Reply to Borst: Randomized clinical trials of high-dose intravenous vitamin C in cancer patients are warranted

Borst (1) may question our equipoise (2), but the fact remains that “the war on cancer,” launched in 1971 by President Richard M. Nixon, is far from won. Novel, promising approaches to cancer therapy, therefore, should be examined with an open mind.

To specifically address Borst's points (1):

1. No preclinical animal model of cancer is without limitations. To focus on xenografts versus tumors driven by oncogene manipulation misses the point of Chen's work (3), which linked in vivo measures of pharmacologic doses of ascorbate to both mechanism and outcome. Although the effect of vitamin C on tumor growth may be “modest” compared with standard chemotherapy, vitamin C—unlike chemotherapy—has no toxic side effects (4). Chen et al. (3) specifically emphasized that ascorbate may have potential as an adjunct to standard chemotherapy by enhancing tumoricidal effects while lessening toxicity toward normal cells. Combining the strengths of several treatment modalities is a proven tactic in fighting cancer.

2. “Superiority” of generating oxidative stress is not the only criterion to consider in evaluating treatment strategies. Plumbagin and menadione, even though they are vitamin K precursors, are drugs that have pharmacokinetic behavior, tissue distribution, and metabolism different from those of vitamin C. Unlike vitamin C, menadione has dose-limiting side effects (5), and therefore the rationale was absent for comparing the two compounds in Chen's model system (3).

3. Cameron's intravenous vitamin C dose ranged from 5 to >45 g/day (6–8), well within the range “now considered necessary for a substantial antitumor effect.” Regarding “biased patient selection,” Cameron and Pauling (8) noted that “the original group of matched controls was selected [randomly] by senior members of the Medical Records Staff in Vale of Leven Hospital.” An independent physician then analyzed the case records without knowledge of survival times of the ascorbate-treated patients. Cameron and Pauling believed that bias in patient selection was thus minimized, if not eliminated.

4. We cited the original case reports (9). The accompanying editorial (10) concluded that “continued rigorous preclinical and early phase clinical trials of high-dose intravenous vitamin C therapy will provide a scientifically sound basis from which to accept or reject this approach,” which Borst (1) fails to mention and which is precisely the point of our commentary (2). Intravenous vitamin C is remarkably safe in prescreened cancer patients regardless of clinical outcome (4). Phase I trials are designed and statistically powered to assess safety, not efficacy. For Borst (1) to call the lack of an anticancer response “the conclusion of this trial” (4) is inappropriate and misleading.

Borst's letter (1) misses the point of Levine's work and our commentary—that “Chen et al. (3) now have convincingly demonstrated biologic plausibility and are poised to explore the potential value of ‘pharmacologic ascorbate in cancer treatment’ in humans” (2). We want to avoid hype and disdain when unwarranted, but not to pursue ascorbate's potential anticancer value in randomized, double-blind, placebo-controlled clinical trials of high-dose intravenous vitamin C would be a tragedy given the human suffering caused by cancer and by the toxic side effects of current standard chemotherapy.