage of ER/PR pos. cells (immunohistochemistryy). (21 On the other hand, there is no increase in hormone receptor concentration induceable by chemotherapy and no increase in ER/PR pos. cells immunohistochemically. Endometrial

cancer;

Chemotherapy;

Estrogen

Receptor;

Progesterone

Introduction The potential influence of chemotherapy on the steroid receptor concentration (in hormonedependent cancers) is controversially discussed in literature. There is some evidence that cytotoxic therapies may influence the hormone receptor concentration in human carcinomas. This is of great importance to the therapeutic management of metastatic disease after adjuvant chemother-

Correspondence to: Dr. med. A. Vering, UniversitatsFrauenklinik, Abt. fiir Gynakologie und Onkologie, TheodorStern-Kai 7, D-6000, Frankfurt/Main.

Receptor

apy. An investigation on xenotransplanted breast carcinomas reports a significant decrease in estrogen receptor (ER) concentration after 3-5 weeks of high-dose chemotherapy [19,20]. Lippman [17] and Levine [18] assumed that a selection of ER pos. cells by cytotoxic therapy takes place. ER pos. tumor cells might proliferate more slowly and are therefore potentially more resistant to chemotherapy. This would be a way to enhance the steroid receptor capacity of a tumor. In some clinical studies there is evidence, that receptor positivity of metastases can be increased by intermittent chemotherapy when compared to primary tumors [1,11,12]. Other clinical studies with sequential receptor determinations found no influence of intermittent chemotherapy [8,10,21].

132

The above-mentioned controversial results suggest to demonstrate the influence of chemotherapy on receptor capacity in the standardized model of nude mice bearing human endometrial-cancer xenografts. We used two methods, biochemical assay and immunohistochemistry, in our study, both established for clinical routine analyses of receptor capacity.

the specific chemotherapy was applied as a single injection intraperitoneally. The growth development of the tumors was measured for a period of 14 days after initial chemotherapy. The intraperitoneal injection was chosen because of its easy applicability in nude mice. The body surface of human individuals is 300 times larger than the body surface of nude-mice. By dividing the dosages of single agents in standard polychemotherapeutic regimen by factor 300 we get the basis on which we tested the optimal dosage for these kinds of experiments, which should induce a no-change behavior of the tumors. The following dosages were applied to six different treatment groups as single injections per each animal. The drugs were diluted in 5 ml 0.9% NaCI. 1. 2.5 mg cyclophosphamide 2. 50 lug bleomycine 3. 500 pg epirubicine 4. 70 pg mitomycine

Materials and Methods The endometrial cancer was a poorly differentiated, postmenopausal adenocarcinoma of the uterus (grading III). It was serially xenotransplanted in many passages; its ER and PR concentration varied between 50 and 250 fmol/mg in the different passages (castrated animals). The tumors grew on female animals and the longest diameter of the tumor transplants and its perpendicular was measured once a week. After 4 weeks the tumors reached a size of about 1 cm’. Then

40,, tumordiameterin mm*

r-----CYC lophos-

I

I

phamiden=12

300-

200-

100-controls n=15

0’

I

I

12

3

4

5

therapy Fig. 1. Tumor

growth

of endometrial

cancer

6 weeks

chemotherapy

vs. controls.

133

5. 4 mg carboplatinum

6. 5 mg fluorouracil In four of the six treatment groups (carboplatinum, bleomycine, epirubicine, endoxane) the number of the animals was that high that we were able to kill some mice starting on day 7 to 12 after chemotherapy in order to measure receptor concentration. This did not at all influence the determination of the therapeutic effect on tumor growth. We put together the data of this time period in each of the four treatment groups for statistical evaluation. In two treatment groups only one animal was killed for immunohistochemistry. Tumor size was measured 14 days after chemotherapy in all animals still alive in our six treatment groups. Then all the animals were killed in the time period from day 14-16. In order to accumulate enough data for statistical evaluation, we collected the data from this observation period together in all therapy groups. The tumor tissue was extracted (l-3g), frozen in liquid nitrogen and kept at -70°C. The biochemical recep-

tor analysis was performed by the DCC method [6,7] with a Scatchard plot [24]. ER-ICA and PR-ICA served as a method for the immunohistochemical receptor determination (Abott laboratories). The percentage of receptor pos. tumor cells was determined by microscopy on the entire histologic specimen. The results were expressed in rank percentages (O%, -lo%, -20% ... - 100%). Necrotic areas were neglected. We did not use the staining intensity for interpretation because of methodological problems. The Kruskal-Wallis Test [15,16] with multiple comparisons [5] and corrections [9] served for statistical evaluation in case of biochemical assay. Statistical analysis of our immunhistochemical results proved not to be useful due to methodological reasons. Results

The tumor growth of the xenotransplanted human endometrial cancer in six treatment groups vs. controls is shown in Fig. 1. The tumors grew

350fmol/ms 3001

T

250

Itiyrouramitomrcine Pzlwcwc~~~o5-

200 150

EX8eWhitine

100

bleomycine ZkarboPlatinum

50

ZEEEcontrols

0 14-16

days Fig. 2. Biochemical ER concentration after chemotherapy in an endometrial cancer.

