Vitamin D and Breast Cancer Survival

Institution: Cancer Prevention Institute of California
Investigator(s): Esther John, Ph.D. -
Award Cycle: 2010 (Cycle 16) Grant #: 16IB-0039 Award: $220,132
Award Type: IDEA
Research Priorities
Etiology and Prevention>Prevention and Risk Reduction: ending the danger of breast cancer



Initial Award Abstract (2010)

There are notable racial/ethnic disparities in survival following breast cancer diagnosis in U.S. women. African Americans and, to a lesser extent, Hispanics have poorer survival than non-Hispanic Whites. The proposed study aims to address these racial/ethnic disparities in relation to vitamin D, a hormone that has been linked to breast cancer progression, metastasis and survival in experimental and some ecological and epidemiological studies. Racial/ethnic differences in vitamin D status in U.S. women are well described and appear to parallel the racial/ethnic differences in breast cancer survival, with African Americans and Hispanics in general having lower serum vitamin D levels and higher prevalence of vitamin D deficiency than non- Hispanic Whites.

The goals of the study are to better substantiate whether an individualís vitamin D status and vitamin D receptor (VDR) gene polymorphisms are associated with survival following breast cancer; and to determine whether racial/ethnic differences in vitamin D status can explain part of the racial/ethnic disparities in breast cancer survival in U.S. women.

The specific aims of our project include:

  1. To determine whether breast cancer-specific survival and all-cause survival are associated with a) measures of vitamin D status, including serum 25(OH)-vitamin D levels, sun exposure, dietary and supplemental vitamin D intake; and b) VDR polymorphisms.
  2. To determine how much of the differences in breast cancer-specific survival and all-cause survival between African Americans and non-Hispanic Whites observed in our study can be accounted for by their differences in serum vitamin D levels.
  3. To determine the extent to which associational effects determined in Aims 1 and 2 compare to those determined by causal analyses.

To address these aims we will build upon existing data and biospecimen resources for 2,244 female invasive breast cancer cases aged 35-79 years from a multiethnic population-based case-control study that ascertained newly diagnosed cases through the Greater Bay Area Cancer Registry. By linking to the Cancer Registry, we will obtain updates on vital status and cause of death. We will perform survival analysis to determine whether vitamin D-related exposures are associated with death after breast cancer diagnosis after adjusting for known breast cancer prognostic factors. We will apply rigorous statistical methods to address the question of how much of the observed difference in survival between African Americans and non-Hispanic whites can be accounted for by their difference in serum vitamin D levels.

Having a suboptimal serum vitamin D level is a widespread health problem facing U.S. women and particularly those with breast cancer. If improvement of vitamin D status can indeed extend breast cancer survival, it will benefit an enormously large number of breast cancer patients. Moreover, because modifying vitamin D status through supplementation can be achieved in a simple, safe, and low cost manner, it can provide a conceivably cost effective means to achieve our goal of reducing disparities and improving breast cancer survival in all women.




Final Report (2012)

This study addressed two main questions:
1) For women with breast cancer, are those who have lower levels of vitamin D in their bodies, and/or those who have a less efficient vitamin D receptor (VDR), more likely to die after being diagnosed with the disease?
2) It is known that in the U.S., African American women on average have lower vitamin D levels than non-Hispanic Whites. Can the lower vitamin D levels in African American women explain in part why they also have a worse survival after breast cancer diagnosis?

First, we examined whether post-diagnostic blood vitamin D levels, vitamin D related exposures (sun exposure, diet) or VDR genetic changes are associated with survival. In Aims 2 and 3 we used different statistical approaches to assess to what extent differences in blood vitamin D levels account for the survival difference between African American and non-Hispanic White women. We measured blood vitamin D levels in 882 women, and obtained data on vitamin D related exposures (sun exposure, diet) assessed by questionnaire and on genetic variants in the VDR gene and other genes in the vitamin D metabolic pathway (genotyping was supported by NCI grant R01 CA14000; PI: Marty Slattery). A manuscript on the environmental and genetic predictors of blood vitamin D levels is currently under review. When analyzed alone, blood vitamin D was a significant predictor of overall survival. Specifically, women with lower vitamin D levels had worse survival after being diagnosed with breast cancer. In addition, time spent in outdoor activities during the year before diagnosis and a genetic variant in the VDR gene were also significantly associated with survival. However, after taking into account clinical factors that are known to be important in the survival of breast cancer patients, only outdoor activities remained a significant predictor of survival.

Next, our data do not support an important role of blood vitamin D levels in explaining the survival difference between African American and non-Hispanic White women with breast cancer. Given this unexpected finding, we revised Aim 3 to use a different statistical method (namely, Random Survival Forests) to analyze the survival data. We discovered evidence of a threshold effect with blood vitamin D levels. Specifically, only in the range that was below 20ng/ml, blood vitamin D was a significant predictor of survival of breast cancer patients. The association remained significant after taking into account known factors that predict survival of breast cancer patients. A manuscript is close to completion.

In future work, we plan to extend the survival analyses and examine the role of genetic predictors of blood vitamin D levels in breast cancer survival.