Inability of Cytoplasmic or Nuclear Androgen Receptor Content or Distribution to Distinguish Benign from Carcinomatous Human Prostate Sydney A. Shain, Lester S. Gorelic, Robert W. Klipper, et al. Cancer Res 1983;43:3691-3695. Published online August 1, 1983. Updated Version Citing Articles E-mail alerts Reprints and Subscriptions Permissions Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/8/3691 This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/43/8/3691#related-urls Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at pubs@aacr.org. To request permission to re-use all or part of this article, contact the AACR Publications Department at permissions@aacr.org. Downloaded from cancerres.aacrjournals.org on July 12, 2011 Copyright © 1983 American Association for Cancer Research [CANCER RESEARCH 43. 3691-3695, August 1983] Inability of Cytoplasmic or Nuclear Androgen Receptor Content or Distribution to Distinguish Benign from Carcinomatous Human Prostate1 Sydney A. Shain,2 Lester S. Gorelic, Robert W. Klipper, Ibrahim Ramzy, Donald E. Novicki, Howard M. Radwin, and Donald L. Lamm Department of Cellular and Molecular Biology, Southwest Foundation for Research and Education [S. A. S., L S. G., R. W. K.}; Division of Urology, Audie L Murphy Memorial Veterans Hospital [S. A. S., D. L. L.¡;Departments of Surgery [H. M. R., D. L. L] and Pathology [I. R.], University of Texas Health Science Center; and Department of Urology, Wilford Hall United States Air Force Medical Center [D. E. N.], San Antonio, Texas 78284 ABSTRACT We used exchange saturation analysis at 15°to quantitate total cytoplasmic and nuclear androgen receptor content of 70 patient specimens. Cytoplasmic androgen receptor contents (fmol/mg DMA) for eight specimens of clinically benign hyperplasia, 14 specimens of historically hyperplastic prostate obtained at cystoprostatectomy, and carcinomatous and noncarcinomatous prostate obtained at radical prostatectomy for prostatic carcinoma, 48 specimens, respectively, were 830 ±165 (mean ±S.E.), 890 ±445,955 ±240, and 750 ±95. Nuclear androgen receptor contents of these same specimens, respectively, were 275 ±40, 235 ±30, 345 ±25, and 350 ±30; whereas, the values of the cytoplasmic/nuclear receptor content, respectively, were 3.25 ±0.55, 3.05 ±0.80, 2.50 ±0.50, and 2.80 ±0.40. Multiway analyses of variance of these cross-sectional data showed that there was no significant difference (p > 0.05) between group mean values. This result principally reflects the fact that the families of values for the four tissue groups were highly heterogenous with broad overlap. The results would not appear to be unduly influenced by carcinomatous epithelial cell content of the specimens, because cytoplasmic and nuclear androgen receptor content were not related to specimen carci nomatous epithelial cell content. Paired analyses of receptor content in carcinomatous and noncarcinomatous prostate spec imens from the same prostate showed enhanced or unchanged receptor content in 58% (cytoplasmic) and 62% (nuclear) of specimens. Our studies show that cross-sectional analyses of androgen receptor content fail to distinguish carcinomatous pros tate from noncarcinomatous prostate. However, paired analyses of these tissues from the same gland identify distinguishing differences. The clinical relevance of these observations remains to be examined. INTRODUCTION Quantitation of estrogen receptor content of human breast carcinoma biopsies is an established diagnostic aid for treatment selection and assessing prognosis (3, 11, 13, 23). A portion of the success of this diagnostic may be attributed to the profound physiological differences which characterize nonlactating and 1Supported in part by USPHS Grant CA 23668 from the National Cancer Institute, Department of Health and Human Services. The views herein expressed are those of the authors and do not necessarily reflect the views of the United States Air Force. 