Severe acute kidney injury after abiraterone vs enzalutamide in advanced prostate cancer: A SEER–Medicare study.

Abstract

120 Background: Acute kidney injury (AKI) is a frequent complication among older adults with cancer, linked to short-term mortality, prolonged hospitalization, and progression to chronic kidney disease (CKD). Traditional androgen deprivation therapy has been associated with AKI, but contemporary evidence for androgen receptor pathway inhibitors (ARPIs) is limited. Abiraterone acetate with prednisone (AAP) may increase renal vulnerability via mineralocorticoid excess and steroid effects, whereas enzalutamide (ENZA) has other mechanisms. We quantified the risk of AKI-related hospitalization following ARPI initiation and, comparing AAP with ENZA overall and within key clinical subgroups. Methods: Using SEER–Medicare data, we identified men with advanced prostate cancer (PCa), with no prior AKI, who initiated AAP (2013–2018) or ENZA (2013–2019). AKI hospitalizations were identified using inpatient ICD-9/10 codes. Follow-up ran from ARPI start to first AKI hospitalization, death, loss of Medicare coverage, or December 31, 2020. Fine–Gray sub-distribution hazard models were used to estimate adjusted hazard ratios (HRs). Results: We included 5,574 patients with PCa (2,948 AAP; 2,626 ENZA); 78% were ≥75 years. The 12-month cumulative incidence of AKI hospitalization was higher with AAP vs ENZA (17.9% vs 13.6%). At 12-months post treatment initiation, cumulative risk of AKI was substantially higher in patients with pre-existing CKD vs without CKD (27.2% vs 9.2%; p<0.05) and with congestive heart failure (CHF) vs without CHF (23.9% vs 11.7%; p<0.05). Risk factors associated with increased hospitalization with AKI included older age, higher CCI, pre-existing CKD and CHF (Table 1). Conclusions: AAP was associated with a higher adjusted hazard of AKI hospitalization than ENZA. Clinically, these data support (i) preferring ENZA over AAP when oncologically acceptable in patients at elevated renal risk (CKD, CHF, high comorbidity, older age, Black race), and (ii) structured renal monitoring after ARPI initiation, with tighter surveillance for AAP users and those with CKD/CHF. Key predictors of AKI hospitalization after ARPI initiation (adjusted) with selected baseline differences. Risk factor Comparison Adjusted HR 95% CI (lower) 95% CI (upper) Age group ≥75 vs 65–74 1.24 1.09 1.42 Race Black vs White 1.30 1.05 1.62 Marital status Unmarried vs Married 1.33 1.09 1.63 Charlson comorbidity index 1 vs 0 1.23 1.06 1.42 ≥2 vs 0 1.61 1.27 2.04 ARPI agent AAP vs ENZA 1.25 1.10 1.41 Treatment year 2013–2016 vs 2017–2019 1.56 1.37 1.79 Pre-existing CKD Yes vs No 2.54 2.20 2.94 Pre-existing CHF Yes vs No 1.53 1.25 1.88 Charlson Comorbidity Index did not include AMI, CHF and DM in its calculation. Model was adjusted for age group, race, marital status, education, income, SEER region, state buy-in, dual eligibility, CCI, ARPI agent, treatment year, pre-existing CKD, DM, CHF and AMI.