Page 315 - Vitamin D and Cancer

302 M. Fakih et al.

Table 13.2 Case–control studies (1989–2007): vitamin D status and colorectal
cancer
Year Author Population Risk reduction
1989 Garland [40] Men in Maryland ~80%
1995 Braun [41] ATBC Study ~40%
1997 Tangrea [42] Finnish Study 40%
2004 Feskanich [44] NHS 47%
2006 WaktawskiWende [45] WHI 60%
2007 Wu [43] HPHS plus NHS 34%
2007 Otani [46] JPHC Neg
ATBC Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study, JPHC
Japan Public Health Centre-base Prospective Study, WHI Women’s Health
Initiative, NHS Nurses Health Study, HPHS Health Professionals Health
Study

(HPFS) [43]. One hundred and seventy-nine of the patients enrolled on the HPFS
study were diagnosed with colorectal cancer between 1993 and 2002. These cases
were matched to 356 controls by age and by month and year of blood collection.
Blood was collected between 1993 and 1995 pre-diagnosis. Serum 25-D was
3
assayed by RIA. An inverse association between higher levels of 25-D and risk of
3
colorectal cancer was noted in the case-control population but did not reach statisti-
cal significance. However, the association was highly significant when the analysis
was limited to colon cancer (OR 0.46; CI 0.24–0.89). A nested case-control study
was also performed in the Nurse’s Health Study (NHS) [44]. One-hundred and
ninety-three cases of colorectal cancer were identified within 11 years of their initial
blood draw. Three-hundred and fifty-six controls were selected in a 2:1 ratio from
the same cohort as the case. Controls had to be cancer free at the time of diagnosis
and were matched to cases by age and month of blood draw. 25-D was assayed by
3
RIA. A significant inverse association was noted between 25-D and risk of colorec-
3
tal cancer (p = 0.02). The risk reduction in colorectal cancer for the highest 25-D
3
quintile when compared to the lowest quintile was 47%, close to statistical signifi-
cance (OR = 0.53; CI 0.27–1.04). Another nested case-control study, from the
Women’s Health Initiative (WHI), matched 317 women with colorectal cancer to
317 non-cancer controls by age, center, race or ethnic group, and date of blood sam-
pling [45]. 25-D was assayed using chemiluminescent RIA. A significant inverse
3
association between 25-D and risk of colorectal cancer was confirmed (p = 0.02).
3
The risk of colorectal cancer among the lowest quartile of 25-D was 2.53-fold
3
higher than the highest quartile (CI 1.49–4.32). A recent Japanese case-control study
from two large male and female cohorts failed to support the above findings [46].
Three-hundred and seventy-five colorectal cancer cases were identified within
11.5 years from blood collection. Two controls were matched to each case by age,
sex, study area, date of blood draw, and fasting time. 25-D was assayed by a com-
3
petitive protein-binding assay. Although no association was found between 25-D
3
levels and risk of colorectal cancer, low 25-D levels were associated with a statisti-
3
cally significant increased risk of rectal cancer in both males and females.