A pathological morphology parameter-based prognostic nomogram for high-risk prostate cancer patients treated with neoadjuvant therapy followed by radical prostatectomy: a retrospective study Authors: Hua Liang, Junjie Fan, Ke Xu, Hao Song, Chaosheng Gan, Guojing Wang, Qiyuan Kang, Jianzhou Liu, Dalin He, Jinhai Fan, Kaijie Wu Published: 2026 ABSTRACT BackgroundThe prognostic value of these pathological morphology alterations induced by neoadjuvant therapy in high-risk prostate cancer (PCa) patients is still unclear. Hence, this study retrospectively reviewed the data of 124 patients with high-risk PCa who underwent neoadjuvant therapy followed by radical prostatectomy (RP), and aimed to explore the prognostic value of pathological morphology alterations.MethodsData from 124 patients with high-risk PCa who underwent neoadjuvant therapy followed by RP were retrospectively reviewed. Pathological morphology alterations observed in RP specimens were independently recorded by two uropathologists, and the primary endpoint was biochemical progression-free survival (bPFS). Cox regression analyses were performed to explore independent predictors of bPFS, and a nomogram was developed. The C-index, calibration curves and decision curve analysis (DCA) were used to evaluate the performance of the nomogram.ResultsAmong 124 patients, 66 patients (53.2%) were treated with neoadjuvant hormonal therapy (NHT), and 58 patients (46.8%) were treated with neoadjuvant chemohormonal therapy (NCHT). Moreover, 11 patients (8.9%) had a complete reduction in glandular density and diameter, and intraductal carcinoma of the prostate (IDC-P) was observed in 8 patients (6.5%). Cox regression revealed that NHT, pelvic lymph node (LN) metastasis, positive surgical margins, negative reduction in glandular density and diameter, and IDC-P were associated with worse bPFS (all P < 0.05), and those factors were subsequently selected to develop a nomogram. The C-index was 0.813 (95% CI: 0.626–0.863), and time-dependent C-index were 0.902 (6 months), 0.882 (12 months), and 0.951 (24 months). The calibration curves showed a high consistency between the predicted and observed bPFS probabilities, and DCA confirmed the nomogram’s clinical utility across a range of threshold probabilities.ConclusionsPelvic LN metastasis, positive surgical margins and IDC-P were independent prognostic factors for high-risk PCa, and NCHT might significantly improve bPFS compared with the NHT. In addition, the degree of reduction in glandular density and diameter was also associated with bPFS, and the nomogram based on pathological morphology parameters-intended for postoperative risk stratification-might be helpful in clinical decision-making.Trial registrationRetrospectively registered.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12957-026-04442-z. Backgroud Prostate cancer (PCa) is the second most common cancer in men worldwide [1], and its incidence has increased annually because of the increased use of active prostate-specific antigen (PSA) screening and accurate diagnostic methods [2]. Approximately 15% of newly diagnosed patients with localized PCa are at high risk [3] and are more likely to have micrometastatic disease and early recurrence after local treatment. Neoadjuvant therapy followed by radical prostatectomy (RP) may be an option for these patients and has been recently demonstrated to improve oncological outcomes, especially for patients receiving neoadjuvant chemohormonal therapy (NCHT) [4–7]. Nevertheless, a subset of patients still experience treatment failure within 5 years. Thus, early identification and salvage or adjuvant treatment are crucial to improve the prognosis of these patients. Neoadjuvant therapy provides a unique opportunity to evaluate the in situ effects of therapies and consequently guide personalization following therapy. Several published studies have demonstrated that the Gleason score of samples after treatment is no longer accurate in assessing the disease severity and patient prognosis because the morphologic phenotype of the tumor has changed [7–10]. Previously, Wang et al., reviewed the data of 85 localized high-risk PCa patients who were treated with neoadjuvant hormonal therapy (NHT) followed by RP and reported that three pathological morphology variables, named a cribriform growth pattern, macronucleoli, and ductal adenocarcinoma differentiation, were independent predictive factors for the effectiveness of neoadjuvant hormonal therapy [11]. Furthermore, on the basis of hierarchical clustering analysis, Neil et al., categorized pre-treated PCa into three morphologically distinct groups after detailed morphologic