The risk of cancer among adults with metabolically healthy obesity (MHO) has not yet been established. We systematically searched from inception to 15 March 2020. We included prospective cohort studies that compared participants with MHO and participants with metabolically healthy non-obesity (MHNO) for incidence of any type of cancer. Benign tumors, cancer mortality or cancer prognosis were not in the scope of our analysis. The Newcastle-Ottawa Scale was used for quality assessment. Ultimately, eight studies with a total of 12 542 390 participants were included. The pooled meta-analysis using random effect model showed participants with MHO demonstrated a significantly increased risk of developing cancer (odds ratio [OR], 1.14; 95% confidence interval [CI], 1.05 to 1.23; and I = 39%) than those with MHNO. The subgroup analysis revealed a higher pooled estimate (OR, 1.17; 95% CI, 1.01-1.35; and I = 56%) in comparison with metabolically healthy normal weight. No evidence of effect modification by age, sex, ethnicity, smoking, sample size or length of follow-up was found. In conclusion, the present study reports a positive association between MHO and cancer incidence. All individuals with obesity, even in the absence of metabolic dysfunction, should be encouraged to lose weight.