Development of metastatic endometrial endometrioid adenocarcinoma while on progestin therapy for endometrial hyperplasia

Gynecologic Oncology 99 (2005) 472 – 476 www.elsevier.com/locate/ygyno

Case Report

Development of metastatic endometrial endometrioid adenocarcinoma while on progestin therapy for endometrial hyperplasia Jennifer M. Rubatt a, Brian M. Slomovitz b, Thomas W. Burke b, Russell R. Broaddus c,* b

a Department of Obstetrics and Gynecology, VCUMC Medical College of Virginia, VA 23287, USA Department of Gynecologic Oncology, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, USA c Department of Pathology, Box 85, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, USA

Received 21 April 2005 Available online 15 August 2005

Abstract Background. Conservative treatment with progestins is a reasonable treatment option for endometrial complex atypical hyperplasia and, in the experimental setting, for some women with grade 1 endometrial endometrioid adenocarcinoma. The risk of progression to a high-stage endometrial cancer is quite low, with only two previously reported cases in the English literature. Case. A 40-year-old woman with endometrial complex atypical hyperplasia diagnosed by dilatation and curettage was managed conservatively with progestin therapy (initially, megesterol acetate; then, a combination oral contraceptive). More than 2 years after her original diagnosis, she developed endometrial endometrioid adenocarcinoma, FIGO grade 2, with lymph node metastasis. The tumor was microsatellite instability-high due to methylation of MLH1 and loss of MLH1 protein. Conclusion. Currently, there are no good criteria for predicting which patients with complex atypical hyperplasia/grade 1 endometrioid adenocarcinoma will optimally respond to progestin therapy. There is some evidence that endometrial complex hyperplasia demonstrating loss of MLH1 protein by immunohistochemistry is strongly related to subsequent or concurrent endometrial cancer, especially tumors of higher grade and stage. In a woman with a biopsy diagnosis of endometrial hyperplasia, evaluation of MLH1 protein status by immunohistochemistry may provide useful information when medical management is being considered. D 2005 Elsevier Inc. All rights reserved. Keywords: Endometrial hyperplasia; Endometrial cancer; Progestin therapy; MLH1; Microsatellite instability

Introduction Endometrial complex atypical hyperplasia (CAH) is a precursor to endometrioid adenocarcinoma, the most common histological subtype of endometrial cancer. Untreated, CAH carries a risk of progression to carcinoma of approximately 25% [1]. The treatment standard of care for women with CAH is usually hysterectomy. In women who desire to maintain fertility, progestin-based medical therapy is an alternate option. Progestational treatment has been shown to decrease progression to carcinoma as well as cause regression of hyperplasia in the majority of patients [2,3]. In

* Corresponding author. Fax: +1 713 792 5532. E-mail address: [email protected] (R.R. Broaddus). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.06.050

addition, the conservative management of grade 1 endometrial endometrioid adenocarcinoma has also been attempted, with documented regression in several small series [2– 5]. The decision to conservatively manage CAH or grade 1 endometrioid adenocarcinoma with progestins is felt to be safe because the risk of progressing to a higher stage cancer is so low. We have identified only two reported patients with extrauterine disease at the time of hysterectomy [6]). Both of these patients initially responded to progestin therapy. Here, we report a young woman initially diagnosed with CAH who was managed conservatively with progestin therapy. More than 2 years after her initial biopsy diagnosis, she again developed vaginal bleeding. Pathological evaluation of the uterus and pelvic lymph nodes revealed a FIGO grade 2 endometrial endometrioid adenocarcinoma with nodal metastasis.

