To the Editor,

Results from the meta-analysis by Chen et al. in 2010 [1] showed an interesting result that an inverse relation was found between dietary calcium, vitamin D intake, and breast cancer risk [compared to the lowest quantile, the relative risk (RR) for the highest quantile of calcium, vitamin D intake, and serum vitamin D (measured as 25(OH)D levels) is 0.81 (95 %CI = 0.72–0.90), 0.91 (95 %CI = 0.85–0.97), and 0.55 (0.38–0.80), respectively]. And, Yin et al. in 2010 [2], assuming a linear relationship without performing a formal test, also found that the RR for an increase of 25(OH)D by 20 ng/ml is 0.73 (0.60–0.88). This investigation has much practical implication regarding the prevention of breast cancer among women. However, we would like to draw attention to the type of dose–response analysis between dietary calcium intake, vitamin D levels, and breast cancer and the possibility that calcium intake and vitamin D levels might have a threshold effect on the risk of breast cancer because a linear association in epidemiologic research can rarely be assumed a priori [3]. Besides, categories of calcium intake and vitamin D levels differed between studies, which might complicate the interpretation of the pooled results across study populations with different categories. In this respect, a dose–response meta-analysis with restricted cubic spline functions provides a solution to the problem [3] from which a summary risk estimate can be derived for a standardized increase and specific exposure values for calcium intake and vitamin D levels. And, many studies were published for the association of calcium intake and vitamin D levels with risk of breast cancer after 2010.

A comprehensive literature search was performed to Jun 2012 using the databases of Pubmed, Web of Knowledge, China Biology Medical literature database (CBM), Database of Chinese Scientific and Technical Periodicals (VIP), China National Knowledge Infrastructure (CNKI), and Google scholar. Medical Subject Headings (MeSH) were used as the search terms without restriction to MeSH Major Topic, and the search strategy was as follows: ((((“1,25-dihydroxyvitamin D” [Mesh]) OR “Calcium”[Mesh]) OR “Vitamin D”[Mesh]) OR “25-hydroxyvitamin D 2”[Mesh]) AND “Breast Neoplasms”[Mesh]. In addition, we reviewed the reference lists of all identified relevant publications and relevant reviews.

The number of cases and participants (person-years), and RR 95 %CI for each category of calcium intake and vitamin D levels were extracted. We extracted the RR (95 %CI) which reflected the greatest degree of control for potential confounders. The median or mean calcium intake and vitamin D levels for each category were assigned to each corresponding RR for every study. If the results were reported for both dietary and total intake of calcium and vitamin D intake, we used the results for total intake. If the upper boundary of the highest category was not provided, we assumed that the boundary had the same amplitude as the adjacent category.

A 2-stage random-effects dose–response meta-analysis was performed taking into account the between-study heterogeneity proposed by Orsini et al. [4] to compute the trend from the correlated log RR estimates across categories of calcium intake and vitamin D levels. Briefly, a restricted cubic spline model, with 4 knots at the 5th, 35th, 65th, and 95th percentiles [5] of the calcium intake and vitamin D levels, was estimated by generalized least square regression taking into account the correlation within each set of published RRs [6]. Then, we combined the study-specific estimates by the restricted maximum likelihood method in a multivariate random-effects meta-analysis [7]. A P value for nonlinearity and overall significance was calculated by the method proposed by Greenland and Longnecker [8]. All statistical analyses were performed by means of STATA version 12 (Stata Corporation, College Station, Texas, USA). All reported probabilities (P values) were two-sided with P ≤ 0.05 considered statistically significant.

For dietary calcium intake, data from 10 publications [9–18] were used including 14450 breast cancer cases. A linear relationship was found for calcium intake with risk of breast cancer (P for non-linearity = 0.23, P for overall significance < 0.00), and the RRs (95 %CI) of breast cancer were 0.96 (0.93–0.99), 0.95 (0.91–0.99), 0.93 (0.89–0.98), 0.92 (0.87–0.98), 0.91 (0.86–0.97), 0.91 (0.86–0.96), 0.91 (0.86–0.96), 0.91 (0.86–0.96), and 0.91 (0.86–0.96) for 250, 350, 450, 550, 650, 750, 850, 950 and 1100 mg/day, respectively (Fig. 1).

Fig. 1
figure 1

The dose–response analysis between calcium intake and risk of breast cancer with restricted cubic splines in a multivariate random-effects dose–response model. The solid line and the long dash line represent the estimated relative risk and its 95 % confidence interval. Short dash line represents the linear relationship

Full size image

For dietary vitamin D intake, data from 13 publications [9–13, 15, 16, 19–24] were used including 20,343 breast cancer cases. A nonlinear relationship was found for vitamin D intake with risk of breast cancer (P for non-linearity < 0.01, P for overall significance < 0.00), and the RRs (95 %CI) of breast cancer were 0.98 (0.95–1.00), 0.95 (0.92–0.99), 0.93 (0.89–0.98), 0.92 (0.88–0.96), 0.91 (0.87–0.96), 0.91 (0.87–0.85), 0.90 (0.86–0.95), 0.90 (0.85–0.95), and 0.90 (0.85–0.95) for 50, 100, 150, 200, 250, 300, 350, 500, and 600 IU/day, respectively (Fig. 2).

Fig. 2
figure 2

The dose–response analysis between vitamin D intake and risk of breast cancer with restricted cubic splines in a multivariate random-effects dose–response model. The solid line and the long dash line represent the estimated relative risk and its 95 % confidence interval of the nonlinear relationship. Short dash line represents the linear relationship

Full size image

For serum vitamin D levels, data from 4 publications with 1,25-dihydroxyvitamin D [23, 25–27], and 8 publications with 25-hydroxyvitamin D [25, 28–34] were used including 8,716 breast cancer cases. A nonlinear relationship was found of serum vitamin D levels with risk for breast cancer (P for non-linearity < 0.01, P for overall significance < 0.00), and the RRs (95 %CI) of breast cancer were 0.97 (0.93–1.02), 0.95 (0.88–1.03), 0.92 (0.84–1.02), 0.87 (0.79–0.96), 0.82 (0.75–0.89), 0.78 (0.71–0.85), 0.77 (0.71–0.84), 0.80 (0.75–0.85) for 5, 10, 15, 20, 25, 30, 35, and 40 ng/ml, respectively (Fig. 3).

Fig. 3
figure 3

The dose–response analysis between serum vitamin D levels and risk of breast cancer with restricted cubic splines in a multivariate random-effects dose–response model. The solid line and the long dash line represent the estimated relative risk and its 95 % confidence interval of the nonlinear relationship. Short dash line represents the linear relationship

Full size image

Overall, this dose–response analysis suggested that women might suffer from the lowest risk of breast cancer with dietary calcium intake of about 600 mg/day, dietary vitamin D intake of 400 IU/day, and serum vitamin D levels of 30 ng/ml. Further researches deserve to confirm the findings and to determine the utility of dietary calcium intake and vitamin D levels for the primary prevention of breast cancer.