Skip to Main Content

An official website of the United States government

Vitamin D

Purpose

Since the beginning of Etiologic and Early Marker Studies (EEMS), vitamin D, as measured in blood serum, has been an extremely popular analyte. Investigators have used it in examinations of several different cancer sites. For CDAS, these measurements are available in both a combined summary dataset, as well as separately, contained in the study for which they were originally selected.

Population

These studies were each selected at different time points, using a different set of exclusions and matching criteria. For all studies, the common requirements for inclusion were etiologic consent, serum availability, intervention arm, a baseline questionnaire, some form of follow-up within the trial, and no history of the site-specific cancer prior to blood collection. All studies used blood collected at T0, except for bladder and the second prostate study, where the first available pre-diagnostic collection was used, and the lung study, where the closest collection prior to diagnosis was used. For all studies except for breast and prostate, the set of cases were matched directly to the controls. For breast and prostate, however, the populations were the subsets of the Cancer Genetic Markers of Susceptibility (CGEMS) study who had available serum. The pancreas population was selected in two phases; the first was selected in 2007, while the second was added in 2009. For the adenoma study, cases were required to have a left-sided adenoma at baseline (adenoma ≥ 1 cm, or containing high-grade dysplasia or villous elements), while controls were required to have a negative screen (no polyp or other suspect lesion). For the second prostate study, African American cases were first matched to African American controls at a 1:2 ratio, and Caucasian controls were then matched to half of the African American controls at a 1:1 ratio.

Site Non-Common Inclusion Criteria Matching Variables Cases Non-Cases
Colorectal
Adenoma
  • Randomized before 9/30/99
  • Adequate Flexible Sigmoidoscopy at Baseline
  • No Cancer at Baseline
  • No Colorectal Co-Morbities at Baseline (Ulcerative Colitis, Crohn’s Disease, Familiar Polyposis, Colorectal Polyps, or Gardner’s Syndrome)
  • Gender
  • Race
 388 395
Breast
  • Females
  • No non-breast cancers during trial
  • Age at Randomization (5 Years)
  • Fiscal Year of Randomization (<=1997, 1998+)
 1000 1002
Prostate
  • White
  • Males
  • With a Prostate Cancer Screen
  • Age at Randomization (5 Year)
  • Fiscal Year of Blood Draw
  • Study Year of Dx/Selection
 748 728
Pancreas
(Phase 1 + Phase 2)		 None
  • Gender
  • Race
  • Age at Randomization (5 Years)
  • Year and Month of Entry (2 Months)
  • Year of Birth (5 Years)
  • Study Year of Dx/Selection
 234 (178+56) 416 (354+62)
NHL No Cancer at Baseline
  • Gender
  • Race
  • Age at Randomization (5 Years)
  • Year and Season of Entry (3 Months)
  • Year of Birth (5 Years)
  • Study Year of Dx/Selection
 295 295
Endometrial
  • No Cancer at Baseline
  • Females
  • Confirmed Post-Menopausal
  • No Bilateral Oophorectomy at Baseline
  • No Hysterectomy at Baseline
  • Race
  • Age at Randomization (+/- 1 Year)
  • Date of Blood Draw (+/- 30 days)
  • Date of Dx/Selection
 147 146
Ovarian
  • No Cancer at Baseline
  • Females
  • Confirmed Post-Menopausal
  • No Bilateral Oophorectomy at Baseline
  • Race
  • Age at Randomization (+/- 1 Year)
  • Date of Blood Draw (+/- 30 days)
  • Date of Dx/Selection
 74 74
Renal No Cancer at Baseline
  • Gender
  • Race
  • Age at Randomization (+/- 1 Year)
  • Date of Blood Draw (+/- 30 days)
  • Date of Dx/Selection
 160 160
Upper GI No Cancer at Baseline
  • Gender
  • Race
  • Age at Randomization (+/- 1 Year)
  • Date of Blood Draw (+/- 30 days)
  • Date of Dx/Selection
 99 97
Bladder
  • No Cancer at Baseline
  • No Cancer Prior to DX/Selection
  • Gender
  • Race
  • Age at Randomization (+/- 1 Year)
  • Date of Blood Draw (+/- 30 Days)
  • Days on Study until Dx/Selection
 371 362
Colorectal
  • No Cancer at Baseline
  • No History of Colon Disease
  • Gender
  • Race
  • Age at Blood Draw (+/- 1 Year)
  • Date of Blood Draw (+/- 30 Days)
  • Date of Dx/Selection
 476 476
2nd Prostate
  • No Cancer at Baseline
  • Males
  • Black/White
  • Age at Randomization (+/- 5 Years)
  • Study Year of Blood Draw
  • Date of Blood Draw (+/- 210 Days for case-control match; +/- 30 Days for control match)
  • Days on Study until Dx/Selection
  • Geographic Region (control match only)
 226 678
Lung
  • No Cancer at Baseline
  • Black/White
  • Controls Cancer-free through Follow-up
  • Specific Smoking History
  • Gender
  • Race
  • Date of Birth (+/- 5 Years)
  • Smoking History
  • Study Year of Blood Draw
  • Date of Blood Draw (+/- 6 Months)
  • Time of Day of Blood Draw (3 groups)
  • Days on Study until Dx/Selection
 450 450
Total     4488 4660

Data Collected

All studies measured 25(OH) Vitamin D. Second prostate and lung measured 25(OH) Vitamin D2 and D3 separately. Adenoma, breast and prostate measured 1,25(OH)2 Vitamin D, though prostate only measured it on a subset of 176 controls. 25(OH) measures are presented in ng/mL and 1,25(OH)2 measures are presented in pg/mL.

Vitamin D data can be obtained in the Vitamin D datasets

Main Findings