Abstract
Background: We have previously reported a negative correlation between the effect of chemotherapy and 25-hydroxy vitamin D3 (25-D3) levels in patients with colorectal cancer. Based on this finding, we hypothesized that the response to vitamin D3 supplementation may be attenuated in patients with colorectal cancer. Aim: To determine 25-D3 response to 2000 IU/day vitamin D3 supplementation in patients with colorectal cancer. Materials and Methods: Fifty evaluable colorectal cancer patients were treated with vitamin D3 at 2000 IU/day for 6 months. Serum 25-D3 levels were measured at baseline, 3, and 6 months of supplementation. Results: The mean 25-D3 level was 17.5 ng/ml at baseline, 31.6 ng/ml at 3 months, and 33.8 ng/ml at 6 months. The most important factor in determining 25-D3 response was chemotherapy status. A rise in 25-D3 of ≥10 ng/ml at the 3-month interval was observed in 92% of chemotherapy-free patients vs. 39% of chemotherapy patients. Similar differences in response were noted at the 6-month interval. Conclusion: Depressed 25-D3 levels are common in patients with colorectal cancer. Active chemotherapy is associated with an attenuated response to 2000 IU of D3 supplementation in this patient population. Alternative vitamin D3 dosing schedules need further investigation in colorectal cancer patients undergoing chemotherapy.
The serum level of 25-hydroxy vitamin D3 (25-D3) is accepted as the best indicator of vitamin D status in humans (1, 2). The utility of the 25-D3 level in assessing vitamin D status is based on its long serum half-life (ranging from 2-6 weeks), its unregulated synthesis, and its reflection of the overall supply of vitamin D3 precursors (3). The association between vitamin D deficiency, as assessed by 25-D3 level, and an increased risk of developing colorectal neoplasia has been documented in several large case control studies (4-13). However, the incidence and clinical significance of vitamin D deficiency in patients with established colorectal cancer, especially those with metastatic disease, has not been adequately investigated. Two reports suggest that vitamin D deficiency may impact negatively on the outcome of patients with established colorectal cancer. The first study evaluated the outcome of patients with colorectal cancer based on season of diagnosis and region in Norway (14). Colorectal cancer mortality was higher when the diagnosis of colon cancer was made during winter or spring (lower 25-D3) compared to summer and autumn (higher 25-D3) (14). The second study analyzed the outcome of colorectal cancer patients who had a baseline 25-D3 level at least two years prior to colorectal cancer diagnosis through a retrospective analysis of the Nurse's Health Study (NHS) and Health Professionals Follow-up Study (HPFS) (15). Three hundred and four colorectal cancer cases were identified. Stages I-IV of colorectal cancer were equally distributed among all four quartiles of 25-D3. Yet, the mortality rate was the lowest in the highest quartile of 25-D3. Compared to the lowest quartile, the highest quartile had an adjusted hazard ratio (HR) for overall mortality of 0.52 (95% confidence interval, CI=0.29-0.94). The HR for colorectal cancer mortality was 0.61 (95% CI=0.31-1.19) for the highest 25-D3 quartile compared to the lowest (15). This study suggests a correlation between vitamin D status and the risk of death from colorectal cancer, even in patients with advanced metastatic disease. Therefore, a better understanding of vitamin D status (as measured by 25-D3 levels) and response to cholecalciferol (D3) supplementation is imperative prior to proceeding with prospective clinical trials that address the value of D3 supplementation in colorectal cancer outcome. We have previously shown through a retrospective study of 315 patients with various stages of colorectal cancer that severe vitamin D deficiency (25-D3 <15 ng/ml) is highly prevalent in this population (16). On multivariate analysis, we showed, for the first time, that chemotherapy treatment was the most important predisposing variable (HR=3.74; 95% CI=1.9-7.3) for severe vitamin D deficiency in colorectal cancer patients (16). Given the severe deficiency that was encountered, we proceeded to evaluate, prospectively, the response of patients with various stages of colorectal cancer with or without chemotherapy to D3 supplementation at 2000IU per day. The 2000 IU/ day dose-level is considered a safe dose by the Food and Nutrition Board (Institute of Medicine) and has been associated with a mean increase in 25-D3 levels of 15-25 ng/ml in non-cancer populations (17-21).