2 To whom requests for reprints should be addressed, at Southwest Foundation for Research and Education, West Loop 410 and Military Drive, San Antonio, Texas 78284. Received April 19,1983; accepted May 9,1983. carcinomatous breast tissue. The former contains a scant epi thelial cell population which, in human (2,7), mouse (16,22), and rat (5), is essentially devoid of estrogen receptors. By contrast, many breast carcinomas contain significant quantities of estro gen receptors (3, 11, 13, 23). Consequently, the diagnostic evaluates a positive signal in breast carcinoma, receptor present, against a negative signal in normal nonlactating breast, receptor absent. Many investigators have established that noncarcinomatous (1, 4, 6, 8, 15, 17, 19-22) and carcinomatous (1, 6, 8, 15, 17, 21) human prostatic tissues contain androgen receptors. Inci dental comparisons of nuclear (1, 6, 8,15, 21 ) or cytoplasmic (1, 6, 17, 21) androgen receptor content in carcinomatous and noncarcinomatous human prostate specimens have been de scribed. Detailed comparative analysis of cytoplasmic and nu clear androgen receptor content and distribution in well-charac terized carcinomatous and noncarcinomatous human prostate specimens has not been described. As part of studies to develop a detailed profile of the properties of carcinomatous and noncar cinomatous human prostate, we have performed such an anal ysis. This report details the results of our systematic evaluation. MATERIALS AND METHODS Patient Specimens. All tissues were collected at the time of diagnostic or palliative surgery after obtaining informed consent. Tissue was dis sected by a pathologist immediately following excision and was grossly separated into noncarcinomatous (benign) and carcinomatous prostate. Tissues were processed and transported to the research laboratory as described previously (21). Representative samples of all specimens were obtained prior to freezing and were processed (21) for histological analysis. The studies described in this report were performed with tissues obtained from 47 patients: benign hyperplastic prostate obtained at prostatic enucleation, 8 patients; histologically hyperplastic prostate ob tained at cystoprostatectomy for bladder carcinoma, 14 patients; and carcinomatous and noncarcinomatous prostate obtained at radical pros tatectomy for prostatic carcinoma, 25 patients. Histological Evaluation. All tissues were step-serial sectioned, and slides were prepared and evaluated at 3 levels, the top, middle, and bottom of the block. Diagnoses were made by reading at least one slide at each level for each specimen. Epithelial cell content and carcinomatous cell content of all specimens were visually estimated. Evaluations were performed simultaneously by a surgical pathologist (I. Ramzy) and an experienced, trained nonpathologist (S. A. Shain). Chemicals. 17a-[mef/>y/-3H3]R1881 (specific activity, 87 Ci/mmol) and radioinert R1881 were obtained from New England Nuclear (Boston, Mass.). Sodium molybdate, dithiothreitol, phenylmethylsulfonyl fluoride, and human -y-globulin (Fraction II) were from Sigma Chemical Co. (St. Louis, Mo.). Norit was from Fisher Scientific Co. (Fairtown, N. J.), and dextran T-70 was from Pharmacia Fine Chemicals, Inc., (Piscataway, 3691 AUGUST 1983 Downloaded from cancerres.aacrjournals.org on July 12, 2011 Copyright © 1983 American Association for Cancer Research S. A. Shain et al. N. J.). All other chemicals were reagent grade and were used as obtained from the manufacturer. Solutions were prepared in water that had been distilled, detonized, and redistilled from glass. Tissue Fractionation and Androgen Receptor Quantitation. Cytoplasmic and nuclear tissue extracts were prepared at 2-4° precisely as 3490 A MEDIAN •MEAN 2500 described previously (21). Cytoplasmic and nuclear androgen receptor content were determined by exchange saturation analysis during incu bation at 15°(21). The exact incubation protocols and ligand concentra o 0< tions used