Abstract
Endometrial carcinoma (EC) is the most common gynecological malignancy in Sweden and accounts for about 6 % of all female malignancies. The risk of EC increases with age and the majority of cases are diagnosed between age 50 and 60. Ninety percent of cases occur in women older than age 50. About 2% of EC may have a familial association related to Lynch syndrome (LS). About 80% of women with Cowden syndrome have the PTEN mutation, which increases their lifetime risk of developing EC. The benefit of EC surveillance among LS patients remains undetermined. Available studies are controversial concerning optimal age at which to initiate screening and screening modalities.
Our first study explored the prevalence of LS, Cowden syndrome (CS) and hereditary breast ovarian cancer syndrome in consecutively diagnosed women with EC. In addition, we explored the possibility of a familial association between uterine cancer and other specific malignancies. In all, we included 481 consecutively diagnosed cases of endometrial cancer. We used the Swedish Cancer Registry to confirm all diagnoses, as well as for a reference population for the years 1970 and 2010. We conducted mutation analyses on all families who met criteria for the syndromes referred to above to identify potential causal genes. Our study demonstrated familial clustering among relatives of our index EC cases; EC prevalence was twice as high in our study population as in the cancer population in Sweden at large (6% vs 4% and 3%). In addition, we identified LS in 1.5% of all women. No BRCA1 and BRCA2 mutations were identified. No families fulfilled the CS criteria. Among all the LS families in which mutation could be verified, only one had previously been diagnosed. Moreover, we found that onset of cancer at a young age in family members of EC patients and diagnosis of multiple malignancies in the same patient lend support to the concept of hereditary uterine cancer syndromes.
In study II, 54 Cowden syndrome-like families were identified from consecutively diagnosed EC patients. PCR and DNA sequencing analysis were carried out on genomic DNA to amplify all nine PTEN gene exons. Since we identified no germline mutations or polymorphisms, the implication is that these must be rare among CS-like families. Therefore strict Cowden syndrome criteria should be applied to identify CS patients.
Our third study involved a retrospective nationwide study of 170 women with Lynch syndrome. We gathered data on all diagnostic methodology employed for gynecological screening of LS and prophylactic surgery, including age at surgery. In all, 86 of the 117 women who were eligible for screening complied with the screening program. Gynecological surgery was carried out on 43 women prior to diagnosis with LS, for which reason they were inappropriate for screening. A lower incidence of cancer was found in the screened group than in the non-screened group. EC was diagnosed by endometrial biopsy in a large number of cases. In addition, the incidence of cancer was significantly reduced by prophylactic surgery.
In conclusion, the results from our studies will improve both characterization of EC and family screening while expanding genetic counseling, and thereby help to prevent endometrial cancer in high-risk patients by enrolling them in EC prevention programs before endometrial cancer develops. Our results will improve routine procedures used to investigate families and surveillance programs for patients at high risk. Such patients can then be offered a choice between participation in screening programs or surgery for prophylactic purposes. According to our results, LS patients should be screened for gynecological cancer with transvaginal ultrasound (TVUS) and endometrial biopsy (EB) by age 30-35. Once high-risk women have completed childbearing, prophylactic surgery should be made available.
