Development and Internal Validation of a Prediction Model For Biochemical Recurrence Following Radical Prostatectomy

Summary

This retrospective study developed and internally validated a predictive model for biochemical recurrence after radical prostatectomy in 240 patients. Pathological stage, lymph node metastasis, positron emission tomography–positive lesions, and SUV_max were independent predictors. The model demonstrated strong discrimination, calibration, and clinical utility for postoperative risk stratification.

Abstract

This study aimed to develop and validate a predictive model for biochemical recurrence (BCR) after radical prostatectomy in patients with prostate cancer, incorporating clinical, pathological, inflammatory, and 18F-PSMA-1007 positron emission tomography/computed tomography (PET/CT) imaging parameters. This retrospective study included 240 patients with histopathologically confirmed prostate adenocarcinoma who underwent radical prostatectomy between June 2022 and July 2025. BCR was defined as a postoperative serum prostate-specific antigen (≥0.2 ng/mL). Preoperative clinical variables, systemic immune-inflammation index (SII), PET/CT lesion status, and maximum standardized uptake value (SUV_max) were collected along with postoperative pathological staging. Patients were divided into BCR (n = 64) and non-BCR (n = 176) groups. Univariate and multivariate logistic regression analyses identified independent predictors of BCR. A multivariable predictive model was developed and internally validated using a 70/30 training–validation split. Model performance was evaluated using receiver operating characteristic curves, area under the curve (AUC), calibration, and decision curve analysis. Patients with BCR showed more advanced pathological stage (pT3–4: 96.9% vs. 52.3%, p < 0.001), higher lymph node metastasis rates (59.4% vs. 29.5%, p < 0.001), higher SII (682.45 ± 118.23 vs. 637.08 ± 92.40, p = 0.002), and higher SUV_max values (6.59 ± 1.34 vs. 4.92 ± 1.49, p < 0.001). Multivariate analysis identified pathological stage (OR = 36.814, p < 0.001), lymph node status (OR = 7.286, p < 0.001), SUV_max (OR = 2.732, p < 0.001), and PET-positive lesions (OR = 27.929, p < 0.001) as independent predictors of BCR. The combined predictive model achieved excellent discrimination in training (AUC = 0.952) and validation cohorts (AUC = 0.927). Calibration curves showed agreement between predicted and observed outcomes, and decision curve analysis demonstrated superior net clinical benefit compared with treat-all and treat-none strategies. The integrated model demonstrates excellent discrimination and clinical utility, supporting individualized postoperative risk stratification.

Introduction

Prostate cancer remains one of the most common malignancies affecting men worldwide, and radical prostatectomy represents a key treatment option for patients with localized disease^1,2. Despite improvements in surgical techniques and patient selection, biochemical recurrence (BCR) occurs in approximately 20%–40% of cases, indicating persistent or recurrent disease and increased risks of metastasis and mortality^2,3. Accurate prediction of BCR after prostatectomy remains challenging, necessitating more effective prognostic tools that combine clinical, immunological, and advanced imaging parameters to guide postoperative surveillance and adjuvant treatments.

Inflammation and immune responses play essential roles in tumor progression and recurrence. The systemic immune-inflammation index (SII), a novel inflammatory biomarker calculated from peripheral neutrophil, lymphocyte, and platelet counts, has demonstrated prognostic significance in various cancers^4,5,6. Additionally, 18F-PSMA-1007 positron emission tomography/computed tomography (PET/CT), a prostate-specific membrane antigen-based imaging modality, has shown promise in detecting recurrent disease earlier than conventional imaging methods^7,8, thereby potentially enhancing postoperative prognostication. This integrative approach extends beyond conventional clinicopathological models by incorporating both systemic inflammatory biomarkers and advanced molecular imaging parameters. Unlike established risk stratification tools such as the CAPRA-S score or D’Amico classification, which rely primarily on baseline clinicopathological parameters, the proposed approach integrates dynamic systemic inflammatory indices with functional molecular imaging findings.

However, the clinical value of integrating inflammatory markers such as SII, pathological tumor staging, and advanced imaging modalities, including 18F-PSMA-1007 PET/CT, in predicting prostate cancer recurrence remains unclear. Therefore, this study aimed to construct and internally validate a comprehensive predictive model for BCR following radical prostatectomy by combining these factors to improve risk stratification, optimize postoperative patient management, and facilitate early intervention strategies. Clinically, this model is intended to assist clinicians in tailoring personalized surveillance protocols and identifying high-risk candidates who may benefit from early adjuvant interventions or intensified follow-up.

