Phase II Study of Medroxyprogesterone Acetate Plus Tamoxifen in Advanced Endometrial Carcinoma: a Gynecologic Oncology Group Study

Home , Medroxyprogesterone, Megestrol acetate

Gynecologic Oncology 92 (2004) 4–9 www.elsevier.com/locate/ygyno

Phase II study of medroxyprogesterone acetate plus tamoxifen in advanced endometrial carcinoma: a Gynecologic Oncology Group study

Charles W. Whitney,a,* Virginia L. Brunetto,b Richard J. Zaino,c Samuel S. Lentz,d,1 Joel Sorosky,e Deborah K. Armstrong,f and Roger B. Leeg

a Christiana Care Health System, Newark, DE 19713, USA b Gynecologic Oncology Group Statistical and Data Center, Roswell Park Cancer Institute, Buffalo, NY 14263, USA c Department of Pathology, Milton S. Hershey Medical Center of Pennsylvania State University, Hershey, PA 17033, USA d Section on Gynecologic Oncology, Wake Forest University School of Medicine, Winston-Salem, NC 27157, USA e Division of Gynecologic Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA 52242, USA f Sidney Kimmel Cancer Center at Johns Hopkins, Baltimore, MD 21231-1000, USA g Division of Gynecologic Oncology, University of Washington, Seattle, WA 98195, USA

Received 29 January 2003

Abstract

Objectives. The objectives of this study were to estimate the clinical response rate and toxicity of daily tamoxifen combined with intermittent weekly medroxyprogesterone acetate (MPA). Methods. This study reports the results of 61 patients with measurable advanced or recurrent endometrial carcinoma enrolled on this study to be treated with tamoxifen 40 mg p.o. daily plus alternating weekly cycles of MPA 200 mg p.o. daily. Results. One patient was excluded and two patients did not receive study treatment. The percent of patients responding (6 complete and 13 partial) was 33% (95% confidence interval [CI]: 21–46%) among 58 eligible patients who received therapy. Median progression-free survival (PFS) was 3 months and median overall survival (OS) was 13 months. Conclusion. The combination of daily tamoxifen and intermittent weekly medroxyprogesterone acetate is an active treatment for advanced or recurrent endometrial carcinoma. Further investigation of this combination is appropriate. D 2003 Published by Elsevier Inc.

Introduction be on the order of 15–20% [7,8]. In any event, such responses are usually of short duration, with a median of A variety of progestational agents has long been known approximately 4 months, though an occasional prolonged to be effective in the treatment of recurrent and metastatic response is seen [11]. Tamoxifen alone has produced modest endometrial carcinoma [1,2]. These include hydroxyproges- response rates of 10–20% [12–14]. terone caproate, response rates of 9–34% [3–5]; medrox- Estrogenic compounds such as tamoxifen have been yprogesterone acetate (MPA), response rates of 14–53% shown to increase progesterone receptors (PRs) in human [6–8]; and megestrol acetate, response rates of 11–56% endometrial cancers [15,16] and, therefore, could theoreti- [9,10]. The more recent Gynecologic Oncology Group cally increase the effectiveness of progestational agents in (GOG) experience suggests that this proportion may only the treatment of endometrial carcinoma. In order to test this hypothesis, in 1991 the GOG activated a phase II study of daily tamoxifen and intermittent weekly dosing of the progestational agent medroxyprogesterone acetate in the * Corresponding author. Christiana Gynecologic Oncology, 4923 outpatient management of recurrent or metastatic endome- Ogletown-Stanton Road, Newark, DE 19713. Fax: +1-302-683-0277. E-mail address: [email protected] (C.W. Whitney). trial carcinoma. The goals of the study were to evaluate the 1 Present address: Catawba Women’s Center, Hickory, NC 28603, activity and assess the side effects of this combination of USA. treatments.

