Circulating vitamin C concentration and risk of cancers: a Mendelian randomization study

Two-sample bidirectional Mendelian randomization using circulating vitamin C genetic instruments (10 SNPs; European ancestry) to assess causal effects on five site-specific cancers. Primary analysis used inverse-variance weighted MR; sensitivity analyses employed six additional MR methods. Cancer outcomes sourced from UK Biobank (456,348 participants) with replication in ILCCO (lung cancer), PRACTICAL (prostate cancer), and BCAC (breast cancer). A parallel meta-analysis summarized prospective cohort studies and randomized controlled trials of dietary or supplemental vitamin C intakes, including up to 1,992,894 participants.

No evidence that higher circulating vitamin C causally affects risk of lung/bronchus, breast, prostate, colon, or rectal cancer. Breast cancer showed an initial UK Biobank association (OR 1.34 per 1-SD increase; 95% CI 1.14-1.57) but could not be replicated in BCAC (OR 1.05; 95% CI 0.94-1.17). Replication datasets (ILCCO, PRACTICAL) showed null results for lung and prostate cancers. Sensitivity analyses across six MR methods yielded similar null results. The prospective vitamin C intake meta-analysis found lower lung cancer risk with higher dietary vitamin C (RR 0.84; 95% CI 0.71-0.99) but not with supplements. Conclusion: Physiological-level circulating vitamin C does not have a large protective effect on these cancers in European populations; very small effects cannot be ruled out; vitamin C supplementation is unlikely to prevent these cancers; observational dietary associations may reflect confounding by other components of fruit- and vegetable-rich diets.

Power to detect very small effects; heterogeneity and potential pleiotropy; predominantly European ancestry; colorectal cancer subtypes not independently replicated; analysis focused on physiological circulating vitamin C exposure, not pharmacological high-dose vitamin C.