134

for 4 weeks, than chemotherapy was applied in all of the treatment groups. The therapeutic effects could be measured during a time period of 14 days. With 4 mg carboplatinum and 70 ,ug mitomycine and 2.5 mg cyclophospamide, the tumors showed a no-change status and remained at the size of 1 cm’. After treatment with 5 mg fluoruracil and 500 pg epirubicine, tumor growth decreased to about 65% of the controls [p 0.011. Tumor growth decreased only slightly in the first week with 50 pg bleomycine. The influence of six chemotherapies with regard to the biochemically determined ER concentration in our endometrial cancer is demonstrated in Fig. 2. In four of the six therapy groups the ER concentration could be measured from day 7 to 12 after a single injection of chemotherapy (carboplatinum, bleomycine, epirubicine, cyclophosphamide). 7-12 days after treatment the mean ER concentration in the four treatment groups did not differ significantly from our controls. In all six groups (including our experiments with mitomycine and flourouracile) the ER was

7-12

determined 14 to 16 days after therapy. Compared to our controls (x = 166 fmol/mg protein) the mitomycine treated groups showed a decrease (x = 70 fmol/mg protein) in this observation period, although this difference did not prove to be significant. In our study, mitomycine is the most effective agent on tumor growth. On the other hand, we found an increase in the ER after flourouracil (x = 200 fmol/mg protein) and cyclophosphamide (x = 250 fmol/mg protein) treatment. According to our results there is no apparent tendency that the investigated chemotherapies when looked upon together induce any significant influence on the ER concentration. In our six treatment groups we were able to determine the percentage of ER pos. cells within the tumors by using immunohistochemical ER determination during the periods from day 7 to 12 and from day 14 to 16 after treatment with chemotherapy. As shown in Fig. 3 20% of the control tumor cells are ER pos. The median of ER pos. cells in the carboplatinum, mitomycine

14-16 days

Fig. 3. Immunohistochemical

ER values after chemotherapy in an endometrial cancer.

135

veS4 Fig. 4. Biochemical

PR concentration

and flourouracil group showed a decrease in the time period from day 7-12, whereas the bleomycine group showed a slight increase.

7-12

days after chemotherapy

in an endometrial

cancer.

Bleomycin has been of minor effect on tumor growth, as is shown in Fig. 1. In the time period of day 14-16 we did not

14-16

days Fig. 5. Immunohistochemical

PR values after chemotherapy

in an endometrial

cancer,

136

observe a significant tendency on the receptor concentration induced by one of the injected chemotherapeutic drugs. Fig. 4 shows results of the biochemical PR concentration after chemotherapy. Seven to 12 days after a single injection of chemotherapy, four of the treatment groups (carboplatinum, bleomycine, epirubicine, cyclophosphamide) could be analysed for the above-mentioned reasons. The PR concentration of our controls after 7 to 12 days was about 70 fmol/mg protein, only the epirubicine treated tumors reached elevated values with a mean of 220 fmol/mg protein. 14-16 days after treatment in all of the six chemotherapy groups (including mitomycine and fluorouracil) PR concentration could be measured. The mean value in our controls was 75 fmol/mg protein. We found elevated values (x = 130 fmol/mg protein) after bleomycine treatment, decreased levels (x = 35 fmol/mg protein) after mitomycine treatment. As shown in Fig. 2 this corresponds very well with the decrease in ER concentration 14 days after the application of mitomycine. The differences of these treatment groups compared to our control groups were statistically not significant. The data cannot be interpreted as proof for a systematic tendency of chemotherapy on the PR concentration when all the data of all the treated groups are taken together and compared to our controls. Fig. 5 shows the immunhistochemical PR values with regard to the six treatment groups. The control level of 20% PR pos. cells was exceeded in the epirubicine-treated group during the period from day 7 to 12(35%). No major differences could be found from the 14-16 day after treatment. However, when considering all the data of the treated tumors there is no significant effect on the percentage of PR pos. cells during the observed time period. Discussion Many patients with metastasized carcinomas have already been pretreated with adjuvant chemotherapy. Whether chemotherapy has any influence on the hormone receptor concentration of tumor cells and on the effectiveness of en-

