In general, there was no evidence found for positive or negative interactions between the drugs used in the treatment of cancer and vitamin D in cancer patients. Several studies report gastrointestinal cramps and ulcerations after the administration of high dose 1,25(OH)₂D₃ [20,21,22]. The use of calcifediol and thiazide medications in the elderly may present a cause for concern as one case report was found reporting hypercalcemia in two individuals [19].

Hypercalcemia, an expected side effect of high dose vitamin D therapy alone, was also the most frequently reported side effect that occurred in conjunction with the various pharmaceutical drugs included in the review.

In vivo studies have identified that a calcitriol peak plasma concentration of 10 nmol/L has significant anti-tumor activity [23,24]. In several studies, as a strategy to increase the serum vitamin D levels to parallel those peak plasma concentrations and AUC that, in vivo studies, suggested induced anti-tumor activity, dexamethasone was used to reduce the incidence of hypercalcemia at these higher doses of vitamin D [25]. The maximum tolerable dose (MTD) of calcitriol was found to be 74 µg/week; [23] however, with the addition of dexamethasone, the MTD was increased to 125 µg/week [26].

Hypophosphatemia was seen in two studies where docetaxel was used for prostate cancer but not in all studies that used this drug combination. In the Petrioli et al. study, 32 µg of calcitriol was administrated orally, once per week, in three divided doses, and most prostate cancer patients experienced hypophosphatemia [27]. While in the Tiffany et al. study, 60 µg of calcitriol was administered orally, once per week and 16.7% of the prostate cancer patients experienced hypophosphatemia [28].

In a case report, Boulard et al. reported elevated calcium levels, mental confusion, asthenia, constipation with fecal impaction with the use of calcifediol (vitamin D₂) and thiazide medications in two elderly women over the age of 75 years [19]. All medications were halted and 45 mg/day of prednisone was administered; both cases resolved within one week. One of prednisone’s mechanisms of action is to reduce intestinal calcium absorption, and this seemed to help to resolve these women’s symptoms [29]. However, it is unclear as to the exact role of vitamin D in these cases, as some of these side effects reported are known risks of thiazide medications.

Several studies examined the pharmacokinetics of vitamin D during the course of treatment. calciferol was not found to impact the pharmacokinetics of gefitinib, or docetaxel [22,23,26]. Studies reporting on the pharmacokinetics are summarized in Appendix Tables A3.

Beer et al. evaluated the pharmacokinetics of 5 µg/kg of calcitriol by mouth (p.o.) and 36 mg/m² of docetaxel intravenous (i.v.) alone, and in combination in five patients. They found no difference between the pharmacokinetics of calcitriol alone or with docetaxel [22]. The pharmacokinetics of orally administered calcitriol, in escalating doses, in combination with paclitaxel is presented in Appendix Tables A4 [30]. The pharmacokinetics were determined as part of a maximum tolerable dose finding study; which was halted when evidence of a reduction in calcitriol oral bioavailability become evident at the higher dose.

Studies on the pharmacokinetics of iv administered 1,25(OH)₂D₃ were also identified. Appendix Tables A4 summarizes the pharmacokinetics of i.v. administered calcitriol in increasing doses from two different studies [23,26]. Fakih et al. found that gefitinib did not have any impact on calcitriol pharmacokinetics; this finding was also confirmed in a pharmacokinetic study conducted by Muindi et al. [23,26], who compared the serum calcitriol plus dexamethasone concentration versus time plots from this study with the calcitriol only results from the Fakih et al. study [23], and reported that dexamethasone had no impact on calcitriol PKs [26].

Several studies reported on the impact that pharmaceutical drugs had on vitamin D metabolism. Investigations into the impact of statin medication on vitamin D metabolism were conducted in two studies. Rosuvastatin was found to increase both serum 25(OH)D₃ and 1,25(OH)₂D₃ levels in both studies [31,32]. This may not be a drug class effect, since fluvastatin did not have the same impact on serum vitamin D parameters [32].

Odes and colleagues investigated the impact of cimetidine on vitamin D metabolism in nine participants during the spring months where there was increasing sun exposure [33]. They found that the anticipated increase in serum 25(OH)D3 levels from the increased sunlight did not occur in these individuals. There was no impact on 24,25(OH)D₃ nor 1,25(OH)₂D₃.

A small study investigated the impact of prednisone on vitamin D metabolism in four healthy subjects. Avioli et al. found that 30 mg/day of prednisone altered vitamin D metabolism, reducing the half-life of 25(OH)D₃ by 40–60%, and also reduced the vitamin D metabolite responsible for intestinal calcium absorption [29]. Briefly in this study, the pharmacokinetics of 1,2³H-vitamin D₃ were established in four normal healthy adult volunteers over a five day period. Ten µCi of radiochemically pure 1,2³H-vitamin D₃ was administered orally after a 16 h fast and blood samples were obtained at 5, 15, 30 and 45 min and at 1, 2, 4, 8, 12, 16 and 24 h for the first 24 h period and then every 12 h for an additional four days. After a two week wash-out period, each volunteer was given 30 mg of prednisone for 10 days. On Day 5, 1,2³H-vitamin D₃ was once again administered and blood samples were obtained according to the previously described schedule for the remaining five days of the study.

A small study involving four patients with gynecological malignancies examined the vitamin D metabolites before, during and after various chemotherapy regimens that included cisplatin. They found that, while there was variation in 24,25(OH)D₃ and 25(OH)D₃ levels during the study, 1,25(OH)₂D₃ levels were significantly reduced by the chemotherapy [34]. The authors suggested that the reduction in 1,25(OH)D₃ levels may be attributed to cisplatin’s nephrotoxic profile that results in the kidney’s reduced ability to convert 25(OH)D₃ to 1,25(OH)₂D₃.

A second study compared the change in serum vitamin D metabolites between colorectal cancer patients undergoing chemotherapy versus those who were not receiving chemotherapy through a retrospective chart review. The study found that patients receiving chemotherapy were more likely to have lowered 25(OH)D₃ levels than those not receiving chemotherapy [35]. A recently published paper found the same result, lower serum 25(OH)D₃ levels in breast cancer patients during and after chemotherapy treatment [36].

Our review found that high dose calcitriol used in conjunction with several different pharmaceutical drugs used in patients with cancer did not result in adverse events beyond what could be expected from the use of high dose vitamin D alone, i.e., hypercalcemia. There were, however, several areas identified that warrant further investigation.

Dexamethasone was used to reduce the incidence of hypercalcemia and achieve a higher maximally tolerated dose (MTD) in prostate cancer patients, while prednisone was used to manage the side effects of hypercalcemia [19,26]. Both medications act by reducing intestinal absorption of calcium as a means to reduce the hypercalcemic state [19,26]. Avoili et al. found that prednisone reduced the half-life of 25(OH)D₃ whereas dexamethasone had no impact on 1,25(OH)₂D₃ levels [26,29]. Work in animal models suggests that dexamethasone may impact 1,25(OH)₂D₃ levels through up regulation of CYP24A1 transcription resulting in increased catabolism of 1,25(OH)₂D₃, indicating that more research into dexamethasone’s potential impact on vitamin D metabolism is warranted given its wide spread use in cancer patients and similar mechanism of action to prednisone [37]. Further, there is evidence that many cancer tumor types: breast, lung, colon and cervical as examples, over expresses CYP24A1 mRNA which may result in increased catabolism of vitamin D [37]. Several studies pointed to a reduction in serum 25(OH) D₃ levels in patients undergoing chemotherapy when exposed to a broad base of different chemotherapy drugs [34,35,36]. Whether this is as a result of alterations in lifestyle from undergoing cancer treatments or from the chemotherapy drugs themselves is not entirely clear. Cisplatin does induce nephrotoxicity and reduced vitamin D levels have been reported in patients exposed to cisplatin. However, other chemotherapy drugs that do not induce nephrotoxicity have also induced these phenomena [34,35,36]. Vitamin D’s role in immune system modulation, potential chemopreventative role, and in bone metabolism suggests that monitoring of vitamin D serum levels during the course of treatment for cancer may be important in this vulnerable population.

While the search parameters of this review did not directly reveal an impact of vitamin D on P450 system nor drugs that have an impact on vitamin D levels through alternation of the enzymatic activity through the P450 system, there are several drugs that are known to impact vitamin D levels. The azole class of antifungal drugs, such as ketoconazole, flucinazole have been shown to inhibit the activity of CYP24A1 [37]. Genistein, a plant isoflavone found in Glycine max (soybeans) and in other plant products, has been shown, in vitro, to in inhibit transcription of both the CYP24A1 and CYP27B1 genes [37,38]. Work by Wang and colleagues have demonstrated that 1,25(OH)2D₃ can be catabolized by a CYP3A4 dependent pathway, which is inducible by rifampin [39]. This CYP3A4 pathway maybe responsible for the osteomalacia-inducing aspects of several pharmaceutical drugs [37,39].

This review not only demonstrates the minimal evidence amassed relating to direct correlations between vitamin D and pharmaceutical agents employed in people with cancer, but also the limited collection of evidence relating to vitamin D metabolism in situations where normal organ or physiological function may be compromised due to pharmaceutical agents.

A limitation of this review is that the studies were all conducted in small patient populations, limiting both the power and the generalizability of the results. Despite the rigorous design of some of the existing studies, more robust studies with larger sample sizes might help address individual variations which may impact vitamin D and chemotherapeutic regimes. This relative lack of research points to opportunities for further exploration on the impact of pharmaceutical drugs on vitamin D metabolism.

Using an iterative process, a sensitive search strategy was developed and executed using the OVID platform. We searched OVID MEDLINE^® (1948 to March Week 4, 2012), OVID MEDLINE^® In-Process & Other Non-Indexed Citations (April 10 2012), Embase (1980 to 2012 Week 14), and AMED (Allied and Complementary Medicine, all years to 11 April 2012). The search employed both controlled subject headings (e.g., vitamin D, Cholecalciferol, Cytochrome p-450 enzyme system) and text words (e.g., vitamin D3, Calcitrol, CYP). The drug formulary of Cancer Care Ontario was used to identify cancer-related drugs, the names of which were also incorporated into the search. When appropriate, floating subheadings for concepts such as adverse drug reaction, drug interaction, and drug toxicity were included in the search strategy. We also searched the Cochrane Library on Wiley (including CENTRAL, Cochrane Database of Systematic Reviews, DARE, HTA, and NHS EED). No language or study group limits were applied to any of the searches. However, where possible, results were restricted to the human population. Previous reviews were hand-searched to identify other potentially relevant publications. A search of the WHO International Clinical Trials [40] and the MetaRegister of Controlled Trials databases [41] were also conducted to ensure that all relevant publications had been identified.

The strategy was peer reviewed prior to execution by an experienced information specialist using the PRESS Checklist [42].

Data extraction was carried out by one reviewer and independently checked for accuracy by a second reviewer. Data collected included the study location, year of publication, type of cancer, study design, participant numbers, drugs and dosage, form of vitamin D used, endpoints, study protocol and relevant reported outcomes. Additional information for pharmacokinetics studies was extracted and included: number of observations, dose, route of administration, half-life, C_max, Area under the Curve (AUC_{0–24 and} AUC_0–72) and clearance.