Patients and Methods
Objectives. The primary endpoint of this study was to determine the impact of oral cholecalciferol (D3) at 2000 IU/day on 25-D3 levels after 3 and 6 months of supplementation. Secondary endpoints included the evaluation of the impact of various patient characteristics on 25-D3 levels at 3 and 6 months.
Study design and treatment administration. Patients with colorectal cancer treated at a single institute (Roswell Park Cancer Institute) were offered to participate in a prospective pilot vitamin D supplementation study. Patients received 2000 IU of vitamin D orally once daily. Patients' compliance was measured by performing a pill count at 3 and 6 months. A total of 50 evaluable patients, defined as receiving at least 3 months of vitamin D supplementation, were enrolled. Vitamin D was purchased from Vital Nutrients (Middletown, Connecticut, USA) and was supplied to patients free of charge. Patients' sera were assayed for 25-D3 levels before treatment, at 3 months, and at 6 months of therapy. A standard commercially available Immunochemiluminometric Assay (ICMA) was used for all collected samples (22). The normal recommended 25-D3 levels using this assay are 32 ng/ml to 100 ng/ml (22).
Patients' demographic data were collected at the time of enrollment and during treatment. Collected clinical data included age, gender, body mass index (BMI), stage of disease, prior surgeries, type and duration of chemotherapy received during study, and the type of prior intestinal surgery.
The study and consent form were approved by the Institutional Scientific and Review Committee (SRC) and the Institutional Review Board (IRB) prior to study activation. The study was registered at clinicaltrials.gov as NCT01198548.
Patient eligibility. Only patients with a diagnosis of colorectal cancer were eligible to participate. Other eligibility criteria included age ≥18 years, Eastern Cooperative Oncology Group (ECOG) performance status of ≤2, life expectancy >6 months, serum creatinine <2.0 mg/dL, total bilirubin <2mg/dL, alanine transaminase (ALT) and aspartate transaminase (AST) <5× upper limit of normal, and a baseline 25-D3 level <40 ng/ml. Patients who had prior D3 supplementation ≥800 IU/day were not allowed to take part unless they discontinued vitamin D3 supplementation two months or more prior to enrollment. No vitamin D supplements were allowed beyond those supplied in the study. Other exclusion criteria included history of hypercalcemia or history of genitourinary stones. All patients provided signed informed consent before study entry. The study was conducted in accordance with the Good Clinical Practice Guidelines as issued by the International Conference on Harmonization and the Declaration of Helsinki.
Clinical evaluation and follow-up. Physical examination and toxicity assessment were performed prior to study initiation, and at 3 and 6 months from study enrollment. Serum creatinine, calcium, and electrolyte levels were obtained at these same time points. No dose modification was allowed on study. Patients with intolerance to study treatment or with grade 2 or above hypercalcemia were to be taken off treatment.
Variables. Vitamin D response was assessed at 3 and 6 months of treatment. Vitamin D response was categorized using two different definitions: i) absolute increase of ≥10 ng/ml from baseline, and ii) achievement of a level of 25-D3 ≥32 ng/ml. Patients who received ≥4 weeks of chemotherapy during study treatment were considered to have a meaningful chemotherapy exposure and were labeled as the ‘chemo group’ while patients with <4 weeks of chemotherapy exposure or no chemotherapy exposure were labeled the ‘no chemo’ group.
Statistics. We investigated the associations between vitamin D response and gender, age, BMI, stage, type of surgery, season of initiation of treatment, and chemotherapy status. Analysis of quantitative levels of 25-D3 included t-test, ANOVA, and (multiple) linear regression as appropriate. Fisher exact test was used for 2×2 contingency tables. Analysis was performed in the R statistical computing environment (http://www.R-project.org/).