docrine therapies is not evident. There may be a difference in efficiacy of cytotoxic therapy on ER pos. or ER neg. cell clones which causes a selection of resistant cells. Assuming that cytotoxic therapy would have a greater effect on ER neg. tumor cell clones, one could possibly increase the relation to ER pos. cell clones. This may ultimately lead to an improved effectiveness of endocrine therapy. This question has already been discussed controversially by Lippman [17] and Kiang [13] in 1978. Lippman found remission after chemotherapeutic treatment in 76% of ER neg breast cancer patients, while only 12% of patients with ER pos. breast carcinomas responded to the same therapy. If chemotherapy has a selective effect on receptor neg. tumor cells, then it should be possible to induce receptor shifting from ER poor to ER pos. tumors. Jonat supports this perspective in his clinical study. He determined the hormone receptor capacity in primary tumors and later on in metastases. After an intermittent chemotherapy a change from ER neg. to ER pos. was observed in 36%, the contrary was observed in only 18% of the cases. In two further clinical studies there is evidence which supports this hypothesis. 11,121. Kiang [13] published exactly the opposite. He found a significantly improved remission rate after chemotherapy in ER pos. breast carcinomas. An overview on 385 reported breast cancer cases treated with chemotherapy presented at the consensus meeting in Bethesda in 1979 showed no difference in the remission rates of ER neg. and ER pos. tumors. Lippman [17] and Kiang [14] started their controversy once more and pointed out two different hypotheses. Lippman reported an enhanced effectiveness on fast-growing and ER neg. tumor cell clones, especially when treated with doxorubicin. Kiang found in his study that there is equal effectiveness of chemotherapy on ER pos. and neg. tumors, but that the time period to relapse is longer for ER pos. tumors. Therefore, treatment results in ER pos. cases should be better when compared with ER neg. ones. A decrease in ER capacity to 25% after 3 weeks of treatment was found in previously pub-

137

lished studies on two xenotransplanted breast cancers [18,19]. A weekly therapy with relatively high doses of cyclophosphamide and adriamycine was applied in these particular settings. It is possible that these therapies induce a necrotic tumor demarkation and this could be a possible explanation for the differences in the results. Although the calculated treatment dosages were selected according to published polychemotherapeutic regimens, they cannot be compared directly with intravenous application in patients due to different application modalities. Our aim in this study was to inhibit tumor growth nearly completely and demonstrate whether or not this may influence the steroid hormone receptor concentrations or the percentage of receptor pos. cells in the particular tumors. As shown in Fig. 1, we observed distinctive effects on tumor growth. Therefore the resorption of the chemotherapeutic drugs when given intraperitoneally was sufficient for our purposes. We would expect a higher percentage of receptor pos. cells or a higher receptor concentration after the applied therapies, if chemotherapy would have a selective effect on tumor growth of receptor neg. tumor cell clones. We could not find any distinctive effects of chemotherapeutic drugs on ER pos. or ER neg. tumor cell clones when looking upon the data of all therapies using biochemical or immunhistochemical ER and PR determination. This was also the case when we looked upon the most effective therapies concerning tumor growth. A selection of receptor pos. cells in our xenotransplanted endometrial cancer by chemotherapy could not be achieved. According to these results it seems that chemotherapies are effective on receptor pos. as well as on receptor neg. tumor cell clones. This is in accordance with Kiang and such clinical studies [7,8,19] who found no effect of intermittent chemotherapy on receptor concentration. It remains questionable whether these results in the nude-mice model can be transferred to human tumors. Many tumor variables remain constant after xenotransplantation, as is well documented in many publications [2]. Although a selection of tumor cell clones by xenotransplantation cannot be excluded, the results may be of

some interest. The constant frame conditions of this model allows to study the time course of parameters after a certain therapy. These results should be investigated in other xenotransplanted tumors.

