Is There an Association between Type II Endometrial Carcinomas and Breast Cancers? A Critical Appraisal

“Several studies have now raised the possibility of an association between breast cancer, the most prevalent noncutaneous malignancy in American women, and type II endometrial carcinomas.”

Over 47,000 endometrial carcinomas are diagnosed annually in the USA, making them the most prevalent gynecologic tract malignancy [1]. For several decades, endometrial carcinomas have been classified into two broad groups based on epidemiologic, pathologic and clinical factors [2]. In this dualistic model, type I endometrial carcinomas, the prototype that is the endometrioid histotype, constitute the vast majority of endometrial cancers, are associated with hyperestrogenic states, are predominantly uterus-confined at presentation, and are generally associated with favorable patient outcomes [2]. Type II endometrial carcinomas are more heterogeneous and typically display the opposite of the aforementioned features for type I cancers [2]. The appropriate ‘typing’ and classification of some histotypes, including clear cell carcinomas and high-grade endometrioid carcinomas, remains a matter of debate in this model. However, at present, endometrial serous carcinomas (ESC), clear cell carcinomas and carcinosarcomas are all classified as type II cancers. The risk factors associated with type II relative to type I cancers are not entirely clear, and include nonwhite race, later age at menarche, multiparity, older age and nonobesity [3]. Several studies have now raised the possibility of an association between breast cancer, the most prevalent noncutaneous malignancy in American women [1], and type II endometrial carcinomas. The evidentiary basis for this association is appraised herein.

The first series of studies posed the two related questions of whether a personal history of breast cancer increases the risk of type II cancers (relative to the general population and/or type I cancers), and whether there is an increased risk of breast cancer in patients with a personal history of type II cancers. In their study of 1166 consecutive cases of endometrial carcinoma, Gehrig et al. found that 54 patients had a previous diagnosis of breast cancer [4]. In total, 13 of these 54 carcinomas were found to be ESC, as compared with 120 out of the 1112 endometrial carcinomas in patients without a history of breast cancer. The authors concluded that women with a history of breast cancer who subsequently developed an endometrial cancer had a 2.6-fold increased risk of developing ESC as compared with endometrioid carcinoma [4]. These findings were confirmed in a population-based study of 52,109 women diagnosed with corpus cancer in the Surveillance Epidemiology and End Results database [5]. The authors reported that 9.4% of the 1922 women with a history of breast cancer had ESC, compared with the 6.3% of those without such a history. In their analysis of 1178 endometrial cancer patients, results from Liang et al. [6] largely mirrored those of the aforementioned studies [4,5]: 19.4% of women with ESC had a history of breast cancer, compared with 3% of women with endometrioid carcinoma. Furthermore, the frequency of a personal history of breast cancer was noted to be markedly elevated in the <55 years age group (41 vs 16%; p < 0.0001) [6]. Finally, in a recent analysis of 539 patients with type II carcinomas from the Gynecologic Oncologic Group, the authors reported that 13% had a personal history of breast cancer [3]. Single-institution studies provide some additional insight on this issue. In two such studies, a personal history of breast cancer was reported by 11–12.4% of patients with ESC [7,8], and this figure rises to 35% in datasets enriched for Jewish patients [9,10]. In these and other studies, a family history of breast cancer was reported in 16–29% of ESC patients [7–9]. However, the significance of the latter may not necessarily be specific to type II cancers, as it has previously been demonstrated in endometrial carcinomas (of all histotypes) that a family history of breast cancer in multiple first-degree relatives increases the risk of endometrial cancer [11].

The related corollary is whether there is an increased risk of breast cancer in patients with a personal history of type II endometrial cancers. Two large population-based studies have demonstrated an increase in the frequency of breast cancers as a secondary malignancy after uterine corpus cancers, but the latter were not specifically noted to be type II cancers, and the increased risk was generally bidirectional, suggestive of shared risk factors between these cancers [12,13]. Indeed, in one analysis, although the relative risk of breast cancers after endometrial cancer was greater than the expected rate in the general population, this increased risk was eliminated when shared risk factors (e.g., obesity, which were primarily related to type I cancers) were controlled for [14]. Very few studies have specifically examined the question as it relates to type II cancers. In one such study, Geisler et al. evaluated the records of 592 endometrial carcinoma patients and found that although only 25 (4.2%) patients were diagnosed with breast cancer either concurrently or subsequent to their diagnosis of endometrial cancer, 25% of the endometrial cancers in those patients were of the serous histotype, with only 3.2% being endometrioid [15].

