In this study, we analyzed Hereditary EC cases as part of two syndromes, LS and HBOC. Specifically, we report the results of mutation analysis of the main susceptibility genes to such syndromes, MLH1, MSH2, BRCA1 and BRCA2, performed on patients affected with EC and belonging to LS and HBOC families.
Approximately 5% of all ECs are caused by mutations in MMR genes. The lifetime risk of developing EC is approximately 2.9% in the general population compared with the 21‐54% lifetime risk for mutated women with LS, depending on the mutation type [18].
MSH2 is the most frequently mutated gene in women with EC associated with LS and it is reported in 50–66% of EC cases with a mutation. Mutations in MLH1 occur in 24–40%, and in MSH6 in 10–13%, of cases [19].
Our study reports a higher mutation rate in the MMR genes (42.1%), compared to other studies (18–22.7%) [16, 20]. Particularly, in our cohort of patients, we found no differences in the percentage of pathogenic variants in the analyzed genes, as we revealed a percentage of 50% in both MLH1 and MSH2. A previous study on EC patients found variants most frequently in the MSH2 gene (43%) followed by MSH6 (24%) and MLH1 (22%) [20]. Therefore, the percentage of mutations in MMR genes changes in the various populations. The mutated LS patients developed EC at mean age of 35.6 years, particularly patients carrying MLH1 pathogenic variants developed EC at a younger age (mean age = 31) than did MSH2 pathogenic variant carriers (mean age = 40.2). The mean age of EC onset in families with LS was 42 years. This finding confirms that in LS patients EC tends to occur at a younger age than in sporadic cases. In the previous report, patients carrying MLH1 variants developed EC at a younger age (mean age = 46.4) than did MSH2 variant carriers (mean age = 51.9) [20]. Thus, in our patients the observed EC onset was even earlier.
In EC patients with MMR mutation, a personal history of LS-related cancers was described in 50% of MLH1 mutation carriers and in 45% of MSH2 mutation carriers [20]. In our cohort, we have found a personal history of LS-related cancers in 50% of patients with MLH1 pathogenic variants and in 25% of patients with MSH2 pathogenic variants. Patients carrying MLH1 pathogenic variants were affected by EC and CC, the patient with a pathogenic variant in MSH2 was affected by EC, CC and gastric cancer (GC). Moreover, a personal history of LS-related cancers was also observed in 10.3% of LS family members (Supplementary Figures S1-S8).
Besides, in families of the mutated patients not only other EC cases occur, but also other LS-related cancers, as shown in Fig. 2a. In addition to EC, CC is the most frequent (31.3% of LS families) confirming that CC risk is high in LS patients and varies according to the involved gene.
GC occurs in 7.5% of our cases. This cancer type is reported in approximately 5–13% of LS individuals. Risks are higher in MLH1 and MSH2 than other mutation carriers, and higher in males than in females [21].
Bladder cancer is present in 6.3% of patients. Recent data suggest a two- to four-fold elevated risk of bladder cancer, with the highest risk occurring in men with MSH2 mutations [22].
The relationship between breast cancer (BC) and LS remains unresolved. Studies have not consistently demonstrated a higher incidence of BC among individuals with LS than expected [23]. Likewise, in our patients BC occurs in 2.5% of cases. As BC is fairly common in the general population, larger studies are needed to determine whether BC is indeed part of the LS cancer spectrum.
Approximately 2% of ovarian cancers (OC) are due to LS. In the families of our mutated patients OC was observed in 2.5% of cases. Reported lifetime risks for OC in women with LS fall primarily within the range of 3‐20% and appear highest for carriers of MSH2 mutations, followed by MSH6 and MLH1 [18].
Individuals with LS have up to a 3% lifetime risk of developing cancers of the brain [24], which was present in 2.5% of our families.
Prostate cancer has been also associated with LS; several studies have found the lifetime risk for prostate cancer in LS to be increased by two- to five-fold [25]. In our families, prostate cancer was reported by 1.25% of patients. Additional studies are needed to determine whether LS-associated prostate cancers occur at an earlier average age or are more aggressive.
Up to 4% of people with LS develop liver cancer by age of 70 years another rare cancer in the general population [21]. It was diagnosed in 1.25% of our patients.
The spectrum of LS-associated tumors is wide, and several very rare cancers in the general population are seen more frequently in this syndrome. Although the risks for these rare tumors are greatly increased above the general population risks, the absolute risks are low. Additionally, LS-related cancers in family members were more common in MLH1 (70%) than MSH2 variant carriers (65%) [20].
In EC patients belonging to HBOC families we found a high mutation rate of 42.8%, with 66.7% in BRCA1 and 33.3% in BRCA2.