evaluation of specimens from 115 patients with high-risk PCa who underwent neoadjuvant therapy followed by RP and noted that the presence of cribriform or intraductal spread morphology was a stronger predictor of biochemical recurrence (BCR) [12]. Thus, pathological morphology variables might be promising factors for predicting the effectiveness of neoadjuvant therapy and evaluating patient prognosis. Nowadays, most published studies have only focused on the value of pathological characteristics in PCa tissues before treatment, but the prognostic value of these pathological morphology alterations after neoadjuvant therapy is still unclear. To address this void, we retrospectively reviewed the clinicopathological characteristics of high-risk PCa patients who underwent neoadjuvant therapy followed by RP at our institution and aimed to explore the prognostic value of the pathological morphology alterations induced by neoadjuvant therapy. A prognostic nomogram based on pathological morphology parameters was subsequently developed to help urologists make clinical decisions after RP. Methods Study population We retrospectively reviewed the medical records of patients who were initially diagnosed with high-risk localized PCa at our institute. The inclusion criteria were as follows: (1) clinical stage greater than cT2c or a biopsy Gleason score (GS) ≥ 8, or a serum PSA ≥ 20 ng/ml, or pelvic metastatic lymph node involvement; (2) PCa was initially diagnosed with a resectable tumor; (3) patients received neoadjuvant therapy before RP; and (4) Eastern Cooperative Oncology Group (ECOG) score between 0 and 1. Patients were excluded if they had prior invasive malignancy, any serious comorbidities, incomplete data, or were withdrawn from neoadjuvant therapy. All patients provided written informed consent, and this study was approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, P. R. China. Neoadjuvant therapy and radical prostatectomy Patients received different neoadjuvant therapies according to their physical state, the surgeon’s suggestion and the patient’s choice. The NHT protocol involved 3 months of maximal androgen blockade, while four cycles of docetaxel-based chemotherapy within 3 months were added to the NCHT protocol. The maximal androgen blockade consisted of the subcutaneous administration of 10.8 mg goserelin acetate every 3 months and the oral administration of bicalutamide at 50 mg per day. Docetaxel was administered intravenously at a dose of 75 mg/m2 every 3 weeks, along with the oral administration of prednisone (5 mg) twice a day. RP was performed via a robot-assisted laparoscopic transabdominal approach with extended pelvic lymph node dissection 2 weeks after the last cycle of neoadjuvant chemotherapy. Pathological morphology evaluation All biopsy and surgical specimens were independently reviewed by two experienced pathologists, and any discordant case was re‑reviewed to reach a consensus before analysis. The pathological morphology alterations were observed in RP samples as follows: (1) a reduced degree of glandular density and diameter; (2) foamy cell infiltration; (3) cytoplasmic vacuolization; (4) stromal mucin and mucinous breakdown; (5) lymphocytic infiltration; (6) stromal fiber hyperplasis; (7) nuclear pyhnosis; and (8) intraductal carcinoma of the prostate (IDC-P). The degree of reduction in glandular density and diameter was evaluated as a single composite parameter, reflecting the overall degree of treatment‑induced glandular atrophy, which was assessed using a two-step method as previous described [11]. First, in diagnostic pretreatment needle biopsy, available tumor tissue or up to 4 biopsy cores with tumor were selected to calculate the average percentage of overall tumor volume and cellularity. Second, in RP specimens, the largest cross-section dimension in whole-mount slides containing residual tumor was identified. A cellularity hotspot was chosen under ×200 magnification, and 4 additional random fields were evaluated on the same slide. The mean cellularity percentage was calculated from a total of 5 fields (1 hotspot + 4 random fields). Then, the degree of reduction in glandular density and diameter was classified into three tier grades as follows: negative: ≤30% reduction in tumor cells; moderate: >30% reduction in tumor cells; and complete: pathological complete response (pCR) or minimum residual disease (largest cross-sectional dimension of residual tumor measuring 1–5 mm) in RP specimens. The status of lymphocytic infiltration was defined as positive when the extent of lymphocytic infiltration in RP samples was ≥ 10%. Follow-up and outcome measures Patients were