Results
Patient demographics. Fifty-three patients were enrolled on study between September 2007 and May of 2008. Three patients were non-evaluable for vitamin D3 response, leaving a total of 50 evaluable patients. The reason for lack of response evaluation in these three patients were study withdrawal related to non-compliance in one patient, major protocol violation in the second patient (lack of baseline 25-D3 level), and worsening renal function prior to study drug administration in the third patient. All patients carried a diagnosis of colorectal cancer. Stage II, III, and IVdisease constituted 6%, 40%, and 54% of the evaluable population, respectively. Fifty-six percent of the patients were male, and 92% were white. Seventy-four percent received 4 or more weeks of chemotherapy during the 3 months of vitamin D3 supplementation (‘chemo group’). Other patient characteristics including age, season at the time of study treatment initiation, type of chemotherapy used, BMI, primary tumor site and type of colonic surgery are detailed in Table I.
Baseline 25-D3 levels and patients' characteristics. The mean 25-D3 level at the time of enrollment was 17.5 ng/ml. Mean baseline 25-D3 levels differed by season (Figure 1) but not by other measured covariates. Levels of sera drawn in the summer or fall were more likely to be elevated than levels drawn in the winter or spring (p<0.01). The association between mean 25-D3 levels and season did not reach statistical significance when all four seasons were categorized separately (p=0.06). After adjusting for season, no other baseline variables were found to be predictive of 25-D3 levels on multivariate analysis.
25-D3 Status after 3 months of cholecalciferol supplementation. Four patients did not undergo their scheduled 3-month blood draw for 25-D3 level, resulting in 46/50 patients with an evaluable 3-month response to cholecalciferol. The mean level of 25-D3 after 3 months of daily replacement with cholecalciferol at 2000 IU/day was 31.6 ng/ml. This corresponded to an increase of 14.1ng/ml from baseline. Sixty-five percent of patients had an increase of ≥10 ng/ml and in 52% this reached the recommended range of ≥32 ng/ml.
25-D3 levels after 3 months of cholecalciferol supplementation did not correlate with baseline 25-D3 levels. This observation is contrary to recently published reports showing that baseline 25-D3 levels to significantly influence serum 25-D3 level response to cholecalciferol supplement (23). This suggests that patients with low levels of vitamin D are as likely to respond to supplementation as those with higher levels. Contrary to baseline levels, season had no impact on 25-D3 levels on day 90, suggesting that 2000 IU/day supplementation overwhelms the seasonal effect on 25-D3 levels. 25-D3 levels on day 90 were significantly associated with chemotherapy status between baseline and day 90 (p<0.01). Patients who received chemotherapy between baseline and day 90 had a mean 25-D3 level of 28.7 (±10.3) ng/ml while those without chemotherapy had a mean 25-D3 level of 39 (±7.9) ng/ml. Chemotherapy attenuated the response to cholecalciferol supplementation irrespective of which regimen was used, but this did not reach statistical significance (Figure 2). When categorizing responders and non-responders to 3 months of cholecalciferol by an increase of ≥10 ng/ml, a significant difference was also seen between the chemo group (39% responders) and no chemo group (92% responders). Similar results were reproduced when responses were categorized by day 90 levels ≥32 ng/ml. On multivariate analysis, and upon controlling for chemotherapy effect on 25-D3, none of the baseline patient characteristics were found to influence day 90 25-D3 levels. There was a trend for an association between higher BMI and lower 25-D3 levels on day 90 (p=0.057), suggesting that a higher BMI may attenuate the response to cholecalciferol replacement.
