Vitamin D (VitD) supplementation decreases the risk of osteoporotic fractures in the elderly; however, its extraskeletal benefits, especially in the prevention and treatment of breast cancer, are less well-established. Many studies have shown an association of low VitD levels with higher cancer incidence, including breast cancer and poorer outcomes, but whether this association merely reflects a selection bias to healthier lifestyles, remains an area of controversy. We hypothesized that women with more aggressive HER2+ breast cancer would have improved clinical outcomes while on VitD supplements.


We performed a retrospective review of all patients (n = 300) given trastuzumab chemotherapy between 2006 and 2012 at UM/SCCC. We identified two groups of patients for comparison - those who received VitD supplementation (VD) during adjuvant chemotherapy (n = 130) or none (NVD) during adjuvant chemotherapy (n = 123). Patients who lacked sufficient records to clarify VitD supplement use, men, patients with de-novo-metastatic breast cancer, bilateral breast cancers, and patients without follow-up were excluded. Five-year disease-free survival (DFS) and overall survival (OS) were calculated. Univariate and multivariate analyses were performed using a Cox proportional hazards (CPH) model to evaluate the relationship between VD supplementation and death.


The median age at diagnosis was 54 and 50 in the VD and NVD groups. In the VD group, the average VitD dose was 10,890 IU/wk, and the baseline and post-25-H VitD serum level was 35 and 41ng/ml, respectively. Descriptive analysis of the VD and NVD groups were as follows: postmenopausal (55.4%, 43.9%), tumor <2cm (42.3%, 36.6%), no lymph node involvement (42.3%, 36.6%), LVI (46.4%, 33.3%), high nuclear grade (60%, 61.5%), HR+ (66.2%, 54.5%), African American race (4.6%, 9.8%), and BMI>30 at end of chemotherapy (26.2%, 31.7%). At a median follow-up of 31 and 23 months, the estimated five-year DFS (69.4% vs. 44.7%, p = 0.009) and OS (97.5% vs. 85.6%, p = <0.0001) were significantly superior in the VD group versus the NVD group. These differences remained significant after adjustment for age, ethnicity, menopausal state, tumor size, node positivity, LVI, high-grade tumor, HR+, and BMI>30. Analysis showed an interaction between OS and ethnicity (African American = 0.008) and node positivity (p = 0.02) and near-significance for LVI (p = 0.07). Despite those confounding variables, VitD use remained significantly associated with improved OS (p = 0.003) and had a HR or 0.10 with a 95% CI of 0.02-0.45.


Our study suggests that Vitamin D supplementation in non-metastatic HER2 breast cancer patients is associated with improved DFS and OS, and the relationship remains significant after adjusting for potential confounding factors. It is unclear whether vitamin D supplementation might have pre-selected for HER2+ breast cancers with more favorable prognosis or synergized with anti-HER2 therapy. To our knowledge, this is the first study reporting improved outcomes associated with relatively high dose Vitamin D supplementation in the HER2+ breast cancer population. Further research is warranted to define the role of Vitamin D in breast cancer treatment.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-09-02.