The idea of using vitamin C to treat and prevent cancer was first proposed in 1949 and later supported by Cameron et al.who, in a controversial study, showed that administration of high-dose ascorbic acid improved the survival of patients withterminal cancer (148–150). Their results led to the proposal of using megadoses of vitamin C to combat degenerative diseases, including cancer and CVD.
One of the most important modifiable determinants of cancer risk is diet. Several research panels and committees have independently concluded that high fruit and vegetable intake decreases the risk of many types of cancer (151,152). Because vitamin C is present in large quantities in these foods, it is plausible that the reduction in cancer risk associated with the consumption of fruits and vegetables may be, at least in part attributable to dietary vitamin C. This is supported by 2 large prospective studies that showed that plasma vitamin C concentration is inversely related to cancer mortality in human subjects (153,154). However, contradictory results have also been reported (155,156). The inconsistency of the vitamin C-cancer correlation and lack of validated mechanistic basis for its therapeutic action has critically undermined the feasibility of using vitamin C in clinical treatment or prevention of cancer (157).
One of the most critical findings that has cast doubt over the effectiveness of vitamin C in treating cancer is the Moertel study (158), a randomized, placebo-controlled clinical study in which a high dose of vitamin C was given orally to advanced cancer patients with no effect detected. It contradicted the findings of early studies conducted by Cameron et al. (148–150) in which clear improvements in the health status of terminal cancer patients were shown after high-dose i.v. vitamin C treatment. The discrepancy between these studies may be explained by thedifferences in the plasma vitamin C concentrations achieved by different administration methods. The former administered vitamin C exclusively orally, whereas the latter used both oral and i.v. administrations. Maximum plasma vitamin C concentrations achievable by oral administration are limited by the kidney, which eliminates excess ascorbic acid through renal excretion. In contrast, because i.v. injection bypasses the renal absorptive system, it results in elevated plasma concentrations to high levels (6). This pharmacokinetic property of ascorbic acid wasdemonstrated recently in healthy subjects. I.v. administration resulted in substantially higher (70-fold) plasma vitamin C levels than those attainable by oral dose (6). In light of these results, it is likely that higher plasma concentrations were achieved in the Cameron study (148–150), which used both i.v. and oral administrations, but not in the Moertal study (158), in which only oral administration was used. The difference in effective vitamin C concentrations may have, in turn, contributed to the observed discrepancy in therapeutic outcomes reported. Indeed, a recent case study examining the clinical history of 3 cancer patients and the treatment they received supports the notion that high-dose vitamin C administration through i.v. injection has potential anti-tumor effects for certain types of cancer (157).
Newly available pharmacokinetic data, improved understanding of the regulation of vitamin C transport, and the growing evidence on the therapeutic efficacy of vitamin C have stimulated interest to reassess the feasibility of using vitamin C in the prevention and treatment of cancer. Though different in their methodologies, most recent studies on vitamin C and cancer have been conducted around 2 central themes: 1) the effects of high-dose ascorbic acid on the development and progression of tumors; and 2) the mechanisms of action that may contribute to the anti-cancer effect of this vitamin.
High-dose i.v. vitamin C administration
Because achieving high levels of ascorbic acid by i.v. injection are feasible in vivo (157), research has refocused on the implications and applicability of high-dose i.v. vitamin C administration in cancer therapy. Pharmacological concentrations of ascorbic acid (0.3–20 mmol/L) that are comparable to those attained by i.v. administration selectively target and kill tumor cells in vitro (159). In contrast, physiological concentrations of ascorbic acid (0.1 mmol/L) do not have any effect on either tumor or normal cells (159). This tumor-killing phenomenon isattributable to the pro-oxidant property of vitamin C, which, at high concentrations, mediates the production of hydrogen peroxide (159). This provides a potential mechanism of action for the anti-tumor effect of vitamin C and implicates it as a pro-drug in cancer treatment (6,156,159). The manifestation of this effect in a real clinical setting has also been examined (157). A case study examined the treatment effects of i.v. vitaminC administration on cancer progression in patients with well-documented case histories (157). In all 3 cases, high-dose i.v. vitamin C therapy effectively reduced the progression of malignant tumor and improved the health status of these patients (157). Unfortunately, the information on the plasma vitamin C concentrations of these patients is not available to establish a causal relationship between the route of administration, the resultant effective concentrations, and the observed therapeutic effect. Nonetheless, this association can be reasonably assumed based on findings of a previous pharmacokinetic study, which shows that i.v. injection leads to high concentrations of serum vitamin C (6). However, it is difficult to assess the precise contribution of vitamin C in the clinical outcome, because all subjects under examination were receiving other forms of therapeutic treatments concurrent with high-dose vitamin C therapy (160). Moreover, alternative explanations for this outcome cannot be readily ruled out. As pointed out by the authors (6) and others (160), the observed remission of cancer in these cases may be attributable to spontaneous remission or as the consequence of prior treatments rather than ascorbic acid administration (160). Therefore, the therapeutic value of high-dose vitamin C administration in cancer progression or remission is not unequivocally supported by this study.
