Prostate cancer epidemiological studies of risk factors

University dissertation from Umeå : Umeå universitet

Abstract: In spite of the fact that prostate cancer is the most common male cancer in both Sweden and many other countries in the developed world, little is known of risk factors and predisposing conditions. The only well recognized risk factors are age, race and familial aggregation. More knowledge about risk factors could lead to better preventive measures together with better treatments.One way to evaluate this is to study second primary cancers; the connection between two different cancers can give valuable insight in etiology or clues to shared risk factors. This thesis aims at evaluating risk factors for prostate cancer.We constructed a cohort of 135,713 men diagnosed with prostate cancer and reported to the Swedish Cancer Registry 1958-1996. The cohort was followed for second primary cancers and a doubled risk of male breast cancer was found. We also noted increased risks for small intestine cancers and melanoma.As a follow-up on the increased risk of male breast cancer, we performed a nested case – control study. Included cases were men with first prostate and then breast cancer (n = 41) matched to men with only prostate cancer (n =81). For these men, we collected medical records and extracted data regarding treatment. Furthermore, all men diagnosed with both prostate and breast cancer irrespective which came first (n = 83) were used as probands. To both these sets of cases with breast and prostate cancer, we identified first degree relatives and grandchildren from parish offices throughout Sweden. Linking to the Cancer Registry retrieved all cancer diagnoses amongst relatives. Results from this study show a relation between estrogen treatment of prostate cancer and the risk of developing breast cancer. We also found that a small part of the cases with both cancers appeared in families with inheritance patterns possibly attributed to BRCA2.As estrogen treatment seemed involved in increased risk of breast cancer after prostate cancer, we wanted to investigate the newly discovered Estrogen receptor ? and the relation to prostate cancer risk. Previous reports have shown that ER? acts as a negative regulator of proliferation. ER? expression occurs mainly in prostatic epithelial cells and the expression gradually diminishes when cancer develops and aggravates. We used a single nucleotide polymorphism (SNP) association study approach to evaluate genetic variation in ER? as a risk factor for prostate cancer. One SNP, located in the promoter region associated with a small increased risk of prostate cancer whereas variation in the rest of the gene did not.In the last paper, we investigated trans-urethral resection (TURP) of the prostate due to benign prostate hyperplasia (BPH) as a risk factor for later development of prostate cancer. Evidence has gathered that both BPH and prostate cancer are associated to inflammation. By comparing incidence and mortality in a cohort of 7,901 men with the general population there appeared to be an increased risk of prostate cancer but decreased mortality. Analyzing this increased risk further, we conducted a nested case - control study with men extracted from the cohort. Cases had a TURP and later developed prostate cancer and controls just had a TURP. We then evaluated the specimens from TURP regarding extent of inflammation, degree of androgen receptor down regulation and expression of p53, all factors previous associated with prostate cancer. None of these parameters differed between cases and controls and they can therefore not explain the increased risk. Decreased mortality but increased risk might be explained by surveillance bias, which means more medical attention to these patients, resulting in diagnosing clinically non-significant cancers.In summary, our results show a doubled risk of male breast cancer following prostate cancer. A risk that can be attributed to the use of estrogen to treat prostate cancer or to some extent a possible mutation in BRCA2. We also propose that a SNP change in the ER? promoter confer a small increased risk of prostate cancer. A small risk elevation of prostate cancer following TURP most probable could depend on surveillance bias.