were as described previously (21). Saturation data were analyzed both by Scatchard (18) representation, in order to monitor for possible contribution of secondary binding components to measured receptor content, and by double-reciprocal plot analysis (20), for deter 2000 mination of receptor site content by simple linear regression analysis (14). Other Methods. Protein was determined by the method of Lowry ef a/. (9) using bovine serum albumin as standard. Interference due to dithiothreitol was eliminated by a modification of the iodogen procedure of McClard (10), which we have described previously (21). Salmon testis DMA was used as standard for quantitation of DNA as described previ ously (21). Statistical Analysis of Data. Because replicate saturation analyses indicated that no definable relationship characterized the interaction of the dependent variable (fmol bound)"1 and the independent variable (IR18811) '. we used unweighted 4730 6643 2 a* 1500 I e •o 1000 « Hl ° t linear regression analyses of double- 500 reciprocal plots to determine receptor site content and R1881 binding affinity. We used analysis of co vari ance to determine the significance of the slope of regressed lines (27). Differences among group means were evaluated by one-way analysis of variance, and a posteriori contrasts were evaluated by the Student-Newman-Keuls procedure (12). BPH Cystoprosto- Radical tectomy RESULTS Prostatectomy Benign Androgen Receptor Content and Distribution in Cross-Sec tional Specimens of Human Prostatic Tissues. Cytoplasmic androgen receptor content of human prostatic tissues varied from zero (one case) to 6640 fmol per mg DNA (Chart 1). The families of values for cytoplasmic receptor site content were similar with broad overlap. As anticipated from inspection of the data, mean receptor site content of each of the 4 families (Chart 1) was not significantly different (p > 0.05). Interestingly, cyto plasmic receptor content of a significant number of benign histologically hyperplastic specimens, obtained at radical prostat ectomy, clustered at values of less than 500 fmol per mg DNA. The same result characterized cytoplasmic androgen receptor site content of histologically hyperplastic prostate obtained at cystoprostatectomy for bladder carcinoma (Chart 1). By contrast, most specimens of clinically benign hyperplastic prostate con tained greater than 500 fmol cytoplasmic androgen receptor per mg DNA. The mean and median values for these analyses are summarized in Table 1. Nuclear androgen receptor content of human prostatic tissues ranged between 65 and 700 fmol per mg DNA (Chart 2). As was the case for cytoplasmic androgen receptor content, the families of values for nuclear androgen receptor content were similar with broad overlap. Mean nuclear androgen receptor site content (Chart 2) for each of the 4 families was not significantly different (p > 0.05). There was a tendency for nuclear receptor site content to be higher in radical prostatectomy specimens when compared to enucleation (benign prostatic hyrjerplasia) or cys toprostatectomy specimens (Chart 2). Mean and median values for these analyses are summarized in Table 1. The value of the ratio of cytoplasmic androgen receptor con tent/nuclear androgen receptor content ranged from 0.35 to Carcinoma Chart 1. Cytoplasmic androgen receptor content of human prostatic tissues. BPH, benign prostatic hyperplasia. Table 1 Cytoplasmic and nuclear androgen receptor content and distribution in human prostatic tissues Tissues were obtained and analyzed for receptor content as described in the text. Receptor content (fmol/mg DNA) Procedure/tissue type Enucleation Clinical BPH Cystoprostatectomy Histológica! BPH Radical prostatectomy Benign" Carcinoma *C/N, cytoplasmic Cytoplasmic Receptor distribution (C/N)a Nuclear 830 ±165°(755)° 275 ±40 (240) 3.25 ±0.55 890 ±445 (440) 3.05 + 0.80 235 + 30(230) 955+ 240 (615) 345±25(310) 2.50±0.05 750+ 95 (650) 350±30(345) 2.80±0.40 receptor content/nuclear receptor content; BPH, benign prostatic hyperplasia, principally mixed glandular and straniai. 