Vitamin D status after 6 months of cholecalciferol supplementation. Nine patients were missing day 180 25-D3 levels, limiting the population assessable for day 180 response to 41 patients. The mean and median levels for 25-D3 levels after 6 months of cholecalciferol replacement were 33.8 ng/ml and 34.9 ng/ml, respectively. The mean, median, mean increase in 25-D3, and the percentage of responders were similar for the D180 and D90 25-D3 levels (Table II). Patients receiving chemotherapy between day 90 and 180 had lower 25-D3 levels compared to those without chemotherapy. The mean 25-D3 level was 29 ng/ml for patients receiving chemotherapy vs. 37.5 ng/ml for those without chemotherapy. On multivariate analysis, only chemotherapy status between day 90-180 and day 90 25-D3 levels were significant predictors of day 180 25-D3 levels.
Discussion
Several studies have reported a correlation between a higher baseline 25-D3 level and improved colorectal cancer outcome (15, 24). Yet a recent analysis of the N9741 study of three different chemotherapy regimens in the first-line treatment of metastatic colorectal cancer did not substantiate any significant association between baseline 25-D3 levels and mortality (25). However, no assessment of vitamin D status during treatment was made in this study (25). Vitamin D may exert a favorable impact on colorectal cancer outcome through a direct antitumor activity or through indirect mechanisms such as normal tissue protection or immunomodulation (26-28). To conclusively determine the impact of D3 supplementation on colorectal cancer outcome, effective supplementation strategies should be compared in randomized blinded studies to non-supplementation strategies. Prior to the conduct of such studies, optimal dosing and scheduling of vitamin D supplementation has to be determined. A recent retrospective analysis reported a suboptimal response to vitamin D3 supplementation of 8000 IU/day in cancer patients (23). The response rate (>32 ng/ml) at first follow-up (median of 11 weeks) in patients with colorectal cancer was 46.7% (23). Response in that study was not stratified according to chemotherapy administration. On the other hand, significantly more robust responses have been associated with 2000 IU/day supplementation in deficient healthy individuals (21, 29). Therefore, we embarked on a prospective pilot study to determine the efficacy of 2000 IU/day supplementation on achieving adequate 25-D3 levels (exceeding 32 ng/ml). Consistent with our prior findings of an association between chemotherapy and vitamin D deficiency, chemotherapy was the only significant factor predicting for a reduced response to 2000 IU of vitamin D3 supplementation in our study (30). The association between chemotherapy administration and a poor response to vitamin D3 (2000 IU/day) was significant both at the 3 and 6-month endpoints. Our data suggest that colorectal cancer patients who are not receiving active chemotherapy can be adequately supplemented with 2000 IU/day, while patients receiving active chemotherapy may require substantially higher dosing. The reason for the attenuated response to vitamin D supplementation has not been elucidated. Potential contributing factors include dietary changes, limited absorption secondary to clinical or subclinical enteritis, increased 25-D3 metabolism, and an increased inflammatory response triggered by chemotherapy. The investigation of the impact of chemotherapy on inflammation and 25-D3 levels would be of particular interest given the recent association between elevated post-surgical C-reactive protein levels and a decline in 25-D3 levels (31). Whether the poor response to vitamin D supplementation is related to limited absorption due to subclinical chemotherapy-induced enteritis, a chemotherapy-induced pro-inflammatory process, or to disease burden or its response is unclear at this time. These hypotheses may need testing in future supplementation studies.
We acknowledge that our study is limited by the small sample size and clinical stage as well as treatment heterogeneity. However, our results stress the need to study chemotherapy and vitamin D status interactions in larger studies. Future prospective studies should address optimal vitamin D3 dosing in patients receiving chemotherapy. We are currently conducting a phase II clinical trial in patients with metastatic colorectal cancer receiving first-line 5-fluorouracil, oxaliplatin, leucovorin (FOLFOX) to determine the efficacy of 5000 IU of D3 in achieving optimal 25-D3 levels.
- Received January 15, 2012.
- Revision received March 17, 2012.
- Accepted March 19, 2012.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved