Purpose: To determine whether concurrent pyridoxine therapy can prevent the development of hand-foot syndrome (HFS) in patients being treated with capecitabine. Methods: Chemotherapy-naive patients with GI tract cancers scheduled for capecitabine-containing chemotherapy were randomly assigned to concurrent oral pyridoxine (200 mg/d) or placebo. Patients were stratified by chemotherapy regimen and monitored until development of National Cancer Institute Common Toxicity Criteria grade 2 or worse HFS or capecitabine- containing chemotherapy ended. Patients in the placebo group who developed grade 2 or worse HFS were randomly assigned again to receive pyridoxine or placebo in the next chemotherapy cycle to determine whether pyridoxine could improve HFS. Results: The median number of chemotherapy cycles to grade 2 or worse HFS was three in both groups. Grade 2 or worse HFS developed in 55 (30.6%) of 180 placebo-treated patients and in 57 (31.7%) of 180 pyridoxine patients. The cumulative dose of capecitabine to grade 2 or worse HFS was not different between the two groups (median not reached in either group; hazard ratio [HR] = 0.95; P = .788). Randomization of the 44 patients in the placebo group with grade 2 or worse HFS to placebo or pyridoxine for the next cycle resulted in no significant difference in the proportion showing improvement of HFS (42.9% v 47.8%; HR = 1.12; P = .94). By multivariate analysis, age ≥ 56 years (HR = 1.768; 95% CI, 1.190 to 2.628; P = .005) was an independent risk factor for grade 2 or worse HFS, and combined use of docetaxel (HR = 2.046; 95% CI, 0.880 to 4.755; P = .096) was of borderline significance. Conclusion: Pyridoxine is not effective in prevention of capecitabine-associated HFS.