followed up every month for the first 3 months and every 3 months thereafter with PSA tests and imaging examinations. The imaging examinations included chest and abdominal computed tomographic imaging, pelvic multiparameter magnetic resonance imaging (mp-MRI), and bone scintigraphy. Biochemical recurrence was defined as 2 consecutive rising PSA levels of > 0.2 ng/mL confirmed 2 weeks apart [13]. The primary endpoint was biochemical progression-free survival (bPFS), which was defined as the time from surgery to the date of biochemical recurrence, last follow-up, or death from any cause, whichever occurred first. Statistical analysis Data for continuous variables are presented as the mean ± standard deviation or median (range) as appropriate, and categorical data are displayed as numbers (percentages). Inter-observer variability was assessed using kappa coefficients for all key morphological features. Univariate and multivariate Cox regression analyses were performed to explore the independent predictors of bPFS. Survival curves were generated via the Kaplan‒Meier method, and log-rank tests were used to compare the differences. Variables with P < 0.05 in the multivariate Cox regression were considered potential candidate factors for nomogram building, and discrimination (C-index) and calibration (calibration curves) were used to evaluate the performance of the nomogram. Time-dependent C-indices were calculated at 6, 12, and 24 months. Decision curve analysis (DCA) was performed to evaluate the clinical application value of the nomogram. Statistical analysis was performed via SPSS 23.0 for Windows (SPSS, Chicago, IL). All tests were 2-tailed, and P < 0.05 was considered statistically significant. Nomogram construction, time-dependent C-indices, DCA and calibration curve analysis were performed via R i386 3.5.1 software. Results Baseline characteristics of patients According to the inclusion and exclusion criteria, 124 patients were ultimately included in this study. As shown in Table 1, the mean age of the 124 patients was 67.32 ± 5.72 years, and the median PSA level at diagnosis was 52.67 (22.89–92.97). A total of 64 patients (51.6%) had biopsy GSs of 8–10, and 99 patients (79.8%) were diagnosed with clinical T stage ≥ 3. Sixty-six patients (53.2%) were treated with the NHT, and 58 patients (46.8%) were treated with the NCHT. After treatment, 11 patients (8.9%) achieved a pCR, and 73 patients (58.8%) had a pathological T stage ≥ 3. Moreover, 21 patients (16.9%) were positive for pelvic lymph node (LN) metastases, 69 patients (55.6%) were positive for extraprostatic extension, 38 patients were positive for seminal vesicle invasion, and 27 patients (21.8%) were positive for surgical margins. The median follow-up and bPFS were 21.57 ± 9.65 and 15.97 ± 8.54 months, respectively. Table 1Clinicopathological parameters of 124 patients in the present studyVariableOverall (N = 124)Age, years (mean ± SD)67.32 ± 5.72PSA at diagnosis, ng/mL (median, range)52.67 (22.89–92.97)GS, n (%) 616 (12.9%) 744 (35.5%) 8–1064 (51.6%)Clinical T stage, n (%) cT225 (20.2%) cT371 (57.3%) cT428 (22.6%)Neoadjuvant treatment, n (%) NHT66 (53.2%) NCHT58 (46.8%)Pathological stage, n (%) pT011 (8.9%) < pT340 (32.3%) ≥pT373 (58.8%)Pelvic LN Negative103 (83.1%) Positive21 (16.9%)EPE Negative55 (44.4%) Positive69 (55.6%)SV Negative86 (69.4%) Positive38 (30.6%)Margins Negative97 (78.2%) Positive27 (21.8%)Median Follow-up, months (mean ± SD)21.57 ± 9.65Median bPFS, months (mean ± SD)15.97 ± 8.54PSA Prostate-specific antigen, GS Gleason score, NHT Neoadjuvant hormonal therapy, NCHT Neoadjuvant chemohormonal therapy, LN Lymph node, EPE Extraprostatic extension, SV Seminal vesicle, bPFS Biochemical progression-free survival Table 1Clinicopathological parameters of 124 patients in the present studyVariableOverall (N = 124)Age, years (mean ± SD)67.32 ± 5.72PSA at diagnosis, ng/mL (median, range)52.67 (22.89–92.97)GS, n (%) 616 (12.9%) 744 (35.5%) 8–1064 (51.6%)Clinical T stage, n (%) cT225 (20.2%) cT371 (57.3%) cT428 (22.6%)Neoadjuvant treatment, n (%) NHT66 (53.2%) NCHT58 (46.8%)Pathological stage, n (%) pT011 (8.9%) < pT340 (32.3%) ≥pT373 (58.8%)Pelvic LN Negative103 (83.1%) Positive21 (16.9%)EPE Negative55 (44.4%) Positive69 (55.6%)SV Negative86 (69.4%) Positive38 (30.6%)Margins Negative97 (78.2%) Positive27 (21.8%)Median Follow-up, months (mean ± SD)21.57 ± 9.65Median bPFS, months (mean ± SD)15.97 ± 8.54PSA Prostate-specific antigen, GS Gleason score, NHT Neoadjuvant hormonal therapy, NCHT Neoadjuvant chemohormonal therapy, LN Lymph node, EPE Extraprostatic extension, SV Seminal vesicle, bPFS Biochemical progression-free survival