J.M. Rubatt et al. / Gynecologic Oncology 99 (2005) 472 – 476

Case report A 40-year-old Caucasian G0P0 woman presented to the outpatient Gynecologic Oncology Clinic with a 2-month history of heavy vaginal bleeding. Prior to this, she had normal, regular menstrual cycles since the age of 9. She was not infertile. The patient denied a history of dysmenorrhea, dyspareunia, postcoital bleeding, or a change in bowel habits. She had used oral contraceptives 15 years earlier. She denied a history of diabetes mellitus, hypertension, or tobacco use. She had gallstone pancreatitis in the past. Her mother had a history of breast cancer. She denied a family history of endometrial cancer, ovarian cancer, or colon cancer. On physical examination, the patient was morbidly obese (BMI 51.6 kg/m2). Significant findings included a small polyp protruding from the cervical os, which biopsy showed to be benign endocervix with squamous metaplasia. Otherwise, there were no significant findings on physical examination. Hysteroscopy demonstrated a polypoid overgrowth throughout the endometrial cavity. Pathological evaluation of the dilatation and curettage specimen demonstrated endometrial complex atypical hyperplasia (Fig. 1A). A chest X-ray and abdominal MRI with and without contrast demonstrated normal findings and no evidence of metastatic disease. A pelvic MRI with and without contrast showed a small amount of fluid in the endometrial cavity and cervix with normal-sized ovaries. No masses or enlarged lymph nodes were detected. Given her desire to preserve fertility, the patient opted for conservative management and was started on megesterol acetate (Megace) 40 mg twice daily. After taking Megace for 2 weeks, the patient developed severe headaches that were not well-tolerated, so therapy was changed to continuous combination contraceptive levonorgestrel and ethinyl estradiol (Alesse) with which she was compliant. Three months later, the patient returned for a follow-up endometrial biopsy, which revealed weak secretory endometrium and fragments of benign endocervical tissue. The patient continued on Alesse therapy and returned to the clinic every 6 months for routine follow-up. At each visit, she denied abnormal vaginal bleeding. She did have normal withdrawal bleeding during the placebo week of Alesse. Two and a half years after her original diagnosis of hyperplasia, she began having irregular vaginal spotting while still on Alesse. Changes in medical history over this time period included a 24 kg weight loss and a diagnosis of hypertension, controlled with an oral anti-hypertensive agent. An endometrial biopsy revealed FIGO grade 2 endometrial endometrioid adenocarcinoma. The patient underwent an exploratory laparotomy, pelvic washings, total abdominal hysterectomy, bilateral salpingooophorectomy, pelvic lymph node dissection, omentectomy, and appendectomy. Intraoperative findings included a small uterus and normal-sized ovaries. There were some slightly enlarged lymph nodes in the right pelvis. The upper abdomen was normal. Pathologic review revealed an

473

endometrial endometrioid adenocarcinoma, FIGO grade 2, with papillary architecture (Fig. 1B). The depth of myometrial invasion was 1.0 cm, with a total myometrial thickness of 2.5 cm. There were numerous foci of lymphatic/vascular invasion within the myometrium. The endocervical stroma was focally involved by adenocarcinoma. One of nine lymph nodes from the right pelvis revealed metastatic adenocarcinoma (Fig. 1C). The left pelvic lymph nodes, omentum, and appendix were free of disease. Pelvic cytology revealed metastatic adenocarcinoma. The final pathological stage was therefore stage IIIC. Interestingly, the uninvolved endometrium microscopically demonstrated pre-decidualization of endometrial stromal cells with inactive glands, consistent with exogenous progestin effect (Fig. 1D). Because of the patient’s young age, the possibility of endometrial cancer arising in the hereditary non-polyposis colorectal cancer (HNPCC) family cancer syndrome was evaluated. Immunohistochemistry for the DNA mismatch repair gene products MLH1 and MSH2 was performed as previously described [8]. The tumor showed diffuse and strong nuclear expression for MSH2 (Fig. 2A). However, the tumor was negative for MLH1 (Fig. 2B). Microsatellite instability (MSI) analysis was performed using formalinfixed, paraffin-embedded sections of uterine tumor and corresponding normal uterus as previously described [8]. Briefly, DNA was extracted from microdissected areas of tumor and normal tissue from sections on glass slides. Following DNA amplification using fluorescent labeled primers, a panel of six microsatellites recommended by the NCI [9], BAT25, BAT26, BAT40, D2S123, D5S346, and D173250 was analyzed for allelic shift using a multiplex fluorescence-based PCR assay. Tumors with allelic shift in two or more markers are considered MSI-high. The uterine tumor demonstrated allelic shift in all six markers examined, indicating that the tumor was MSI-high (Fig. 3). Loss of MLH1 immunohistochemical staining and MSI-high may be due to germline mutation, as in HNPCC, or due to MLH1 promoter methylation, as in sporadic endometrial cancer. Methylation-specific PCR for MLH1 was performed as described previously [8]. Briefly, DNA was isolated from formalin-fixed, paraffin-embedded tissue sections that were microdissected to provide a relatively pure tumor sample. Following treatment with bisulfite, the DNA was amplified by PCR using primers that were specific for either the methylated (M) or the unmethylated (U) versions of MLH1 (MLH1-M forward, 5V-gatagcgatttttaacgc-3V and MLH1-M reverse, 5V-tctataaattactaaatctcttcg-3V; MLH1-U forward, 5Vagagtggatagtgatttttaatgt-3V and MLH1-U reverse, 5V-actctataaattactaaatctcttca-3V). PCR products were separated on 6% polyacrylamide gels and visualized after staining with ethidium bromide. The uterine tumor demonstrated methylation of MLH1 (Fig. 4). Thus, the negative family history for colon, endometrial, or ovarian cancer and the presence of tumor MLH1 methylation suggested that the possibility of HNPCC was unlikely.

