Premium
The interaction of CYP3A5 polymorphisms along the androgen metabolism pathway in prostate cancer
Author(s) -
Vaarala Markku H.,
Mattila Henna,
Ohtonen Pasi,
Tammela Teuvo L.J.,
Paavonen Timo K.,
Schleutker Johanna
Publication year - 2008
Publication title -
international journal of cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.475
H-Index - 234
eISSN - 1097-0215
pISSN - 0020-7136
DOI - 10.1002/ijc.23425
Subject(s) - prostate cancer , cancer , medicine , oncology , genotype , prostate , carcinogenesis , odds ratio , population , androgen , testosterone (patch) , cancer research , endocrinology , biology , gene , genetics , environmental health , hormone
Prostate cancer is a leading solid tumor among men in the Western world. Androgens play an important role in the carcinogenesis and treatment of prostate cancer. CYP3A5 is a cytochrome P450 superfamily member which also has activity in testosterone metabolism. In this study, we looked for two‐gene interactions associated with clinical characteristics of prostate cancer in the Finnish population. We used multifactor‐dimensionality reduction for the identification of the two‐gene interactions in androgen metabolism pathway genes together with clinical characteristics of prostate cancer among 754 genotyped prostate cancer patients. The CYP3A5 *3/*3 and SRD5A2 A49T GG genotype interaction was associated with the clinical tumor stage T2–T4 (T‐stage, TNM classification) with odds ratio (OR) 2.14, 95% confidence interval (CI) 1.35–3.40. Patients with CYP3A5 *3/*3 and KLK3 I179T CC/TC genotypes had increased OR 2.30, 95% CI 1.16–4.58 for metastatic disease. Further, two‐gene interaction CYP3A5 *3/*3 and KLK3 −252A > G AA was associated with Gleason scores ≥7 with OR 1.52, 95% CI 1.11–2.09. Prostate cancer patients with CYP3A5*3/*3 and KLK −252A > G GG/AG genotypes had decreased OR of 0.70 with 95% CI 0.50–0.98 for high prostate‐specific antigen levels at diagnosis. For prostate cancer patients aged below 65 years, the OR for interaction of CYP3A5 *1/*3 or *1/*1 and AKR1C3 Q5H CC genotypes was 1.84 with 95% CI 1.03–3.28. For prostate cancer, the best two‐gene interaction included genotypes SRD5A2 V89L GG and AKR1C3 Q5H CC with OR 1.30, 95% CI 1.01–1.66. It remains to be clarified whether these polymorphism associations identified here are also present in other populations. © 2008 Wiley‐Liss, Inc.