Protocol

This study protocol was reviewed and approved by the Ethics Committee of First Affiliated Hospital of Xinjiang Medical University in accordance with the Declaration of Helsinki (Approval No.: 230608-7). Informed consent was waived for this retrospective study due to the exclusive use of de-identified patient data, which posed no potential harm or impact on patient care.

Study Design and Patient Selection
This retrospective study included 240 patients with histopathologically confirmed prostate adenocarcinoma who underwent radical prostatectomy between June 2022 and July 2025. Patients were identified through a systematic search of the electronic medical record system at the First Affiliated Hospital of Xinjiang Medical University. A consecutive sampling method was employed to minimize selection bias. Patients were categorized into a BCR group (n = 64) and a non-BCR group (n = 176) based on postoperative serum prostate-specific antigen (PSA) levels, with BCR defined as a serum PSA level of ≥0.2 ng/mL confirmed by two consecutive measurements. The inclusion criteria were as follows: (1) histopathologically confirmed prostate adenocarcinoma; (2) radical prostatectomy performed; and (3) availability of complete clinical, pathological, and imaging data. These criteria were independently applied by two experienced researchers (Yubin Li and Qizhou Zhang). Any disagreements regarding patient eligibility were resolved through discussion or consultation with a third senior investigator. The exclusion criteria were: (1) Gleason score of <7; (2) receipt of preoperative hormonal therapy, chemotherapy, or radiotherapy, or incomplete medical records; and (3) receipt of neoadjuvant therapy, including androgen deprivation therapy, chemotherapy, or radiotherapy, prior to PET/CT or surgery. Exclusion criteria were applied sequentially. Patients who received neoadjuvant or preoperative therapies were excluded first, followed by patients with Gleason scores of <7 and finally those with missing key data points. For patients meeting multiple exclusion criteria, the primary reason for exclusion was documented. Patients with missing or incomplete clinical, pathological, or imaging data were excluded from the final analysis to ensure the robustness of the predictive model. No data imputation methods were applied owing to the retrospective study design.

Clinical and Pathological Data Collection
Baseline clinical data, including patient age, Gleason score, preoperative serum PSA levels, pathological tumor stage, lymph node involvement, and postoperative follow-up details, were retrospectively extracted from electronic medical records. Data extraction was performed through manual review of the electronic medical records by two independent investigators using a standardized data collection form to ensure consistency. Any discrepancies between the two investigators were resolved through consensus or consultation with a third senior investigator who reviewed the original medical records. Pathological tumor staging was classified according to the 8th edition of the American Joint Committee on Cancer TNM staging system. Staging data were obtained directly from the original postoperative pathology reports issued by the hospital pathology department. To ensure accuracy, all pathological reports and representative slides were secondarily reviewed by an experienced uropathologist who was blinded to the imaging findings and clinical outcomes.

Systemic Immune-Inflammation Index (SII)
Preoperative peripheral blood samples were obtained from all patients. Blood samples were collected via venipuncture within 7 days prior to surgery. All patients were required to fast for at least 8 h before sample collection to ensure baseline stability. Laboratory parameters, including neutrophil, lymphocyte, and platelet counts, were processed and analyzed in a single centralized clinical laboratory at our institution using standardized automated hematology analyzers. The SII was calculated using the following equation:

The optimal cutoff value for SII was determined using r  eceiver operating characteristic (ROC) analysis and Youden’s index. The cutoff value was data-driven and derived from the entire dataset using the “pROC” package in R statistical analysis software (version 4.2.2). ROC analysis was performed to maximize the combined sensitivity and specificity, and the resulting threshold was used to categorize patients into high-SII and low-SII groups.

