Investigating the Adverse Effects of Hormone Therapy in Patients with Prostate Cancer

The incidence rate of prostate cancer in Taiwan is increasing and the 5-year overall survival is also increasing. However, many patients treated with androgen deprivation therapy (ADT) may notice treatment-related adverse effects lasting far beyond the end of treatment. These adverse events included cardiovascular diseases, diabetes and lipid-metabolic complication, bone-related disorders, depression, dementia/Alzheimer's disease, erectile dysfunction. However, it is hard to draw a conclusion about the increased adverse effects of ADT among patients with prostate cancer because of limited covariates in patient database, unrepresentative patient population or troublesome methodology for statistical analysis. To resolve these drawbacks, we are going to perform secondary data analysis on the Taiwan National Health Insurance Research Database (NHIRD), Taiwan Cancer Registry (TCR) and Taiwan Death Record (TDR) and aim to: 1. To investigate the association between ADT and dementia/Alzheimer's disease among patients with prostate cancer. 2. To investigate the impact of ADT related adverse effects on non-cancer death of patients with prostate cancer. For aim 1, two cohorts (ADT ever use and ADT never use) with newly-diagnosis prostate cancers from 2000-2013 will be enrolled. The index date will be defined as the date of orchiectomy or injection of the first dose of gonadotropin-releasing hormone (GnRH) agonist and followed till an occurrence of dementia/Alzheimer’s disease, the date of death, withdrawal from the NHI, or Dec. 31, 2015, whichever came first. It is anticipated about 60,000 patients in these two groups. Survival analysis with inverse probability of treatment weighting (IPTW) of propensity score and without weighting will be used to examine the robustness of the results. For aim 2, a nested case-control study will be used. The cases are those who died due to non-cancer after the diagnosis of prostate cancer. The index date will be defined as the date of dying due to non-cancer. Each case will be matched with survivor-control by age of prostate diagnosis and index year. The matching ratio of case-control is dependent upon number of case and controls in NHIRD. One advantage of this nested case-control design is that both cases (non-cancer death) and controls (survivor) have the similar exposure duration. Structural equation modelling (SEM) will be used to examine the complicate relationship between demographic characteristics, multiple comorbidities, treatment (ADT or not), ADT-related adverse effect and non-cancer death.

Project ID:PC10708-1223
External Project ID:MOST107-2314-B182-050