Comparison of mpMRI and ⁶⁸Ga-PSMA-PET/CT in the Assessment of the Primary Tumors in Predominant Low-/Intermediate-Risk Prostate Cancer

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Overall Assessment
This study compares the diagnostic performance of mpMRI and PSMA-PET/CT for intraprostatic lesion localization in low-/intermediate-risk prostate cancer patients undergoing focal HDR brachytherapy. The manuscript is well-structured, methodologically sound, and addresses a clinically relevant question. However, the results should be approached carefully due to a few limitations: the small sample size and possible biases. Below are detailed comments.

Given the growing interest in focal therapies, the comparison of mpMRI and PSMA-PET/CT for intraprostatic lesion detection is timely. The 24-segment prostate model enhances spatial precision, critical for brachytherapy planning.

Since the study has a small sample size with only 27 patients (19 with biopsy data), that limits statistical power and generalizability. Also, the predominance of low-risk cases (Gleason ≤3+4) may not reflect higher-risk cohorts. 
The suggestion is that the manuscript might emphasize this as a pilot study and advocate for larger, multi-center validation.
The potential bias: Biopsy targeting was MRI-guided, which may favour MRI's performance. While the authors cite meta-analyses suggesting no difference between cognitive/software fusion, this remains a confounder. The authors might discuss whether standalone PSMA-PET/CT-targeted biopsies could be explored in future work.
The other bias is significant differences between PET readers (AUC 0.684 vs 0.608; *p*<0.001), suggesting subjectivity in PET interpretation. Standardization (e.g., PSMA-RADS) might be proposed to reduce variability.

Also, the spatial resolution limitations of PET's inferior resolution (vs. MRI) may be a disadvantage in a 24-segment model, which is a technical constraint, and whether fewer segments (e.g., 12) would yield different results is debatable.

Consolidation of key findings in a summary table (e.g., AUCs, congruence scores, p values).
Does the 24-segment model explain discrepancies in that there are contrast results with the PRIMARY trial (PSMA-PET/CT vs MRI)? It also should be addressed why PSMA-PET/CT underperformed despite high PSMA expression in Gleason 7 disease.

The abstract should clarify if "447 segments" include all patients or only the 19 with biopsy. Using sensitivity and specificity at optimal thresholds  (not just AUC) is advisable.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors analysed the two modalities mpMRI and Ga-PSMA-Pet/CT in a group of 27 patients. All patients had a histopathologically diagnosed prostate carcinoma. The majority of the 27 patients showed an intermediate risk according to the D'Amico classification, while only one patient had a high risk. The mpMRI and PSMA-PET/CT were compared in terms of specificity, sensitivity and congruence with the localisation of the intraprostatic malignant lesion of the biopsy. PSMA-PET/CT is primarily used for extraprostatic staging only and is not recommended for intraprostatic prostate cancer diagnosis. In this study, the biopsy was performed as a fusion biopsy.

A total of 19 patients showed a positive biopsy result. A comparison of the mpMRI between the two readers showed AUC values of 0.770 and 0.781 (DeLong p = 0.75) and for the PSMA-Pet/CT of 0.684 and 0.608 (DeLong p < 0.001). The combination of both readers showed an advantage for the mpMRI compared to the PSMA-PET/CT (AUC: mpMRI: 0.815, AUC PSMA-PET/CT: 0.690, p = 0.006). In their conclusion, the authors mentioned that mpMRI was superior to PSMA-PET/CT in terms of intraprostatic detection. However, PSMA-PET/CT showed extraprostatic detection of metastases in two patients.

Disadvantages of the study are a small number of patients, the number of examiners, a low number of biopsy-positive prostate segments and the inclusion of patients with a low or intermediate risk according to D'Amico (only one patient with high risk).

The high variance between PET1 and PET2 of 72.17% is striking, while the variance for MRI was 21.79%. There was also a significant difference in the AUC values between PET1 and PET2 (0.684 vs. 0.608, p < 0.001). This could explain the difference between mpMRI and PSMA-PET/CT compared to the intraprostatic lesion of the biopsy. It is unclear why the results of PET1 and PET2 differ so significantly. The authors could explain this in the discussion.

Author Response

Please see the attachment

Author Response File: Author Response.pdf