Background Obesity increases cancer risk, whereas surgery-induced weight loss is associated with reduced risk. Risk-based patient stratification may be needed to better understand and maximize benefits of weight loss interventions in individuals with obesity. To this end, comprehensive data from high-quality studies with extended follow-up are imperative. This study examines the link between bariatric surgery and long-term cancer outcomes, focusing on patient subgroups defined by previously suggested predictors of treatment benefit, such as sex and baseline insulin levels. Methods and findings This post-hoc analysis used data from the Swedish Obese Subjects (SOS) study, a prospective, controlled intervention trial, designed to investigate the long-term effects of bariatric surgery-induced weight loss (ClinicalTrials.gov, NCT01479452). The study was conducted at 25 public surgical departments and 480 primary healthcare centers across Sweden. Between Sept 1, 1987, and Jan 31, 2001, 2,007 per-protocol patients with obesity who underwent bariatric surgery (gastric bypass, n = 266; gastric banding, n = 376; vertical banded gastroplasty, n = 1,365) and 2,040 matched controls, receiving standard nonsurgical obesity-related care, were recruited. Inclusion criteria were age 37–60 years and a body mass index (BMI) ≥34 kg/m 2 for men and ≥38 kg/m 2 for women. The primary outcome measures were cancer events and cancer-related deaths, captured through nearly complete data sourced from national Swedish health registries. Female-specific cancers were defined as gynecologic and breast cancers. Analyses were adjusted (adj) for baseline age, sagittal diameter, alcohol consumption, smoking, and serum insulin levels. The study was closed on December 31, 2022. Median follow-up was 26.8 years (interquartile range (IQR) 22.9, 29.6) in the surgery group and 24.9 years (IQR 18.7, 28.8) in the control group. Bariatric surgery was associated with a lower overall cancer incidence rate in women (adjusted hazard ratio (HRadj) = 0.78; 95% confidence interval (CI) 0.67, 0.90; p = 0.001), but not in men (sex–treatment interaction p = 0.013). The HRadj for overall cancer mortality rate in women was 0.78 (95% CI 0.61, 1.00; p = 0.050). In women, surgery was associated with a lower incidence rate of both obesity-related cancers (HRadj = 0.70; 95% CI 0.58, 0.85; p < 0.001) and female-specific cancers (HRadj = 0.60; 95% CI 0.47, 0.75; p < 0.001). Importantly, subgroup analyses showed that the associations between surgery and female-specific cancer incidence, as well as female-specific cancer-related mortality, were stronger in women with high baseline insulin levels (insulin-treatment interaction p = 0.021 and 0.039, respectively). The main limitation is that cancer was not a predefined study outcome. Conclusions Bariatric surgery is associated with a lower risk of cancer and cancer-related mortality in women with obesity, with the strongest association observed for female-specific cancers in women with elevated baseline insulin levels. In men, bariatric surgery was not associated with overall cancer incidence or mortality. These findings support incorporating risk-based stratification to better tailor cancer prevention strategies in obesity care.
Sjöholm et al. (Mon,) studied this question.