Abstract 2791: Genetic variations associated with breast cancer-related lymphedema in a prospective multiethnic cohort

Abstract

Abstract Background: Lymphedema is a common chronic condition secondary to surgical removal of lymph nodes in breast cancer (BC) patients. Currently, there is no cure for breast cancer-related lymphedema (BCRL), and early diagnosis is critical for symptom management. Because primary lymphedema is hereditary, genetic variations may predispose BC patients to secondary BCRL and may thus be predictive markers for risk stratification. Methods: In this analysis of 2,756 participants in the Pathways Study, a prospective cohort of BC survivors, we analyzed 134 single nucleotide polymorphisms (SNPs) selected to represent common variations in 8 genes implicated in primary lymphedema (VEGFC, VEGFD, FLT4, GJA12, FOXC2, CCBE1, GATA2 and SOX18). Up to 48 months post-BC diagnosis, 324 women reported development of BCRL. We derived hazard ratios (HR) and 95% confidence intervals (CI) of BCRL for each SNP, under additive genetic effects from Cox proportional hazards models adjusting for genetic ancestry and known BCRL risk factors, including age at BC diagnosis, BMI, physical activity, and clinical factors. Analyses were performed separately in each of four racial/ethnic groups (European American (EA) (n = 1,789), African American (AA) (n = 208), Asian (n = 344), Hispanic (n = 344)). Results: Among EA women, the most significant SNP associated with BCRL was an intronic variant, rs11659526 in CCBE1. Each copy of the variant C allele was associated with a 27% reduced multivariable-adjusted hazard of BCRL (HR = 0.73, 95% CI: 0.57-0.92, p = 0.009). Two other SNPs in CCBE1 were also associated with risk of BCRL (p<0.05). Other SNPs in CCBE1 were significantly associated with BCRL risk in the three minority groups: 7 in AA women, 2 in Asian women, and 9 in Hispanic women, but with essentially no SNP overlap across the four groups. SNPs in several other genes were also significantly associated with BCRL risk in minority groups, including GATA2 and VEGFC in AAs, FLT4 and FOXC2 in Asians, and VEGFC and VEGFD in Hispanics. Most associations remained unchanged after excluding BCRL within 6 months after BC diagnosis. As SNPs were selected for a priori hypotheses from associations with primary lymphedema, these associations were not adjusted for multiple testing. However, in conservative analyses to account for multiple testing, associations were no longer statistically significant. Conclusion: To our knowledge, this is the largest genetic association study of BCRL conducted to date. We found that SNPs in genes previously implicated in hereditary primary lymphedema, including CCBE1, were associated with BCRL, suggesting common biological mechanisms underlying primary and secondary lymphedema. The relatively small sample sizes for minority groups and moderate significance level of the associations points to the need for future replication in independent studies and the potential search for additional genes at a genome-wide level. Citation Format: Song Yao, Yali Zhang, Valerie S. Lee, Janise M. Roh, Isaac J. Ergas, Lwarence H. Kushi, Christine B. Ambrosone, Marilyn L. Kwan. Genetic variations associated with breast cancer-related lymphedema in a prospective multiethnic cohort. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2791. doi:10.1158/1538-7445.AM2015-2791