University of Vienna Medical School, Vienna, Austria

FERTILITY AND STERILITY VOL. 81, NO. 4, APRIL 2004 Copyright 2004 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A. Estrogen and progesterone receptor expression in patients with uterine smooth muscle tumors Klaus Bodner, M.D., a Barbara Bodner-Adler, M.D., a Oliver Kimberger, M.D., b Klaus Czerwenka, M.D., c and Klaus Mayerhofer, M.D. a University of Vienna Medical School, Vienna, Austria Objective: To investigate the expression of estrogen and progesterone receptor in uterine leiomyomas, smooth muscle tumors of uncertain malignant potential (STUMP), and leiomyosarcomas (LMS), and to assess the correlation between steroid receptor expression and clinicopathologic parameters in LMS. Design: Estrogen/progesterone receptor expression was investigated by immunohistochemistry. Setting: Department of Gynecology and Obstetrics of the University Hospital of Vienna. Patient(s): Twenty-six women with leiomyoma, 24 with STUMP, and 21 with LMS of the uterus. Intervention(s): Formalin-fixed, paraffin-embedded tissues were sectioned and stained. Main Outcome Measure(s): Number of tumor cells stained. Result(s): Significant differences regarding the frequency of estrogen receptor expression were observed between LMS and leiomyoma and STUMP and leiomyoma (P.05). The progesterone receptor expression did significantly differ between LMS and STUMP (P.05), and LMS and leiomyoma (P.05). In uterine LMS, the relationship between estrogen/progesterone receptor expression and clinicopathologic parameters did not reach statistical significance (P.05), and neither of the markers studied revealed prognostic significance (P.05). Conclusion(s): The present study observed significant differences of steroid receptor expression between uterine leiomyoma, STUMP, and LMS. Our data indicate that the progesterone receptor may be an especially useful marker to distinguish cases of malignant smooth muscle tumors in which histological features are ambiguous or borderline. (Fertil Steril 2004;81:1062 6. 2004 by American Society for Reproductive Medicine.) Key Words: Estrogen receptors, progesterone receptors, uterine smooth muscle tumors Received April 30, 2003; revised and accepted August 26, 2003. Reprint requests: Barbara Bodner-Adler, M.D., Department of Gynecology and Obstetrics, University of Vienna Medical School, A-1090 Vienna, Währinger Gürtel 18-20, Austria (FAX: 43-1-40400-2911; E-mail: klausbodner@yahoo.com). a Department of Gynecology and Obstetrics. b Department of Anesthesiology. c Department of Gynecopathology. 0015-0282/04/$30.00 doi:10.1016/j.fertnsert.2003. 08.043 Based on mitotic count, nuclear atypia, and other morphologic features, uterine smooth muscle tumors can be classified as leiomyomas, smooth muscle tumors of uncertain malignant potential (STUMP), and leiomyosarcomas (LMS) (1, 2). Uterine leiomyomas are very common benign smooth muscle tumors that occur in nearly 40% of women older than 35 years (3, 4). Itis widely accepted that they grow hormone dependent with estrogen and progesterone receptor expression being reported in the literature (5 8). This recognition has been used as the basis for their medical treatment with gonadotropin-releasing hormone (GnRH) agonists (5). Tumors with a lower mitotic count or less nuclear atypia than uterine leiomyosarcomas are classified as smooth muscle tumors of uncertain malignant potential (STUMP) (1, 3). Until recently, little was known about the status of steroid receptors in these tumors (1, 9). Previous studies reported an excellent prognosis in patients with STUMP (1, 3). However, the distinction between STUMP and leiomyosarcoma and the assessment of the malignant potential may sometimes be problematic, especially when a difficulty in recognizing mitotic figures occurs or clumped and degenerative nuclei are misinterpreted as mitotic figures (1). Thus, the need for the detection of additional markers to distinguish these tumors is evident. Uterine LMS are rare tumors accounting for only about 1.3% of all uterine malignancies with a high risk of local recurrence and metas- 1062