0090-8258/$ - see front matter D 2003 Published by Elsevier Inc. doi:10.1016/j.ygyno.2003.09.018 C.W. Whitney et al. / Gynecologic Oncology 92 (2004) 4–9 5

Materials and methods Table 2 Extent of disease at study entry (n = 60) To be eligible for this study, patients must have had any Involved site* Number of patients % grade of histologically confirmed recurrent or metastatic with site involved endometrial carcinoma considered incurable by local therapy. Vagina 37 61.7 Tissue samples from recurrent or metastatic lesions were re- Pelvis 11 18.3 Liver 2 3.3 quired for quantitative assessment of estrogen and progester- Abdomen— 7 11.7 one receptor status. Required slides were reviewed by the other than liver GOG Pathology Committee to confirm primary site, histo- Lung 11 18.3 logical type, and tumor grade. All patients were to have Bone 1 1.7 measurable disease defined as lesions measurable in two Other 8 13.3 dimensions by palpation or imaging. Lesions measurable *Patients may have two or more sites involved. by CT scan or ultrasound had to be greater than 3 cm in diameter. The lesion(s) being followed must not have been treated with radiation within 3 months before the entry date. forming to all federal, state, and local regulations was ob- Normal hematologic, renal, and hepatic functions were tained from each participant before enrollment onto the study. required. Patients had to be free from infection and have a Eligible patients were to receive tamoxifen 20 mg p.o. GOG performance status of 0–2 to be eligible. Patients with twice daily. On alternating (even-numbered) weeks, they other previous or concomitant cancers except non-melanoma also received MPA 100 mg p.o. twice daily. This schedule cancer of the skin were not eligible for this study. Patients was to be repeated until disease progression or adverse with a history of systemic therapy for endometrial cancer, in- effects precluded further therapy. cluding cytotoxic drugs and hormone therapy, were ineligi- Response was recorded and graded according to standard ble. Eligibility was confirmed by the review of submitted GOG criteria. A complete response was defined as the records, forms, and operative reports by the GOG Gyneco- disappearance of all gross evidence of disease for at least 1 logic Oncology Committee. Written informed consent con- month. A 50% or greater reduction in the product of the largest diameter and its perpendicular of each measurable lesion on two separate examinations at least 2 weeks apart in Table 1 the absence of deteriorating performance status or new Patient characteristics lesions was considered a partial response. Any condition Characteristics Number % not meeting these two criteria was considered no response. Age (years) Disease progression was defined as a 50% or greater increase V60 15 25.0 in the product of the largest dimension and its perpendicular 61–70 22 36.7 of any measurable lesion or the appearance of any new lesion. 71–80 16 26.7 Reported adverse effects were graded according to GOG z81 7 11.7 toxicity criteria [17]. Any grade 4 (life threatening) or Race White 50 83.3 Black 8 13.3 Other 2 3.3 GOG performance status Table 3 0 26 43.3 Adverse effects (n = 57) 1 26 43.3 Adverse effect Grade 2 8 13.3 1 234 Cell type Endometrioid 32 53.3 Gastrointestinal 9 3 0 0 Adenocarcinoma, unspecified 6 10.0 Fatigue 2 2 0 0 Serous 14 23.3 Thrombophlebitis 0 1 0 0 Mixed epithelial 3 5.0 Thromboembolic episode 2 1 1 0 Adenoca with sq. diff. 3 5.0 Hepatic 0 1 1 1 Villoglandular 1 1.7 GU 3310 Squamous 1 1.7 Pulmonary 0 1 0 0 Histological grade Dermatologic 3 1 0 0 1 15 25.0 Edema 5 4 0 0 2 17 28.3 Anemia 0 2 2 0 3 27 45.0 Neurological 6 0 0 0 Not specified 1 1.7 Pain 1 2 0 0 Stage Hypertension 0 0 2 0 IV 6 10.0 Cardiac 0 1 0 0 Recurrent 54 90.0 Weight gain or loss 10 8 2 0 Prior RT 36 60.0 Hyperglycemia 0 1 0 0 6 C.W. Whitney et al. / Gynecologic Oncology 92 (2004) 4–9

