Introduction

Prostate cancer (PCa) poses a substantial health challenge and ranks as the second-most common cause of cancer death among men. In 2022, an estimated 137,308 new cancer cases were diagnosed in Texas.¹ Notably, prostate, breast, lung, and colorectal cancers collectively account for approximately 48% of all cancer diagnoses, which necessitates prompt diagnosis and appropriate treatment strategy.^1,2 Standard treatment options for prostate cancer encompass radiation, surgery, or androgen deprivation therapy (ADT).³ However, ADT carries a range of possible adverse effects, including cardiovascular diseases, neurologic disorders, and rheumatic autoimmune diseases (RAD).^4-6
Rheumatic autoimmune diseases (RAD) comprise a group of debilitating conditions that lead to chronic inflammation, such as rheumatoid arthritis, systemic lupus erythematosus, ankylosing spondylitis, and Sjogren's syndrome.^7-11 Given the autoimmune origins of RAD, it is essential to investigate the mechanisms and correlation with ADT to improve patient care outcomes and minimize adverse events.

Objective: To investigate whether ADT is associated with an increased risk of rheumatic autoimmune diseases in men with prostate cancer.

Methods

A cohort of patients aged >=66 years who were first diagnosed with stages I–III prostate cancer between 2010-2016 was identified and followed up until 2019, using the Texas Cancer Registry data (TCR) linked to Medicare. We excluded patients who died <90 days after PCa diagnosis, and men who were previously diagnosed with RAD at any time before cohort entry. Exposure to ADT was defined as the first dose of a GnRH or orchiectomy within six months following PCa. All patients were followed until the diagnosis of RAD or censored by the end of the study follow-up period, death, or loss of coverage, whichever occurred first.
Our analyses included Cox proportional hazards, inverse probability treatment weighting (IPTW), and propensity score matching. Propensity score approaches are used to control for variations in the scope of ADT vs non-ADT use by balancing the effect of age at diagnosis, race, stage, and year of diagnosis intrinsically. Time to RADs was calculated by using the Kaplan-Meier curve, and Cox proportional hazards analyses were used to estimate hazard ratios (HR) with 95% CI of RAD.

Results

A total of 16,401 patients were included in the analysis. Patients who received ADT showed distinct characteristics in age at diagnosis, race or ethnicity, cancer stage, and year of diagnosis, compared to non-ADT. Propensity score matching and weighting methods effectively reduced these differences. Kaplan-Meier curves for time to RAD indicated notably higher failure rates for those who received ADT compared to non-ADT group. The five-year failure rates were 0.11 (95% CI: 0.1–0.12) and 0.09 (95% CI: 0.08–0.1) for the ADT and non-ADT groups, respectively, with a p-value <0.001.
In the multivariable Cox proportional analyses, the receipt of ADT was associated with an increased risk of RAD (HR: 1.28; 95% CI: 1.15–1.48). Comparable results were obtained using conditional logistic regression models with the propensity matching method (HR: 1.29; 95% CI: 1.17–1.56), while our analyses with IPTW yielded smaller reductions (HR: 1.21; 95% CI: 1.09–1.43), all of which demonstrated statistical significance.

Conclusion

Based on the findings of this population-based cohort with non-metastasized prostate cancer, we observed that patients who underwent ADT (agonists, antagonists, orchidectomy) showed a heightened risk of developing rheumatic autoimmune diseases (RAD). This association remained significant even after adjusting for various sociodemographic and clinical confounders.
The significance of these findings lies in the clinical and epidemiological aspects, highlighting that androgen deprivation therapy (ADT) encompasses a broader spectrum of autoimmune diseases compared to alternative treatments, and it will be crucial to evaluate therapy approaches.