Background: Vitamin C at high concentrations is toxic to cancer cells in vitro. Early clinical studies of vitamin C in patients with terminal cancer suggested clinical benefit, but 2 double-blind, placebo-controlled trials showed none. However, these studies used different routes of administration.

Objective: To determine whether plasma vitamin C concentrations vary substantially with the route of administration.

Design: Dose concentration studies and pharmacokinetic modeling.

Setting: Academic medical center.

Participants: 17 healthy hospitalized volunteers.

Measurements: Vitamin C plasma and urine concentrations were measured after administration of oral and intravenous doses at a dose range of 0.015 to 1.25 g, and plasma concentrations were calculated for a dose range of 1 to 100 g.

Results: Peak plasma vitamin C concentrations were higher after administration of intravenous doses than after administration of oral doses (P < 0.001), and the difference increased according to dose. Vitamin C at a dose of 1.25 g administered orally produced mean (±sd) peak plasma concentrations of 134.8 ± 20.6 µmol/L compared with 885 ± 201.2 µmol/L for intravenous administration. For the maximum tolerated oral dose of 3 g every 4 hours, pharmacokinetic modeling predicted peak plasma vitamin C concentrations of 220 µmol/L and 13 400 µmol/L for a 50-g intravenous dose. Peak predicted urine concentrations of vitamin C from intravenous administration were 140-fold higher than those from maximum oral doses.

Limitations: Patient data are not available to confirm pharmacokinetic modeling at high doses and in patients with cancer.

Conclusions: Oral vitamin C produces plasma concentrations that are tightly controlled. Only intravenous administration of vitamin C produces high plasma and urine concentrations that might have antitumor activity. Because efficacy of vitamin C results cannot be judged from clinical trials that use only oral dosing, the role of vitamin C in cancer results should be reevaluated.