Table 4 Results Response (n = 58) Response Number of patients Between June 1991 and February 1996, 61 patients were Complete 6 entered into this study. One patient was ineligible because at Partial 13 central pathology review, the primary was not felt to be an No response* 39 endometrial carcinoma. The remaining 60 eligible patients *Three patients were not evaluable for response. constitute the basis of this report. Patient characteristics are displayed in Table 1. Most unexpected adverse effects were to be reported immediately patients were postmenopausal and had good performance to the GOG Administrative Office and to the study chair. status. Tumor characteristics were predominantly adeno- Progression-free interval (PFI) was defined as the time carcinomas of high histological grade (23% were papil- from study entry to clinical or radiographic evidence of lary serous carcinomas) and most were from recurrent disease progression or to the date the patient was last seen. lesions. Sixty percent of the patients had received prior Overall survival (OS) was defined from study entry to death radiation therapy. The sites of measurable disease at entry or date last seen. Progression-free survival (PFS) is equal to are outlined in Table 2. Thirty-seven of the 60 patients the PFI if the patient progressed; otherwise, it is equal to the had disease within the pelvis, of which 25 had received survival time. Life tables were computed using the method prior pelvic radiotherapy. Moreover, there were 32 with of Kaplan and Meier. vaginal disease only. Extra-pelvic disease occurred in 23 This study utilized a standard two-stage phase II design patients, with the most common site of involvement being to allow for early termination due to inactivity. For the lung. purpose of study design, agents with a true probability of Table 3 depicts toxicity for the 57 patients who received response (complete or partial) 0.15 or less were considered study therapy and were evaluated for toxicity. The reported to have insufficient activity to warrant further investiga- toxicities were typical of those encountered with standard tion. On the other hand, a true probability of response dose hormonal therapy. None of the deaths was attributed to 0.30 or higher would be clinically significant and indicates the study treatment. Two patients, who did not receive any that further investigation of the regimen is appropriate. study treatment, are not included in the summary of re- The interim decision rule for this study required three or sponse or toxicity. One patient was never treated and the more responses out of 20 patients to continue to the other received 4 days of MPA only and had no baseline second stage of accrual. The sample size of 58 patients measurements of measurable disease. One patient refused provides an 85% chance of correctly classifying an active further therapy after 1 day and was not evaluated for agent while type I error was limited to 5% [18].A toxicity. confidence interval (CI) adjusted for the two-stage design As shown in Table 4, 19 of 58 patients (33%; 95% CI: is reported [19]. 21–46%) with study drug exposure achieved a response,

