Associations between body mass index and bladder cancer survival: Is the obesity paradox short-lived?

Introduction: We investigated the associations of pre-surgical body mass index (BMI) with bladder cancer outcomes in patients treated with radical cystectomy. Methods: We retrospectively analyzed data from 488 bladder cancer patients treated with radical cystectomy between 1994 and 2007 and followed up until 2016. Cox regression with step function (time-segment analysis) was conducted for overall survival because the proportional hazard assumption was violated. Results: Of 488 bladder cancer patients, 155 (31.8%) were normal weight, 186 (38.1%) were overweight, and 147 (30.1%) were obese. During the median followup of 59.5 months, 363 (74.4%) patients died, including 197 (40.4%) from bladder cancer. In adjusted Cox regression analyses, BMI was not significantly associated with bladder cancer-specific survival for overweight (hazard ratio [HR] 0.79, 95% confidence interval [CI] 0.57-1.10, p=0.16) or obese (HR 0.76, 95% CI 0.52-1.09, p=0.13) patients. In the Cox regression with step function for overall survival, the time interaction was significant overall (p=0.020) and specifically for overweight patients (p=0.006). In the time-segment model, the HR for overweight during the first 63 months was 0.66 (95% CI 0.49-0.90, p=0.008), whereas it was 1.41 (95% CI 0.89-2.23, p=0.14) after 63 months. Although not statistically significant, a similar pattern was observed for obese patients. Conclusions: Our findings suggest that overweight and obese bladder cancer patients had better outcomes within the first five years after radical cystectomy; however, there were no differences in longer-term survival. These data suggest that the obesity paradox in bladder cancer patients treated with radical cystectomy may be short-lived.