474

J.M. Rubatt et al. / Gynecologic Oncology 99 (2005) 472 – 476

Fig. 1. (A) Endometrial complex hyperplasia with atypia detected in the initial biopsy. (B) Endometrial endometrioid adenocarcinoma, FIGO grade 2, approximately 2 years following the initial diagnosis of hyperplasia. (C) Lymph node with metastatic endometrial cancer (arrow). This lymph node had a number of similar foci of metastasis. (D) Non-neoplastic endometrium from the hysterectomy specimen showing histological changes characteristic of progestin therapy. All photomicrographs H&E, 200.

The original endometrial biopsy with CAH was retrospectively evaluated for MLH1 and MSH2 immunohistochemical staining. For MSH2, the hyperplastic glands diffusely showed strong nuclear staining (Fig. 2C). However, for MLH1, the hyperplastic glands were negative.

Endometrial stromal cells, an internal positive control, were positive (Fig. 2D). Post-operatively, the patient had external beam radiation (50 Gy) to the pelvis followed by 2 high dose intracavitary vaginal dome and cylinder treatments. This was followed by

Fig. 2. MSH2 and MLH1 immunohistochemistry. Both the endometrioid adenocarcinoma (A) and the hyperplasia biopsy (C) demonstrated strong nuclear expression of MSH2. However, both the adenocarcinoma (B) and the hyperplasia biopsy (D) were entirely negative for MLH1. For MLH1, the few positively staining cells are benign stromal cells, which serve as an internal positive control. All photomicrographs 200.

J.M. Rubatt et al. / Gynecologic Oncology 99 (2005) 472 – 476

475

Fig. 3. Representative results of microsatellite instability analysis for the endometrial endometrioid adenocarcinoma. The uterine tumor had allelic shift (exhibited in the analysis as additional abnormal peaks in tumor DNA compared to normal DNA) in all 6 microsatellites examined (results for D5S346 and D173250 not shown).

chemotherapy with 4 cycles of single-agent carboplatin. One year following surgery, she continues to be free of recurrence.

Discussion Our review of the English literature revealed two other patients who developed extrauterine disease following the conservative management with progestin therapy of FIGO grade 1 endometrial endometrioid adenocarcinoma. We found no other reported cases of high stage endometrial cancer following progestin therapy of CAH. The first patient, 36 years old, was initially diagnosed with grade 1 endometrioid adenocarcinoma and was treated with medroxyprogesterone acetate, 600 mg/day, for 9 months [6]. At hysterectomy, she had a non-invasive grade 1 endometrioid adenocarcinoma. However, 13 months later, she developed metastasis in a left obturator lymph node. This woman was free of metastatic disease 2 years after the second surgery to remove the positive lymph node. The second patient, 31 years old, was initially diagnosed with grade 1 endometrioid adenocarcinoma and was treated with progestins [7]. Two

Fig. 4. Analysis of MLH1 methylation by methylation-specific PCR. Methylation of MLH1 is present in the positive control (RKO colon carcinoma cell line) and the uterine tumor, indicated by the presence of a band in the M (methylated) lane. The negative control (HEC-1A endometrial cancer cell line) shows no evidence of MLH1 methylation, indicated by the presence of a band only in the U (unmethylated) lane.

