Iris de Nie

127 PROSTATE CANCER INCIDENCE UNDER ANDROGEN DEPRIVATION 8 Table 2. Standardized Incidence Ratios for prostate cancer in trans women using hormone treatment Follow up time (years) Observed cases Dutch incidence rates (per 100,000 persons, per year) 139 Expected cases Standardized Incidence Ratio (95% confidence interval) Age categories: < 30 years 7,190 0 0 0 ·· 30-44 years 14,800 0 0.32 0 ·· 45-59 years 13,429 1 58.56 7 0.14 (0.01-0.70) 60-74 years 4,428 4 490.54 20 0.20 (0.06-0.48) > 75 years 535 1 567.68 3 0.33 (0.02-1.64) Overall (n=2,281) 37,117 6 30 0.20 (0.08-0.42) Subgroup analyses: Hormone treatment with orchiectomy (n=1572) 26,048 4 ·· 24 0.17 (0.05-0.40) Hormone treatment without orchiectomy (n=709) 6,796 2 ·· 5 0.44 (0.07-1.47) We were unable to perform analyses on different hormone treatment protocols, since on the one hand, many trans women change often between different types of prescribed estrogens over time and, on the other hand, they mostly use cyproterone acetate as anti- androgenic treatment. DISCUSSION Our study shows a five-fold decrease in prostate cancer risk in trans women using hormone treatment compared to the general male population of similar age. This observation provides new insight in the relationship between testosterone and prostate cancer risk. Where previously no association was found between serum testosterone concentrations and the incidence of prostate cancer, our results show that very low serum testosterone concentrations have a substantial preventive effect on the initiation and development of prostate cancer. In this study, we linked the cohort of trans women with nationwide registries on prostate cancer and mortality. Therefore, we feel that the incidence we found of 16.2 prostate cancer cases per 100,000 years is a reliable estimate. In 2014, Gooren and Morgentaler reported an incidence of 2.0 prostate cancer cases per 100,000 person years in trans women. 141 This is lower than the reported incidence in our current study and is likely due to the absence of information on prostate cancer cases diagnosed in other centers and the lackof mortalitydata leading to an overestimation of follow-up time. Another cohort study showed a much higher incidence of 72 prostate cancer cases per 100,000 person years for trans women. 142 However, the higher incidence of prostate cancer in this American cohort may be explained by the fact that 38% of their study population consisted of trans women who had not undergone gender-affirming hormone treatment and were, therefore, not androgen deprived. 143