Main Session
Sep 29
PQA 06 - Genitourinary Cancer, Gynecological Cancer, and Health Care Access and Engagement

3405 - Diagnostic Performance of mpMRI and PSMA PET In Local Intraprostatic Radiorecurrent Prostate Cancer: A Radiologic-Pathologic Correlation Study

02:15pm - 03:30pm ET
Poster Hall - Exhibit Hall A
Screen: 20
POSTER

Presenter(s)

Esther Yu, MD Headshot
Esther Yu, MD - UCLA David Geffen School of Medicine/UCLA Medical Center, Los Angeles, CA

E. D. Yu1, L. Valle2, V. Ludwig3, E. R. Felker4, J. Calais5, L. S. Marks6, W. Brisbane6, S. Raman7, R. E. Reiter8, M. Rettig9, J. Czernin5, M. L. Steinberg10, N. Nichols11, A. J. Chang2, R. R. Huang6, and A. U. Kishan2; 1UCLA David Geffen School of Medicine/UCLA Medical Center, Los Angeles, CA, 2Department of Radiation Oncology, University of California, Los Angeles, Los Angeles, CA, 3UCLA, Los Angeles, CA, 4Department of Radiology, David Geffen School of Medicine, UCLA, Los Angeles, CA, 5Department of Nuclear Medicine and Theranostics, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, 6University of California, Los Angeles, Los Angeles, CA, 7Department of Diagnostic and Interventional Radiology, University of California, Los Angeles, Los Angeles, CA, 8Department of Urology, University of California, Los Angeles, Los Angeles, CA, 9Department of Medical Oncology, University of California, Los Angeles, Los Angeles, CA, 10Department of Radiation Oncology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, 11VA and UCLA, Los Angeles, CA

Purpose/Objective(s):

Accurate localization of intraprostatic radiorecurrent prostate cancer is clinically challenging but remains critical for biopsy planning and target delineation for local salvage therapy. Multiparametric MRI (mpMRI) and prostate specific membrane antigen (PSMA) PET are widely used, yet their relative and combined diagnostic performance in the post-radiation setting requires clarification.

Materials/Methods:

This study evaluated 116 biopsy-sampled prostate “octants” across 21 patients with suspected radiorecurrence discussed in a tumor board setting. Imaging findings from mpMRI and PSMA PET as well as their combined interpretation (positive on any imaging modality), and a consensus expert combined interpretation (positive as determined by an expert in both PET and MRI radiology) were correlated with targeted and systematic histopathology, which served as the reference standard. Diagnostic performance was assessed using sensitivity, specificity, positive and negative predictive values (PPV/NPV), and area under the curve (AUC). Additional analyses explored the utility of PIRADS and Primary scoring frameworks in the recurrent setting.

Results:

All imaging strategies demonstrated strong overall performance (AUC > 0.84). Expert combined interpretation yielded the highest accuracy (AUC 0.885), followed by combined modality imaging (AUC 0.874), mpMRI (AUC 0.861), and PSMA PET (AUC 0.848). For single modality, mpMRI was more sensitive (77.8% vs 74.1%) and provided the highest NPV (93.3% vs 92.4%), whereas PSMA PET was more specific (95.5% vs 94.4%) with the highest PPV (83.3% vs 80.8%). Combining modalities improved sensitivity (81.5%) and NPV (94.3%) but reduced specificity (93.3%) and PPV (78.6%). Expert integration of both modalities improved specificity (95.5%) and PPV (84.6%) while maintaining high sensitivity (81.5%) and NPV (94.4%). Site-specific analysis revealed high false positive rates in the SVs (20-25%), and high false negative rates at the mid gland (20-40%) for both modalities even with expert interpretation. Exploratory analyses suggested that traditional PIRADS and Primary scoring systems may have mixed relevance in the radiorecurrent context, with primary scores faltering specifically with recurrent disease in the SVs.

Conclusion:

In local intraprostatic radiorecurrent prostate cancer, both mpMRI and PSMA PET perform well, with complementary strengths. Expert combined interpretation offers the highest diagnostic accuracy, supporting its potential role in guiding targeted biopsy and treatment planning, but false positives/negatives on both mpMRI and PSMA PET confirm the need for biopsy confirmation of recurrence prior to initiation of local salvage therapy especially at SV and prostate mid gland. Further work is needed to refine imaging criteria specific to the post-radiation environment.