Optimal body mass index cutoff point for predicting survival of colorectal cancer in Asian population: A National Health Information Database Analysis.

Abstract

e13098 Background: A U-shaped relationship between body mass index (BMI) and colorectal cancer mortality has been revealed. However, an optimal BMI range for predicting survival in Asian colorectal cancer patients is unknown. We established the most appropriate cut-off point of BMI for better survival in Asian colorectal cancer patients. Methods: Two cohorts of colorectal cancer patients were included in this study; 5,815 from the Seoul National University Hospital, Korea for development cohort and 54,043 colorectal cancer patients from the Korean National Health Information Database for validation cohort. To determine the optimal cut-off point of BMI at diagnosis, the systematic dichotomization based on log-rank test was performed by the method of Contal and O’Quigley. We evaluated the association between BMI at diagnosis and overall survival (OS) using the Cox proportional hazard model. Results: During a median follow-up of 5.7 years for hospital-based development cohort and 4.3 years for nation-wide validation cohort, 1,180 (20.3%) deaths and 10,244 (19.0%) deaths occurred, respectively. The optimal cut-off point of BMI at diagnosis was identified at 20.2 kg/m2 (log-rank statistic = 96.69, Q-statistic = 3.07, P< 0.01) for differentiating between the poorer and better OS in development cohort. When compared to the patients with BMI < 20.2 kg/m2, the patients with BMI ≥20.2 kg/m2 were significantly associated with the better OS in the Cox proportional hazard regression analysis (HR = 0.62, 95% CI = 0.54-0.72, P= 1.1×10-10). The association was validated in the nation-wide cohort showing the better OS in the patients with BMI ≥20.2 kg/m2. (HR = 0.64, 95% CI = 0.60-0.67, P< 0.01). Conclusions: We suggest the use of a BMI value of 20.2 kg/m2 to predict survival in Asian colorectal cancer patients.