Effect of high-dose intravenous vitamin C on inflammation in cancer patients

Inflammation plays a key role in tumour development, affecting tumour proliferation, angiogenesis, metastasis, and resistance to therapy[1‐6]. Key features of cancer-related inflammation (CRI) include leukocyte infiltration, cytokine build-up, tissue remodelling, and angiogenesis. Infiltrating leukocytes such as tumour associated macrophages (TAMs), neutrophils, dendritic cells, and lymphocytes establish an inflammatory microenvironment[7] and are key components in tumours of epithelial origins[8]. These leukocytes secrete pro-inflammatory cytokines such as IL1, IL6, TNFα, TGFβ, FGF, EGF and HGF21, as well as chemokines such as CCL2 and CXCL8[9]. While immune cells may repress tumour growth in some cases[10‐12], there is increasing concern that inflammatory microenvironments caused by infiltrating leukocytes can facilitate cancer development[13‐16]. In clinical studies, TAMs are associated with poor prognosis, while the use of anti-inflammatory agents is associated with reduced instances of certain cancers[14, 17].

Several studies indicate that inflammation is a marker of high cancer risk and poor treatment outcome[18‐22]. In response to systemic inflammation, and in particular in response to elevated IL-6 levels, the liver produces CRP[23], a protein that binds to dead or dying cells to activate the complement system. CRP can be used as a marker of systemic inflammation. It correlates with disease progression and can be used to monitor infection[18, 24‐27]. For example, subjects with highly elevated CRP concentrations (above 80 mg/L, as compared to values below 5 mg/L) showed 3.5 times the risk for “all-cause” mortality compared to other subjects[28]. There are particularly strong negative correlations between CRP levels and cancer survival[14, 24, 28‐38] in a wide variety of cancer types. For example, cancer patients with highly elevated CRP showed increased mortality by a factor of 28. Thus, CRP concentration data confirm a correlation between cancer progression and inflammation.

Serum CRP concentrations in human subjects are, according to one report, inversely correlated with antioxidant nutrient concentrations[39]. Vitamin C (ascorbate, ascorbic acid, AA) is a water-soluble antioxidant and essential nutrient for immune cells and extracellular matrix production[40, 41]. These properties, as well as the correlations between ascorbate depletion in cancer patients and prognosis[42], suggest that vitamin C may have a beneficial effect on inflammation in cancer patients. Experiments by Hartel et al.[43] indicate that 20 mM ascorbate inhibited production of IL-6 and TNF-α in monocytes without affecting IL-1 or IL-8 levels. For lymphocytes, the same ascorbate concentrations inhibited IL-2 production without affecting TNF-α of IFN-γ levels. Ascorbate, at milli molar concentrations, may also inhibit NF-κB activation in endothelial cells[44]. NF-κB is an important transcription factor that mediates changes in gene expression during inflammation. The effect may be concentration dependent, as a different study indicated that low ascorbate doses (0.2 mM) actually enhanced NF-κB in Jurkat T-cells[45]. Other studies show that ascorbate inhibits TNF-α activation of NF-κB in human cell lines in vitro in a concentration dependent fashion, and can also inhibit GM-CSF, IL-3, and IL-5 production[46].

While millimolar ascorbate concentrations are not usually considered ‘physiological’, they can be achieved if the vitamin is administered intravenously at high dose. Intravenous vitamin C therapy has been used in the treatment of cancer[47, 48]. Rationales for IVC therapy include preferential toxicity of ascorbate toward cancer cells[49, 50], potential benefits of ascorbate for immune cells[48], and ascorbate inhibitory effect on angiogenesis[51, 52]. In a study with guinea pigs, tumour growth was significantly reduced in cases where intra-tumour ascorbate concentrations reached the millimolar level[53].

In addition, inflammation and oxidative stress can cause down-regulation of immune system associated with T cell dysfunction, which has been described in cancer, infectious, and autoimmune diseases. In cancer and other inflammatory conditions, the T cell receptor (TCR) zeta chain is cleaved resulting in T-cell and Natural Killer (NK) cell inability to activate[54]. Several studies have shown that chronic inflammation is mandatory for the induction of an immunosuppressive environment, due to the high concentrations of reactive oxygen species formed by myeloid suppressor cells[55, 56]. Clinically, this occurrence is associated with poor prognosis[57, 58]. As the in vitro tissue cultures have shown a reversal of TCR cleavage by antioxidants[59, 60], the result of treatment by ascorbic acid may augment T-cell and NK cell immunity.

Over the course of 35 years, the Riordan Clinic has administered IVC to cancer patients. For some of these subjects, blood chemistry data before, during, and after IVC treatments provide measurements of plasma ascorbate, CRP levels, and the levels of cancer markers such as PSA and CEA. This manuscript details an analysis of these clinical data to determine if a correlation exists between IVC treatments, inflammation (measured by CRP, cytokine levels), and tumour progression (measured by PSA, CEA and other tumour markers).

