Development and Validation of a Preoperative Nomogram for Predicting Positive Surgical Margins After Laparoscopic Radical Prostatectomy
Prostate cancer remains a significant health concern globally, particularly as the third leading cause of cancer-related deaths among men. Laparoscopic radical prostatectomy (LRP) is a primary curative intervention for localized prostate cancer. However, positive surgical margins (PSM), defined as tumor cells extending to the inked surface of resected specimens, occur in approximately 20–35% of cases and are strongly associated with adverse outcomes, including biochemical recurrence, local recurrence, and distant metastasis. Despite the clinical relevance of PSM, preoperative tools to predict its likelihood remain limited. This study aimed to develop and validate a preoperative nomogram to stratify patients based on their risk of PSM after LRP, leveraging preoperative variables to guide surgical planning and postoperative management.
Study Design and Patient Cohort
The retrospective analysis included 418 patients who underwent LRP without neoadjuvant therapy at Peking University Third Hospital between January 2010 and March 2016. Exclusion criteria comprised incomplete clinical or pathological data, resulting in 81 exclusions. The cohort’s median age was 70 years (interquartile range [IQR]: 65–75), with 34% (142/418) exhibiting PSM on final pathology. Among PSM cases, 18.3% had solitary apical margins, 34.5% solitary non-apical margins, and 47.2% multiple margins.
Preoperative variables analyzed included age, body mass index (BMI), total prostate-specific antigen (tPSA), free PSA (fPSA) ratio (fPSA/tPSA), digital rectal examination (DRE) findings, biopsy Gleason scores, clinical stage (2012 TNM system), percentage of positive biopsy cores, and surgeon experience. Pathological staging followed the 2005 International Society of Urological Pathology criteria.
Statistical Methodology
Univariable logistic regression identified predictors of PSM, followed by multivariable analysis using backward stepwise selection. Continuous variables, such as age and tPSA, were modeled with restricted cubic splines to capture nonlinear relationships. The Akaike Information Criterion (AIC) guided spline knot selection. Missing data were addressed via expectation-maximization algorithms for continuous variables or treated as separate categories for categorical variables.
The final nomogram incorporated four variables: percentage of positive biopsy cores, clinical stage, fPSA/tPSA ratio, and age. Model performance was evaluated using the concordance index (C-index), calibration curves, and decision curve analysis (DCA). Bootstrap validation with 1,000 resamples assessed overfitting.
Key Findings
Predictors of Positive Surgical Margins
- Percentage of Positive Biopsy Cores: Each 1% increase in positive cores raised the PSM risk by 2% (odds ratio [OR]: 1.02; 95% CI: 1.01–1.03; P<0.001). This aligns with prior studies linking higher tumor burden to margin involvement.
- Clinical Stage: Advanced clinical stages significantly increased PSM risk. Compared to stage T1, stage T2 (OR: 3.91; P=0.03), T3 (OR: 5.49; P=0.01), and T4 (OR: 15.08; P=0.004) showed escalating risks, underscoring tumor extent as a critical determinant.
- fPSA/tPSA Ratio: A lower fPSA/tPSA ratio correlated with higher PSM risk (OR: 0.96; P=0.01), potentially reflecting aggressive tumor biology or larger tumor volume.
- Age: A U-shaped relationship emerged, with PSM risk peaking in younger (75 years) patients. Younger patients may harbor more aggressive tumors, while older patients face technical challenges due to anatomical constraints or comorbidities.
Model Performance
The nomogram demonstrated robust discrimination, with a C-index of 0.722 in the development cohort and 0.700 after bootstrap validation. Calibration curves revealed close alignment between predicted and observed PSM probabilities, with a mean absolute error of 2.0%. Decision curve analysis confirmed clinical utility, showing net benefits across threshold probabilities of 20–70%, making the tool applicable for a broad patient spectrum.
Clinical Implications
The nomogram provides a practical, preoperative tool to stratify PSM risk, enabling tailored surgical approaches. For high-risk patients (e.g., advanced clinical stage, extensive biopsy positivity), surgeons may opt for wider excision margins or adjuvant therapies. Conversely, low-risk patients could benefit from nerve-sparing techniques to preserve functional outcomes without compromising oncologic safety.
Limitations and Contextual Considerations
The study’s PSM rate (34%) exceeds Western cohorts (15–20%), likely reflecting differences in surgical technique (non-robotic LRP), later-stage diagnoses in Chinese populations, and potential genetic or anatomic variations. Notably, the nomogram excludes intraoperative factors like nerve-sparing status and surgeon experience, which were nonsignificant in multivariable analysis. Additionally, reliance on retrospective data introduces selection bias, and external validation is needed to generalize findings across diverse populations.
Conclusion
This study successfully developed and validated a preoperative nomogram incorporating percentage of positive biopsy cores, clinical stage, fPSA/tPSA ratio, and age to predict PSM after LRP. With strong discrimination and calibration, the tool enhances personalized surgical planning and risk communication. Future studies should validate the model in robotic-assisted cohorts and integrate imaging biomarkers (e.g., MRI) to refine predictive accuracy.
doi.org/10.1097/CM9.0000000000000161
Was this helpful?
0 / 0