“The observed association between type II endometrial cancers and breast cancers is likely to have a multifactorial basis.”

Ultimately, deciphering whether breast cancer predisposes to the endometrial cancer, or vice versa, may be of secondary significance to the larger point of why this apparent association exists: shared environmental risk factors, genetic risk factors, iatrogenic factors, or some combination of these. The few risk factors shared by type II endometrial carcinomas and breast carcinomas are broad and nonspecific, and include gender, age, and ironically, a family history of breast cancer. It is well known that a family history of endometrial cancer is an independent risk factor for endometrial cancer [11]. A genetic and heritable component to endometrial carcinoma, possibly as a component of a syndrome that is separate from the Lynch syndrome, has been suggested in two small pedigree studies, one of which found that the cancers that arise in this setting tend to be poorly differentiated and anaplastic-appearing, which suggests that they are type II cancers [16,17]. In addition, a recent study identified an excess of ovarian, endometrial and pancreatic cancers in relatives of patients with serous or mixed endometrial cancer [18]. Furthermore, germline mutations in the TP53 and CHEK2 genes, which typify the Li–Fraumeni syndrome and which increase the risk of breast cancer, have been identified in 1.3% of ESC patients [19]. Despite these findings, most investigations have centered on determining whether type II cancers are a component of the hereditary breast cancer/ovarian cancer syndrome. Combined data from two studies, in which all endometrial cancers were analyzed in datasets enriched for, or composed exclusively of Jewish patients, found the BRCA1/2 founder mutations (185delAG and 5382insC in BRCA1, and 6174delT in BRCA2) to be present in only eight (1.6%) of 488 cases, which was lower than the mutational frequency in the general population for this group, and all mutations were in type I cancers [20,21]. However, two studies that specifically focused on the serous histotype in Jewish patients found rates of BRCA1/2 mutations of up to 27% [9,10]. Studies of ESC in the nonfounder population have found low mutational rates: 0% in a study of 56 cases [8], and 2% in a study of 151 cases [19]. The latter mutational rate (2%) is higher than would be expected in a nonfounder population [19]. The aforementioned findings are somewhat conflicting, but do generally suggest that a small subset of ESC patients have germline BRCA1/2 mutations and, when combined with a history of breast cancer, may indeed represent a poorly characterized manifestation of a BRCA1/2-related syndrome that includes both cancers; and very large datasets will be required to demonstrate the precise magnitude of the contribution of BRCA1/2 mutations to the association between them. Nevertheless, this contribution is likely to be minor. One probable contributor to this association is the use of tamoxifen in the management of patients with breast cancer. Pooled data from the three largest case–control studies of endometrial cancer after breast cancer have clearly confirmed the findings of previous studies that showed an excess of type II cancers after prolonged tamoxifen exposure [22]. Furthermore, a recent Gynecologic Oncologic Group study of 3499 endometrial carcinoma patients confirmed that patients with type II cancers more frequently had histories of breast cancer with tamoxifen treatment compared with patients with type I cancers [3]. However, most ESC patients have no history of tamoxifen exposure [3], and the association between ESC and breast cancer has been found to be independent of tamoxifen exposure in at least two studies [4,6].

In summary, 9.7–32% of patients with type II endometrial carcinomas, especially ESC, have a personal history of breast cancer and up to 29% have a family history of breast cancer [3,4,6–9]. These figures significantly exceed the frequency of these histories in patients with type I endometrial cancers, and certainly the general population. The endometrial carcinomas that develop in patients subsequent to breast cancer are more likely to be type II than type I cancers. In addition, preliminary evidence suggests that when breast cancers develop subsequent to endometrial carcinoma, the latter are more likely to be type II than type I cancers. The observed association between type II endometrial cancers and breast cancers is likely to have a multifactorial basis. Minor contributors may include BRCA1/2 mutations and tamoxifen use; however, the major and primary contributors remain unclear.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.