Abstract

Given prior evidence that an affected woman conveys a higher risk of ovarian cancer to her sister than to her mother, we hypothesized that there exists an X-linked variant evidenced by transmission to a woman from her paternal grandmother via her father. We ascertained 3,499 grandmother/granddaughter pairs from the Familial Ovarian Cancer Registry at the Roswell Park Cancer Institute observing 892 informative pairs with 157 affected granddaughters. We performed germline X-chromosome exome sequencing on 186 women with ovarian cancer from the registry. The rate of cancers was 28.4% in paternal grandmother/granddaughter pairs and 13.9% in maternal pairs consistent with an X-linked dominant model (Chi-square test X2 = 0.02, p = 0.89) and inconsistent with an autosomal dominant model (X2 = 20.4, p<0.001). Paternal grandmother cases had an earlier age-of-onset versus maternal cases (hazard ratio HR = 1.59, 95%CI: 1.12–2.25) independent of BRCA1/2 status. Reinforcing the X-linked hypothesis, we observed an association between prostate cancer in men and ovarian cancer in his mother and daughters (odds ratio, OR = 2.34, p = 0.034). Unaffected mothers with affected daughters produced significantly more daughters than sons (ratio = 1.96, p<0.005). We performed exome sequencing in reported BRCA negative cases from the registry. Considering age-of-onset, one missense variant (rs176026 in MAGEC3) reached chromosome-wide significance (Hazard ratio HR = 2.85, 95%CI: 1.75–4.65) advancing the age of onset by 6.7 years. In addition to the well-known contribution of BRCA, we demonstrate that a genetic locus on the X-chromosome contributes to ovarian cancer risk. An X-linked pattern of inheritance has implications for genetic risk stratification. Women with an affected paternal grandmother and sisters of affected women are at increased risk for ovarian cancer. Further work is required to validate this variant and to characterize carrier families.

Highlights

  • A history of ovarian cancer among first-order relatives remains the strongest and best-characterized predictor of ovarian cancer risk [1,2,3] and a main determinant of genetic testing referral [4, 5]

  • Our article uses the largest familial study of ovarian cancer to argue that there exists an ovarian cancer susceptibility gene on the X-chromosome acting independently of BRCA1 and BRCA2

  • We isolated a candidate gene, MAGEC3, that may be associated with earlier onset of ovarian cancer

Read more

Summary

Introduction

A history of ovarian cancer among first-order relatives remains the strongest and best-characterized predictor of ovarian cancer risk [1,2,3] and a main determinant of genetic testing referral [4, 5]. The evidence for a monogenic, autosomal dominant mode of inherited risk dates to the pre-BRCA era where studies focused on assessing heritability [6,7] using affected firstorder and second-order [8] female relatives. Genetic evidence of X-linkage has appeared in cytogenetic studies where loss of X-chromosome inactivation (XCI) can be visualized by loss of heterochromatin based Barr bodies [9]. Studies investigating a mechanistic connection between BRCA1 and XCI [15,16], especially in tissue after transformation [17], are mixed but tend to conclude that XCI dysregulation is BRCA independent [9,18,19]

Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call