FIGURE 1 The staining intensity for estrogen receptor expression in uterine leiomyosarcoma (red bar), leiomyoma (green bar), and smooth muscle tumor of uncertain malignant potential (yellow bar) was evaluated to the percentage of stained cells: negative (less than 10% were stained), weak positive (11% 50% of the cells were stained), moderate positive (51% 80% of the cells were stained), and strong positive (more than 80% of the tumor cells were stained). Bodner. Uterine smooth muscle tumors and steroid receptor expression. Fertil Steril 2004. FIGURE 2 The staining intensity for progesterone receptor expression in uterine leiomyosarcoma (red bar), leiomyoma (green bar), and smooth muscle tumor of uncertain malignant potential (yellow bar) was evaluated to the percentage of stained cells: negative (less than 10% were stained), weak positive (11% 50% of the cells were stained), moderate positive (51% 80% of the cells were stained), and strong positive (more than 80% of the tumor cells were stained). Bodner. Uterine smooth muscle tumors and steroid receptor expression. Fertil Steril 2004. FERTILITY & STERILITY 1063

tasis (10). In these tumors, several prognostic parameters have been established; the most important prognostic factor is the extent of tumor at diagnosis (10 14). The prognostic impact of other factors, such as age at diagnosis, vascular space involvement, histological grade, and myometrial invasion, is still discussed controversially (12 16). Because uterine LMS can either arise de novo or in otherwise benign leiomyomas (17), it might be possible that their growth is similar to leiomyomas controlled by ovarian steroids. Until recently, only a few studies investigated this important topic (1, 9, 18 22). The aim of the present study was to investigate the expression of estrogen and progesterone receptor in uterine leiomyomas, STUMP, and LMS. Furthermore, we determined the correlation between steroid receptor expressions and various clinicopathologic parameters in LMS to assess the prognostic significance. MATERIALS AND METHODS Tissue Collection This retrospective study included 26 cases of leiomyoma, 24 cases of STUMP, and 21 cases of LMS of the uterus treated between 1990 and 2000 at the Department of Gynecology and Obstetrics of the University Hospital Vienna. The diagnosis of LMS and STUMP was based on previously published criteria (3, 23). Tumors with marked atypia were classified as leiomyosarcoma if they had 5 or more mitoses per 10 high-power fields (HPF) and as STUMP if they had 1 to 4 mitoses per 10 HPF. Tumors with moderate nuclear atypia were classified as leiomyosarcoma if they had 10 or more mitoses per 10 HPF and as STUMP if they had 5 to 9 mitoses per 10 HPF. Tumors with mild nuclear atypia and 5 or more mitoses per 10 HPF were classified as STUMP. Tumors without nuclear atypia or fewer mitoses than previously outlined were diagnosed as leiomyomas. All patients with leiomyosarcomas were staged retrospectively, according to a modified staging system of the International Federation of Gynecology and Obstetrics (FIGO) for endometrial cancer. Serial sections were prepared for hematoxylin and eosin staining and immunohistochemistry. All slides were reviewed by an experienced pathologist. Clinical information, including follow-up data, was obtained from the database of the Department of Gynecology and Obstetrics. This retrospective data analysis was approved by the institutional review board. Immunohistochemistry Immunohistochemical staining for estrogen and progesterone was performed on formalin-fixed and paraffin-embedded sections by the avidin-biotin-immunoperoxidase complex method. Two samples of each tumor were used for the immunohistochemical staining procedure. A mouse monoclonal antibody to human estrogen receptor (ER1D5, dilution 1:50; Dako Corporation, Carpinteria, CA) and to human progesterone receptor (PGR-1A6, dilution 1:40; BioGenex, San Ramon, CA) was used as the primary antibody. In brief, the sections were deparaffinized in xylene, rehydrated through graded alcohols, and treated in a 0.01-M citrate buffer for 15 minutes (ph 6) in a microwave oven. Then they were incubated with 0.03% hydrogen peroxide to block endogenous peroxidase activity and with normal goat serum to reduce nonspecific binding. The sections were incubated with the specific primary antibodies or control nonimmunized mouse serum at 4 C overnight. Biotinylated goat antimouse immunoglobulin G was used as a linker. After washing, the streptavidin-biotin complex