Re: Prediagnosis Body Mass Index, Physical Activity, and Mortality in Endometrial Cancer Patients

Abstract

Body mass index (BMI) is an established risk factor for endometrial cancer, a hazard that is exponential with increasing BMI (1), and when risk estimates are standardized, endometrial cancer is ranked highest among obesity-associated cancers in women (2). Therefore, the findings from Arem and colleagues (3), using the large National Institute of Health–AARP Diet and Health Study, of the adverse impact of prediagnosis BMI on survival in women who developed endometrial cancer are potentially important. The authors used the date of cancer diagnosis as zero in their time-to-event analysis, such that the results might be interpreted as prognostic and, in principle, modifiable. However, before one extrapolates to the impact of intervention strategies, the distinction between the effects of pre- and postdiagnosis BMI (the latter not measured in the AARP study) warrants closer examination. In the Cancer Prevention Study II (CPS-II) (4) and Women’s Health Initiative study (5) combined, among those who developed colorectal cancer (CRC), increasing prediagnosis BMI was associated with decreased survival—similar to the Arem study in endometrial cancer—but these associations were null for postdiagnosis BMI. Similar analytical plans were used—namely, cancer diagnosis as time zero. So why the differences? Table 1 shows additional analyses from the CPS-II of mean weights in patients with CRC at pre- and postdiagnosis questionnaire assessments. The first observation is that postdiagnosis weight data are unavailable in 15% of all patients who developed CRC. These data are unlikely to be missing at random and reflect deaths in the immediate diagnosis and initial treatment phase. Second, overall, there is an approximate 2 kg loss in patients from pre- to postdiagnosis questionnaire. Importantly, the amount of weight loss correlates with prediagnosis BMI, such that heavier patients lost appreciably more body weight than did lighter patients. Weight loss is not limited to advanced-stage cases. These weight changes represent migration from a higher to lower BMI category, analogous to tumor stage migration seen, for example, with improved image detection of nodal involvement. The net effect is an attenuation of the influence of postdiagnosis BMI on survival as “bad prognosis” patients migrate from