18F-PSMA-1007 PET/CT Imaging Protocol and Analysis
All patients underwent preoperative imaging with 18F-PSMA-1007 PET/CT within four weeks prior to surgery. The timing of imaging relative to surgery was standardized, and patients with an imaging-to-surgery interval exceeding 30 days were excluded to ensure consistency between imaging findings and pathological status. Imaging was performed following intravenous administration of 4.0 MBq/kg of 18F-PSMA-1007. The radiotracer was synthesized onsite using a cyclotron and an automated synthesis module. Quality control procedures were performed before each injection to ensure a radiochemical purity >95%. Radiotracer dosing was standardized according to body weight, and no major dosing deviations were recorded. PET/CT scans were acquired from the skull base to the mid-thigh region approximately 60 min post-injection using a dedicated PET/CT system. Scans were performed using a Discovery VCT PET/CT scanner. CT acquisition parameters included 120 kV, automatic tube current modulation (150–200 mA), and a slice thickness of 3.0 mm. PET images were reconstructed using an ordered subset expectation maximization algorithm with a 192 × 192 or 256 × 256 matrix size, incorporating time-of-flight and point spread function corrections.

PET/CT images were independently analyzed by two experienced nuclear medicine physicians who were blinded to the clinical and pathological outcomes. Interobserver agreement was assessed using Cohen’s kappa coefficient. Any discrepancies in lesion localization or categorization were resolved through consensus review or consultation with a third senior nuclear medicine physician. Visual analysis categorized lesions as positive (local recurrence, pelvic lymph nodes, or distant metastases) or negative. Lesions were classified as positive according to the Prostate Cancer Molecular Imaging Standardized Evaluation criteria. Specifically, focal tracer uptake exceeding the surrounding background activity and not attributable to physiological distribution (e.g., ureter or bladder activity) was considered indicative of recurrence or metastasis. Additionally, the maximum standardized uptake value (SUV_max) of suspected lesions was recorded. Regions of interest were manually defined using a three-dimensional spherical volume encompassing the focal uptake on PET images. In patients with multiple lesions, the lesion with the highest uptake was selected for SUV_max measurement. Quantitative image analyses were performed using Advantage Workstation software (version 4.7).

Follow-up and Biochemical Recurrence Definition
Follow-up included regular clinical visits every three to six months postoperatively, with serum PSA measurements. The total follow-up duration for the entire cohort ranged from 8 to 19 months. Follow-up intervals were generally consistent across all patients. For patients who missed scheduled visits, follow-up information was updated through telephone interviews or review of the most recent outpatient records available at our institution. Patients who were lost to follow-up before the minimum 6-month threshold or before documentation of a BCR event were excluded from the final analysis to maintain data integrity. BCR was defined as a postoperative serum PSA level of ≥0.2 ng/mL confirmed by two consecutive measurements. The confirmatory PSA measurements were typically obtained 4–8 weeks apart. This definition of BCR was applied consistently throughout the manuscript and served as the primary endpoint of the predictive model.

Subgroup Analysis
To assess the robustness of the predictive model and reduce potential bias associated with short follow-up durations, a subgroup analysis was performed in patients with follow-up durations >12 months. A total of 133 patients met this criterion, whereas 107 patients with follow-up durations of ≤12 months were excluded from the subgroup analysis. Multivariate logistic regression analysis was subsequently performed in this subgroup using the same statistical methodology applied in the primary analysis.

Statistical Analysis
Continuous variables were presented as median and range, whereas categorical variables were expressed as frequencies and percentages. Data distribution was assessed using the Shapiro–Wilk test. Continuous variables with a normal distribution were presented as mean ± standard deviation (SD) and compared using Student’s t-test. Variables with a non-normal distribution were expressed as median and range and compared using the Mann–Whitney U test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Prior to analysis, assumptions for the chi-square test were verified, including confirmation that no more than 20% of cells had expected frequencies <5. When these assumptions were violated, Fisher’s exact test was applied to ensure statistical validity.

A multivariate logistic regression model was developed to predict BCR, incorporating clinically relevant factors such as SII, PET/CT SUV_max, imaging positivity, and pathological tumor stage. Variables with p-values <0.05 in univariate analysis, together with variables considered clinically relevant based on prior literature, were entered into the multivariate model. Multicollinearity among predictors was assessed using the variance inflation factor, with values of <5 considered acceptable.

Internal validation of the predictive model was performed using a 70/30 train–test split method. The dataset was randomly divided into training (70%) and testing (30%) cohorts using stratified sampling to preserve the BCR event distribution across both cohorts. A fixed random seed was applied to ensure reproducibility. Model performance was evaluated using ROC curves and corresponding area under the curve (AUC) values. Calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test. Decision curve analysis was additionally performed to evaluate the clinical net benefit of the integrative model.

