Year: 2017
Type: dissertation
Study design: Observational Study
Abstract: Prostate cancer (PCa) is the most common cancer in Swedish men and approximately 16% of all men will receive a PCa diagnosis before his 75th birthday.Testosterone (T) is crucial for the growth of both the healthy prostate and PCa, and is also important in the activation of the expression prostate specific antigen (PSA).The central role of T in PCa, and in the making of PSA, a clinical marker for prostatic disease, has led to the common belief that high T is a risk factor for PCa.T and its metabolite 5-alpha dihydrotestosterone (DHT) acts through the androgen receptor (AR), and genetic variants of the AR have been associated with PCa in several studies.The aims of this thesis were to investigate if T concentrations or the combination of genetic variants in the AR could influence the risk of being diagnosed with PCa and if AR-variants could affect the risk of having PSA above clinical thresholds on the suspicion of PCa in men without PCa.Additionally, the long term effect of extreme levels of T on mortality risk was also investigated, as previous studies have indicated an association between low T and mortality, but without determining the direction of the association.For the association studies regarding AR-variants and PSA concentrations, the European Male Ageing study (EMAS) was used (n after exclusion=1,804).For the association studies regarding AR-variants and PCa risk the EMAS cohort was combined with men from the Swedish Osteoporotic fractures in men (MrOS; n after exclusion=1,120) and a nested case-control sample of the Malmö diet and cancer study (MDCS; n after exclusion=883) to a final dataset of n=689 PCa cases and n=3214 men without PCa.Odds ratios (OR) and 95% confidence intervals (95%CI) were calculated for the risk of PSA above 3 or 4 ng/mL in the n=1,744 men in EMAS without PCa, and for the risk of PCa diagnosis in the combined dataset.For the associations between T concentration and risk of PCa and mortality, all men from whom T measurements were analyzed (n=4278) at the Department of Clinical Chemistry, Skåne University Hospital, Malmö, Sweden 1987-1992 were linked to national Cancer Registry and Death Registry.Hazard ratios (HR) and 95%CIs were calculated for the risk of mortality or being diagnosed with prostate cancer after more than 20 years of follow-up, for the men with the 5% highest and 5% lowest T concentrations compared to the men in the >10% to <90% range.All analyses were adjusted for age.The results revealed two dominant AR haplotypes, H1 and H2 with a frequency of ~85 and ~15%, respectively, in European men.The H2 haplotype had statistically significantly shorter CAG-repeat lengths (p<0.001),statistically significantly lower risk of PCa: OR (95%CI) = 0.70 (0.54-0.91; p=0.007), and statistically significantly increased risk of having PSA above 3 and 4 ng/mL; OR (95%CI) = 1.69 (1.13-2.52,p=0.011) and 1.99 (1.21-3.29;p=0.007), respectively.After 20 years of follow-up, no differences in PCa risk were detected for the men with the 5% lowest T (p=0.122) or 5% highest T (p=0.282).An increased risk of all-cause mortality was detected in the lowest 5% group, HR (95%CI) = 1.39 (1.10-1.75;p=0.006) and when divided into younger (<50 years) and older (>50 years) at T measurement, the increased risk was found to be statistically significant only in the younger men; HR (95%CI) = 2.31 (1.48-3.60;p<0.001).In conclusion, this thesis show a haplotype difference in PCa risk and risk of having PSA above clinical cutoffs, which might indicate a subgroup of men for which the PSA-test should be haplotype-adjusted.In addition, high T was not a risk factor for PCa, but low T might be an indicator of future risk of all-cause mortality, calling for closer monitoring of T by clinicians.
Author: Magdalena Bentmar Holgersson
Institution: Lund University
Language: English
Subfield: Cancer Research
Domain: Life Sciences
Keywords: prostate cancer risk, testosterone, dihydrotestosterone, prostate-specific antigen, PSA screening, cancer mortality, European Male Ageing Study, MRO, Malmö Diet and Cancer Study, cancer registry linkage, Swedish men, genetic susceptibility to cancer, hormone replacement therapy, androgen deprivation therapy
Open Access status: green
OpenAlex ID: W2767182957
