Abstract Background Classical anthropometric traits may fail to fully represent the relationship of weight, adiposity, and height with cancer risk. We investigated the associations of body shape phenotypes with the risk of overall and site-specific cancers. Methods We derived four distinct body shape phenotypes from principal component (PC) analysis on height, weight, body mass index (BMI), waist (WC) and hip circumferences (HC), and waist-to-hip ratio (WHR). The study included 340,152 men and women from 9 European countries, aged mostly 35–65 years at recruitment (1990–2000) in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Cox proportional hazards regression was used to estimate multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs). Results After a median follow-up of 15.3 years, 47,110 incident cancer cases were recorded. PC1 (overall adiposity) was positively associated with the risk of overall cancer, with a HR per 1 standard deviation (SD) increment equal to 1.07 (95% confidence interval 1.05 to 1.08). Positive associations were observed with 10 cancer types, with HRs (per 1 SD) ranging from 1.36 (1.30–1.42) for endometrial cancer to 1.08 (1.03–1.13) for rectal cancer. PC2 (tall stature with low WHR) was positively associated with the risk of overall cancer (1.03; 1.02–1.04) and five cancer types which were not associated with PC1. PC3 (tall stature with high WHR) was positively associated with the risk of overall cancer (1.04; 1.03–1.05) and 12 cancer types. PC4 (high BMI and weight with low WC and HC) was not associated with overall risk of cancer (1.00; 0.99–1.01). Conclusions In this multi-national study, distinct body shape phenotypes were positively associated with the incidence of 17 different cancers and overall cancer.