Fig. 1. Progression-free survival and survival. C.W. Whitney et al. / Gynecologic Oncology 92 (2004) 4–9 7 with 6 patients achieving a complete response and 13 species-dependent. In the human endometrium, it does patients attaining a partial response. Thirty-six patients increase progesterone receptors. At a dose of 40 mg/day, had no evidence of response during protocol therapy. Three tamoxifen was also shown by Mortel et al [15] to increase patients were not evaluable for response for the following PR concentrations in endometrial carcinoma in vivo. Long- reasons: one patient refused further therapy after 1 day, term treatment of human endometrial carcinoma in nude another patient refused treatment after one course and was mice has also been shown to significantly increase PR in the never evaluated for response, and the third patient discon- tumor tissue [25]. Additionally, the combination of MPA tinued therapy after 3 weeks due to hospitalization and and tamoxifen was found superior to MPA alone in the surgery for a previously reported bowel obstruction. These treatment of endometrial cancer in the nude mouse model patients were classified as having no response. At the time [26]. Based on these clinical and laboratory factors, the of this report, 54 patients are dead from their disease, 3 are current study was undertaken. The overall response rate of alive with disease, and only 1 is alive without any clinical or 33% in this study is consistent with the initial hypothesis. radiographic evidence of disease. Among four patients who Whether this result is actually superior to the previous GOG, died while on study, two were classified as not responding experience with MPA [7,8] will require a randomized trial to and died from endometrial cancer 9 and 24 days, respec- assess. Although pooled data indicated a 22% response rate tively, into study therapy. One patient whose best response for tamoxifen [27], a GOG study of 68 patients reported by was stable disease died 60 days into protocol therapy from Thigpen et al. [14] demonstrated a response rate of only endometrial cancer, and another patient died as a result of 10% for patients treated with single agent tamoxifen at a Alzheimer’s disease after completing nearly 18 courses of dose of 20 mg p.o. twice daily. Pandya et al. [28] reporting study therapy and achieving a PR. Median PFS is 3 months for the Eastern Cooperative Oncology Group (ECOG) did and median OS is approximately 13 months. Survival not find sufficient evidence to suggest a difference between curves are shown in Fig. 1. megestrol acetate as a single agent versus tamoxifen and megestrol acetate. In that phase II study of 42 patients, response rates were similar for the combination and meges- Discussion trol acetate alone: 19% and 20%, respectively. Fiorica et al. [29] in the GOG studied alternating tamoxifen and meges- Progestational agents have been the most commonly trol in the treatment of advanced or recurrent endometrial used form of hormonal therapy for patients with advanced carcinoma. That study of 56 women first treated patients or recurrent endometrial carcinoma [1–8]. Although re- with megestrol 80 mg twice daily for 3 weeks, followed by sponse rates have been reported in the range of 30–35%, tamoxifen 20 mg twice daily for 3 weeks, using the rationale more recent studies suggest even lower frequencies of that the initial dose of megestrol would destroy cancer cells objective response, on the order of 15–25% [7,8]. and the subsequent doses of tamoxifen would recruit further Because the uterus is a sex steroid responsive organ, cancer cells that would then be susceptible to further doses many investigators have explored the usefulness of measur- of megestrol. The overall response rate was 26% (21% ing progesterone receptor (PR) concentrations in endome- complete and 5.4% partial responses). That study had a trial cancers to predict hormone responsiveness in patients much smaller percentage of papillary serous carcinomas with these tumors [20–24]. In fact, the few studies of (7% vs. 23%) and perhaps a more favorable patient popu- receptor status in metastatic endometrial cancer and hor- lation than the current study. Historically, papillary serous monal response do indeed point to a strong predictive value tumors are not considered hormonally responsive. However, for the level of progesterone receptor concentration a requirement for tissue from a metastatic site in this expressed in the tumor tissue. However, the responses are protocol could have introduced a potential bias toward