Nomogram Development and Guideline Evaluation for Metastatic Prostate Cancer Prediction

Overview

This study developed nomograms using SEER data to predict distant and bone metastases in prostate cancer patients and evaluated bone imaging recommendations from NCCN, AUA, and EAU guidelines. Key risk factors such as PSA level, Gleason score, and tumor stage were identified as significant predictors of metastasis. The study also assessed the diagnostic performance of current guideline criteria for bone imaging in metastatic prostate cancer.

Background

Prostate cancer (PCa) has high survival rates when localized but poor prognosis once metastasized, particularly to bone, the most common metastatic site. Early identification of patients at risk for metastasis is critical to guide follow-up and treatment strategies. Existing guidelines from NCCN, AUA, and EAU recommend bone imaging primarily for high-risk patients, but consensus and validation of these recommendations remain limited. This study aimed to analyze risk factors for metastasis, create predictive nomograms, and evaluate guideline performance using a large contemporary dataset.

Data Highlights

Key Findings

• PSA level ≥20 ng/ml, higher Gleason grade groups (≥4), and advanced T stage (≥T3) are significant independent predictors of distant metastasis in prostate cancer.
• Nomograms incorporating clinical variables demonstrated good predictive accuracy for distant metastasis and bone-only metastasis, validated by concordance indices and calibration curves.
• Bone imaging recommendations from NCCN, AUA, and EAU guidelines show variable sensitivity and specificity in detecting bone metastases, with no consensus on imaging for high-risk localized disease.
• Patients with unfavorable intermediate or high-risk disease per guideline criteria are recommended for bone imaging, but the number needed to image (NNI) and predictive values differ among guidelines.
• Exclusion of patients with ambiguous clinical T2 subclassification and lymph node metastasis was necessary for guideline evaluation, focusing analysis on bone metastasis detection.

Clinical Implications

Clinicians can use the developed nomograms to better stratify prostate cancer patients by metastasis risk, potentially guiding decisions on imaging and treatment intensification. Awareness of the strengths and limitations of NCCN, AUA, and EAU bone imaging guidelines can inform individualized patient management, especially in high-risk localized disease where imaging recommendations vary. Incorporating these predictive tools may improve early detection of metastasis and optimize resource utilization.

Conclusion

This study provides validated nomograms for predicting metastatic prostate cancer and highlights differences in bone imaging guideline performance. These findings support more tailored approaches to metastasis risk assessment and imaging decisions in clinical practice.

References

1. SEER Database/Various Years -- Prostate Cancer Survival and Epidemiology Data
2. NCCN Guidelines Version 1.2020 -- Prostate Cancer Imaging Recommendations
3. AUA/ASTRO/SUO Guidelines 2018 -- Prostate Cancer Bone Imaging
4. EAU-EANM-ESTRO-ESUR-SIOG Guidelines 2020 -- Prostate Cancer Bone Imaging