Androgen deprivation therapy (ADT) increases the risk of adverse cardiovascular events in patients with prostate cancer. ADT can induce body composition change; however, the association between body composition change and cardiovascular outcomes remains unclear.

This study aimed to determine the association between ADT-induced body composition change and cardiovascular outcomes in patients with prostate cancer. This retrospective study included 681 patients with prostate cancer (410 in the derivation cohort and 271 in the external validation cohort) who underwent radiotherapy and ADT between 2008 and 2021.

Computed tomography (CT) scans at baseline and 6 months post-ADT were used to assess the changes in skeletal muscle index (ΔSMI), subcutaneous adipose tissue index (ΔSATI) and visceral adipose tissue index (ΔVATI) at the L3 vertebral level. The primary outcome was major adverse cardiovascular events (MACEs), defined as a composite of myocardial infarction, stroke, heart failure, arterial revascularization and cardiovascular death.

The association between body composition changes and MACE was analysed using the SHapley Additive ExPlanations (SHAP) method and validated using the Cox proportional hazards model. The median age was 72 years (interquartile range: 66-77).

With a median follow-up of 6.0 years (derivation cohort) and 6.6 years (validation cohort), 62 (15.1%) and 39 (14.4%) patients experienced MACE, respectively. Six months post-ADT, there was a reduction in SMI, whereas SATI and VATI had increased (p < 0.001).

The ΔSMI and ΔSATI were the most important features for predicting MACE in both cohorts, whereas ΔVATI and baseline body composition parameters were less influential. SMI loss was inversely correlated with MACE risk, whereas increases in SATI and VATI were positively correlated.

The identified thresholds for MACE prediction were SMI loss ≥ 4.7% and SATI gain ≥ 8.2%. On multivariable Cox regression analysis, ΔSMI (hazard ratio: 1.27 per 1% decrease, p < 0.001) and ΔSATI (hazard ratio: 1.07 per 1% increase, p < 0.001) were independently associated with MACE risk; however, ΔVATI was not (hazard ratio: 0.99, p = 0.23).

Stratification based on thresholds confirmed that SMI loss ≥ 4.7% (hazard ratio: 10.58, p < 0.001) and SATI gain ≥ 8.2% (hazard ratio: 3.03, p < 0.001) were independently associated with increased MACE risk. ADT-induced muscle loss and increased subcutaneous adipose tissue were associated with an increased MACE risk in patients with prostate cancer.

These findings highlight the need to monitor and address body composition changes in patients undergoing ADT to reduce cardiovascular risk. Future research should explore interventions to mitigate these metabolic effects and improve patient outcomes.