Clinicopathological parameters of 124 patients in the present study PSA Prostate-specific antigen, GS Gleason score, NHT Neoadjuvant hormonal therapy, NCHT Neoadjuvant chemohormonal therapy, LN Lymph node, EPE Extraprostatic extension, SV Seminal vesicle, bPFS Biochemical progression-free survival A detailed comparison of clinicopathological parameters between the NHT and NCHT groups is presented in Supplemental Table 1. Baseline characteristics including age, PSA at diagnosis, and clinical T stage were comparable between the two groups. However, the biopsy Gleason score was significantly higher in the NCHT group (70.7% with GS 8–10 vs. 34.8% in NHT, P = 0.001). Regarding pathological morphology, the degree of glandular density and diameter reduction also differed significantly between groups (P = 0.007), with more complete responses observed in the NCHT group (13.8% vs. 4.5%). Importantly, the rates of most other morphological features—including IDC‑P, foamy cell infiltration, cytoplasmic vacuolization—were not significantly different between the two treatment groups. As shown in Tables 2, 11 patients (8.9%) had complete reductions in glandular density and diameter, and 71 patients (57.3%) had moderate reductions in glandular density and diameter in RP samples. A review of the RP samples revealed that 60 patients (48.4%) were positive for foamy cell infiltration, 58 patients (46.8%) were positive for cytoplasmic vacuolization, 41 patients (33.1%) were positive for stromal mucin and mucinous breakdown, 53 patients (42.7%) were positive for lymphocytic infiltration, and 74 patients (59.7%) were positive for stromal fiber hyperplasis. Also, IDC-P was observed in 8 patients (6.5%). Notably, the inter-observer agreement for these morphological features was good-to-excellent (κ range: 0.79–0.91), supporting the reliability of our pathological assessment. Table 2Pathological morphology appearance in radical prostatectomy specimens of 124 patients in the present studyVariableOverall (N = 124)Reduced glandular density and diameter Negative42 (33.9%) Moderate71 (57.3%) Complete11 (8.9%)Foamy cell infiltration Negative64 (51.6%) Positive60 (48.4%)Cytoplasmic vacuolization Negative66 (53.2%) Positive58 (46.8%)Stromal mucin and mucinous breakdown Negative83 (66.9%) Positive41 (33.1%)Lymphocytic infiltration Negative71 (57.3%) Positive53 (42.7%)Stromal fiber hyperplasis Negative50 (40.3%) Positive74 (59.7%)Nuclear pyhnosis Negative27 (21.8%) Positive97 (78.2%)IDC-P Absence116 (93.5%) Presence8 (6.5%)IDC-P intraductal carcinoma of the prostate Table 2Pathological morphology appearance in radical prostatectomy specimens of 124 patients in the present studyVariableOverall (N = 124)Reduced glandular density and diameter Negative42 (33.9%) Moderate71 (57.3%) Complete11 (8.9%)Foamy cell infiltration Negative64 (51.6%) Positive60 (48.4%)Cytoplasmic vacuolization Negative66 (53.2%) Positive58 (46.8%)Stromal mucin and mucinous breakdown Negative83 (66.9%) Positive41 (33.1%)Lymphocytic infiltration Negative71 (57.3%) Positive53 (42.7%)Stromal fiber hyperplasis Negative50 (40.3%) Positive74 (59.7%)Nuclear pyhnosis Negative27 (21.8%) Positive97 (78.2%)IDC-P Absence116 (93.5%) Presence8 (6.5%)IDC-P intraductal carcinoma of the prostate Pathological morphology appearance in radical prostatectomy specimens of 124 patients in the present study IDC-P intraductal carcinoma of the prostate Prognostic factors for bPFS Univariate Cox regression revealed that neoadjuvant treatment modality, pathological T stage, pelvic LN metastases, extraprostatic extension, seminal vesicle invasion, positive surgical margins, the degree of reduction in glandular density and diameter, and IDC-P were associated with bPFS (Table 3). After all eight variables were reintegrated into the Cox regression model, multivariate analysis suggested that NHT, pelvic LN metastases, positive surgical margins, and IDC-P in RP specimens were also associated with a worse bPFS (all P < 0.05). Furthermore, the risk of disease progression decreased nearly 70% in patients with moderate reductions in glandular density and diameter, and more than 80% in patients with complete reductions in glandular density and diameter. Table 3Cox regression analysis for associations of clinicopathological parameters with bPFSVariableCategoriesUnivariableMultivariableHR (95%CI) P HR (95%CI) P AgeContinuous1.008 (0.961–1.058)0.74PSA at diagnosisContinuous1.001 (0.997–1.003)0.989GS6 vs. 71.762 (0.661–4.698)0.2576 vs. 8–101.325 (0.508–3.459)0.565Clinical T stagecT2 vs. cT31.011 (0.483–2.113)0.977cT2 vs. cT41.851 (0.833–4.111)0.131Neoadjuvant treatmentNHT vs. NCHT0.403 (0.225–0.721)0.0020.310 (0.157–0.613)0.001Pathological stage< pT3 vs. ≥pT37.450 (2.923–18.988)0.0011.364 (0.226–8.240)0.735Pelvic LNNegative vs. Positive3.237 (1.754–5.974)0.0013.153 (1.490–6.673)0.003EPENegative vs. Positive6.563 (3.129–13.763)0.0013.850 (0.730-20.316)0.112SVNegative vs. Positive2.870 (1.634–5.040)0.0011.606 (0.826–3.123)0.163MarginsNegative vs. Positive2.743 (1.560–4.823)0.0012.997 (1.568–5.731)0.001Reduced glandular density and diameterNegative vs. Moderate0.220 (0.116–0.418)0.0010.296 (0.150–0.585)0.001Negative vs. Complete0.087 (0.022–0.336)0.0010.188 (0.044–0.811)0.025Foamy cell infiltrationNegative vs. Positive0.763 (0.441–1.320)0.334Cytoplasmic vacuolizationNegative vs. Positive0.981 (0.566–1.699)0.944Stromal mucin and mucinous breakdownNegative vs. Positive0.623 (0.337–1.152)0.132Lymphocytic infiltrationNegative vs. Positive0.716 (0.408–1.257)0.245Stromal fiber hyperplasisNegative vs. Positive0.576 (0.331–1.002)0.051Nuclear pyhnosisNegative vs. Positive0.628 (0.328–1.203)0.161IDC-PAbsence vs. Presence4.218 (1.863–9.549)0.0012.938 (1.190–7.256)0.019PSA Prostate-specific antigen, GS Biopsy Gleason score, NHT Neoadjuvant hormonal therapy, NCHT Neoadjuvant chemohormonal therapy, LN Lymph node, EPE Extraprostatic extension, SV Seminal vesicles, bPFS Biochemical progression-free survival, IDC-P intraductal carcinoma of the prostate Table 3Cox regression analysis for associations of clinicopathological parameters with bPFSVariableCategoriesUnivariableMultivariableHR (95%CI) P HR (95%CI) P AgeContinuous1.008 (0.961–1.058)0.74PSA at diagnosisContinuous1.001 (0.997–1.003)0.989GS6 vs. 71.762 (0.661–4.698)0.2576 vs. 8–101.325 (0.508–3.459)0.565Clinical T stagecT2 vs. cT31.011 (0.483–2.113)0.977cT2 vs. cT41.851 (0.833–4.111)0.131Neoadjuvant treatmentNHT vs. NCHT0.403 (0.225–0.721)0.0020.310 (0.157–0.613)0.001Pathological stage< pT3 vs. ≥pT37.450 (2.923–18.988)0.0011.364 (0.226–8.240)0.735Pelvic LNNegative vs. Positive3.237 (1.754–5.974)0.0013.153 (1.490–6.673)0.003EPENegative vs. Positive6.563 (3.129–13.763)0.0013.850 (0.730-20.316)0.112SVNegative vs. Positive2.870 (1.634–5.040)0.0011.606 (0.826–3.123)0.163MarginsNegative vs. Positive2.743 (1.560–4.823)0.0012.997 (1.568–5.731)0.001Reduced glandular density and diameterNegative vs. Moderate0.220 (0.116–0.418)0.0010.296 (0.150–0.585)0.001Negative vs. Complete0.087 (0.022–0.336)0.0010.188 (0.044–0.811)0.025Foamy cell infiltrationNegative vs. Positive0.763 (0.441–1.320)0.334Cytoplasmic vacuolizationNegative vs. Positive0.981 (0.566–1.699)0.944Stromal mucin and mucinous breakdownNegative vs. Positive0.623 (0.337–1.152)0.132Lymphocytic infiltrationNegative vs. Positive0.716 (0.408–1.257)0.245Stromal fiber hyperplasisNegative vs. Positive0.576 (0.331–1.002)0.051Nuclear pyhnosisNegative vs. Positive0.628 (0.328–1.203)0.161IDC-PAbsence vs. Presence4.218 (1.863–9.549)0.0012.938 (1.190–7.256)0.019PSA Prostate-specific antigen, GS Biopsy Gleason score, NHT Neoadjuvant hormonal therapy, NCHT Neoadjuvant chemohormonal therapy, LN Lymph node, EPE Extraprostatic extension, SV Seminal vesicles, bPFS Biochemical progression-free survival, IDC-P intraductal carcinoma of the prostate Cox regression analysis for associations of clinicopathological parameters with bPFS PSA Prostate-specific antigen, GS Biopsy Gleason score, NHT Neoadjuvant hormonal therapy, NCHT Neoadjuvant chemohormonal therapy, LN Lymph node, EPE Extraprostatic extension, SV Seminal vesicles, bPFS Biochemical progression-free survival, IDC-P intraductal carcinoma of the prostate Survival analysis As shown in Fig. 1 A, the median bPFS of patients who underwent NCHT was 38 months, which was significantly longer than that of patients who underwent NHT (38 months vs. 15 months, P = 0.002). The median bPFS of patients with pelvic LN metastases and patients with positive surgical margins was significantly shorter than that of patients without pelvic LN metastases and patients with negative surgical margins (9 months vs. 38 months, P = 0.001; 12 months vs. 41 months, P = 0.001) (Fig. 1B C). Moreover, the median bPFS of patients with IDC-P in RP specimens was only 9.5 months, which was significantly greater than that of patients without IDC-P (9.5 months vs. 38 months, P = 0.001) (Fig. 1D). With respect to the degree of reduction in glandular density and diameter, patients with a negative reduction in glandular density and diameter had a bPFS of only 11 months, which was significantly shorter than that of patients with moderate and complete changes in glandular density and diameter (11 months vs. 27 months; 11 months vs. 38 months, all P < 0.01). In addition, a significant difference in bPFS between patients with moderate