Our database contained forty-five subjects who received IVC therapy (ranging from one to 100 treatments, with a median of nine treatments and an inter-quartile range, IQR, of five to eighteen treatments) and for whom CRP data were available before or after treatment (Table 1). Figure 1 shows CRP values before and after treatment.

Most of the subjects, 76 ± 13% (95% confidence) showed a reduction in CRP values during IVC therapy. If only subjects who had initially elevated CRP (> 10 mg/L) levels are considered, 86 ± 13% (95% confidence) showed a reduction in CRP values during IVC therapy. This is also indicated by the fact that twenty-eight subjects had elevated CRP prior to therapy, but only 14 had elevated CRP levels after treatment. Among the twenty-eight subjects who started with elevated CRP levels, the median CPR reduction was 80% with an IQR of 39% to 94%.

Example of the changed levels of CRP with indicated periods of IVC treatments by 15 g and 50 g of IVC is shown in Figure 2.

Unfortunately, our data covers patients who varied in the number of treatments they received. For patients with prostate cancer, the percentage decrease in CRP concentration after treatment was higher when the patient had more frequent treatments. Median CRP reductions of 95% were attained in subjects who were treated frequently (intervals between IVC infusions being fewer than 10 days) while those treated less frequently. The negative response to IVC treatment with dosages 15 g-50 g was found in patients with gastric cancer (CRP was increased from 1.0 mg/L to 60 mg/L in 160 days, 15 IVC treatments). Progression was measured for three patients with pancreatic cancer. One patient had metastasis in liver and bones, and for this patient CRP was increased from 3.8 mg/L to 54 mg/L in 42 days (13 IVC treatments by dosages 25 g-50 g). For another patient without metastasis, the increase in CRP was from 2 mg/L to 25 mg/L in 400 days. The third patient with pancreatic cancer had 24 treatments during 90 days and CRP was increased from 15 mg/L to 89 mg/L. The patient with cholangiocarcinoma stage IV had 33 IVC treatments in 68 days and CRP was increased from 19 mg/L to 144 mg/L. Negative outcome was found for a patient with lung cancer with metastasis. This patient had 37 IVCs during 170 days and CRP was increased from 19 mg/L to 59 mg/L.

Progression was found for 2 patients with breast cancer (one with infiltrating tubular carcinoma, poorly differentiated duct cell carcinoma with second diagnosis pancreatic cancer in malignant stage 3, non-resectable), and another with breast cancer, category 5. For these patients the increase was mild from 3 mg/L and 8 mg/L to 10 mg/L and 13 mg/L frequently showed median CRP reductions of 64%.

The effects of IVC therapy on PSA tumor marker levels is shown in Figure 3. We had twenty patients for whom PSA levels were assessed before and after IVC therapy. Fifteen of these subjects, 75% with a 95% confidence interval of ± 19%, showed reduced PSA levels during IVC treatment. The examples of the positive response to IVC treatments for patients with prostate cancer are presented in Figures 4 and5.

We also had data from nineteen other subjects (mostly breast cancer patients, with a few lung, pancreatic, bladder, and colon cancer patients) for cancer markers CEA, CA27.29, or CA15.3. There was statistically no benefit here, with only 53 ± 22% (95% confidence) of subjects showing reduction in these tumor markers post treatments.

Table 2 summarizes the percentage of patients who saw improvement in various parameters. The fraction of subjects seeing reduced inflammation, as measured by decreases in CRP, was significantly above fifty percent in the prostate cancer group, the breast cancer group, and the entire study group. IVC seemed to have a beneficial effect in reducing PSA levels, but we could not see a significant effect in breast cancer subjects or other cancer markers.

In patients for whom both tumor marker and CRP data are available, we found an interesting correlation: patients who had reduced CRP levels tended to also have reduced tumor marker values. This is shown in Figure 6: linear regression shows a positive correlation with r² = 0.62. Note that we did not observe a correlation between CRP concentrations and tumor marker values pre-treatments (r² = 0.02)

To study the effects of IVC on inflammation in more detail, we measured serum cytokine levels in eleven subjects before and after IVC therapy. These subjects were given six treatments (at doses of 15 g for the first treatment, 25 g for the second treatment, and 50 g for the next four treatments). Plasma ascorbate levels in these subjects were measured to verify that sufficient concentrations were attained for cytokine effects. At the fifty gram treatments, plasma ascorbate concentrations immediately post infusion reached average values of roughly 18 mM. The effects of these treatments on serum concentrations of key pro-inflammatory cytokines are shown in Figure 7. This figure shows two endpoints for each cytokine. The first is the change in cytokine concentration at the point just before the patients are given their sixth IVC dose and after this single 50 g IVC. The second is the change in cytokine concentration (relative to pre-therapy value) at the point just after the patients are given all six IVC dosages. The first endpoint perhaps indicates sustained reductions in cytokine levels due to one treatment while the second indicates an effect of ascorbate after several dosing. After six treatments, patients had noticeably lower levels of IL-2, TNF-α, and eotaxin. All cytokine levels were reduced after the last IVC injection. For six of these subjects, CRP levels were measured before and after the last injection of IVC. In all six cases, CRP concentrations decreased, typically by ten percent, after the infusion.