years after the initial diagnosis, a hysterectomy revealed grade 1 endometrioid adenocarcinoma with extrauterine disease. Twenty-seven months after this surgery, the patient was alive but had developed additional regional metastases. It is important to note that both of these patients, as well as the patient presented in our case report, initially showed biopsy-proven complete responses to conservative progestin therapy. In fact, the non-neoplastic endometrium in the hysterectomy specimen for our patient microscopically showed features associated with exogenous progestin therapy, indicating a continued response of at least a portion of the endometrium. Unfortunately, there are no established, accepted criteria for pathologically distinguishing grade 1 endometrial cancer from complex hyperplasia, which causes poor diagnostic reproducibility [10]. Furthermore, the decision to treat CAH conservatively should be weighed against the considerable rate (43%) of concurrent adenocarcinomas in follow-up hysterectomy specimens [10]. The decision to conservatively treat is further complicated by the fact that there are no established guidelines for choosing a progestational agent, dose of progestin, and duration of progestin use. Additionally, there are no good pathologic indicators that are predictive of response. Because of the patient’s young age, we evaluated MLH1 and MSH2 by immunohistochemistry and found that the tumor had lost MLH1 expression. This loss of MLH1 protein expression was associated with MSI-high and methylation of the MLH1 gene. In a recent study, loss of MLH1 immunohistochemical expression in endometrial hyperplasia biopsies was significantly associated with the subsequent development of endometrial cancer [11]. Furthermore, there is evidence from a small recent study that MSI-high endometrial cancers are more likely to be higher grade and higher stage than microsatellite stable

476

J.M. Rubatt et al. / Gynecologic Oncology 99 (2005) 472 – 476

endometrial cancers [12]. Of course, the results of these studies need to be confirmed by larger studies. In aggregate, the data suggest that assessment of MLH1 expression by immunohistochemistry may be useful in the treatment planning of women with CAH. When considering conservative therapy, caution should be exercised for cases in which the CAH has lost immunohistochemical expression of MLH1; such cases may be at higher risk for developing a higher grade or higher stage endometrial cancer. These women might be medically managed more effectively with higher dose progestational agents or followed in the clinic at more frequent time intervals.

Acknowledgments This work was supported by NIH N01-CN-05127 (Chemoprevention of Endometrial Cancer in HNPCC) and NIH 1P50CA098258-01 (SPORE in Uterine Cancer). Dr. Rubatt was a Felix Rutledge Fellow in the Department of Gynecologic Oncology, M.D. Anderson Cancer Center.

References [1] Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia: a long term study of untreated hyperplasia in 170 patients. Cancer 1985;56:403 – 12. [2] Randall TC, Kurman RJ. Progestin treatment of atypical hyperplasia and well-differentiated carcinoma of the endometrium in women under age 40. Obstet Gynecol 1997;90:434 – 40.

[3] Jadoul P, Donnez J. Conservative treatment may be beneficial for young women with atypical endometrial hyperplasia or endometrial adenocarcinoma. Fertil Steril 2003;80:1315 – 24. [4] Kim YB, Holschneider CH, Ghosh K, Nieberg RK, Montz FJ. Progestin alone as primary treatment of endometrial carcinoma in premenopausal women. Cancer 1997;79:320 – 7. [5] Ramirez PT, Frumovitz M, Bodurka DC, Sun CC, Levenback C. Hormonal therapy for the management of grade 1 endometrial adenocarcinoma: a literature review. Gynecol Oncol 2004;95:133 – 8. [6] Kaku T, Yoshikawa H, Tsuda H, Sakamoto A, Fukunaga M, Kuwabara Y, et al. Conservative therapy for adenocarcinoma and atypical endometrial hyperplasia of the endometrium in young women: central pathologic review and treatment outcome. Cancer Lett 2001;167:39 – 48. [7] Greenblatt RB, Gambrell RD, Stoddard LD. The protective role of progesterone in the prevention of endometrial carcinoma. Pathol Res Pract 1982;174:297 – 318. [8] Broaddus RR, Lynch PM, Lu KH, Luthra R, Michelson SJ. Unusual tumors associated with the hereditary nonpolyposis colorectal cancer syndrome. Mod Pathol 2004;17:981 – 9. [9] Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshelman JR, Burt RW, et al. A National Cancer Institute workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998;58: 5248 – 57. [10] Trimble CL. Atypical endometrial hyperplasia: a tough call. Int J Gynecol Cancer 2005;15:401. [11] Orbo A, Nilsen MN, Arnes MS, Pettersen I, Larsen K. Loss of expression of MLH1, MSH2, MSH6, and PTEN related to endometrial cancer in 68 patients with endometrial hyperplasia. Int J Gynecol Pathol 2003;22:141 – 8. [12] Hirasawa A, Aoki D, Inoue J, Imoto I, Susumu N, Sugano K, et al. Unfavorable prognostic factors associated with high frequency of microsatellite instability and comparative genomic hybridization analysis in endometrial cancer. Clin Cancer Res 2003;9:5675 – 82.