• Division of Epidemiology, Vanderbilt Ingram Cancer Center, Department of Medicine, Vanderbilt University School of Medicine, Vanderbilt University Medical Center, Nashville, TN, 37203, USA.
• Master of Public Health Program, School of Medicine, Vanderbilt University, Nashville, TN, 37232, USA.
• Department of Molecular Physiology and Biophysics, Vanderbilt University, Nashville, TN, USA.
• Division of Gastroenterology, Hepatology and Nutrition, Department of Medicine, Vanderbilt University Medical Center, Nashville, TN, 37232, USA.
• Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, MA, 02115, USA.
• Division of Epidemiology, Vanderbilt Ingram Cancer Center, Department of Medicine, Vanderbilt University School of Medicine, Vanderbilt University Medical Center, Nashville, TN, 37203, USA. Electronic address: qi.dai@vanderbilt.edu.

BACKGROUND & AIMS: In contrast to many observational studies, large-scale randomized trials do not support the protective role of vitamin D for the prevention of colorectal neoplasia. However, in previous studies, individuals with blunted parathyroid hormone (PTH) response to vitamin D insufficiency/deficiency (BPRVID), were not differentiated from those with high PTH response to vitamin D insufficiency/deficiency (HPRVID). Individuals with BPRVID are responsive to magnesium treatment, particularly treatment of magnesium plus vitamin D while those with HPRVID are responsive to vitamin D treatment. We prospectively compared these two distinct groups (i.e. BPRVID and HPRVID) for risk of incident adenoma, metachronous adenoma, and incident colorectal cancer (CRC) METHODS: Three nested case-control studies in the Prostate, Lung, Colorectal and Ovarian Cancer (PLCO) Screening Trial.

RESULTS: We found optimal 25(OH)D levels were associated with a significantly reduced risk of CRC, primarily among women. The associations between 25(OH)D and CRC risk significantly differed by PTH levels, particularly among women. Compared to individuals with optimal levels for both 25(OH)D and PTH, all others were at an elevated risk of incident CRC, primarily in women. We found those with BPRVID had 2.56-fold significantly increased risk of CRC compared to 1.65-fold non-significantly increased risk for those with HPRVID. Among women, we observed those with BPRVID had 4.79-6.25-fold significantly increased risks of incident CRC and adenoma whereas those with HPRVID had 3.65-fold significantly increased risk of CRC.

CONCLUSIONS: Individuals with BPRVID are at higher risks of incident adenoma and CRC compared to those with HPRVID, particularly among women.