Pennington et al. in 2013 sequenced 30 candidate tumor suppressor genes in 151 patients with uterine carcinoma and found the prevalence of germline mutations in BRCA1 to be 2% (26). Afterwards, four other studies involving only Jewish patients found an increased mutation rate in BRCA1 between 14 and 27%, which is significantly higher than the 2% [27].
In our study, the percentage of pathogenic variants in these two genes was very high (42.8%); thus, this is the first study with such a high percentage among the few studies correlating EC with pathogenic variants in BRCA genes. In addition, the age of EC onset was relatively low, 51.5 years in mutated patients and 54.5 years in HBOC family members.
A personal history of BC was described in 50% and 100% of patients with BRCA1/2 mutations [27]; particularly, a personal history of BC was found in 16.4% of EC patients [26]. In our study, 6/21 of HBOC patients (28.6%) had a personal history of BC and EC; 3 of them were mutated, suggesting that EC is not only one of the HBOC-related tumors, but patients with BC risk can develop EC other than OC.
Several studies suggest that BRCA mutation carriers display an increased risk of papillary serous carcinoma of the endometrium [28]. It has been highlighted that HBOC-associated EC tend to be of serous papillary type, an aggressive histologic subtype [29] that accounts for less than 10% of EC (20) whereas HNPCC-associated EC typically are of endometrial type [30]. This finding was confirmed in other studies focused on the incidence of BRCA founder mutations in patients with uterine serous carcinoma. A first report on 22 cases of uterine serous cancer found that 6 patients, accounting for 27%, had a germline mutation in BRCA1 or BRCA2 [31]. In other studies, a significantly higher mutation rate in BRCA1 (11.9%) than in BRCA2 mutations (1.7%) was found [5, 26, 32]. These results indicated that uterine serous cancer is more commonly associated with mutations in BRCA1 than BRCA2.
Unfortunately, we do not have information on the specific histologic subtype of EC of our patients. These data suggest that in HBOC patients the oncological prevention path should be performed not only for OC onset but also for EC.
In our cohort BC was observed in 38.2% of cases (Fig. 2b), which is a very high percentage; indeed, a family history of BC was found in 29.9% of EC patients [26]. As shown, BC is the most frequent cancer, confirming that the BC risk in BRCA1 and BRCA2 mutation carriers is 45–80% [33].
Moreover, BRCA1 and BRCA2 mutation carriers have a risk of OC onset of 45–60% and 11–35% respectively; accordingly, in the families of our patients, OC was reported in 13.2% of cases (Fig. 2b).
In a smaller percentage, we also found the presence of other tumors, among them prostate cancer occurred in 7.4% of cases and GC was found in 3% of cases. It has been highlighted that BRCA1/2 mutation carriers present an increased risk for prostate cancer (3.4-fold in BRCA1, 8.6-fold in BRCA2) [34]. Moreover, an increased frequency of other malignancies, such as gastro-intestinal tumors, has been reported in families with mutations in the BRCA2 gene [34].
Moreover, in 1.5% of cases, we found Hodgkin lymphoma, laryngeal, kidney and thyroid cancers that are known to be part of the HBOC (Fig. 2b).
Finally, 1/21 (4.8%) patient reported a variant of unknown significance (UV), c.599 C > T (p.T200I), in BRCA2 gene. This variant was observed in a family affected with breast and ovarian cancer. Algorithms developed to predict the effect of missense changes on protein structure and function (SIFT, PolyPhen-2, Align-GVGD) all suggest that this variant is likely to be disruptive, but these predictions have not been confirmed by published functional studies. Experimental studies on the effect of this variant on mRNA splicing are contradictory [35, 36]. The available evidence is currently insufficient to determine the role of this variant in disease. Therefore, we classified as a UV.
Furthermore, in 1/21 (4.8%) patient we have identified a novel missense variant in BRCA2, c.9541A > T (p.Met3181Leu), which has not been described yet. In-silico analysis indicates that this variant may have a benign effect; however, the evaluation of the pathogenetic significance needs to be corroborated by further experimental evidence.
Our data suggest that patients with hereditary EC have a high percentage of mutations in the main susceptibility genes to LS and HBOC. To our knowledge, there are no current published studies that have found hotspot mutations in these genes correlating with EC; therefore, it would be interesting to carry out further studies that evaluate a possible genotype–phenotype correlation.
Moreover, since EC occurs in mutation carriers at an early age, these at-risk individuals should undergo cancer prevention routes not only for the most frequent tumors but also for EC. The screening for EC among LS patients has been recommended by numerous experts; indeed, there is evidence that EC is often a sentinel cancer for women with LS. This implementation of cancer prevention should also be extended also to HBOC patients.