Abstract
BackgroundAndrogen deprivation therapy (ADT) is the mainline treatment for men with advanced prostate cancer (PCa), with some men remaining on ADT for up to two decades. Prolonged use of Gonadotropin Releasing Hormone (GnRH) agonists may be associated with survival benefits, but also with potential side-effects in men with PCa. One of the more recently investigated side-effects of ADT is an increased risk of cardiovascular disease (CVD). Observational studies that have explored CVD effects following GnRH agonists have found consistent positive associations whereas GnRH antagonists have shown less metabolic characteristics of CVD in preclinical models. Moreover, patterns of non-adherence to GnRH agonists among men with PCa may be associated with worse prognosis. This thesis used real world data to investigate risk of CVD following GnRH agonists and antagonists and to explore patterns and factors influencing adherence to GnRH agonists in men with PCa.
Methods
Data from six countries (United Kingdom (UK) excluding Scotland, Scotland, Belgium, the Netherlands, France and Canada) was extracted to evaluate the association between GnRH agonists or GnRH antagonists and the risk of CVD. Country-specific hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using multivariable adjusted Cox proportional hazards models and then pooled using a random effects meta-analysis model. Meta-analytical models included stratifications by history of CVD indicator (HCVDi) and age. In order to identify patterns affecting non-adherence to GnRH agonists, data from Sweden and UK were collated considering determinants at 3 years following GnRH agonists’ initiation. Non-adherence was determined by a medication possession ratio (MPR) of < 80%. Odds ratios (ORs) and 95% confidence intervals (CI) were calculated using logistic regression. Factors contributing to adherence in men with PCa on GnRH agonists in the UK were also thematically analysed using qualitative data from interviews with men with PCa on GnRH agonists and focus groups with their clinicians.
Results
Men with PCa on GnRH antagonists had an increased risk of developing any CVD (HR = 1.22; 95% CI = 1.03-1.45), arrhythmia (HR = 1.39; 95% CI = 1.13-1.72) and heart failure (HF) (HR = 1.33; 95% CI = 1.12-1.58) compared to men on GnRH agonists. In men on GnRH antagonists and with a HCVDi, there was an increased risk of developing arrhythmia (HR = 1.48; 95% CI = 1.03-2.13), HF (HR = 1.06; 95% CI = 1.05-1.07) and stroke (HR = 1.04; 95% CI = 1.03-1.05). Stratification by age showed an increased risk of developing any CVD (HR = 1.24; 95% CI = 1.04-1.48), ischaemic heart disease (IHD) (HR = 1.22; 95% CI = 1.03-1.45), arrhythmia (HR = 1.43; 95% CI = 1.19-1.73) and HF (HR = 1.39; 95% CI = 1.12-1.73) in those aged ≥ 75 years. MPRs showed an increased adherence both for men with PCa on primary (Sweden = 88%; UK = 75%) and secondary (Sweden = 84%; UK = 70%) GnRH agonists after 3 years on the treatment. Analysis from both countries showed that an increased age and longer injection intervals were associated with increased adherence to primary and secondary GnRH agonists. In Sweden, increased adherence was also observed in men with PCa given anti-androgens (OR = 1.53; 95% CI = 1.21-1.93) and radiotherapy (OR = 1.77; 95% CI =1.39-2.27) as prior PCa treatment before GnRH agonists compared to deferred PCa treatment. Qualitative analysis of interviews and focus groups in the UK showed that some multi-factorial reasons such as side-effects, strong patient belief system and quality over quantity of life contributed to non-adherence in some men.
Conclusion
Men with PCa and a HCVDi who were on GnRH antagonists may be at an increased risk of developing certain CVD subtypes compared to men on GnRH agonists. Pooling data from different countries can be challenging in the real world setting and results from both real world data and randomised controlled trials may be useful to better understand adverse effects of a drug. Therefore, results from the PRONOUNCE trial are required to fully address the potential of indication bias in this observational setting. Factors such as age, injection intervals and prior PCa treatments can influence adherence patterns to GnRH agonists in the PCa population. Moreover, employing different strategies by clinicians to support non-adherent men and keeping them engaged with the health care system may lead to the eventual acceptance of treatment whilst also acknowledging their reasons for non-adherence.
| Date of Award | 1 Feb 2020 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Mieke Van Hemelrijck (Supervisor) & Hans Garmo (Supervisor) |