0 Mean ±S.E. '' Numbers in parentheses, median value. " Benign tissues were free of carcinoma and principally consisted of histotogical hyperplasia as described for benign prostatic hyperplasia. greater than 10.0 (Chart 3). The value of this ratio was equal to or greater than one for 85% (58 of 68) of the tissues analyzed (Chart 3). Mean values of the receptor ratio for the 4 families of specimens were not significantly different. Within a family, there was a broad distribution of values for the receptor ratio, and there was also marked overlap of values between families (Chart 3). Mean and median values for these analyses are reported in Table 1. Paired Analyses of Cytoplasmic and Nuclear Androgen 3692 CANCER Downloaded from cancerres.aacrjournals.org on July 12, 2011 Copyright © 1983 American Association for Cancer Research RESEARCH VOL. 43 Androgen Receptors of Benign and Carcinomatous Prostate . .A 1 o o MEDIAN• MEAN."•••• 600 o "o. O) o Û o> -*• Ë 400 05 o> o •oo E C _.**" matched carcinomatous specimen. Analysis of nuclear androgen receptor content in paired specimens revealed 5 of 13 with diminished, 6 of 13 with enhanced, and 2 of 13 with unchanged receptor content when benign tissue was compared to the paired carcinomatous specimen (Chart 5). Effect of Relative CarcinomatousEpithelial Cell Content on Specimen Cytoplasmicand Nuclear Androgen Receptor Con tent. Becausecarcinomatousepithelialcell contentof the tissue »3» -A-.JÉ?5•-...5, *i 200 2000- i " *A•I*. o O BPH Cystoprosta- Radical tectomy Prostatectomy Benign Carcinoma Chart 2. Nuclear androgen receptor content of human prostatic tissues. BPH, benign prostatic hyperplasia. iA 10.08.06.04.02.0 MEDIAN MEAN•• o .5! o -. • -••*• . u o "E </> .•**•-• "o. o Benign o o Corcinomo Chart 4. Cytoplasmic androgen receptor content of paired specimens of carci nomatous and benign (noncarcinomatous) tissues of prostate glands from 12 patients. ~0 (T *-.;• • CL 0> *- -A-BPH O 0> cr •% 750 § o 0---**-A-•• si CystoprostaRadicaitectomy 500 ü O ProstatectomyBenign Carcinoma Charts. Value of the ratio of cytoplasmic receptor content/nuclear content in human prostatic tissues. BPH, benign prostatic hyperplasia. receptor ^ o e I < 3- 350 1 Receptor Content of Carcinomatous and Benign Prostatic Tissue Obtained at Radical Prostatectomy. Pairedanalysesof cytoplasmic androgen receptor content in benign (noncarcinom atous, histological hyperplasia) and carcinomatous prostate specimens from the same gland (Chart 4) revealed 5 of 12 with diminished, 5 of 12 with enhanced, and 2 of 12 with unchanged receptor content when benign tissue was compared to the AUGUST Benign Carcinoma ChartS. Nuclear androgen receptor content of paired specimens of carci nomatous and benign (noncarcinomatous) tissues of prostate glands from 13 patients. 3693 1983 Downloaded from cancerres.aacrjournals.org on July 12, 2011 Copyright © 1983 American Association for Cancer Research S. A. Shain et al. Nucleor y=-0.627X+393 r =0.159 500 z o H h + Cytoplasmic y =-l.884X + 878 r =0.134 ' 2000 « O 1500 o er IODO e 500 20 40 60 80 100 Carcinomotous Epithelium Content (% total epithelium) Chart 6. Relationship between prostate specimen carcinomatous epithelial cell content and cytoplasmic or nuclear androgen receptor content. specimen conceivably could influence receptor content, we per formed a preliminary evaluation of this relationship. Both cyto plasmic and nuclear androgen receptor contents of carcinoma containing prostate specimens were independent of relative carcinomatous epithelial cell content (Chart 6). The slopes of these regression lines (Chart 6) were not significantly different (p > 0.05) from zero. In all cases, epithelial cell content was estimated to be at least 40% of total specimen cell content. In the majority of cases, epithelial cell content was 50 to 75%. DISCUSSION We have