and complete glandular density and diameter was also observed (27 months vs. 38 months, P < 0.05) (Fig. 1E). Fig. 1Kaplan‒Meier estimates of biochemical progression-free survival (bPFS) in different subgroups. A Neoadjuvant treatment modality. B Pelvic lymph node (LN) metastases. C Surgical margins. D IDC-P. E Degree of reduction in glandular density and diameter Kaplan‒Meier estimates of biochemical progression-free survival (bPFS) in different subgroups. A Neoadjuvant treatment modality. B Pelvic lymph node (LN) metastases. C Surgical margins. D IDC-P. E Degree of reduction in glandular density and diameter Nomogram construction and internal validation Based on the above results, we selected neoadjuvant treatment modality, the status of pelvic LN metastases, surgical margins, IDC-P, and the degree of reduction in glandular density and diameter as variables to develop a prognostic nomogram (Fig. 2). The C-index value of this nomogram was 0.813 (95% CI: 0.626–0.863), and The time-dependent C-indices were 0.902 (95% CI: 0.825–0.979) at 6 months, 0.882 (95% CI: 0.819–0.944) at 12 months, and 0.951 (95% CI: 0.901–0.991) at 24 months. The calibration curves showed a robust consistency between the predicted and observed bPFS survival probabilities, and DCA demonstrated that the nomogram provided a positive net benefit across a wide range of threshold probabilities compared with treat-all or treat-none strategies at 6 months, 12 months and 24 months (Fig. 3). Fig. 2Nomogram for biochemical progression-free survival (bPFS) in high-risk prostate cancer (PCa) patients Nomogram for biochemical progression-free survival (bPFS) in high-risk prostate cancer (PCa) patients Fig. 3Calibration plots. The consistency of the predicted potentiality and actual values at 6 months (A), 12 months (B) and 24 months (C) is shown. Decision curve analysis for the nomogram. The y-axis represents the net benefit, and the x-axis represents the threshold probability. The nomogram (red line) shows a higher net benefit than the “treat-all” (gray line) or “treat-none” (black line) strategies across a wide range of reasonable threshold probabilities at 6 months (D), 12 months (E) and 24 months (F) Calibration plots. The consistency of the predicted potentiality and actual values at 6 months (A), 12 months (B) and 24 months (C) is shown. Decision curve analysis for the nomogram. The y-axis represents the net benefit, and the x-axis represents the threshold probability. The nomogram (red line) shows a higher net benefit than the “treat-all” (gray line) or “treat-none” (black line) strategies across a wide range of reasonable threshold probabilities at 6 months (D), 12 months (E) and 24 months (F) Discussion The management of high-risk PCa is a challenge for urologists, and neoadjuvant therapy is increasingly applied in clinical practice to reduce the local tumor burden, treat possible micrometastases, and improve patient survival. Owing to the presence of castration-resistant cells in the early stages of disease, the NHT has always failed to significantly improve the outcomes for high-risk PCa patients [14–16], which has led to a renewed interest in adding chemotherapy to androgen deprivation therapy (ADT) in the neoadjuvant setting. Previously, Pan et al., reviewed the data of 177 patients who received neoadjuvant therapy and reported that NCHT resulted in a longer bPFS time after RP compared with the NHT [5]. In a prospective randomized phase III trial, NCHT was also associated with improved overall bPFS compared with RP alone [6]. In this study, we also found that the median bPFS of NCHT patients was significantly greater than that of NHT patients, and Cox regression demonstrated that NCHT was associated with a longer bPFS. In addition, patients with higher biopsy Gleason scores in this present study were more likely to receive NCHT, and the degree of glandular density and diameter reduction also differed significantly between groups, with more complete responses observed in the NCHT group (13.8% vs. 4.5%), consistent with the intensified treatment effect. Thus, we believe that NCHT might be an effective neoadjuvant regimen to improve the outcome of patients with high-risk PCa. However, it is important to note that the NHT regimen used in our study represents traditional maximal androgen blockade, whereas contemporary neoadjuvant trials are increasingly evaluating androgen receptor pathway inhibitors (ARPIs) such as apalutamide, enzalutamide, and darolutamide (e.g., PROTEUS, ARNEO, SUGAR). So, the optimal neoadjuvnt therapy regimens still needed further investigation. Although most adverse events associated with NCHT were well