Do selective estrogen receptor modulators treat cervical precancer and cancer? Time to pool data from relevant trials

Abstract

Based on the nearly absolute etiologic link between carcinogenic human papillomavirus (HPV) and cervical cancer, two new approaches for the prevention of cervical cancer have emerged: (1) HPV vaccination for primary HPV prevention in younger women and (2) carcinogenic HPV detection for secondary prevention via identifying and treating cervical precancer and early cancers. Both have demonstrated high degrees of efficacy with maximum effectiveness guided by an understanding of the causal model and application of these technologies in an age appropriate manner.1 Despite the advent of these promising prevention tools, there is a real need to develop nonsurgical methods for treating cervical precancerous lesions and even early cancer for three reasons. First, current HPV vaccines are prophylactic and do not treat pre-existing HPV infections.1 Second, surgical excision of screen-detected precancerous lesions, while a highly efficacious (90–95%) treatment1 increases the risk of preterm delivery and infant morbidity and mortality.2 Third, most women who screen positive by a Pap and/or HPV do not have clinically actionable disease, yet these women are at risk of cervical precancer and cancer in the future and must be followed up intensively. In addition, in the low-resource settings, surgical methods are not commonly available and cryotherapy, while moderately efficacious when its use is limited to smaller lesions, causes significant watery discharge and requires sexual abstinence for several weeks to permit healing of treated epithelium. Despite the knowledge that HPV infection is the obligate cause of cervical cancer, targeted immunological approaches for therapy such as therapeutic vaccines have been unsuccessful.3 Chung and Lambert4 present a novel approach of using estrogen receptor antagonists (selective estrogen receptor modulators or SERMs) to treat cervical precancer and cancer. As noted by the authors, there are strongly supportive laboratory5 and epidemiological data6, 7 that estrogen plays contributing role in cervical carcinogenesis. In this study, Chung and Lambert4 successfully treated 7 E6/E7 transgenic mice that spontaneously manifested cervical cancer with Raloxifene, an estrogen receptor antagonist that reduces the risk of breast cancer. To investigate whether Raloxifene might reduce the incidence of cervical cancer and its immediate precursor, carcinoma in situ (CIS), previously unpublished data on these outcomes from two randomized clinical trials to evaluate two SERMs, tamoxifen and raloxifene, for breast cancer risk reduction were analyzed. The NSABP (National Surgical Adjuvant Breast and Bowel Project), an NCI funded Cooperative Group, has designed and conducted the two large breast cancer prevention trials, Breast Cancer Prevention Trial P-1 (BCPT-P1)8 and Study of Tamoxifen vs. Raloxifene (STAR).9 While both tamoxifen and raloxifene are estrogen receptor antagonists in breast tissue, raloxifene is an estrogen receptor antagonist whereas tamoxifen is an agonist in the human reproductive tract (e.g., Tamoxifen increases the risk of endometrial cancer while raloxifene does not.).4, 10, 11 However, as shown in the Table 1, despite small numbers, there was no evidence that long-term raloxifene use reduces, or long-term tamoxifen use increases, the risk of cervical CIS and cancer (n.b., results were combined because SERMS were reported to treat both in the mouse models, and CIS has a high risk of invading if left untreated12) (BCPT P-1: Placebo vs. Tamoxifen, p = 1.0, two-sided Fisher's exact; STAR: Tamoxifen vs. Raloxifene, p = 1.0, two-sided Fisher's exact). This post-hoc analysis was limited by small numbers. Data from other clinical trials of SERMs13, 14 should be pooled to examine the impact of SERMs on outcomes of cervical precancer and cancer. Until evidence is forthcoming on the positive effects of SERMs on cervical precancer and cancer in humans, extreme caution in interpreting data from mouse models is warranted. Despite the probable role of estrogens in cervical carcinogenesis, the protective effects of SERMs against cervical cancer cannot be inferred based on mouse models. Importantly, the therapeutic effectiveness of SERMs would need to be exceeding high to be acceptable, given the efficacy of current methods of treatment for cervical intraepithelial neoplasia15 and early cervical cancer.16 Dr. Castle acknowledges Dr. Joseph Costantino (University of Pittsburgh, Pittsburgh, PA, USA) for providing data from Breast Cancer Prevention Trial P-1 (BCPT-P1) and the Study of Tamoxifen vs. Raloxifene (STAR). Dr. Castle was supported by the Intramural Research Program of the NIH, National Cancer Institute and he reports no conflicts of interest. Yours sincerely, Philip E. Castle