# The Use of PI-FAB Score in Evaluating mpMRI After Focal Ablation of Prostate Cancer: Is It Reliable? Inter-Reader Agreement in a Tertiary Care Referral University Hospital

## TL;DR

This study shows that the PI-FAB score is reliable for evaluating MRI scans after prostate cancer focal therapy, with strong agreement among radiologists of varying experience.

## Contribution

The study confirms the high reproducibility of the new PI-FAB scoring system for post-focal therapy MRI assessments.

## Key findings

- The inter-reader agreement for PI-FAB scoring was excellent with a Gwet’s AC2 value of 0.941.
- Most patients were scored as PI-FAB 1 at 6 months, with no recurrence observed at later follow-ups.
- HIFU appears to be a reliable treatment for selected prostate cancer patients.

## Abstract

Focal therapies are very promising techniques for prostate cancer treatment, but their use in post-therapy MRI assessment, both in image acquisition and reporting, requires standardization. Prostate Imaging after Focal Ablation (PI-FAB) scoring, first described in May 2023, is a new scoring system for interpreting prostate multiparametric magnetic resonance imaging (mpMRI) results after focal therapies. The aim of our study was to assess the inter-reader agreement of PI-FAB scores among radiologists in a single large cohort of patients treated with focal therapy (in particular, high intensity focused ultrasound, HIFU) in a tertiary care referral University Hospital. To our knowledge, only two studies have evaluated the inter-reader agreement of PI-FAB scoring so far. In our experience, this new score significantly improves the inter-reader agreement of radiologists with different levels of experience. Furthermore, HIFU seems to be a very promising tool for the treatment of selected prostate cancer patients.

Background/Purpose: to assess the inter-reader agreement of the PIFAB (Prostate Imaging after Focal Ablation) score, a new MRI-based standardized system for evaluating post-focal therapy prostate mpMRI, among radiologists in a single large cohort of patients treated with focal therapy (HIFU) in a tertiary care referral University Hospital. Methods: In total, 68 consecutive patients who underwent HIFU were included in this single-center retrospective observational study. A total of 109 post-HIFU follow-up mpMRIs were evaluated by three radiologists with varying levels of experience (12, 8, and 3 years, respectively). All patients underwent their first follow-up mpMRI at 6 months post-treatment, with 30 patients receiving additional evaluations at 18 months and 11 at 30 months. Results: The patients had a mean age of 70.6 ± 8.31 years, a mean pre-treatment PSA (prostate-specific antigen) of 7.85 ± 1.21 ng/mL, and a mean post-treatment PSA of 4.64 ± 4.2 ng/mL. The inter-reader agreement for PI-FAB among the three radiologists showed a Gwet’s AC2 value of 0.941 (95% confidence interval: 0.904–0.978, p < 0.0001). For the most experienced radiologist, at the 6-month follow-up 64 (94.14%) patients were scored as PI-FAB 1, 1 (1.47%) as PI-FAB 2, and 3 (4.41%) as PI-FAB 3. At the 18-month and 30-month follow-ups all patients were scored as PI-FAB 1 (no suspicion of recurrence). Conclusions: Our study demonstrates excellent inter-reader agreement among radiologists with varying levels of experience, confirming that the PI-FAB score is highly reproducible when evaluating post-treatment mpMRI scans. The low rate of PI-FAB 2 and PI-FAB 3 lesions observed at the first follow-up, coupled with the absence of significant recurrence in subsequent evaluations, suggests that HIFU is a reliable technique for prostate cancer treatment in selected patients.