Body mass index (BMI) and risk of lung cancer: a systematic review and meta-analysis of studies using directly measured and genetically proxied measures of BMI
A partir d'une revue systématique de la littérature (46 études), cette méta-analyse évalue l'association entre le risque de cancer du poumon et l'indice de masse corporelle, qu’il soit mesuré directement (cliniquement ou autodéclaré) ou estimé génétiquement
Purpose: Many studies using measured Body Mass Index (BMI) report an inverse association with lung cancer, but a few recent Mendelian randomization (MR) studies using genetically proxied BMI suggest a possible risk association. The aim of this study was to systematically evaluate and synthesize evidence on the association between lung cancer risk and both directly measured (i.e., clinically measured or self-reported) and genetically proxied BMI.
Methods: We systematically searched four bibliographic databases to identify prospective studies that examined the relationship between BMI and lung cancer. Random effects meta-analyses were used to pool relative risks (RRs) and odds ratios (ORs). Dose–response trends were assessed.
Results: A total of 46 studies met inclusion criteria (41 directly measured BMI, three genetically proxied BMI, two with both measures). In studies that used directly measured BMI (n = 43), compared with normal weight, lung cancer risk was higher in the underweight category (RR: 1.43; 95% CI 1.25–1.64) and lower in the overweight/obese category (RR: 0.79; 95% CI 0.74–0.83). Each 1 kg/m2 increase in measured BMI was associated with a 4% reduction in lung cancer risk [RR: 0.96; 95% CI 0.95–0.97). In five studies, genetically proxied BMI was associated with a non-significant increased risk of lung cancer (OR: 1.05; 95% CI 0.99–1.12).
Conclusion: A large body of consistent evidence from studies that directly measured BMI provides strong evidence of an inverse association between BMI and lung cancer risk. This evidence meets some of the key epidemiologic causal criteria such as temporality, strength of the association (including dose–response), and consistency of the association, but without a mechanism does not meet the criterion of biologic plausibility and thus causality is not established. A small body of MR evidence yielded non-significant findings that trended toward a risk association, indicating additional MR studies are warranted to clarify this issue.
Cancer Causes & Control , résumé, 2026