Purpose: Reports in the osteoporosis literature demonstrating the increased activity of specific alleles of the vitamin D receptor and epidemiological data linking vitamin D levels with prostate cancer have stimulated research into possible associations between vitamin D receptor genotype and the development of prostate cancer. Recent studies showed that patients homozygous for a less active vitamin D receptor have a 4 to 5 times increased risk of localized prostate cancer. In 1 study this association was strongest in patients with advanced disease. To understand better the relationship between advanced disease and the vitamin D receptor we compared the vitamin D receptor genotype of 41 patients who died of prostate cancer to 41 controls with no clinical evidence of prostate cancer.
Materials and methods: Noncancerous deoxyribonucleic acid was isolated from lethal prostate cancer and control cohorts. To determine the TaqI restriction fragment length polymorphism a 740 base pairs (bp) segment of the vitamin D receptor was amplified by PCR, digested with TaqI endonuclease and resolved on an agarose gel. Depending on the presence or absence of a TaqI restriction site at codon 352 in each allele, products were digested into 2 fragments of 495 and 245 bp (T allele) or 3 fragments of 290, 245 and 205 bp (t allele). Individuals were classified as TT, Tt or tt. To determine the size of the poly-A microsatellite repeat an approximately 410 bp fragment was amplified by polymerase chain reaction using [gamma-32P] labeled primers, resolved by gel electrophoresis and sized by autoradiography. Fragments 410 bps or greater corresponded to repeats 18 bps or greater (L allele) and fragments less than 410 bps corresponded to repeats less than 18 bps (S allele). Individuals were classified as LL, LS or SS.
Results: The TaqI restriction fragment length polymorphism and poly-A microsatellite repeat were in strong linkage disequilibrium, indicating that both polymorphic markers identified the same vitamin D receptor genotype in the majority of cases. Using the TaqI restriction fragment length polymorphism 33 of 41 controls (80.5%) and 35 of 41 cases (85.3%) were heterozygous (Tt) or homozygous (TT) for the less active vitamin D receptor allele, while 8 of 41 controls (19.5%) and 6 of 41 cases (14.6%) were homozygous (tt) for the more active vitamin D receptor allele. Using the poly-A microsatellite repeat 32 of 40 controls (80%) and 32 of 38 cases (84.2%) were heterozygous (LS) or homozygous (LL) for the less active vitamin D receptor allele, while 8 of 40 controls (20.0%) and 6 of 38 cases (15.8%) were homozygous (SS) for the more active vitamin D receptor allele. There was no statistically significant difference in the distribution of either marker between cases and controls.
Conclusions: We failed to demonstrate an association between vitamin D receptor genotype and lethal prostate cancer. Our data do not support the hypothesis that specific vitamin D receptor genotypes are associated with an aggressive prostate cancer phenotype. Further studies that take into account cofactors important in vitamin D activity and/or a better definition of prostate cancer phenotype may explain the discrepancy between our findings and those of a previous study.