often of short duration, approximately 4 months. Because patients with more responsive disease. In this study, the progestins have been shown to have a negative effect on majority (52%) of patients had measurable disease limited to progesterone receptor concentration (downregulation), Mor- the vagina. This is much higher than that reported by Fiorica tel et al [15] proposed that the reduced effectiveness of these et al., which was not more than 25%. agents and the short duration of response may, at least in In this study, the median PFS was 3 months and median part, be attributed to depletion of PR in these tumors treated survival was 13 months. These results are similar to those with progestational agents. They also postulated that any reported for progestins alone [4,5,7]. In the GOG study agent that augments PR concentrations within the tumor reported by Fiorica et al., the patients exhibited a median (upregulation) might be expected to potentiate the effective- PFS of 2.7 months and a median overall survival of 14 ness of progestin therapy. months. Adverse effects were similar to other series of Estrogenic compounds are known to increase progester- patients treated with hormonal therapy for advanced or one receptor concentration in normal and malignant endo- recurrent endometrial cancer. metrium. Tamoxifen binds to estrogen receptors and Hormonal therapy for advanced or recurrent endometrial translocates with the receptor complex to the nucleus where carcinoma remains an attractive option for selected patients. its estrogenic and antiestrogenic activity is tissue- and Low toxicity with the potential for response makes proges- 8 C.W. Whitney et al. / Gynecologic Oncology 92 (2004) 4–9 tational agents a suitable therapeutic first choice for many of advanced or recurrent endometrial carcinoma: a dose– response patients, particularly for those with hormone receptor study by the Gynecologic Oncology Group. J Clin Oncol 1999;17: 1736–44. tumors. A future manuscript will address the relationship [9] Geisler HE. The use of megestrol acetate in the treatment of ad- between steroid receptors, their isoforms, and response to vanced malignant lesions of the endometrium. Gynecol Oncol 1979; treatment, using this regimen. The 33% response rate 1:340–5. reported herein is among GOG’s most positive experiences [10] Quinn MA, Cauchi M, Fortune D. Endometrial carcinoma: steroid with hormonal therapy for women with advanced or recur- receptors and response to medroxyprogesterone acetate. Gynecol On- col 1985;21:314. rent endometrial carcinoma. Indeed, it approaches that seen [11] Barakat RR, Grigsby PW, Sabbatini P, Zaino RJ. Corpus epithelial with active single agent cytotoxic chemotherapy [11,30,31]. tumors. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and Practice of Gynecologic Oncology. Second ed. Philadelphia, PA: Lip- pincott-Raven; 1997. p. 880. Acknowledgments [12] Swenerton K. Treatment of advanced endometrial adenocarcinoma with tamoxifen. Cancer Treat Rep 1980;64:805. [13] Slavik M, Petty WM, Blessing JA, Creasman WT, Homesley HD. This study was supported by National Cancer Institute Phase II clinical study of tamoxifen in advanced endometrial adeno- grants to the Gynecologic Oncology Group (GOG) Admin- carcinoma: a Gynecologic Oncology Group study. Cancer Treat Rep istrative Office (CA 27469) and the GOG Statistical and 1984;68(5):809–11. Data Center (CA 37517). The following member institutions [14] Thigpen JT, Brady MF, Homesley HD. Tamoxifen in the treatment of participated in this study: advanced or recurrent endometrial carcinoma: a Gynecologic Oncol- ogy Group study. J Clin Oncol 2001;19:364–7. Duke University Medical Center, Abington Memorial [15] Mortel R, Levy C, Wolff JP, Nicolas JC, Robel P, Baulieu EE. Female Hospital, Walter Reed Army Medical Center, University of sex steroid receptors in postmenopausal endometrial carcinoma and Minnesota Medical School, University of Mississippi biochemical response to an antiestrogen. Cancer Res 1981;41(3): Medical Center, Milton S. Hershey Medical Center, 1140–7. University of Cincinnati, University of Iowa Hospitals and [16] Carlson Jr JA, Allegra JC, Day Jr TG, Wittliff JL. Tamoxifen and endometrial carcinoma: alterations in estrogen and progesterone re- Clinics, University of Texas Southwestern Medical Center ceptors in untreated patients and combination hormonal therapy in at Dallas, Wake Forest University School of Medicine, advanced neoplasia. Am J Obstet Gynecol 1984;149(2):149–53. University of California Medical Center at