Obesity and Cancer Treatment Outcomes: Interpreting the Complex Evidence.

A wealth of epidemiological evidence, combined with plausible biological mechanisms, present a convincing argument for a causal relationship between excess adiposity, commonly approximated as body mass index (BMI, kg/m), and incident cancer risk. Beyond this relationship, there are a number of challenges posed in the context of interpreting whether being overweight (BMI 25.0-29.9 kg/m) or obese (BMI ≥ 30.0 kg/m) adversely influences disease progression, cancer mortality and survival. Elevated BMI (≥ 25.0 kg/m) may influence treatment selection of, for example, the approach to surgery; the choice of chemotherapy dosing; the inclusion of patients into randomised clinical trials. Furthermore, the technical challenges posed by an elevated BMI may adversely affect surgical outcomes, for example, morbidity (increasing the risk of surgical site infections), reduced lymph node harvest (and subsequent risk of under-staging and under-treatment) and increased risk of margin positivity. Suboptimal chemotherapy dosing, associated with capping chemotherapy in obese patients as an attempt to avoid excess toxicity, might be a driver of poor prognostic outcomes. By contrast, the efficacy of immune checkpoint inhibition may be enhanced in patients who are obese, although in turn, this observation might be due to reverse causality. So, a central research question is whether being overweight or obese adversely affects outcomes either directly through effects of cancer biology or whether adverse outcomes are mediated through indirect pathways. A further dimension to this complex relationship is the obesity paradox, a phenomenon where being overweight or obese is associated with improved survival where the reverse is expected. In this overview, we describe a framework for evaluating methodological problems such as selection bias, confounding and reverse causality, which may contribute to spurious interpretations. Future studies will need to focus on prospective studies with well-considered methodology in order to improve the interpretation of causality.