Background: As a tumor suppressor and a factor in glucose metabolism, p53 is an alluring point of focus in the study of obesity and cancer. Population-based studies show that single nucleotide polymorphisms (SNPs) at codon 72 of p53 are individually linked with development of breast cancer, high BMI, and type II diabetes. In this study cohort of African American (AA) and Caucasian (CA) women with breast cancer, we assessed the relationships between arginine (Arg) and proline (Pro) phenotypic variants and obesity and type II diabetes, accounting for confounding molecular and demographic influences. We also evaluated the prognostic relevance of these factors.

Methods: DNA samples isolated from breast tumor tissues collected from 190 (AA=90 and CA= 100) women were genotyped to detect SNPs at codon 72 of the p53 gene and were sequenced by using exon-specific primers (4 through 9). These exons were selected because more than 90% of mutations are located in these exons of p53. For all women, information on BMI and type II diabetes status was collected from electronic health records. Obese was defined as BMI ≥ 30. The closest measure of BMI within 5 years of the date of the diagnostic sample was used, and the time difference between BMI and sample date was included as a confounding variable in adjusted analyses. Only pre-tumor sample diagnoses of type II diabetes were considered. Clinical and demographic variables were evaluated for their association to p53 SNPs, using χ2 tests for categorical analysis and F-tests for continuous variables. Linear regression models were used to examine interactions with BMI. Cox proportional hazard regression models were used for all time-to-event analyses. Separate models were built for each BMI-defined category of weight (underweight, normal, overweight, or obese).

Results: Both SNP72 and BMI were strongly associated with race. For AAs, 68% were obese, and 49% were Pro/Pro phenotype. CAs were less obese (24%), and most were Arg/Arg (60%). Most cases of type II diabetes (78%) were AAs, but 23 total patients with diabetes was insufficient for BMI-stratified analysis. As determined in a linear regression model, there was an interaction between race, SNP72 and BMI. AAs with Arg/Arg had significantly higher BMI values than other SNP/race groups (P=0.003). In a Cox model of obese-only patients, adjusted for race, p53 somatic mutation, time of BMI measure, tumor stage, and molecular subtype, Arg/Arg women had 3.81 times higher hazard of death from cancer than Pro/Pro women (95% CI: 1.08,13.48). Type II diabetes apparently did not influence survival, but this observation may be due to insufficient sample size.

Conclusions: The association between SNP 72 and BMI varied according to race. African American women who exhibited Arg/Arg had an average BMI of 35. For obese patients, Arg/Arg had a negative influence upon survival, which was independent of established risk factors, including race.

Citation Format: Michael Behring, Bunyamin Ozaydin, Upender Manne. Codon 72 polymorphisms of p53 are associated with obesity, diabetes, and race in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4209.