Irvine, Tufts- [17] Gynecologic Oncology Group (GOG) Common Toxicity Criteria; New England Medical Center, The Cleveland Clinic October 1988. Foundation, State University of New York at Stony Brook, [18] Schultz JR, Nichol FR, Elfring GL, Weed SD. Multiple-stage proce- dures for drug screening. Biometrics 1973;29:293–300. Columbus Cancer Council, Fox Chase Cancer Center, [19] Atkinson EN, Brown BW. Confidence limits for probability of Medical University of South Carolina, University of response in multistage phase II clinical trials. Biometrics 1985;41: Oklahoma, University of Virginia, University of Chicago, 741–4. and Tacoma General Hospital. [20] Billiet G, DeHertogh R, Bonte J, Ide P, Vlaemynck G. Estrogen receptor in human uterine adenocarcinoma: correlation with tissue differentiation, vaginal, karypyenotic index and effects of progesterone on anti-estrogen treatment. Gynecol Oncol 1982;14(1):33–9. References [21] Creasman WT, McCarty Sr KS, Barton TK, McCarty Jr KS. Clinical correlates of estrogen and progesterone binding proteins in human [1] Kohorn E. Gestagens and endometrial carcinoma. Gynecol Oncol endometrial adenocarcinoma. Obstet Gynecol 1980;55(3):363–70. 1976;4:398. [22] Ehrlich C, Young P, Cleary R. Cytoplasmic progesterone and estra- [2] Kauppila A. Progestin therapy of endometrial, breast and ovarian diol receptors in normal, hyperplastic and carcinomatous endometria: carcinoma. A review of clinical observations. Acta Gynecol 1989; therapeutic implications. Amer J Obstet Gynecol 1981;141:539. 63:441. [23] Freedman M, Hoffman P, Jones W. The clinical value of hormone [3] Kelley RM, Baker WH. Progestational agents in the treatment of receptor assays in malignant disease. Cancer Treat Rev 1981;5:185. carcinoma of the endometrium. N Engl J Med 1961;264:216–9. [24] Janne O, Kauppila A, Kontula K, Syrjala P, Vihko R. Female sex [4] Piver MS, Barlow JJ, Luain JR, Blumenson LE. Medroxyprogeste- steroid receptors in normal, hyperplastic and carcinomatous endome- rone acetate versus hydroxyprogesterone caproate in women with trium. The relationship to serum steroid hormones and gonadotropins metastatic endometrial adenocarcinoma. Am J Clin Oncol 1980;45: and changes during medroxyprogesterone acetate administration. Int J 268–72. Cancer 1979;24(5):545–54. [5] Reifenstein EC. The treatment of advanced endometrial cancer with [25] Sataswaroop PG. Estrogen like effects of tamoxifen on human endo- hydroxyprogesterone caproate. Gynecol Oncol 1974;2:377. metrial carcinoma transplanted into nude mice. Cancer Res 1984; [6] Bonte J. Medroxyprogesterone in the management of primary and 44:4006. recurrent endometrial adenocarcinoma. Gynecol Oncol 1981;11:140. [26] Zaino RJ, Satyaswaroop PG, Mortel R. Hormonal therapy of human [7] Thigpen JT, Blessing J, DiSaia P. Oral medroxyprogesterone acetate endometrial adenocarcinoma in a nude mouse model. Cancer Res in advanced or recurrent endometrial carcinoma: results of therapy 1985;45:535. and correlation with estrogen and progesterone receptor levels. In: [27] Moore TD, Phillips PH, Nerenstone SR, Cheson BD. Systemic treat- Iacobelli S, editor. Endocrinology and Malignancy: Basic and Clin- ment of advanced and recurrent endometrial carcinoma: current status ical Issues, The Proceedings of the First International Congress on and future directions. J Clin Oncol 1991;9(6):1071–88. Cancer and Hormones. CRC Press –Parthenon Publishers; Rome; [28] Pandya KJ, Yeap BY, Davis TE. Phase II study of megestrol and 1986 (April). p. 446–54. mestrol and tamoxifen in advanced endometrial carcinoma: an Eastern [8] Thigpen JT, Brady MF, Alvarez RD, Adelson MD, Homesley HD, Cooperative Oncology Group (ECOG) study. Am J Clin Oncol 2001; Manetta A, et al. Oral medroxyprogesterone acetate in the treatment 24(1):43–6. C.W. Whitney et al. / Gynecologic Oncology 92 (2004) 4–9 9

[29] Fiorica JV, Brunetto VL, Hanjani P, Lentz S, Mannel R, Andersen E. Phase III trial of doxorubicin F cisplatin in advanced or recurrent W. A phase II study of recurrent and advanced endometrial carci- endometrial carcinoma: a Gynecologic Oncology Group study. Proc noma treated with alternating courses of megestrol acetate (Megace) ASCO 12:261 (Abstr #830); 1993 [abstract]. and tamoxifen citrate (Nolvadex). Proc ASCO (Abst# 1499); 2000 [31] Ball HG, Blessing JA, Lentz SS, Mutch DG. A phase II trial of taxol in [abstract]. advanced or recurrent adenocarcinoma of the endometrium: a Gyneco- [30] Thigpen JT, Blessing JA, Homesley H, Malfetano J, DiSaia P, Yordan logic Oncology Group study. Gynecol Oncol 1996;62(2):278–81.