**1. Introduction**

There is increasing evidence that vitamin C (ascorbate) is selectively toxic to some types of tumor cells, functioning as a pro-oxidant [1–3]. Vitamin C at concentrations of 10 nM–1 mM induced apoptosis in neuroblastoma and melanoma cells [4] and was shown to be an important modulator for the growth of mouse myeloma cells in an *in vitro* colony assay [5]. Studies have established that the growth of leukemic progenitor cells from patients with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) can be significantly modulated by vitamin C [6,7]. Intravenous (i.v.) administration of sodium 5,6-benzylidene-l-ascorbate (SBA) to inoperable cancer patients

induced a significant reduction in tumor volume without any adverse side effects [8]. Furthermore, recent clinical studies have reported that manipulation of vitamin C levels *in vivo* can result in clinical benefit for patients with AML and solid tumors [9].

Recent pharmacokinetic studies reported that 10 g of ascorbate given by i.v. produced plasma concentrations of nearly 6 mM, which were 25-fold higher than concentrations resulting from the same oral dose [10–12]. Depending on the dose, as much as a 70-fold difference in plasma concentration was found between oral and i.v. administration [13]. Complementary and alternative medicine practitioners worldwide currently use ascorbate i.v. in patients, in part because there are no apparent harmful effects [14–16]. Physiological concentration of vitamin C is under 0.1 mM, while plasma vitamin C concentrations that cause toxicity to cancer cells *in vitro* (1 mM–10 mM, depending on cell lines) can be achieved clinically by intravenous administration, which means a high dose of vitamin C.

In this review, *in vitro* and *in vivo* studies are summarized, showing that ascorbate killed cancer cells (Table 1). In addition, the mechanisms and physiologic relevance under investigation are also described. The results suggest that doses of vitamin C induce oxidative stress in cancer cells. Our previous results indicated that treatment of malignant lymphocytic cell lines with vitamin C (0.25–1 mM) for 24 h led to a marked dose-dependent decrease of cell proliferation [17]. The responsive cell lines were human myeloid leukemia cell line HL-60, retinoic acid (RA)-sensitive acute promyelocytic leukemia (APL) cell line NB4 and RA-resistant APL cell line NB4-R1. Different types of leukemia cells, such as K562 (chronic myelogenous leukemia cell line) [17] and KG1 (myeloblast cell line) [18], were also responsive to vitamin C. A similar result was obtained with cells containing over 90% of blasts from patients with AML. In these cell lines, induction of apoptosis by vitamin C demonstrated a dose-dependent effect. In addition, vitamin C weakly induced apoptosis in ovarian cell lines, including SK-OV-3, OVCAR-3 and 2774 [17]. For many of the cancer cell lines, ascorbate concentrations caused a 50% decrease in cell survival. The half maximal concentration (IC50) values were less than 5 mM, and all tested normal cells were insensitive to 20 mM ascorbate [13].


**Table 1.** Effects of vitamin C treatment on cell survival [13,17,18].


**Table 1.** *Cont*.

The IC50 values are means ± standard deviations from triplicate experiments. HL-60, human myeloid leukemia; NB4, NB4-R1, NB4-R2, human acute promyelocytic leukemia (APL); KG1, human myeloblast; K562, human chronic myelogenous leukemia; U937, human histiocytic lymphoma; OVCAR, SK-OV3, ovarian cancer; JLP119, human lymphoma; MCF7, MB231, Hs587t, human breast cancer; KLN205, mouse lung cancer; RAG, mouse kidney cancer; CT26, mouse colon cancer; B16, mouse melanoma; LL/2, mouse lung cancer; Hs587Bst, human normal breast cells; CCD34SK, human normal fibroblast cells. \* IC50 value was determined using H<sup>3</sup> incorporation proliferation assay for 24 h.