shown that mean cytoplasmic androgen receptor contents (fmol per mg DNA) of benign clinical hyperplastic pros tate, benign histological hyperplastic prostate (cystoprostatectomy), noncarcinomatous (histologically hyperplastic) prostate from carcinoma-containing glands, and carcinomatous primary prostatic tissues are statistically indistinguishable. This result principally reflects the wide variation in individual values obtained for specimens from the 4 families of tissue (Chart 1). Moreover, the mean values for the family of cystoprostatectomy specimens and benign radical prostatectomy specimens are heavily weighted by the outlier values; compare the value of the mean and median for these groups (Chart 1; Table 1). When the outlier values are deleted, the mean values for these 2 families of samples, respectively, are 450 ±70 and 656 ±130. This analysis supports the impression that cytoplasmic andro gen receptor content of histological hyperplastic prostate tends to be less than that of clinical hyperplastic or carcinomatous prostatic specimens. While this interpretation may have rele vance for the physiology of prostatic dysfunction, it has little relevance as a diagnostic, because androgen receptor content of benign tissue in carcinomatous glands is represented by a family of values including those characteristic of both cystopros tatectomy and carcinomatous tissues (Chart 1). As expected from inspection of the data in Chart 1, deletion of the outlier values for the family of benign radical prostatectomy tissues leaves a family of specimens indistinguishable, with regard to mean value and distribution of family members, from that for carcinomatous tissues. With regard to family mean cytoplasmic androgen receptor values, our results are comparable to pre vious observations (4, 23); however the current studies suggest that analysis of means has limited significance, since all speci mens appear to define a single population consisting of families representing groups within the larger population. The range of values for nuclear androgen receptor content of these specimens was considerably less than that for cytoplasmic androgen receptor content (compare Charts 1 and 2). Although mean and median values for nuclear receptor content of each of the 4 families of specimens were not significantly different, receptor content of specimens from radical prostatectomy tis sues tended to be greater than that of the other 2 families of specimens (Chart 2). The absence of a significant reduction in either nuclear (Chart 2) or cytoplasmic (Chart 1) androgen recep tor content of carcinomatous specimens, when expressed as receptor sites per mg DNA, suggests that increased ploidy of the neoplastic epithelium, in genera!, does not affect receptor dosage. This interpretation is consistent with our observation that cytoplasmic and nuclear androgen receptor contents of carcinomatous prostatic tissues are independent of relative carci nomatous epithelial cell content (Chart 6). There also does not appear to be a correlation between neoplastic epithelial cell content and estrogen receptor content of breast carcinomas (26). These results would not appear to reflect a particular property of early-stage disease represented by the radical prostatectomy specimens. The mean and range of cytoplasmic and nuclear androgen receptor content for lymph node métastasesof pros tatic cancer from 3 untreated patients, respectively, were 935 (300 to 1585) and 415 (55 to 820) fmol per mg DNA.3 These values are comparable to those determined for primary prostatic carcinoma (Table 1). Because nodal prostatic carcinoma is de void of noncarcinomatous tissue and surely contains polyploid cells, these data for primary and metastatic prostatic carcinoma suggest that androgen receptor dosage per cell in prostatic carcinoma frequently may be greater than that of noncarcinom atous prostate. This interpretation is supported by the data of Charts 4 and 5. When cytoplasmic androgen receptor content of paired noncar cinomatous (benign) and carcinomatous tissues, from the same prostate obtained at radical prostatectomy of 12 patients, was evaluated, 42% (5 of 12) and 17% (2 of 12) of the carcinomatous specimens, respectively, showed enhanced or unchanged recep tor content relative to their paired benign specimen. The results for nuclear androgen receptor content in these same paired specimens plus an additional patient specimen for which cyto plasmic data were lost, due to a mechanical failure, were 46% (6 of 13) enhanced and 15% (2 of 13) unchanged. 3 S. A. Shain et al,, unpublished data. 3694 CANCER Downloaded from cancerres.aacrjournals.org on July 12, 2011 Copyright © 1983 American Association for Cancer Research RESEARCH VOL. 43 Androgen Receptors of Benign and Carcinomatous Prostate Our data may help clarify 2 clinical observations. Commonly, 80% (for review, see Ref. 19) or more (1) of prostatic cancer patients have at least limited positive response to ablative hor monal therapy. In the current and other studies (comparable comparative data have been obtained for radical prostatectomy and needle biopsy specimens using a modified assay protocol4), carcinomatous prostate specimens contain levels of cytoplasmic and nuclear androgen receptors representative of the distribution found in noncarcinomatous prostate. Therefore, the majority of these specimens appears to contain functional receptors, and their presence probably accounts for the initially favorable re sponse to therapy. Similarly, the modest prognostic power of androgen receptor analyses of carcinomatous tissues reported by others (6, 25) may reflect the similarity of androgen receptor content and distribution in benign and carcinomatous prostatic tissue characteristic of cross-sectional analysis. Considering the experience of others (6, 25) and the data of the current study, we are inclined to believe that the determination of androgen receptor content of carcinomatous prostatic tissues may provide information of limited prognostic significance regarding response (quality or interval) to ablative hormonal therapy. A more powerful prognostic might be obtained by determination of receptor con tent in matched specimens of carcinomatous and noncarcinom atous prostatic tissue from the same gland. Those determina tions may provide information concerning predicted altered hor monal sensitivity of carcinomatous tissue relative to the individ ual's benign tissue. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. ACKNOWLEDGMENTS 19. We are deeply grateful to Ginny McGrath, R.N.. whose patience and dedication made the close coordination between our surgical collaborators and ourselves possible. We are indebted to the staffs of the Urology Services of The University of Texas Health Science Center at San Antonio, the Audie L. Murphy Memorial Veterans Administration Hospital, and the Wilford Hall United States Air Force Medical Center, without whose participation this study would not have been performed. We are grateful to Cheryl L. Boedeker, at Southwest Foundation, for her assistance and guidance with the statistical analysis of the data. 21. REFERENCES 23. 1. Barrack, E. R., Bujnovszky, P., and Walsh, P. C. Subcellular distribution of androgen receptors in human normal, benign hyperpiastic. and malignant prostatic tissues: characterization of nuclear salt-resistant receptors. Cancer Res., 43:1107-1116,1983. 2. Block, G. E., Jensen, E. V., and Poltey, T. Z., Jr. The prediction of hormonal dependency of mammary cancer. Ann. Surg., 782: 342-351, 1975. 3. DeSombre, E. R., Green, G. L., and Jensen, E. V. Estrophilin and endocrine responsiveness of breast cancer. Prog. Cancer Res. Then, 10: 1-14,1978. 4. Ekman, P., Barrack, E. R., and Walsh, P. C. Simultaneous measurement of 20. 22. 24. 25. 26. 27. 4 S. A. Shain ef al., work in progress. AUGUST progesterone and androgen receptors in human prostate: a microassay. J. Clin. Endocrino!. Metab., 55:1089-1099,1982. Gardner, D. G., and Wittliff, J. L. Specific estrogen receptors in the lactating mammary gland of the rat. Biochemistry, 72: 3090-3096,1973. Ghanadian, R., Auf, G., and Williams, G. Relationship between prostatic cytoplasmic and nuclear androgen receptors in patients with carcinoma of the prostate. Eur. Urol., 7: 39-40, 1981. Leclercq, G., Heuson, J. C., Deboel, M. C., and Mattheiem, W. H. Oestrogen receptors in breast cancer: a changing concept. Br. Med. J., 7: 185-189, 1975. Lehoux, J.-G., Benard, B., and Elhilali, M. Dihydrotestosterone receptors in the human prostate. 1. Nuclear concentration in normal, benign, and malignant tissues. Arch. Andrai., 5:237-248,1980. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem., 793: 265-275, 1951. McClard, R. W. Removal of sulfhydryl groups with 1,3,4,6-tetrachloro-3a,6adiphenylgiyculuni: application to the assay of protein in the presence of thiol reagents. Anal. Biochem., 772:278-281,1981. McGuire, W. L. An update on estrogen and progesterone receptors in prog nosis for primary and advanced breast cancer. Prog. Cancer Res. Ther., 14: 337-343,1980. Nie, N. H., Hull, C. H., Jenkins, J. G., Steinbrenner, K., and Bent, D. H. SPSS: Statistical Package for the Social Sciences. Ed. 2. New York: McGraw-Hill Publishing Co., 1975. Nomura, Y., Yamagata, J., Kondo, H., Kanda, K., and Takenaka, K. Clinical usefulness of estrogen receptor assay in early and advanced breast cancer. Prog. Cancer Res. Ther., 70:15-29, 1978. Ostte, B., and Mensing, R. W. Statistics in Research. Basic Concepts and Techniques for Research Workers. Ed. 3, pp. 165-236. Ames, Iowa: Iowa State University Press, 1975. Rennie, P. S., and Bruchovsky, N. Hormone receptors in prostatic tissue. Res. Steroids, 9:120-125,1981. Richards, J. E., Shyamala, G., and Nandi, S. Estrogen receptor in normal and neoplastic mouse mammary tissues. Cancer Res., 34: 2764-2772,1974. Rio, M. C., and Offner, M. Rapport préliminaired'une étudecomparative des concentrations en récepteurssteroidiens de tissus prostatiques adenomateux et cancéreux.J. Steroid Biochem., 77: 435-441,1982. Scatchard, G. The attractions of proteins for small molecules and ions. Ann. N. Y. Acad. Sci., 57: 660-672,1949. Scott, W. W., Menon, M., and Walsh, P. C. Hormonal therapy of prostatic cancer. Cancer (Phila.), 45:1929-1936,1980. Shain, S. A., and Boesel, R. W. Human prostate steroid hormone receptor quantitation. Current methodology and possible utility as a clinical discriminant in carcinoma. Invest. Urol., 76:169-174,1978. Shain, S. A., Gorelic. L. S., Boesel, R. W., Radwin, H. M., and Lamm, 0. L. Human prostate androgen receptor quantitation: effects of temperature on assay parameters. Cancer Ros., 42: 4849-4854,1982. Shyamala, G., and Nandi, S. Interactions of 6,7-3H-17,8-estradiol with the mouse lactating mammary tissue in vivo and in vitro. Endocrinology, 97: 861867.1972. Trachtenberg, J., Bujnovszky, P., and Walsh, P. C. Androgen receptor content of normal and hyperpiastic human prostate. J. Clin. Endocrino!. Metab., 54: 17-21, 1982. Trachtenberg, J., and Walsh, P. C. Correlation of prostatic nuclear androgen receptor content with duration of response and survival following hormonal therapy in advanced prostatic cancer. J. Urol., 727: 466-471,1982. Wittliff, J. L., Beatty, B. W., Savlov, E. D., Patterson, W. B., and Cooper, R. A., Jr. Estrogen receptors and hormone dependency in human breast cancer. Recent Results Cancer Res., 57: 59-77, 1976. Wittliff, J. L., Lewko, W. M., Park, D. C., Kute, T. E., Baker, D. T., Jr., and Kane, L. N. Steroid binding proteins of mammary tissues and their clinical significance in breast cancer. Prog. Cancer Res. Ther., 70: 325-359,1978. Zar, J. H. Bkjstatistical Analysis, p. 228. Englewood Cliffs, N. J.: Prentice-Hall, Inc., 1974. 3695 1983 View publication stats Downloaded from cancerres.aacrjournals.org on July 12, 2011 Copyright © 1983 American Association for Cancer Research