Statistical analyses were performed using SPSS software (version 26.0) and R statistical software (version 4.2.2). The statistical analysis software packages “pROC” and “rms” were used for ROC and calibration analyses, respectively. Statistical significance was defined as p < 0.05. As the primary outcome was binary BCR status during follow-up, model performance was evaluated using ROC curves and AUC metrics. Time-dependent ROC analysis was not performed because recurrence time points were not uniformly recorded across the cohort. Prior to analysis, data were screened for outliers, and no extreme values requiring transformation were identified. Continuous variables were analyzed in their original form to preserve clinical interpretability.

Results

Baseline Characteristics
Patients with BCR exhibited significantly more advanced pathological tumor stages (pT3–4: 96.9% vs. 52.3%, p < 0.001), higher rates of lymph node metastasis (59.4% vs. 29.5%, p < 0.001), elevated SII values (682.45 ± 118.23 vs. 637.08 ± 92.40, p = 0.002), and higher maximum standardized uptake values (SUV_max) on PET/CT imaging (6.59 ± 1.34 vs. 4.92 ± 1.49, p < 0.001) compared with patients without BCR. PET-positive lesions were also significantly more frequent in the BCR group than in the non-BCR group (73.4% vs. 14.8%, p < 0.001). In contrast, age and Gleason score distributions did not differ significantly between the two groups (Table 1).

Characteristics	BCR recurrence group (n = 64)	Non-recurrence group (n = 176)	Statistical test	P value
Age (years), median (range)	68 (56–78)	67 (55–79)	U = 5034	0.208
Preoperative PSA (ng/mL), mean ± SD	14.24 ± 6.253	16.20 ± 5.502	t = 2.216	0.029
Gleason score, n (%)			χ² = 6.346	0.175
7 (3+4)	4 (6.3)	30 (17.0)
7 (4+3)	15 (23.4)	38 (21.6)
8 (4+4)	19 (29.7)	37 (21.0)
9 (4+5)	10 (15.6)	35 (19.9)
10 (5+5)	16 (25.0)	36 (20.5)
Pathological tumor stage, n (%)			χ² = 40.607	<0.001
pT2	2 (3.1)	84 (47.7)
pT3–4	62 (96.9)	92 (52.3)
Lymph node status, n (%)			χ² = 17.818	<0.001
Positive	38 (59.4)	52 (29.5)
Negative	26 (40.6)	124 (70.5)
SII	682.45 ± 118.23	637.08 ± 92.40	t = 3.112	0.002
PET/CT SUVmax	6.59 ± 1.34	4.92 ± 1.49	t = 8.303	<0.001
PET-positive lesions, n (%)			χ² = 76.317	<0.001
Positive	47 (73.4)	26 (14.8)
Negative	17 (26.6)	150 (85.2)
Follow-up time (months), median (range)	11 (10–19)	13 (8–19)	U = 5422.5	0.66

Table 1: Baseline clinical, pathological, inflammatory, and imaging characteristics of patients with and without BCR. Continuous variables are presented as mean ± standard deviation (SD) or median (range), as appropriate, whereas categorical variables are presented as number (%). Group comparisons were performed using Student’s t-test, Mann–Whitney U test, or chi-square (χ²) test, as appropriate. Abbreviations: BCR, biochemical recurrence; PSA, prostate-specific antigen; SII, systemic immune-inflammation index; SUV_max, maximum standardized uptake value; PET, positron emission tomography.

Univariate and Multivariate Logistic Regression Analysis of Biochemical Recurrence
Table 2 summarizes the results of the univariate and multivariate logistic regression analyses evaluating factors associated with BCR following radical prostatectomy.

Variable	Univariate OR (95% CI)	P value	Multivariate OR (95% CI)	P value
Age (years)	1.025 (0.985–1.067)	0.216	–	–
Preoperative PSA (ng/mL)	0.941 (0.894–0.991)	0.021	0.973 (0.898–1.054)	0.498
Gleason score	1.145 (0.925–1.417)	0.214	–	–
Pathological tumor stage	28.304 (6.714–119.322)	<0.001	36.814 (5.930–228.533)	<0.001
Lymph node status	3.485 (1.923–6.317)	<0.001	7.286 (2.264–23.445)	<0.001
SII	1.005 (1.002–1.008)	0.003	1.001 (0.996–1.006)	0.803
PET/CT SUVmax	2.302 (1.773–2.989)	<0.001	2.732 (1.768–4.221)	<0.001
PET-positive lesions	15.950 (7.972–31.914)	<0.001	27.929 (8.662–90.313)	<0.001

