Correction: Rapid and sustainable deep testosterone reduction predicts effective androgen deprivation therapy for metastatic hormone-sensitive prostate cancer

Correction to: Scientific Reports 10.1038/s41598-025-13480-8, published online 05 August 2025

The original version of this Article contained errors.

The results of the multivariable Cox analysis in Figure 3B were inadvertently misrepresented during the revision process. This error also affected the nomogram in Figure 3C and the Kaplan-Meier (KM) analysis in Figure 3D. The incorrect Figure 3 along with its captions is provided below.

Fig. 3.

KM analysis for CRPC progression stratified by DTR in the first-year ADT (A). Univariable and multivariable Cox analysis for prognostic role of DTR in the first-year ADT in CRPC progression (B). Construction of nomogram model for predicting CRPC progression likelihood based on three independent risk factors including DTR in the first-year ADT, serum PSA levels before ADT treatment and diabetes (C). KM analysis for CRPC progression stratified by risk scores calculated by nomogram model (D).

Correspondingly, the description of the findings related to Figure 3 in the Results section, under subsection, “Rapid and sustainable DTR predicts the efficacy of ADT treatment for mHSPC” now reads:

During the first year of ADT, 107 out of 144 patients achieved a drop of serum TT levels to DTR. Consistent with previous reports12,18 these patients progressed to CRPC at a much slower pace than those who did not achieve DTR (median 1 143 vs. 594 days, P = 0.001) (Fig. 3A). Both univariable and multivariable Cox analyses identified DTR during the first year of ADT as a protective factor for CRPC progression (HR = 0.47, 95% CI 0.29–0.76, P = 0.002; HR = 0.48, 95% CI 0.30–0.78, P = 0.003, respectively) (Fig. 3B). Conversely, serum PSA levels ≥ 100 ng/mL prior to ADT and diabetes were identified as risk factors for CRPC progression (HR = 2.34, 95% CI 1.24–4.43, P = 0.009; HR = 2.03, 95% CI 1.09–3.80, P = 0.027, respectively) (Fig. 3B). Based on these prognostic factors, we developed a nomogram model to score the likelihood of CRPC progression in individual patients (C-index: 0.649) (Fig. 3C). KM analysis showed that patients with higher risk scores progressed to CRPC more rapidly (median 461 vs. 1,361 days, P < 0.001) (Fig. 3D). The results from calibration plots and DCA suggest that the model has good predictive ability and clinical utility (Fig. S2A and B).

The original Article has been corrected.