Effects of Chemotherapy on Vitamin D Levels
A high prevalence of VD deficiency is seen in various cancers. In a retrospective review of 25(OH)D levels in 2098 adults with solid tumors of any stage (6% had metastatic disease [n=124]), suboptimal levels were found in 69% of patients with breast cancer (n=617), 75% with colorectal cancer (n=84), 72% with gynecologic cancer (n=65), 79% with kidney and bladder cancer (n=145), 83% with pancreatic and upper gastrointestinal tract cancer (n=178), 73% with lung cancer (n=73), 69% with prostate cancer (n=225), 61% with skin cancer (n=399), and 63% with thyroid cancer (n=172).5 Suboptimal VD also has been found in hematologic malignancies. In a prospective cohort study, mean serum 25(OH)D levels in 23 patients with recently diagnosed acute myeloid leukemia demonstrated VD deficiency (mean [SD], 18.6 [6.6] nmol/L).9 Given that many patients already exhibit a baseline VD deficiency at cancer diagnosis, it is important to understand the relationship between VD and cancer treatment modalities.5
In the United States, breast and colorectal cancers were estimated to be the first and fourth most common cancers, with 313,510 and 152,810 predicted new cases in 2024, respectively.10 This review will focus on breast and colorectal cancer when describing VD variation associated with chemotherapy exposure due to their high prevalence.
Effects of Chemotherapy on Vitamin D Levels in Breast Cancer—Breast cancer studies have shown suboptimal VD levels in 76% of females 75 years or younger with any T1, T2, or T3; N0 or N1; and M0 breast cancer, in which 38.5% (n=197) had insufficient and 37.5% (n=192) had deficient 25(OH)D levels.11 In a study of female patients with primary breast cancer (stage I, II, or III and T1 with high Ki67 expression [≥30%], T2, or T3), VD deficiency was seen in 60% of patients not receiving VD supplementation.12,13 A systematic review that included 7 studies of different types of breast cancer suggested that circulating 25(OH)D may be associated with improved prognosis.14 Thus, studies have investigated risk factors associated with poor or worsening VD status in individuals with breast cancer, including exposure to chemotherapy and/or radiation treatment.12,15-18
A prospective cohort study assessed 25(OH)D levels in 95 patients with any breast cancer (stages I, II, IIIA, IIIB) before and after initiating chemotherapy with docetaxel, doxorubicin, epirubicin, 5-fluorouracil, or cyclophosphamide, compared with a group of 52 females without cancer.17 In the breast cancer group, approximately 80% (76/95) had suboptimal and 50% (47/95) had deficient VD levels before chemotherapy initiation (mean [SD], 54.1 [22.8] nmol/L). In the comparison group, 60% (31/52) had suboptimal and 30% (15/52) had deficient VD at baseline (mean [SD], 66.1 [23.5] nmol/L), which was higher than the breast cancer group (P=.03). A subgroup analysis excluded participants who started, stopped, or lacked data on dietary supplements containing VD (n=39); in the remaining 56 participants, a significant decrease in 25(OH)D levels was observed shortly after finishing chemotherapy compared with the prechemotherapy baseline value (mean, −7.9 nmol/L; P=.004). Notably, 6 months after chemotherapy completion, 25(OH)D levels increased (mean, +12.8 nmol/L; P<.001). Vitamin D levels remained stable in the comparison group (P=.987).17
Consistent with these findings, a cross-sectional study assessing VD status in 394 female patients with primary breast cancer (stage I, II, or III and T1 with high Ki67 expression [≥30%], T2, or T3), found that a history of chemotherapy was associated with increased odds of 25(OH)D levels less than 20 ng/mL compared with breast cancer patients with no prior chemotherapy (odds ratio, 1.86; 95% CI, 1.03-3.38).12 Although the study data included chemotherapy history, no information was provided on specific chemotherapy agents or regimens used in this cohort, limiting the ability to detect the drugs most often implicated.