Table 2: Univariate and multivariate logistic regression analyses of factors associated with BCR. ORs and corresponding 95% CIs are presented for clinical, pathological, inflammatory, and imaging variables associated with BCR following radical prostatectomy. Variables included in the multivariate logistic regression model were selected based on statistical significance in univariate analysis and/or established clinical relevance. Abbreviations: BCR, biochemical recurrence; OR, odds ratio; CI, confidence interval; PSA, prostate-specific antigen; SII, systemic immune-inflammation index; SUVmax, maximum standardized uptake value; PET, positron emission tomography.

In the univariate logistic regression analysis, several variables were significantly associated with BCR. Higher preoperative PSA levels were associated with a lower likelihood of recurrence (OR = 0.941, 95% CI: 0.894–0.991, p = 0.021). Advanced pathological tumor stage (pT3–4 vs. pT2) demonstrated a strong association with BCR (OR = 28.304, 95% CI: 6.714–119.322, p < 0.001). Lymph node positivity was also significantly associated with recurrence risk (OR = 3.485, 95% CI: 1.923–6.317, p < 0.001). Among the biomarker-related variables, elevated SII values (OR = 1.005, 95% CI: 1.002–1.008, p = 0.003), higher PET/CT SUV_max values (OR = 2.302, 95% CI: 1.773–2.989, p < 0.001), and the presence of PET-positive lesions (OR = 15.950, 95% CI: 7.972–31.914, p < 0.001) were all significantly associated with an increased risk of BCR. In contrast, age and Gleason score were not significantly associated with recurrence.

In the multivariate logistic regression model, four variables remained independent predictors of BCR. Advanced pathological tumor stage remained strongly associated with recurrence risk (OR = 36.814, 95% CI: 5.930–228.533, p < 0.001), as did lymph node metastasis (OR = 7.286, 95% CI: 2.264–23.445, p < 0.001). PET/CT SUV_max also remained independently associated with BCR (OR = 2.732, 95% CI: 1.768–4.221, p < 0.001). PET-positive lesions demonstrated the strongest independent association with recurrence (OR = 27.929, 95% CI: 8.662–90.313, p < 0.001). Preoperative PSA levels and SII values were not statistically significant in the multivariate analysis.

Model Performance and Validation
The combined predictive model demonstrated strong discriminatory performance for the prediction of BCR. In the training cohort, the model achieved an AUC of 0.952 (95% CI: 0.915–0.990), outperforming all individual predictors. Among the single-variable predictors, PET/CT SUV_max (AUC = 0.793, 95% CI: 0.721–0.864) and PET-positive lesions (AUC = 0.790, 95% CI: 0.715–0.864) demonstrated the highest discriminatory performance. In comparison, pathological tumor stage showed moderate predictive performance (AUC = 0.739), whereas lymph node status and SII demonstrated lower predictive accuracy.

In the validation cohort, the combined model maintained excellent discriminative ability, with an AUC of 0.927 (95% CI: 0.861–0.994), confirming the robustness and stability of the predictive model. The relative performance ranking of the individual predictors remained consistent, with PET/CT SUV_max and PET-positive lesions again demonstrating the strongest discriminatory performance.

Calibration plots in both cohorts demonstrated agreement between predicted and observed probabilities of BCR; however, deviations from ideal calibration were observed. The training cohort yielded a calibration intercept of –3.80 and a slope of 7.34, whereas the validation cohort demonstrated an intercept of –3.36 and a slope of 6.37. Although the calibration curves generally followed the reference line, these slope values suggest potential model overfitting or calibration instability and should therefore be interpreted with caution (Figure 1).

Figure 1. Performance evaluation of the combined predictive model in training and validation cohorts. (A) ROC curves comparing the combined model and individual predictors in the training cohort. (B) ROC curves comparing the combined model and individual predictors in the validation cohort. (C) Calibration curve of the combined model in the training cohort, showing agreement between predicted and observed probabilities of BCR. (D) Calibration curve of the combined model in the validation cohort. The diagonal dashed line represents ideal calibration, and the solid curve represents the model-predicted probabilities. Abbreviations: ROC, receiver operating characteristic; AUC, area under the curve; SII, systemic immune-inflammation index; SUV_max, maximum standardized uptake value; LN_pos, lymph node positivity; PET_pos, positron emission tomography-positive lesions; BCR, biochemical recurrence. Please click here to view a larger version of this figure.