When administered in high doses by i.v. injection, vitamin C also improves the health-related quality of life in terminal cancer patients (161). After 1 wk of high-dose therapy, the global health/quality of life on both the functional (such as emotional and cognitive) and the symptom scales (such as fatigue and pain) were significantly improved in 39 terminal cancer patients (161). Though not curative, vitamin C treatment in this case successfully fulfilled an equally important goal in treating cancer—the improvement in the quality of life, which is particularly critical in patients at the terminal stages of this disease. Although showing a direct relationship between vitamin C treatment and therapeutic benefits, the results of this study were not unequivocal. For example, like the case study, the plasma ascorbic acid concentrations that resulted from the treatment were not assessed. In many cancer patients, especially those at the terminal stages, the absorption and excretion of certain drugs, including vitamin C, may be altered due to physiological abnormalities, which in turn may influence bioavailability. Thus, the plasma vitamin C concentrations in these patients may not be comparable to those measured in healthy subjects in the early pharmacokinetic study (6). For this reason, it is imperative to obtain direct information on plasma vitamin C concentrations in future clinical studies, especially when cancer patients are employed as test subjects. Because of a lack of control groups, it is unclear whether the improved status in these patients is a direct result of vitamin C treatment. Nevertheless, the encouraging findings of these clinical (161) and case studies (157) have stimulated new interests for moresystematic research. Phase I trial studies are being conducted to collect preliminary data on the efficacy, safety, and pharmacokinetics of high-dose i.v. therapy and systematically examine its potential application in cancer treatment (160).
Mechanism of action
Parallel to clinical case/prospective studies examining the anti-cancer effects of high-dose vitamin C, experimental studies designed to investigate the mechanisms of action contributing to the therapeutic effect of vitamin C are concurrently being conducted, including its antioxidant or pro-oxidant function, its ability to modulate signal transduction and gene expression, and its potential role in tumor metastasis.
Antioxidant and pro-oxidant. At physiological concentrations, vitamin C is a potent free radical scavenger in the plasma, protecting cells against oxidative damage caused by ROS (162). The antioxidant property of ascorbic acid is attributed to its ability to reduce potentially damagingROS, forming, instead, resonance-stabilized and relatively stable ascorbate free radicals (163). This mechanism is manifest in a number of cytoprotective functions under physiological conditions, including prevention of DNA mutation induced by oxidation (164–167), protection of lipids against peroxidative damage (168,169), and repair of oxidized amino acid residues to maintain protein integrity (168,170,171). The effects of vitamin C on these 3 classes of biological molecules have been reviewed (162). As DNA mutation is likely a major contributor to the age-related development of cancer (172,173), attenuation of oxidation-induced mutations by vitamin C constitutes a potential anti-cancer mechanism. Plasma vitamin C at normal to high physiological concentrations (60–100 µmol/L) decreases oxidative stress-induced DNA damage by neutralizing potentially mutagenic ROS (164–167). Consumption of vitamin C-rich foods is inversely related to the level of oxidative DNA damage in vivo (172,174–176).
Paradoxically, ascorbic acid may also function as a pro-oxidant, promoting oxidative damage to DNA (177). This occurs in the presence of free transition metals, such as copper and iron, which are reduced by ascorbate and, in turn, react with hydrogen peroxide, leading to the formation of highly reactive and damaging hydroxyl radicals (177). However, the relevance of this under normal physiological conditions in vivo has been questioned, as most transition metals exist in inactive, protein-bound form in vivo (178). However, when used at pharmacological concentrations(0.3–20 mmol/L), ascorbic acid displays transition metal-independent pro-oxidant activity, which is more profound in cancer cells and causes cell death (159). This tumor cell-killing response is dependent upon ascorbate incubation time and extracellular ascorbate concentration (159). The findings of this study contradict a view that in vitro cancer killing by vitamin C is a mere artifact due to the presence of free transition metals in the culture medium (179,180). Transition metal chelation had no effect on preventing cell death, indicative of a metal-independent mechanismin effect (159). Extracellular ascorbate is the source of this anti-cancer effect, contrary to the conventionally held view that intracellular vitamin C is a major contributor. Although the mechanism of action for this cancer-killing effect has been identified, the reasons for the selectivity have not yet been confirmed. Nonetheless, the selective toxicity may be attributed to several intrinsic properties of cancer cells, including reducedconcentrations of antioxidant enzymes, such as catalase (181,182) and superoxide dismutase (183,184), increased intracellular transitional metal availability (185), and better accumulation of DHA through GLUT transporter overexpression (186,187), all contributing to the augmented intracellular hydrogen peroxide concentrations. Therefore, a nutritional regimen resulting in increased generation of hydrogen peroxide in vivo may be exploited as a means for inducing tumor-specific cytotoxicity (185).
The effective concentration of vitamin C required to mediate cancer killing can be easily achieved by i.v. injection (6,159) and maintained by repeated dosing in vivo.
Whether vitamin C functions as an antioxidant or pro-oxidant is determined by at least 3 factors: 1) the redox potential of the cellular environment; 2) the presence/absence of transition metals; and 3) the local concentrations of ascorbate (185). The last factor is particularly relevant in treatments that depend on the antioxidant/pro-oxidant property of vitamin C, because it can be readily manipulated and controlled in vivo to achieve desired effects.
Signal transduction, gene expression, and vitamin C. The intracellular redox changes caused by oxidants and antioxidants can modulate the expression of genes involved in signal transduction pathways leading to cell cycle progression, cell differentiation, and apoptosis (188). For example, cells treated with ascorbic acid at low pharmacologic concentration (1 mmol/L) increase expression of apoptotic genes that are induced by UV irradiation and DNA damage, indicating that vitamin C can modulate gene expression (189). Ascorbate enhances the expression of both MLH1, a MutL homolog required for DNA mismatch repair machinery, and p73, a p53 homolog, increasing the cellular susceptibility to apoptosis, especially in the presence of DNA-damaging agents (190). As the induction of MLH1 is a critical determinant in a cell's decision between pathways leading to either accumulation of mutation and subsequent tumorigenic progression or apoptosis (190), these data support an anticancer role for intracellular vitamin C. The therapeutic potential of vitamin C in cancer is further supported by its ability to activate the apoptotic program in DNA-damaged cells independent of the p53 tumor suppressor through an alternative pathway mediated by p73, which, in contrast, is functional in most tumor types (191). Ascorbate also stabilizes p53 and augments the apoptotic response of Hela cells to chemotherapeutic agents (192). At pharmacological concentrations (1 mmol/L), it decreases the Bcl-2:Bax ratio in the cytosol and mediates the mitochondrial release of cytochrome C, leading to the activation of the caspase cascade and apoptotic processes (193). This provides a mechanistic basis for combined therapy of vitamin C and chemotherapeutic drugs, as vitamin C potentiates the effectiveness of such drugs and, consequently, reduces the undesirable collateral damage to healthy cells (190). However, the concurrent use of antioxidants such as ascorbic acid as chemotherapeutic agents is still controversial (194).
Vitamin C, at millimolar intracellular concentrations, inhibits the activation of nuclear factor kappa B, a rapid response transcription factor, by preventing the TNF-mediated degradation of its inhibitor in different human cell lines as well as primary cells through independent mechanisms (195–197). As NFB induces transcription of genes involved in both inhibition of apoptosis and promotion of cell proliferation, its overexpression directly contributes to malignancy (198). Repression of constitutive activation of NFB by vitamin C can induce cell cycle arrest and apoptosis in these cells and attenuate tumor progression in different types of cancer. Moreover, in vitro overexpression of the epidermal growth factor receptor family member Her-2/neu constitutively induces NFB activation, which likely contributes to the transformed phenotype in mammary tumor cells (199). The recent advances in transgenic animal models facilitate the examination of these phenomena in vivo. For example, the availability of Her-2/neu mice overexpressing this receptor (200) and Gulo knockout mice unable to produce vitamin C (146) makes it possible to create a strain of bi-transgenic knockout mice for examining the in vivo effects of vitamin C on breast cancer.