tolerated and Grade 3 or 4 adverse events were relatively rare, we should note that not all high-risk PCa patients could benefit from NCHT, and adding docetaxel in the neoadjuvant setting unavoidably increased the rate of adverse events and imposed additional economic burdens [7]. Pathological morphology parameters, especially morphology alterations induced by neoadjuvant therapy, could serve as the direct predictive factors reflecting treatment efficacy and help urologists make clinical decisions after RP. Most pathological morphology alterations after ADT with or without chemotherapy included tissue fibrosis, reduced tumor cell density, stromal fiber hyperplasis, stromal mucin deposition, vacuolated cytoplasm, nuclear pyknosis, and stromal lymphocytic infiltration [17]. In the present study, we focused on a cohort of high-risk PCa patients who were treated with neoadjuvant therapy followed by RP. Interestingly, in addition to clinical prognostic factors, we found that the degree of reduction in glandular density and diameter and the presence of IDC-P were prognostic factors for oncological outcomes. However, other pathological morphology alterations, such as cytoplasmic vacuolization, stromal mucin and mucinous breakdown, and nuclear pyhnosis, were not associated with bFPS. The multifocality, morphological heterogeneity and high stromal cell density were the main reasons [18]. Currently, the pCR and minimum residual disease (MRD) of a primary tumor to neoadjuvant therapy have been suggested to be important prognostic parameters for patient survival [19]. In a phase II clinical trial conducted by Chen et al., seventy-five patients with high-risk PCa were included, and 25 patients (33.3%) achieved pCR or MRD on the final pathology after neoadjuvant therapy. Moreover, they reported that pCR or MRD on final pathology were the independent variables for the prediction of bPFS (HR: 0.12, 95% CI: 0.02–0.98) [20]. A subsequent pooled analysis conducted by Mckay et al., also revealed that 8% of patients (2/25) with pCR or MRD experienced non-BCR at a median follow-up period of 3.6 years, which was significantly better than that of patients without pCR or MRD [21]. In our study, 11 patients had a complete reduction in glandular density and diameter, and the median bPFS was 38 months for these patients, which was significantly longer than that of patients without a complete reduction in glandular density and diameter. Thus, we deduced that pCR or MRD following neoadjuvant therapy was associated with a favorable impact on the BCR. In clinical practice, pathologists might encounter several problems in assessing pCR and MRD in whole-mount RP specimens, as PCa is multifocal in ~ 60% of cases and has a high heterogeneity [22–24]. In addition, the pCR rates of PCa patients treated with neoadjuvant therapy were relatively low, and most high-risk PCa patients showed the varying degrees of reduction in tumor density and diameter. Thus, the histological grading classification system proposed by Wang et al. might be a promising tool to assist urologists in assessing the pathological response to neoadjuvant therapy concisely [7, 11]. To the best of our knowledge, this is the first study to explore the prognostic role of different degrees of reduction in tumor density and diameter. In the present study, we classified patients into three groups on the basis of the degree of reduction in glandular density and diameter and reported that patients with a ≤ 30% reduction in glandular density and diameter had the poorest outcome. Thus, for these patients, salvage treatment, including radiotherapy or hormonal therapy, should be performed to improve patient prognosis. However, for patients with > 30% reduction in glandular density and diameter but not pCR or MRD, the decision to perform salvage or adjuvant treatment should depend on whether patients have other high-risk factors for poor outcomes, such as IDC-P, pelvic LN metastases, seminal vesicle invasion, and positive surgical margins. IDC-P is defined as the retrograde spread of adenocarcinoma within preexisting prostatic ducts and is a well-established adverse pathological factor associated with high-risk and more advanced disease [25]. A retrospective study including 145 locally advanced high-risk PCa patients revealed that th e presence of IDC-P in RP specimens was associated with the poorest prognosis in terms of disease-free survival, cancer-specific survival, and overall survival (OS) [26]. Furthermore, in our study, IDC-P was observed in 8 patients (6.5%), and the median bPFS was only 9.5 months, which was significantly poorer than that of patients without IDC-P (9.5 