Decision curve analysis further supported the clinical utility of the combined predictive model. In the training cohort, the model achieved a maximum net benefit of 0.246 and consistently outperformed both the treat-all and treat-none strategies across a clinically relevant threshold probability range of 0–0.50, indicating potential usefulness for individualized postoperative risk stratification (Figure 2).

Figure 2. Decision curve analysis of the combined predictive model. Decision curve analysis showing the net clinical benefit of the combined model across a range of threshold probabilities. The solid line represents the combined model, the dashed line represents the treat-all strategy, and the dotted line represents the treat-none strategy. The model demonstrates higher net benefit across a broad range of threshold probabilities, indicating potential clinical utility for individualized postoperative risk stratification. Please click here to view a larger version of this figure.

Subgroup Analysis (Follow-up > 12 Months)
A total of 133 patients were included in the subgroup analysis after excluding 107 patients with follow-up durations of ≤12 months. Among patients with follow-up durations >12 months (n = 133), multivariate logistic regression analysis demonstrated that pathological tumor stage (OR = 36.814, 95% CI: 5.930–228.533, p < 0.001), lymph node status (OR = 22.064, 95% CI: 3.275–288.367, p < 0.001), PET/CT SUV_max (OR = 10.245, 95% CI: 2.905–72.867, p = 0.004), and PET-positive lesions (OR = 51.458, 95% CI: 8.940–530.327, p < 0.001) remained significantly associated with BCR. In contrast, SII was not independently associated with BCR in the subgroup analysis (OR = 1.004, 95% CI: 0.997–1.012, p = 0.255) (Table 3).

Parameters	Odds ratio (95% CI)	P value
Pathological tumor stage	36.814 (5.930–228.533)	<0.001
Lymph node status	22.064 (3.275–288.367)	<0.001
SII	1.004 (0.997–1.012)	0.255
PET/CT SUVmax	10.245 (2.905–72.867)	0.004
PET-positive lesions	51.458 (8.940–530.327)	<0.001

Table 3: Multivariate logistic regression analysis of factors associated with BCR in patients with follow-up durations of >12 months. Multivariate logistic regression analysis was performed in a subgroup of patients with follow-up durations >12 months (n = 133) to evaluate the robustness of predictors associated with BCR following radical prostatectomy. Odds ratios (ORs) with corresponding 95% confidence intervals (CIs) are presented for all variables included in the multivariate model. Abbreviations: BCR, biochemical recurrence; OR, odds ratio; CI, confidence interval; SII, systemic immune-inflammation index; SUV_max, maximum standardized uptake value; PET, positron emission tomography.

Data Availability:
All data generated or analyzed during this study are included in this published article and its supplementary information file (Supplementary Table 1).

Supplementary Table 1: De-identified patient-level dataset used for predictive model development and validation. The supplementary table contains de-identified patient-level clinical, pathological, inflammatory, imaging, follow-up, and outcome variables used for model development and validation. Variables include SII, PET/CT SUV_max, PET/CT lesion positivity status, pathological tumor stage, biochemical recurrence (BCR) status, age, Gleason score, preoperative PSA level, lymph node status, follow-up duration, and time to BCR. All patient identifiers were removed prior to analysis to ensure confidentiality and compliance with institutional ethical standards. Abbreviations: SII, systemic immune-inflammation index; PET/CT, positron emission tomography/computed tomography; SUV_max, maximum standardized uptake value; BCR, biochemical recurrence; PSA, prostate-specific antigen. Please click here to download this file.

Discussion

In this study, we successfully developed a comprehensive predictive model for BCR following radical prostatectomy in prostate cancer patients, integrating the SII, pathological tumor stage, lymph node status, and preoperative 18F-PSMA-1007 PET/CT imaging findings. The model demonstrated excellent discriminative performance, with an AUC of 0.95, underscoring the benefit of combining multiple biological and imaging markers for precise risk stratification. Pathological tumor stage emerged as a powerful and independent predictor of BCR. Patients with advanced-stage tumors (pT3b and pT4) exhibited significantly higher recurrence rates, consistent with established literature that recognizes tumor staging as a critical determinant of prognosis^1,2. Higher tumor stage correlates with extracapsular extension and seminal vesicle invasion, both of which have been linked to poor oncological outcomes³. This reinforces the need for diligent postoperative surveillance and consideration of early adjuvant therapies in patients with high-risk pathological features⁹.

