# CT Radiomics Models Did Not Outperform Experts in Predicting [68Ga]Ga-PSMA-PET Positivity in Prostate Cancer Lymph Node Staging

## TL;DR

This study found that CT radiomics models did not outperform expert radiologists in predicting PSMA-positive lymph nodes in prostate cancer patients, but a simple CT density threshold could still be useful in resource-limited settings.

## Contribution

The study introduces a simple CT-derived metric (median lymph node attenuation) as a practical alternative for predicting PSMA positivity in prostate cancer lymph nodes.

## Key findings

- Radiomics models had high sensitivity and NPV but were outperformed by experts in accuracy and specificity.
- A CT density threshold of 27 Hounsfield units predicted PSMA positivity with 79% accuracy.
- Expert radiologists showed superior performance in specific subanalyses of lymph node assessments.

## Abstract

Contrast-enhanced CT is widely available for prostate cancer staging, whereas [68Ga] PSMA PET/CT—a sensitive and accurate method—is limited by cost and access. We investigated whether CT radiomics can predict PSMA-positive lymph nodes in prostate cancer patients. In 447 patients, 2537 lymph nodes were segmented (425 PSMA-positive) and several feature-selection and classification pipelines were trained and tested; two expert uroradiologists independently assessed a subset of 417 nodes on CT. Radiomics models showed good sensitivity and high negative predictive value, supporting potential use for ruling out nodal involvement, but they did not outperform expert readers overall because reader specificity and accuracy were higher. A simple CT-derived parameter was found to be predictive for PSMA-positive nodes: median lymph node attenuation, with an optimal threshold around 27 Hounsfield units, was associated with PSMA positivity and provided clinically usable discrimination. This pragmatic CT metric may aid nodal risk stratification when PSMA PET/CT or specialist interpretation is not available.

Background: The use of [68Ga]Ga-PSMA-PET/CT for prostate cancer (PCa) staging is limited by cost and availability. This study evaluates whether radiomic features from contrast-enhanced (CE) CT can predict PSMA-positive lymph nodes (LNs) as a surrogate for metastasis. Methods: A retrospective study of 447 patients included 2537 segmented LNs (425 PET-positive, 2112 PET-negative). Two uroradiologists assessed 417 LNs on CE-CT using a four-point Likert scale. Radiomic features were extracted, selected using four algorithms, and analyzed with six model-building methods. Model performance was compared to radiologist ratings. Results: Radiomic models achieved an accuracy of 0.77–0.85, sensitivity of 0.85–0.91, and specificity of 0.74–0.85. Compared to radiologists, models had higher NPV (0.97–0.98 vs. 0.96) and sensitivity (0.85–0.91 vs. 0.76), but radiologists had superior accuracy (0.95 vs. 0.77–0.85) and specificity (0.97–0.98 vs. 0.74–0.85). In a subanalysis of LNs rated as probably benign or malignant, expert radiologists outperformed the algorithm with greater specificity and PPV (p < 0.005). A density threshold of >27 HU predicted PSMA-positive LNs with 0.79 accuracy, 0.87 sensitivity, and 0.78 specificity. Conclusions: While radiomics did not outperform expert radiologists, the single first-order parameter CT density >27 HU was predictive of PSMA-positive LNs. Clinical Relevance Statement: Radiomic models did not outperform expert uroradiologists. However, in high-volume or resource-limited settings lacking access to [68Ga]Ga-PSMA-PET/CT, they may help improve LN assessment in PCa patients with CT alone.