Optimal Vitamin D Status for Colorectal Cancer Prevention: A Quantitative Meta Analysis

Gorham, Edward D.; Garland, Cedric F.; Garland, Frank C.; Grant, William B.; Mohr, Sharif B.; Lipkin, Martin; Newmark, Harold L.; Giovannucci, Edward; Wei, Melissa; Holick, Michael F.

doi:10.1016/j.amepre.2006.11.004

« Previous Next »American Journal of Preventive Medicine
Volume 32, Issue 3 , Pages 210-216, March 2007

Optimal Vitamin D Status for Colorectal Cancer Prevention:

A Quantitative Meta Analysis

Edward D. Gorham, MPH, PhD
- University of California San Diego, Department of Family and Preventive Medicine, School of Medicine, La Jolla, California
- Address correspondence and reprint requests to: Edward D. Gorham, PhD, Research Epidemiologist, Naval Health Research Center (Code 24, Bldg 346), P.O. Box 85122, San Diego CA 92186-5122.
Cedric F. Garland, DrPH
- University of California San Diego, Department of Family and Preventive Medicine, School of Medicine, La Jolla, California
Frank C. Garland, PhD
- University of California San Diego, Department of Family and Preventive Medicine, School of Medicine, La Jolla, California
William B. Grant, PhD
- SUNARC-Sunlight, Nutrition and Health Research Center, San Francisco, California
Sharif B. Mohr, MPH
- University of California San Diego, Department of Family and Preventive Medicine, School of Medicine, La Jolla, California
Martin Lipkin, MD
- Strang Cancer Prevention Center, New York, New York
Harold L. Newmark, ScD
- Susan Lehman Cullman Laboratory for Cancer Research, Rutgers, The State University of New Jersey, Piscataway, New Jersey, and Cancer Institute of New Jersey, New Brunswick, New Jersey
Edward Giovannucci, MD, ScD
- Harvard School of Public Health, Departments of Nutrition and Epidemiology, Boston, Massachusetts
Melissa Wei, BS
- Harvard School of Public Health, Departments of Nutrition and Epidemiology, Boston, Massachusetts
Michael F. Holick, MD, PhD
- Vitamin D Laboratory, Section of Endocrinology, Nutrition and Diabetes, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts

Background

Previous studies, such as the Women’s Health Initiative, have shown that a low dose of vitamin D did not protect against colorectal cancer, yet a meta-analysis indicates that a higher dose may reduce its incidence.

Methods

Five studies of serum 25(OH)D in association with colorectal cancer risk were identified using PubMed. The results of all five serum studies were combined using standard methods for pooled analysis. The pooled results were divided into quintiles with median 25(OH)D values of 6, 16, 22, 27, and 37 ng/mL. Odds ratios were calculated by quintile of the pooled data using Peto’s Assumption-Free Method, with the lowest quintile of 25(OH)D as the reference group. A dose–response curve was plotted based on the odds for each quintile of the pooled data. Data were abstracted and analyzed in 2006.

Results

Odds ratios for the combined serum 25(OH)D studies, from lowest to highest quintile, were 1.00, 0.82, 0.66, 0.59, and 0.46 (p_trend<0.0001) for colorectal cancer. According to the DerSimonian-Laird test for homogeneity of pooled data, the studies were homogeneous (chi²=1.09, df=4, p=0.90. The pooled odds ratio for the highest quintile versus the lowest was 0.49 (p<0.0001, 95% confidence interval, 0.35–0.68). A 50% lower risk of colorectal cancer was associated with a serum 25(OH)D level ≥33 ng/mL, compared to ≤12 ng/mL.