Both studies indicated a complex interplay between chemotherapy and VD levels in breast cancer patients. Although Kok et al17 suggested a transient decrease in VD levels during chemotherapy with a subsequent recovery after cessation, Fassio et al12 highlighted the increased odds of VD deficiency associated with chemotherapy. Ultimately, larger randomized controlled trials are needed to better understand the relationship between chemotherapy and VD status in breast cancer patients.
Effects of Chemotherapy on Vitamin D Levels in Colorectal Cancer—Similar to patterns seen in breast cancer, a systematic review with 6 studies of different types of colorectal cancer suggested that circulating 25(OH)D levels may be associated with prognosis.14 Studies also have investigated the relationship between colorectal chemotherapy regimens and VD status.15,16,18,19
A retrospective study assessed 25(OH)D levels in 315 patients with any colorectal cancer (stage I–IV).15 Patients were included in the analysis if they received less than 400 IU daily of VD supplementation at baseline. For the whole study sample, the mean (SD) VD level was 23.7 (13.71) ng/mL. Patients who had not received chemotherapy within 3 months of the VD level assessment were categorized as the no chemotherapy group, and the others were designated as the chemotherapy group; the latter group was exposed to various chemotherapy regimens, including combinations of irinotecan, oxaliplatin, 5-fluorouracil, leucovorin, bevacizumab, or cetuximab. Multivariate analysis showed that the chemotherapy group was 3.7 times more likely to have very low VD levels (≤15 ng/mL) compared with those in the no chemotherapy group (P<.0001).15
A separate cross-sectional study examined serum 25(OH)D concentrations in 1201 patients with any newly diagnosed colorectal carcinoma (stage I–III); 91% of cases were adenocarcinoma.18 In a multivariate analysis, chemotherapy plus surgery was associated with lower VD levels than surgery alone 6 months after diagnosis (mean, −8.74 nmol/L; 95% CI, −11.30 to −6.18 nmol/L), specifically decreasing by a mean of 6.7 nmol/L (95% CI, −9.8 to −3.8 nmol/L) after adjusting for demographic and lifestyle factors.18 However, a prospective cohort study demonstrated different findings.19 Comparing 58 patients with newly diagnosed colorectal adenocarcinoma (stages I–IV) who underwent chemotherapy and 36 patients who did not receive chemotherapy, there was no significant change in 25(OH)D levels from the time of diagnosis to 6 months later. Median VD levels decreased by 0.7 ng/mL in those who received chemotherapy, while a minimal (and not significant) increase of 1.6 ng/mL was observed in those without chemotherapy intervention (P=.26). Notably, supplementation was not restricted in this cohort, which may have resulted in higher VD levels in those taking supplements.19
Since time of year and geographic location can influence VD levels, one prospective cohort study controlled for differential sun exposure due to these factors in their analysis.16 Assessment of 25(OH)D levels was completed in 81 chemotherapy-naïve cancer patients immediately before beginning chemotherapy as well as 6 and 12 weeks into treatment. More than 8 primary cancer types were represented in this study, with breast (34% [29/81]) and colorectal (14% [12/81]) cancer being the most common, but the cancer stages of the participants were not detailed. Vitamin D levels decreased after commencing chemotherapy, with the largest drop occurring 6 weeks into treatment. From the 6- to 12-week end points, VD increased but remained below the original baseline level (baseline: mean [SD], 49.2 [22.3] nmol/L; 6 weeks: mean [SD], 40.9 [19.0] nmol/L; 12 weeks: mean [SD], 45.9 [19.7] nmol/L; P=.05).16
Although focused on breast and colorectal cancers, these studies suggest that various chemotherapy regimens may confer a higher risk for VD deficiency compared with VD status at diagnosis and/or prior to chemotherapy treatment. However, most of these studies only discussed stage-based differences, excluding analysis of the variety of cancer subtypes that comprise breast and colorectal malignancies, which may limit our ability to extrapolate from these data. Ultimately, larger randomized controlled trials are needed to better understand the relationship between chemotherapy and VD status across various primary cancer types.