was applied and stained with diaminobenzidine. The sections were slightly counterstained with hematoxylin. Specific staining was identified by a brown color of the nucleus. All control slides yielded negative results. The staining of estrogen and progesterone receptor expression was evaluated to the percentage of stained cells: negative (less than 10% were stained), weak positive (11% 50% of the cells were stained), moderate positive (51% 80% of the cells were stained), and strong positive (more than 80% of the tumor cells were stained). Statistical Analysis Chi-square tests were used to compare the frequency distributions of the steroid receptor expression between the analyzed groups (leiomyoma, STUMP, and LMS). Correlation between clinicopathologic parameters (clinical stage, age, vascular space involvement, and recurrence disease) and estrogen/progesterone receptor expression in uterine LMS was tested using the Spearman correlation coefficient. The end point of overall survival was used for analysis. Survival probabilities were calculated by the product limit method of Kaplan and Meier (24). P values of less than.05 were considered statistically significant. The SPSS system (SPSS Inc., Chicago, IL) was used for the calculations. RESULTS Expression of Estrogen Receptor in LMS, STUMP, and Leiomyoma The frequency of estrogen receptor (ER) expression in LMS, STUMP, and leiomyoma is given in Table 1; the staining intensity for ER expression is presented in Figure 1. Estrogen receptor was present in 12 out of 21 LMS, in 13 out of 24 STUMP, and in 24 out of 26 leiomyomas. Significant differences regarding the frequency of estrogen receptor expression were observed between LMS and leiomyoma as well as between STUMP and leiomyoma (P.005 and P.002, respectively), but not between LMS and STUMP (P.05). 1064 Bodner et al. Uterine smooth muscle tumors and steroid receptor expression Vol. 81, No. 4, April 2004

TABLE 1 Estrogen/progesterone receptor expression in uterine smooth muscle tumors. LMS (n 21) Leiomyoma (n 26) STUMP (n 24) Estrogen receptor expression Yes 12 24 13 No 9 2 11 Progesterone receptor expression Yes 9 23 17 No 12 3 7 Note: LMS leiomyosarcomas; STUMP smooth muscle tumors of uncertain malignant potential. Bodner. Uterine smooth muscle tumors and steroid receptor expression. Fertil Steril 2004. Expression of Progesterone Receptor in LMS, STUMP, and Leiomyoma The frequency of progesterone receptor (PR) expression in LMS, STUMP, and leiomyoma is given in Table 1; the staining intensity for PR expression is presented in Figure 2. Progesterone receptor was expressed in 9 out of 21 LMS, in 17 out of 24 STUMP, and in 23 out of 26 leiomyomas. This resulted in a significant difference of the frequency of PR expression between LMS and STUMP, as well as between LMS and leiomyoma (P.05 and P.001, respectively), but not between STUMP and leiomyoma (P 0.05). Clinical and Pathological Findings in Patients with STUMP and LMS All 24 patients with STUMP were alive and well with a median follow-up time of 41 months (range: 13 69 months). Of the 21 patients with uterine LMS, the median age at diagnosis was 51 years (range: 36 78 years). Thirteen patients had stage I, three patients stage II, four patients stage III, and one patient stage IV disease. The median follow-up time was 47 months (range: 1 228 months). All 21 patients had a total abdominal hysterectomy and bilateral salpingooophorectomy as part of their initial surgical treatment. Ten out of 21 patients received an adjuvant therapy; among those, 8 patients received postoperative radiotherapy, and 2 patients received anthracycline-based chemotherapy. The median overall survival of the 21 patients with uterine LMS was 52 months (range: 20 219 months), resulting in an overall survival rate of 41%. The median disease-free survival was 13 months (range: 3 23 months). Neither the relationship between estrogen/progesterone receptor expression nor the relationship between the degree of staining for estrogen/progesterone and clinical stage, age, vascular space involvement and recurrence disease did reach statistical significance in patients with uterine LMS (P.05). Neither of the markers studied revealed prognostic significance in these patients (P.05). Early tumor stage (P.0001), age at diagnosis 50 years (P.02), and the absence of vascular space involvement (P.04) were prognostic factors associated with a lengthened overall survival in uterine LMS. A multivariate analysis could not be done due to the small number of LMS. DISCUSSION Estrogen and progesterone receptors are routinely determined in patients with breast and endometrial cancer (25, 26). Regarding the steroid receptor status of uterine smooth muscle tumors, only leiomyomas have been studied extensively thus far (5, 6). It is generally accepted that these tumors grow hormone dependent, with ER and PR expression being reported in literature (5 8). In accordance with the literature, our data revealed that ERs and PRs are immunohistochemically expressed in nearly all leiomyomas (1, 7 9). However, little is known about the steroid receptor expression of uterine LMS and STUMP (1, 9, 18 22). Previous studies reported that the expression of steroid receptors in uterine leiomyosarcomas ranges between 11% and 60% (9, 19 22). Our data confirm these findings as ERs were expressed in 57% and PRs in 43% of cases. However, as published previously, the presence of steroid receptors did not correlate with any clinicopathologic parameters and had no prognostic impact on the outcome of disease (18). These results are in accordance with the literature (9, 20 22). Until recently, only two studies reported the steroid receptor status of smooth muscle tumors of uncertain malignant potential (1, 9). Zhai and colleagues (9) observed a strong positive staining for progesterone in all eight cases with STUMP. When analyzing ER expression the authors found positive staining in seven out of eight cases with varying staining intensity (9). Mittal and Demopoulos (1) detected positive PRs in all investigated cases of STUMP. Our data, analyzing a much larger series of STUMP, revealed positive ERs in 13 out of 24 cases and positive PRs in 17 out of 24 cases. The frequency and the staining intensity of ERs were significantly reduced in STUMP compared to leiomyomas. However, no significant difference was detected between STUMP and LMS. These findings are in accordance with the data of Mittal and Demopoulos (1) who detected significant differences in estrogen expression only between leiomyoma and STUMP, but not between STUMP and LMS. The present study observed a significantly increased progesterone expression of STUMP compared with LMS, whereas no significant differences between STUMP and leiomyomas could be detected. These data are in accordance with previous reports (1, 9). We suggest that the PR status of STUMP is closer to leiomyomas than LMS. This would be in line with the clinical presentation and the excellent prognosis reported in STUMP. In accordance with previous investiga- FERTILITY & STERILITY 1065

tors, we confirm the favorable outcome in patients with STUMP (1, 3). However, the differential diagnosis of STUMP vs. LMS may sometimes be problematic. Additional information on the steroid receptor status could be helpful to distinguish between cases in which histological findings are ambiguous or borderline (1). In our opinion, the PR appears especially useful as an additional marker in such cases. Though not performed in the present study, the evaluation of the karyotypes of the individual specimens should be a matter for further discussion as this technique could possibly represent an additional tool to distinguish benign from malignant uterine smooth muscle tumors. Due to the limited number of cases in our study, we are aware of the fact that definitive conclusions cannot be drawn, and further studies including larger numbers of patients are necessary to establish progesterone as a routine marker in the differential diagnosis of uterine smooth muscle tumors. In conclusion, we observed significant differences of steroid receptor expression between uterine leiomyoma, STUMP, and LMS. The PR appears to be an especially useful marker to distinguish cases of smooth muscle tumors in which histological features are ambiguous or borderline. Although a significant proportion of uterine LMS exhibited positive ERs and PRs, steroid receptor expression showed no significant association with clinicopathologic parameters and failed to prove its prognostic role. References 1. Mittal K, Demopoulos RI. MIB-1 (KI-67), p 53, estrogen receptor, and progesterone receptor expression in uterine smooth muscle tumors. Human Pathol 2001;32(9):984 7. 2. Clement PB. The pathology of uterine smooth muscle tumors and mixed endometrial stromal smooth muscle tumors: a selective review with emphasis on recent advances. Int J Gynecol Pathol 2000;19:39 55. 3. Zaloudek CJ, Norris HJ. Mesenchymal tumors of the uterus. In: Kurman RJ, ed. Blaustein s pathology of the female genital tract. New York: Springer-Verlag, 1994:488 500. 4. Vollenhoven BJ, Lawrence AS, Healy DL. Uterine fibroids: a clinical review. Br J Obstet Gynecol 1990;97:285 98. 5. Buttram VC, Reiter RC. Uterine leiomyomata: etiology, symptomatology, and management. Fertil Steril 1981;36:433 45. 6. Kawaguchi K, Fujii S, Konishi I, Nanbu Y, Nonokaki H, Mori T. Mitotic activity in uterine leiomyoma during the menstrual cycle. Am J Obstet Gynecol 1989;160:637 41. 7. Kawaguchi K, Fujii S, Konishi I, et al. Immunohistochemical analysis of oestrogen receptors, progesterone receptors and KI-67 in leiomyoma and myometrium during the menstrual cycle and pregnancy. Virchows Arch Pathol Anat 1991;419:309 15. 8. Brandon DD, Bethea CL, Strawn EJ, et al. Progesterone receptor messenger ribonucleic acid and protein are overexpressed in human uterine leiomyomas. Am J Obstet Gynecol 1993;169:78 85. 9. Zhai YL, Kobayashi Y, Mori A, Orii A, Nikaido T, Konishi I, et al. Expression of steroid receptors, KI-67, and p 53 in uterine leiomyosarcomas. Int J Gynecol Pathol 1999;18:20 8. 10. Zaloudek CJ, Norris HJ. Mesenchymal tumors of the uterus. In: Kurman RJ, ed. Blaustein s pathology of the female genital tract. New York: Springer-Verlag, 1994:849 914. 11. Schwartz Z, Dgani R, Lancet M, Kessler I. Uterine sarcoma in Israel: a study of 104 Cases. Gynecol Oncol 1985;20:354 63. 12. Evans HL, Chawla SP, Simpson C, Finn KP. Smooth muscle neoplasms of the uterus other than ordinary leiomyoma. A study of 46 cases, with emphasis on diagnostic criteria and prognostic factors. Cancer 1988; 62:2239 47. 13. Nordal RR, Kristensen GB, Kaern J, Stenwig AE, Pettersen EO, Trope CG. The prognostic significance of stage, tumor size, cellular atypia and DNA ploidy in uterine leiomyosarcoma. Acta Oncol 1995;34:797 802. 14. Bodner K, Bodner-Adler B, Kimberger O, Czerwenka K, Leodolter S, Mayerhofer K. Evaluating prognostic parameters in women with uterine leiomyosarcomas. A clinicopathologic study. J Reprod Med 2003; 48(2):95 100. 15. Van Dinh T, Woodruff JO. Leiomyosarcoma of the uterus. Am J Obstet Gynecol 1982;144:817 823. 16. Silverberg SG. Leiomyosarcoma of the uterus. A clinicopathologic study. Obstet Gynecol 1971;38:613 28. 17. Montague A, Swartz DP, Woodruff JD. Sarcoma arising in leiomyoma of the uterus. Am J Obstet Gynecol 1965;92:421 7. 18. Bodner K, Bodner-Adler B, Kimberger O, Czerwenka K, Leodolter S, Mayerhofer K. Estrogen and progesterone receptor expression in patients with uterine leiomyosarcoma and correlation with different clinicopathologic parameters. Anticancer Res 2003;23:729 32. 19. Lantta M, Kärkkäinen J, Wahlström T, Widholm O. Estradiol and progesterone receptors in gynecologic sarcomas. Acta Obstet Gynecol Scand 1984;63:505 8. 20. Wade K, Hammond I, Williams K, Cauchi M. Uterine sarcoma: steroid receptors and response to hormonal therapy. Gynecol Oncol 1990;39: 364 7. 21. Sutton GP, Stehman FB, Michael H, Young PC, Ehrlich CE. Estrogen and progesterone receptors in uterine sarcomas. Obstet Gynecol 1986; 68:709 15. 22. Soper JT, McCarthy KS, Hinshaw W, Creasman WT, Clar-Pearson DL. Cytoplasmic estrogen and progesterone receptor content of uterine sarcomas. Am J Obstet Gynecol 1984;150:342 8. 23. Bell SW, Kempson RL, Hendrickson MR. Problematic uterine smooth muscle neoplasms. A clinicopathologic study of 213 cases. Am J Surg Pathol 1994;18(6):535 58. 24. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Ass 1958;53:457 81. 25. Kauppilla A, Kujansuu E, Vihko R. Cytosol estrogen and progestin receptors in endometrial carcinoma of patients treated with surgery, radiotherapy and progestin. Clinical correlates. Cancer 1982;20:2157 62. 26. Osborne CK, Yochmowitz MG, Knight WA, McGiure WL. The value of estrogen and progesterone receptors in the treatment of breast cancer. Cancer 1980;46:2884 8. 1066 Bodner et al. Uterine smooth muscle tumors and steroid receptor expression Vol. 81, No. 4, April 2004