1 Allegra JC, Lippmann ME, Simon R, Thompson EB, Barlock A, Green L, Huff K.K. Changes in multiple or sequential estrogen receptor determinations in breast cancer. Cancer 1980;45:792-794. H, Michel RTh, Fortmeyer 2 Bastert G, Schmidt-Matthiesen HP, Sturm R, Nord D, Gerner R. Human mammary cancer in nu/nu mice. A model for testing in research and clinic. Klin Wschr 1977; 55:83-84. A. Monoklonale Antikiirper 3 Beck T, Follow K, Heubner zum immunhistochemischen Nachweis des Gstrogenrezeptor-Status an Gewebschnitt primbrer Mammacarcinome. Geburth Frauenheilk 1986;46:490-494. 4 Bojar H, Ertl D. Comparison of immuncytochemical estrogen receptor assay (ER-ICA), biochemical estrogen receptor enzyme immunoassay (ER-EIA), and radioligand binding assay (ER-RBA) in human breast cancer. Symposion on estrogen receptor determination with monoclonal antibodies. Monte Carlo 1984. using rank sums. Techno5 Dunn OJ. Multiple comparisons metrics 1964;24:241-252. Group: Standards for 6 EORTC Breast Cancer Cooperative the assessment of estrogen receptors in human breast cancer. Eur J Cancer 1973; 9:379-381. Breast Cancer Cooperative Group: revision of 7 EORTC the standards for the assessment of hormone receptors in human breast cancer. Eur J Cancer 1980;16:1513-1515. in receptor status 8 Holdaway IM, Bowditch JV. Variation between primary and metastatic breast cancer. Cancer 1983;52:479-485. 9 Holm S. A simple sequentially rejective multiple test procedure. Stand J Stat 1979;6:65-70. 10 Hull DF, Clark GM, Osborne CK, Chamness GC, Knight WA. Multiple estrogen receptor assays in human breast cancer. Cancer Res 1983;43:413-416. 11 Jonat W. Experimentelle und klinische Erfahrungen mit Steroidhormonrezeptoren beim Mammacarcinom. Wiener Klin Wochenschr 1986;96;499-507. 12 Kiang DT, Kennedy BJ. Factors affecting estrogen receptors in breast cancer. Cancer 1977;40:1571-1576. 13 Kiang DT, Frenning DH, Goldman AI, Ascensao VF, Kennedy BJ. Estrogen receptors and responses to chemotherapy and hormonal therapy in advanced breast cancer. New Engl .I Med 1978;299:1330-1334. 14 Kiang DT. Correlation between estrogen-receptor proteins and response to chemotherapy in patients with breast cancer. Cancer Treat Rep 1984;68:577-579. 15 Kruskal WH, Wallis WA. Use of ranks in one criterion variance analysis. J Am Stat Ass 1952;47:583-621.

138 16 Kruskal WH, Wallis WA. Use of ranks in one criterion variance analysis. J Am Stat Ass 1953;47:907-911. 17 Lippmann ME, Allegra JC, Thompsen EB et al. The relation between estrogen-receptors and response rate to cytotoxic chemotherapy in metastatic breast cancer. New Engl J Med 1978;298:1223-1228. 18 Levine RM, Lippman ME. Relationship between estrogen receptor proteins and responses to chemotherapy in breast cancer. Cancer Treat Rep 1984;68:573-576. 19 Michel RTh, Bastert G, Fortmeyer H, Stegmiiller M, Schmidt-Matthiesen H. Der onkostatische Effekt verschiedener hormoneller und zytostatiscber Behandlungsformen. In: Jonat W, Maass H, eds. Steroidhormonrezeptoren im Karzinomgewebe Stuttgart: Enke,1982;198-208. 20 Michel RTh. Rezeptororientierte hormone11 und hormonell zytostatische Therapieversuche an menschlicben Mammakarzinomen nach Heterotransplantation auf nu/nu Mause. Habilitationsschrift Frankfurt, 1982. 21 Michel RTh, Bastert G. Wichtige Rahmenbedingungen zur Interpretation von Hormonrezeptorbefunden menschlicher Mammakarzinome Medwelt 1983;34:373-377. 22 Mitze M, Jonat W, Braendle T, Kikpe T, Stegner HE. Vergleich immunhistologischer und biochemischer &trogenrezeptor-bestimmungen an normalen, hyperplastischen und neoplastischen Endometrien. Tumor Diagn Ther 1988;9:1-6.

23 Raemaekers JM, Beex LV, Koenders AJ, Pieters GF, Smals AG, Benraad TJ, Kloppenborg PW. Concordance and discordance of estrogen and progesterone receptor content in sequential biopsies of patients with advanced breast cancer. Relation to Survival. Eur J Cancer Clin Oncol 1984;20:1011-1018. 24 Scatchard G. The attraction of proteins for small molecules and ions Ann NY Acad Sci 1949;51:660-672. 25 Schmidt Gollwitzer M. Korrelation zwischen den Sexualsteroiden im Serum und im Endometrium, den ostradiolund progesteronbindenden Reteptorproteinen und der Aktivitlt der 17@HSD wlhrend des menstruellen Zyklus. Habilitationsschrift Berlin, 1978. 26 Vering A, Michel RTh, Fortmeyer HP, Kuhl H, Albrecht M, Stegmiiller M. Estrogen related hormone receptor modulation in human breast- and endometrial cancer xenotransplanted into nude mice. J Steroid Biochem 1987;28 Suppl.: 156. 27 Vering A, Michel RTh, Kuhl H, Stegmiiller M. Estrogen and antiestrogen related short-term receptor modulation in xenotransplanted human breast- and endometrial cancer. Cancer Res Chn Oncol 1988;114 suppl: 91. 28 Vering A, Michel RTh, Stegmiiller M. Veranderungen der Hormonrezeptorkapazitat bei xenotransplantierten Mamma- und Endometriumkarzinomen unter Zytostase. Arch Gynecol Obstet 1989;245:484.