The vitamin C is used in complementary oncology, with intravenous administration being of particular interest since it allows for plasma ascorbate concentrations an order of magnitude beyond those attainable with oral administration[62, 63]. Clinical studies suggest that this approach is feasible and potentially beneficial[47, 48, 64, 65]. Several mechanisms of action for ascorbate efficacy against cancer have been proposed over the years, but to our knowledge its potential role in mediating inflammation has not been previously addressed. In the research described in this manuscript, we used CRP as a clinical inflammation marker. Twenty-eight out of forty-five subjects in our study had sharply elevated CRP levels (above 10 mg/L) prior to IVC therapy, suggesting that inflammation is a prevalent problem for cancer patients. This is especially important since other reports indicate that inflammation, particularly elevated CRP, is a marker of a poor prognosis[66]. In 76 ± 13% of our subjects, IVC therapy reduced CRP levels, with improvements being more prevalent (86 ± 13%) in the sub-population with pre-treatment CRP above 10 mg/L. Post-treatment, the number of subjects with these sharply elevated CRP levels was reduced from 28 to 14.

We found that decreases in CRP during treatment correlated with tumour marker decreases. This is consistent with other work[67] demonstrating that plasma CRP levels are well-correlated with serum PSA levels in prostate cancer patients. In the study[67] it was hypothesized that prostatic infection and inflammation may increase serum prostate-specific antigen levels. In addition, authors suggest that plasma CRP measurements may help to differentiate benign conditions from prostate cancer in patients with elevated serum PSA levels. Inflammation might be fundamental in prostate cancer and chronic inflammation may be a legitimate target for prostate cancer chemoprevention and treatment. Inflammatory processes also play a role in the development of histologic benign prostatic hyperplasia[68, 69]. Acute and chronic inflammatory infiltrates are routinely found in prostate tissue specimens obtained from men with BPH, and greater levels of inflammation have been observed in larger prostates[70‐73].

It is interesting in our study, therefore, that prostate cancer patients showed the most benefit (in terms of reduced PSA and CRP levels) from IVC therapy. A potential effect of IVC in reducing inflammation is also supported by our serum cytokine data, which demonstrate that levels of pro-inflammatory cytokines decrease during IVC therapy. IL-2, TNF-α, and eotaxin appeared to be chronically reduced in patients getting IVC therapy, while all six cytokines studied (IL-1α, IFN-γ, and IL-8, in addition to IL-2, TNF-α, and eotaxin) were acutely reduced after ascorbate infusions of 50 g. Average depression of IL-1 was 20% for six patients and the average decrease for eotaxin was 25%. Interleukin-1 is known to promote inflammatory processes and augment metastasis[74]. It is abundant at tumour sites, where it affects the process of carcinogenesis, tumour growth and invasiveness, and the patterns of tumour-host interactions[75, 76]. IL-1 induces uPA expression and NF-kB activation. TNF-alpha, another key inflammatory cytokines, plays a central role in the tumour progression. Constitutive expression of the TNF-alpha from tumour microenvironment is a characteristic of many malignant tumours and its presence is often associated with poor prognosis[77‐80]. Eotaxin-1 (CLL11) is a chemo-attractant and lymphocyte activator that has been shown to affect tumour cell growth[81‐83]. Chemokine receptor expression in many cancers correlates with poor prognosis, and there is evidence that eotaxin induces angiogenesis and metastasis[84].

In summary, our analysis of data from cancer patients given IVC, along with our tests of cytokine levels, suggest that IVC may reduce inflammation in cancer patients, and that this reduction in inflammation is correlated with reductions in the tumour markers PSA.

The strength of the study is that we used the protocol established in our clinic for the treatment of cancer patients by IVC therapy, lacking at many clinics at least as adjuvant therapy in treatment of malignant disease, and demonstrated the effect of this treatment on the level of inflammation in cancer patients. We analysed the array of patients with various forms of malignant disease, although there were more patients with prostate malignancies than other forms of malignant disease.

The limitations of the study are that the measurements of parameters of inflammation and tumour markers were not very detailed and many patients did not have measurements of the inflammation cytokines during treatment.

Further research in this area and clinical studies of the efficacy of intravenous high dose vitamin C are warranted.