Predictive value of baseline risk factors and early prostate-specific antigen (PSA) kinetics for achieving deep PSA decline in metastatic hormone-sensitive prostate cancer: implications for treatment intensification.

BACKGROUND: Early identification of treatment response is essential for risk-adapted management in metastatic hormone-sensitive prostate cancer (mHSPC). This study aimed to explore the predictive value of baseline clinicopathological factors and prostate-specific antigen (PSA) kinetics for deep PSA decline at 6 months in patients with mHSPC. METHODS: A retrospective cohort study was conducted involving 364 patients with de novo mHSPC. The primary endpoint was defined as PSA at 6 months ≤0.2 ng/mL. Logistic regression analyses were performed to identify independent predictors, and a nomogram was constructed. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) and calibration analysis. RESULTS: Independent predictors of poor PSA response included baseline alkaline phosphatase (ALP) >100 IU/L [odds ratio (OR) =0.363, P=0.02], International Society of Urological Pathology (ISUP) grade group 5 (OR =0.215, P=0.01), perineural invasion (OR =0.345, P=0.006), high disease volume (OR =0.270, P=0.02), and PSA >0.7 ng/mL at 2 months (OR =0.027, P<0.001). A predictive nomogram demonstrated discrimination (AUC = 0.907) and calibration. In a subgroup analysis of patients with PSA >0.7 ng/mL at 2 months, docetaxel addition was associated with a higher likelihood of achieving PSA response (OR =7.810, P=0.01). CONCLUSIONS: PSA at 2 months may serve as a useful biomarker for predicting deep PSA response in patients with mHSPC. The proposed nomogram may facilitate early risk stratification and warrants further validation in prospective multicenter studies.