Socioeconomic, Clinical, and Molecular Features of Breast Cancer Influence Overall Survival of Latin American Women

Molecular profile of breast cancer in Latin-American women was studied in five countries: Argentina, Brazil, Chile, Mexico, and Uruguay. Data about socioeconomic characteristics, risk factors, prognostic factors, and molecular subtypes were described, and the 60-month overall cumulative survival probabilities (OS) were estimated. From 2011 to 2013, 1,300 eligible Latin-American women 18 years or older, with a diagnosis of breast cancer in clinical stage II or III, and performance status ≦̸1 were invited to participate in a prospective cohort study. Face-to-face interviews were conducted, and clinical and outcome data, including death, were extracted from medical records. Unadjusted associations were evaluated by Chi-squared and Fisher’s exact tests and the OS by Kaplan–Meier method. Log-rank test was used to determine differences between cumulative probability curves. Multivariable adjustment was carried out by entering potential confounders in the Cox regression model. The OS at 60 months was 83.9%. Multivariable-adjusted death hazard differences were found for women living in Argentina (2.27), Chile (1.95), and Uruguay (2.42) compared with Mexican women, for older (≥60 years) (1.84) compared with younger (≤40 years) women, for basal-like subtype (5.8), luminal B (2.43), and HER2-enriched (2.52) compared with luminal A subtype, and for tumor clinical stages IIB (1.91), IIIA (3.54), and IIIB (3.94) compared with stage IIA women. OS was associated with country of residence, PAM50 intrinsic subtype, age, and tumor stage at diagnosis. While the latter is known to be influenced by access to care, including cancer screening, timely diagnosis and treatment, including access to more effective treatment protocols, it may also influence epigenetic changes that, potentially, impact molecular subtypes. Data derived from heretofore understudied populations with unique geographic ancestry and sociocultural experiences are critical to furthering our understanding of this complexity.