Ascorbate and its lipophilic derivatives attenuate cell proliferation, arrest cell cycle, and induce apoptosis in human glioblastoma tumor and pancreatic cancer cells by reducing the expression of insulin-like growth factor-I receptor (201,202). Cell cycle arrest induced by vitamin C is also attributable to its ability to prevent the activation and nuclear accumulation of the mitosis-inducing phosphatase Cdc25C, hence providing a mechanism to restore cell cycle checkpoints in p53-deficeint cells (203). The inhibitory effect is more potent in the lipophilic derivatives of ascorbate (201), which may have better intracellular accumulation. Therefore, it is possible that synthetic vitamin C derivatives with increasedlipophilicity may have higher bioavailability in vivo and thus improved therapeutic efficacy.
Can vitamin C attenuate metastasis? The spread of cancer, or metastasis, is initiated by disrupting the physical confinement imposed by the extracellular matrix (ECM) through the primary malignant cell-induced degradation of collagen structure (204). Because vitamin C is essentialfor collagen maturation and stabilization, it has been suggested that ascorbic acid may reduce tumor spreading by potentiating the stability of the ECM, especially since neoplastic invasion exhibits similar pathological manifestations as vitamin C deficiency (185). Unfortunately, the effects of vitamin C deficiency on metastasis caused by reduced collagen stabilization have not yet been examined in vivo due to the lack of appropriate animal models. Interestingly, a vitamin C-independent pathway for collagen biosynthesis may exist in mice, because vitamin C restriction in Guloknockout mice results in no detectable alteration in levels of angiogenesis (205), a prerequisite for en masse tumor growth that requires sufficient collagen deposition. However, whether a similar phenomenon exists in humans is not known. In addition, conflicting results have been reported. For example, in the same mouse model, vitamin C depletion significantly attenuated tumor growth by impairing angiogenesis (206), an observation that has cast some doubt on the anti-tumorigenic property of vitamin C. However, as pointed out by the authors, this finding was based on an implanted tumor that displayed unusual dependence on angiogenesis (206). Whether this mechanism is applicable for other clinical tumors in humans is uncertain. Moreover, blood vessel formation of human endothelial cells, a process that mimics blood vessel formation, is attenuated by ascorbic acid at high physiological concentrations (200 µmol/L) but enhanced in a dose-dependent manner at normal physiologicalconcentrations (<100 µmol/L) (206), indicative of a dual-effect of vitamin C in blood vessel formation. However, the effects of supraphysiological (200 µmol/L) or pharmacological levels (>1 mmol/L) of vitamin C on angiogenesis in vivo, which are more relevant in clinical vitamin C therapy, were not investigated in this study.
Though not fully understood, there are 2 opposing views on the role of the collagen-stabilizing function of vitamin C on tumor growth. First, by stabilizing collagen, ascorbic acid fortifies the ECM and stromal structures, leading to better confinement of neoplastic cells to their primary sites and preventing tumor growth and metastasis (185). Second, the same function may also facilitate the formation of new blood vessels, providing the prerequisite for malignant tumor growth (206). The interplay of these effects in vivo, especially under pharmacological levels of vitamin C, is far from clear. However, with the availability of Gulo knockout mice and a better understanding of collagen biosynthesis, new research is being conducted to understand the mechanistic basis of these phenomena.
In addition to angiogenesis, cancer cells can also modify their energy metabolic pathways to adapt to the low oxygen microenvironment in the interior of a solid tumor (207,208). This is achieved by activation of hypoxia-responsive gene expression networks controlled by hypoxia-inducible factor-1 (HIF-1) (209,210). The activation of HIF-1 by cancer cells is instrumental in both tumor growth and metastasis (208,209,211,212). Ascorbate functions as a cofactor for hydroxylation of HIF-1 (213). Proline hydroxylation targets HIF-1 for ubiquitin-mediated degradation (214,215) and thus decreases HIF-1 levels in the cells. Furthermore, intracellular ascorbic acid can directly attenuate basal or hypoxia-induced expression of HIF-1 in human primary and cancer cells (216). The negative impact of ascorbate on HIF-1 expression raisesthe question of whether intracellular vitamin C can inhibit the hypoxia-induced adaptation of solid tumor and thus restrict tumor growth and metastasis.
The development and availability of new animal models, the increased availability of transcriptome data, and the use of new metabolic approaches will, in the next few years, help to develop a more exhaustive portrait of the manifold roles of vitamin C in human nutrition. These reductionist approaches will reduce reliance on population studies, which are often insufficiently definitive, in confirming or refuting causal roles for vitamin C in chronic degenerative disease, enabling the resolution of the longstanding debate on the value of high levels of vitamin C in human health in normal populations. Future research focused on the potential of high-level therapy in particular cases, including treatment of cancer and in stem cell development, will yield a better understanding of potential vitamin C therapeutic benefit.