The role of preoperative 18F-PSMA-1007 PET/CT imaging in our predictive model is particularly noteworthy. Patients with higher SUV_max values and positive PET/CT scans had significantly increased risks of BCR. These findings align with prior research demonstrating the superior sensitivity and specificity of prostate-specific membrane antigen (PSMA) PET imaging for detecting locoregional and distant recurrences compared to conventional imaging modalities^10,11. A meta-analysis by Alberts et al. further supports the use of PSMA PET tracers, indicating that higher SUV_max values are strongly associated with aggressive disease biology⁷. Notably, the SUV_max threshold in our study (5.46) corresponded closely to those reported in prior multicenter trials, reinforcing its external validity. Additionally, emerging studies have demonstrated the utility of PSMA PET/CT in monitoring response to neoadjuvant therapy in patients with prostate cancer^12,13. However, our cohort was limited to patients who had not undergone any preoperative systemic treatment, allowing us to assess the native prognostic value of PET/CT and inflammatory biomarkers prior to therapeutic intervention. Although 18F-PSMA-1007 PET/CT offers superior imaging sensitivity, it is not yet widely accessible in all healthcare systems, particularly in low-resource settings. To explore the potential adaptability of our model, we conducted an exploratory analysis using a simplified model that incorporated only the SII and pathological tumor stage. While the predictive performance (AUC) of this reduced model was lower than that of the full model, it retained moderate discriminatory power. This finding suggests that a simplified approach may still aid in basic postoperative risk stratification where PSMA PET is unavailable. Future research may focus on optimizing such simplified models or integrating other widely accessible biomarkers for broader clinical utility.

Systemic inflammatory markers, such as the SII, reflect the interplay between host immunity and tumor-promoting inflammation. Elevated SII levels were associated with higher recurrence risk in our cohort, consistent with previous meta-analyses highlighting its prognostic value across genitourinary malignancies^4,6. Although SII did not reach statistical significance in multivariate analysis, it contributed incremental value to the combined model, underscoring its potential role as an adjunctive biomarker. Notably, Li et al. and Lolli et al. previously demonstrated the association between elevated SII and poor survival in bladder and metastatic prostate cancers, respectively⁵. Moreover, we selected SII over other hematologic markers such as the neutrophil-to-lymphocyte ratio (NLR) or platelet-to-lymphocyte ratio (PLR) because SII incorporates three immune components—neutrophils, lymphocytes, and platelets—providing a broader view of the host’s inflammatory and immune response. Several comparative studies and meta-analyses have demonstrated that SII has superior prognostic value over NLR and PLR in various solid tumors, including prostate cancer^6,14,15. Furthermore, SII has shown greater reproducibility and stability across patient populations, making it a promising candidate for risk stratification in clinical settings^16,17.

Although SII demonstrated a significant association with BCR in univariate analysis, it did not remain independently significant in multivariate modeling. To assess whether this was due to collinearity or non-linear effects, we conducted post hoc exploratory analyses using dichotomized SII values, z-score normalization, and interaction terms with PET/CT findings and tumor stage. These approaches did not yield statistically significant improvements, suggesting that while SII reflects systemic inflammation, its incremental prognostic value may be limited in the presence of stronger imaging and pathological predictors. Future studies with larger sample sizes may better delineate the contextual role of inflammatory biomarkers in multimodal risk models. Integrating multiple variables into a single predictive framework enhances prognostic accuracy compared to individual factors alone. Recent advances in oncological research emphasize the superiority of multimodal models, incorporating clinical, imaging, and molecular parameters, to personalize patient care. Our integrated model significantly outperformed the predictive accuracy of individual predictors, supporting this paradigm shift toward holistic risk assessment. Similar findings have been reported in models combining PSMA–PET findings with genomic classifiers¹⁸, which achieved AUC values exceeding 0.90 in predicting early BCR. While traditional clinicopathological factors such as preoperative PSA levels, Gleason score, surgical margin status, and lymph node involvement are well-established prognostic indicators for BCR^19,20, our study intentionally focused on a novel integrative model combining the SII, 18F-PSMA-1007 PET/CT imaging features, and pathological tumor stage. This approach aimed to evaluate the predictive utility of emerging biological and imaging biomarkers beyond conventional variables. Surgical margin status was excluded because of inconsistent documentation in the patient records. Although lymph node status was included in the final multivariate model, potential multicollinearity with pathological tumor stage was carefully assessed during model development. Nonetheless, we acknowledge that incorporating additional traditional clinical predictors into future models could further enhance predictive accuracy and clinical applicability. Comparative studies assessing the additive value of SII and PSMA PET/CT when combined with established clinical variables are warranted.