months vs. 38 months, P = 0.001). Of note, most IDC-P lesions are detectable in pretreatment biopsies, and their persistence in only 8 patients (6.5%) likely reflects treatment-resistant disease. This interpretation is consistent with our findings, as these 8 patients had a markedly shorter median bPFS, underscoring the adverse prognostic impact of residual IDC-P after neoadjuvant therapy. Future studies incorporating paired biopsy-prostatectomy comparisons of IDC-P status would better clarify the dynamic changes induced by neoadjuvant therapy. Moreover, several studies have reported that the prevalence of homologous DNA recombination repair (HRR) mutations and PTEN loss is greater among patients with IDC-P than among those without IDC-P [27]. These results indicated that patients with IDC-P had more aggressive characteristics and poorer prognoses. Thus, more intensive adjuvant treatment and germline genetic testing might be more suitable for these patients. In our study, we also found that pelvic LN metastasis and positive surgical margins were independent prognostic factors for bPFS, which was consistent with the findings of previously published studies [12, 28]. Thus, we next selected neoadjuvant treatment modality, the status of pelvic LN metastases, surgical margins, IDC-P, and the degree of reduction in glandular density and diameter as variables to develop a prognostic nomogram and found that the nomogram based on pathological morphology parameters had considerable discriminative and calibrating ability to help urologists identify patients at high risk of biochemical recurrence. Importantly, the primary intended role of this nomogram is to inform postoperative clinical decision-making by identifying patients at highest risk of early biochemical recurrence based on residual pathological features after neoadjuvant therapy, rather than for preoperative treatment selection. We acknowledge that salvage treatment initiated at the time of PSA rise may not compromise overall survival compared with immediate adjuvant treatment; however, high-risk patients with multiple adverse features may still benefit from closer monitoring or earlier intervention, and the nomogram can help identify this subgroup. Despite these positive results, there are several limitations. First, the retrospective nature of the present study and the small sample size from a single center might cause selection bias. The non-randomized allocation of NCHT, with higher biopsy Gleason scores in that group, is a particular concern; we have therefore avoided causal language and emphasized that residual confounding cannot be fully excluded. Second, we were unable to perform external validation due to the limited sample size, which may lead to overfitting. We have initiated a multicenter collaboration to collect additional cases with standardized pathological review. Third, owing to the short follow-up time, the impact of these pathological morphology parameters on OS was not analyzed, and a longer follow-up duration might produce more positive findings. Fourth, molecular markers (e.g., PTEN loss, HRR mutations) and advanced imaging findings (e.g., mpMRI) were not systematically available; integrating these modalities could further refine prognostic accuracy and align the model with multimodal precision medicine. Fifth, the NHT regimen used (ADT plus bicalutamide) in our study does not represent contemporary ARPI-based neoadjuvant therapy and the morphological changes induced by ARPIs may differ from those observed with conventional NHT. Finally, while we have provided the regression equation and point-scoring algorithm, a static nomogram is less practical for routine clinical use; we plan to develop a publicly available web-based or spreadsheet-based calculator as part of our ongoing validation project. Hence, further prospective multicenter studies are needed to support our findings. Conclusions In the present study, we found that pelvic LN metastasis and positive surgical margins were independent prognostic factors for high-risk PCa patients and that NCHT might significantly improve bPFS compared with NHT. The degree of reduction in glandular density and diameter and the presence of IDC-P in RP patients were associated with worse bPFS. Moreover, a nomogram based on the above factors may help urologists identify patients who might benefit from salvage or adjuvant treatment after RP. Supplementary Information Supplementary Material 1. Supplemental Table 1 The comparion of Clinicopathological parameters between NHT and NCHT. Supplementary Material 1. Supplemental Table 1 The comparion of Clinicopathological parameters between NHT and NCHT.