Gynecologic Manifestations of Hereditary Nonpolyposis Colorectal Cancer

Traditionally, most hereditary nonpolyposis colorectal cancer (HNPCC) syndrome patients have been identified and cared for by gastroenterologists, colorectal surgeons, and gastrointestinal medical oncologists. Hence, the realization that gynecologic tumors actually play a major role in HNPCC has come relatively late. Consequently, much of the clinical and basic science focus of research in HNPCC has concentrated on colorectal cancer. The rationale for such focus is due largely to data such as that presented in Table 2 of the article by Drs. Taylor and Mutch. Table 2 shows that the cumulative incidence of colorectal cancer in HNPCC (men and women) is 82%, while the cumulative incidence of endometrial cancer and ovarian cancer is 60% and 12%, respectively. However, it is instructive to analyze such data more carefully. Two previous studies[1,2] have concluded that women with HNPCC have a 60% lifetime risk of developing endometrial cancer but only a 39% to 54% lifetime risk of developing colorectal cancer. In men, these studies estimated a 74% to 83% risk of colorectal cancer. Therefore, it is important to note that the cancer risks for men and women with HNPCC are strikingly different. Sequence of Cancer Presentations
Most women with HNPCC have been identified as mutation carriers because colorectal cancer was diagnosed in the woman or family member at a relatively young age. In a recently published study,[3] we hypothesized that in a proportion of women with HNPCC, gynecologic cancers presented first, ie, before the development of colorectal cancer. Five different hereditary cancer registries were examined for women with dual colorectal/gynecologic cancers. A total of 117 women with dual primary cancers from 223 Amsterdam families were identified. We found that 51% of these women had an endometrial or ovarian cancer diagnosed first. For these women, the median time between the diagnosis of a gynecologic cancer and the diagnosis of colorectal cancer was 11 years. Therefore, for about one-half of women with HNPCC, gynecologists and gynecologic oncologists may actually be the first members of the health-care team with the opportunity to identify new mutation carriers. Early identification of a patient at the time of initial gynecologic cancer diagnosis is crucial in proactively managing and reducing the patient's risk for subsequent colorectal cancer. Available Data
Clearly, gynecologic cancers are important in HNPCC. What, then, do we know about these tumors? Our knowledge of HNPCC-associated endometrial cancer is not detailed, but it is far greater than our understanding of HNPCC-associated ovarian cancer. As mentioned in the article by Drs. Taylor and Mutch, microsatellite instability (MSI) can result from germline mutation of MLH1 and MSH2 (HNPCC) or from methylation of the MLH1 promoter, resulting in loss of MLH1 protein (sporadic endometrial and colorectal cancer). It has been known for some time that sporadic, MSI-high endometrial cancer due to MLH1 methylation is almost exclusively associated with endometrioid tumors, higher International Federation of Gynecology and Obstetrics (FIGO) grade, and advanced stage.[4-6] We found that HNPCC-associated endometrial cancer, while sharing the common molecular abnormality of MSI, actually includes a broader spectrum of tumor histotypes, such as endometrioid adenocarcinoma, papillary serous carcinoma, clear cell carcinoma, and malignant mixed müllerian tumor. In fact, the endometrial tumor spectrum for HNPCC more closely mirrored that of the general population than that for MLH1 methylation. Importantly, analysis of these HNPCCrelated endometrial cancers revealed that nearly 25% of them had pathologic features (deep myometrial invasion greater than 50% of the myometrial wall thickness; cervix involvement; lymph node or adnexal metastasis) that would necessitate adjuvant therapy following hysterectomy.[7] Screening and Prevention
The literature offers limited information as to screening for endometrial cancer. Clinical guidelines recommend screening for endometrial cancer to begin at age 25 to 35.[8] The appropriate frequency of such screening is entirely unclear. As part of an NCI-sponsored chemoprevention study, we are currently investigating oral contraceptives and medroxyprogesterone(Drug information on medroxyprogesterone) acetate (Depo- Provera) as possible chemopreventive agents for endometrial cancer in HNPCC. For this study, women undergo baseline and 3-month post-treatment endometrial biopsies and transvaginal ultrasounds. A component of this study is to examine, by quantitative polymerase chain reaction (PCR), a series of endometrial tissue biomarkers that may give us clues as to identifying women who are particularly at risk and who should be screened more often. To date, this chemopreventive study has enrolled 32 women, and we anticipate completing this study in about 2 years. Once women have completed childbearing, it might be reasonable to assume that a prophylactic hysterectomy and bilateral salpingo-oopherectomy (BSO) is an option to reduce the gynecologic cancer risk in women with HNPCC. As mentioned in the article by Drs. Taylor and Mutch, we have insufficient evidence to recommend for or against prophylactic surgery. We examined 315 women with documented HNPCC germline mutations from three different cancer registries. Women who had undergone prophylactic hysterectomy with or without BSO were matched to mutation-positive women who did not have surgery. For the prophylactic surgery group, none of the women developed endometrial, ovarian, or primary peritoneal cancer. In the control group, 33% of the women developed endometrial cancer, and 5% developed ovarian cancer. Therefore, this study provides evidence for the protective benefit of prophylactic hysterectomy and BSO in the prevention of endometrial and ovarian cancer in women with HNPCC.[9] Precursor Lesion
Endometrial complex hyperplasia with atypia (CAH) is a well-recognized precursor lesion for endometrioid adenocarcinoma, the most common histologic subtype of endometrial cancer. It is known that approximately 29% of women with CAH detected on endometrial biopsy will progress to endometrial cancer.[10] For HNPCC-associated colon cancer, it has been hypothesized that colon adenomas, especially proximal ones, are more likely to progress to colonic adenocarcinoma (and progress more rapidly) than adenomas in the general population.[11-13] A similar hypothesis could be posed for CAH in HNPCC-associated endometrial cancer, but at this time we do not have sufficient data to address this issue. For the NCI-sponsored endometrial cancer chemoprevention trial that we are conducting at M. D. Anderson Cancer Center, we have encountered two women with HNPCC who had CAH at their baseline endometrial biopsies. At hysterectomy, both of these women had endometrial endometrioid adenocarcinoma, grade 1, associated with complex hyperplasia. Neither tumor was invasive. Therefore, our limited information at this time suggests that CAH is indeed a part of the pathogenesis of endometrial endometrioid tumors in HNPCC. However, the rate of cancer progression in HNPCC-associated CAH remains unknown. Ovarian Cancer
The data regarding ovarian cancer in HNPCC is even more limited. A previous study examined the medical records of 80 ovarian cancer patients from HNPCC families based on germline mutation or clinical criteria.[14] The majority (94%) of these tumors were epithelial cancers. Approximately 56% had papillary serous ovarian cancer, and 18% had endometrioid ovarian cancer. Surprisingly, 84% of the women had stage I or II disease. In contrast, more than 70% of women with sporadic ovarian cancer present with advanced-stage disease. Many of the ovarian cancer cases reported in this study were several decades old and predated the establishment of many of the current guidelines for pathologically distinguishing borderline tumors from invasive cancers. A study involving careful centralized pathologic review of the ovarian tumor slides from women with HNPCC would be useful to further characterize ovarian cancer in HNPCC. In the United States, our understanding of the gynecologic manifestations of HNPCC, and, in fact, HNPCC in general, would be greatly accelerated by the establishment of a centralized study group for HNPCC. We especially encourage the gynecologic community to participate in research efforts, given the substantial risk of gynecologic cancers in women with HNPCC.