Furthermore, the favorable performance of our model complements evolving approaches in precision oncology, where early identification of patients at high risk for recurrence enables tailored adjuvant strategies and closer postoperative monitoring²¹. Incorporating advanced imaging with biological markers such as the SII could pave the way for dynamic risk models that are updated longitudinally as patients progress through their clinical trajectory. Several widely used nomograms, including the Cancer of the Prostate Risk Assessment Post-Surgical (CAPRA-S) score and the Stephenson nomogram, have been validated for predicting BCR after radical prostatectomy based on traditional clinical and pathological factors such as PSA level, Gleason score, surgical margin status, and lymph node involvement. While these models are valuable and well-established, they do not incorporate biological or molecular imaging markers. For instance, recent work has demonstrated the prognostic value of 68Ga-PSMA PET imaging and suggested its potential integration alongside CAPRA-S for enhanced recurrence risk stratification²². Our model complements these approaches by integrating systemic inflammation (via SII) and high-sensitivity imaging (via 18F-PSMA-1007 PET/CT), potentially offering improved granularity for individualized risk stratification. Future research could explore hybrid models that combine established nomograms with emerging biomarkers for enhanced prognostic accuracy.

Nevertheless, this study has limitations. Our analysis was retrospective and confined to a single institution, potentially introducing selection bias. The median follow-up duration of 15 months, while adequate for early recurrence events, limits the assessment of long-term outcomes such as metastasis-free survival and cancer-specific mortality. Larger, prospective, multicenter cohorts are warranted to validate our findings and refine model calibration. Moreover, future integration of genomic risk scores and radiomic features derived from PET/CT could further enhance predictive precision²³. Another key limitation of this study is its retrospective, single-center design with a relatively limited sample size, which may affect the generalizability of the predictive model. Although internal validation using a 70/30 train–test split demonstrated excellent model performance, external validation using larger, multicenter cohorts is essential to confirm its reproducibility and applicability across varied clinical settings. We are currently planning a multicenter prospective study to externally validate this model, which will enable more robust assessment of predictive accuracy, calibration, and real-world utility in guiding postoperative decision-making. In addition, the use of logistic regression to model BCR is also a limitation. While Cox proportional hazards modeling is ideal for analyzing recurrence-free survival, our study focused on predicting BCR status within a fixed follow-up period. Moreover, the exact timing of recurrence was not uniformly recorded for all patients, limiting the feasibility of time-to-event analysis. Future prospective studies with standardized follow-up intervals will allow for survival modeling using Cox regression to further refine risk prediction. We acknowledge that the use of standard ROC curves may not fully capture time-dependent variations in recurrence risk. Time-dependent ROC analysis is better suited for modeling survival outcomes and could be incorporated in future studies with standardized and longer-term follow-up data to evaluate model performance over time.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
18F-PSMA-1007	Shaanxi Zhengze Biotechnology Co., Ltd., China	18F-PSMA-1007	Radiotracer used for PET/CT imaging of prostate cancer lesions
Advantage Workstation software	GE Healthcare, Chicago, IL, USA	Version 4.7	Used for PET/CT image processing, region-of-interest definition, and SUVmax quantification
Blood analyzer	Mindray Bio-Medical Electronics Co., Ltd., China	6000PLUS	Used for measurement of neutrophil, lymphocyte, and platelet counts for SII calculation
PET/CT scanner	GE Healthcare, Chicago, IL, USA	Discovery VCT	Used for whole-body PET/CT image acquisition
ROI/image analysis software	GE Healthcare, Chicago, IL, USA	Advantage Workstation software (version 4.7)	Used for image review and quantitative lesion analysis
R software environment	R Foundation for Statistical Computing, Vienna, Austria	Version 4.2.2	Used for statistical analysis, ROC analysis, and model calibration
SPSS statistical software	IBM Corp., Armonk, NY, USA	Version 26.0	Used for statistical analysis