Gene polymorphisms in LEPR and PTGS2 contribute for prognostic evaluation in breast cancer

Breast cancer is the second most frequent neoplasm worldwide and is considered a heterogeneous disease, with distinct genetic backgrounds and diverse histological and molecular presentations that influence prognosis. Besides tumor-specific characteristics, individual genetic variations may also potentially impact disease outcomes, and need to be assessed in prognostic models. Here, we employed a candidate gene approach, focusing on biologically relevant pathways, such as inflammation, to investigate the effects of single nucleotide polymorphisms (SNPs) on breast cancer outcomes. The investigation involved a 10 year follow-up of a prospective cohort of Brazilian women (N = 1038) with unilateral, nonmetastatic breast cancer treated at the National Cancer Institute (CAAE 55929416.8.0000.5240). Blood samples were genotyped for 14 SNPs from five genes ( LEP , LEPR , PTGS2 , VEGFA and EGFR ) using Real-Time PCR. Linkage disequilibrium and haplotypes were evaluated using Haploview. Survival curves were estimated using the Kaplan-Meier method. The effects of SNPs on disease-free survival (DFS) and breast cancer specific survival (BCSS) were estimated using hazard ratios (HR) and 95% confidence intervals (95% CI). Adjusted hazard ratios (HRadj) were obtained through Cox regression multivariate models. Improved DFS was detected for LEPR rs1137101 when considering the whole population (HR = 0.45; 95%CI 0.26 - 0.77), post-menopausal women (HR = 0.57; 95%CI 0.37 - 0.88), or among patients with luminal tumors (HR = 0.44; 95%CI 0.25 - 0.80). In contrast, PTGS2 rs689466 was associated with worse outcomes of DFS (HR = 1.70; 95%CI 1.21 - 2.37) and BCSS (HR = 1.67; 95%CI 1.13 - 2.47) among post-menopausal women, as well as among obese patients (HR = 1.68; 95%CI 1.08 - 2.64 for DFS and HR = 1.82; 95%CI 1.08 - 3.04 for BCSS). Because LEPR rs1137100 and LEPR rs1137101 showed significant linkage disequilibrium (R 2 = 0,89), they were also evaluated together, in combination with PTGS2 rs689466, considering the three SNPs (A>G, A>G, A>G). The occurrence of at least two variant alleles in either LEPR rs1137100 or LEPR rs1137101, in combination with the reference genotype of PTGS2 rs689466 showed improved DFS (HR = 0.60; 95%CI 0.37 - 0.96) within the whole population of the study. In contrast, the occurrence of PTGS2 rs689466 variant alleles in combination with both LEPR rs1137100 and LEPR rs1137101 reference genotypes showed the worst outcomes even in multivariate models, significantly affecting both DFS (HRadj = 1.75; 95%CI 1.18 - 2.60) and BCSS (HRadj = 1.72; 95%CI 1.07 - 2.75) for the whole population, as well as for post-menopausal women, with DFS (HRadj = 2.09; 95%CI 1.16 - 3.77) and BCSS (HRadj = 2.60; 95%CI 1.35 - 5.0). The present results indicate the importance of tailoring prognostic models for breast cancer considering both tumor and individual biomarkers.