Representation of obese participants in obesity-related cancer randomized trials.

BACKGROUND

Obesity is a risk factor for numerous cancer types, and may influence cancer treatment outcomes. Underrepresentation of obese patients in obesity-related cancer randomized controlled trials (RCTs) may affect generalizability of results. We aimed to assess the reporting of information about eligibility and enrollment of obese participants in obesity-related cancer RCTs.

METHODS

We conducted a systematic review of RCTs of 10 obesity-related cancer types (esophagus, colon/rectum, liver, gallbladder, pancreas, postmenopausal breast, endometrium, ovary, kidney, and thyroid cancer). We selected RCTs published between 2013 and 2016 in five major journals. For each trial, we examined the article, the protocol, and the registration record. We assessed if eligibility criteria limiting the enrollment of obese participants were reported, the proportion of obese participants that were enrolled, and if a subgroup analysis according to obesity status was reported. We systematically contacted corresponding authors and asked for information about eligibility of obese participants and the proportion of obese participants.

RESULTS

We included 76 RCTs. Colon/rectum (n = 20), postmenopausal breast (n = 11), and kidney (n = 11) cancers were the most frequent types. Based on publicly available sources, information on the eligibility of obese participants was available in 5 (7%) trials. The proportion of obese participants could be estimated in 9 (12%) trials only. We found a subgroup analysis in only one RCT. When considering unpublished information, the eligibility of obese participants was explicitly stated in 31 (41%) trials but it was unclear if the remaining 59% trials considered obese participants as eligible and what proportion of obese participants was included. Across 22 trials, the median proportion of obese participants included was 18% (Q1-Q3 11-23).

CONCLUSION

Information on the eligibility and enrollment of obese participants in cancer RCTs is dramatically underreported. More transparency is needed to understand the applicability of obesity-related cancer RCT results to obese patients with cancer.