The Diagnostic Value of PI-RADS v2.1 in Patients with a History of Transurethral Resection of the Prostate (TURP)

Round 1

Reviewer 1 Report

Dear Authors, thanks for the opportunity to review your original paper. This work shows the reduced accuracy of mpMRI in the TZ in patients with previous history of TURP. 

The design of the study is clear, however there are important points that need to be clarified before the paper can be considered for publication:

1. Methods: you consider clinically significant a lesion volume more than 0.5 ml, is this related to mpMRI imaging ? please specify... moreover, in this case, you also consider as cs lesion with gleason 6 (3+3)?

2. The median time interval from the TURP to mpMRI and PSA measurement. It's well known that there is a significant PSA drop after TURP and similarly mpMRI images are biased within the first 6 months after the procedure. Please answer and provide adequate references in Literature.

It would be of value to report also pre-TURP PSA value to see if it was higher among positive cases.

3- as the TZ is mostly removed after TURP, why did you adress 1/3 of the samples in this area without considering anterior zone?

Author Response

Reviewer #1: Dear Authors, thanks for the opportunity to review your original paper. This work shows the reduced accuracy of mpMRI in the TZ in patients with previous history of TURP.

The design of the study is clear, however there are important points that need to be clarified before the paper can be considered for publication:

1. Methods: you consider clinically significant a lesion volume more than 0.5 ml, is this related to mpMRI imaging? please specify... moreover, in this case, you also consider as cs lesion with gleason 6 (3+3)?

Response: Thank you for your valuable comment. The definition of clinically significant prostate cancer (CSPCa) in our study is consistent with the definition of CSPCa in the Prostate Imaging–Reporting and Data System (PI-RADS) V2.1 guidelines (PMID: 26427566 & 30898406). The definition for CSPCa in PI-RADS V2.1 is based on Gleason score, disease tissue volume, and extra prostatic extension. Therefore, in our current study, if prostate cancer with a Gleason score of 3+3=6 exceeds the boundary tissue volume, it would be considered as CSPCa. Since the purpose of our study is to explore the diagnostic value of PI-RADS V2.1, we consider it necessary to keep our CSPCa definition aligned to the definition in PI-RADS V2.1. To emphasize this point, we have revised the CSPCa definition in Line 98 as below:

Clinically significant cancer was defined, following the PI-RADS V2.1 guidelines [7,8], as PCa with a histologic Gleason score ≥7 (including 3+4 with prominent but not predominant Gleason 4 component), and/or volume ≥0.5cc, and/or extra prostatic extension (EPE).

2. The median time interval from the TURP to mpMRI and PSA measurement. It's well known that there is a significant PSA drop after TURP and similarly mpMRI images are biased within the first 6 months after the procedure. Please answer and provide adequate references in Literature.

It would be of value to report also pre-TURP PSA value to see if it was higher among positive cases.

Response: Thank you for your comments and suggestions. The purpose of our study is to identify effective diagnostic tools for assessing post-TURP patients who are suspected of CSPCa in the long-term follow-up. Therefore, the interval between initial TURP and the present prostate biopsy is rather long. All 102 patients went under mpMRI assessment and PSA blood tests after more than 1 year since receiving TURP (Median 8, IQR 4-10.25 years). We have taken into consideration that the mpMRI images and PSA values early after TURP procedures may be biased due to hemorrhage, edema, and early fibrosis at the surgery site. Hence we revised our manuscript to emphasize such considerations in our study protocol in Line 65 as below:

In this retrospective study, consecutive patients who had undergone 12-core transrectal ultrasound (TRUS)-guided prostate biopsy with previous TURP history at our department between October 2014 and August 2020, were recruited. PSA is reported to significantly drop within 3-6 months after TURP [12,13], while hemorrhage, edema, and early fibrosis at the surgical site after TURP may create biases for mpMRI examinations [14,15]. Therefore, patients who received TURP less than 1 year ago were excluded from the study. The total cohort size was 102.

Pre-TURP values would be a valuable complementary marker to our study design. However, due to the long interval between receiving TURP and the present visit, a proportion of patients cannot provide credible medical records to obtain pre-TURP values. Hence, to maintain the integrity of the current study, pre-TURP values were not analyzed in the manuscript. On the other hand, pre-TURP values are not crucial to our study, since we set out to distinguish CSPCa developed in long-term follow-ups after TURP, but pre-TURP values are rather valuable for assessing incidental PCa or missed PCa in the initial procedure instead. We have revised our manuscript and added the lack of pre-TURP value as a shortcoming of our study in Line ? as below:

Pre-TURP PSA levels may provide reference cutoff values for our study. Unfortunately, due to the long interval since TURP and lack of follow-up data in our patient cohort, such information was not available.

3. as the TZ is mostly removed after TURP, why did you adress 1/3 of the samples in this area without considering anterior zone?

Response: Thank you very much for your comment. We apologize for the overly simplistic description of our biopsy protocol. The systematic 12-core TRUS-guided transperineal biopsy procedure used in our study follows a well-designed and widely-used protocol. The 12 biopsy cores distribute bilaterally to cover anterior TZ, posterior TZ, anterior horn of PZ, anterior lateral PZ, posterior lateral PZ, and posterior medial PZ. Therefore, the anterior zone of the prostate is covered in our biopsy procedure. Long-term outcomes of TURP revealed a considerable proportion of patients undergo adenoma regrowth in the TZ of the prostate (PMID: 20184573), which is commonly seen in our cohort of patients with elevated PSA and suspectable prostatic examinations. Therefore, we insisted on the standard biopsy protocol in order not to miss TZ lesions, and slight adjustments were made to adapt to the tissue defect at the surgery site. We have revised the description of our biopsy protocol to provide more details and added a supplementary figure of our biopsy template for illustration.

Line 90:

All patients underwent 12-core transrectal ultrasound (TRUS)-guided transperineal prostate biopsy. The well-designed biopsy template covers bilateral anterior TZ, posterior TZ, anterior horn of PZ, anterior lateral PZ, posterior lateral PZ, and posterior medial PZ. The biopsy template is shown in Figure S1. Core No. 1-8 are targeted to the PZ of the prostate while Core No. 9-12 are targeted to the TZ, slight adjustments were made to adapt to the tissue defect caused by TURP and to cover all suspected lesions found on imaging.

Author Response File: Author Response.docx

Reviewer 2 Report

 Authors presented an interesting preliminary analysis on the accuracy of mpMRI for PCa detection in men previously treated with TURP. Several comments:

-Why were targeted biopsies not performed?

-Why did authors perform only 8 cores of the PZ and 4 cores of the previously removed TZ?

-Subgroup analyses according to PI-RADS score are needed (also dichotomizing PIRADS>=3 vs PIRADS<3)

-Could authors evaluate PSA-D?

-Sensitivity, specificity, PPV and NPV are required

-Clinical implications of these findings are warranted? Possible implication for Active Surveillance (reference PMID: 34687344)?

Author Response

Reviewer #2: Authors presented an interesting preliminary analysis on the accuracy of mpMRI for PCa detection in men previously treated with TURP. Several comments:

1. Why were targeted biopsies not performed?

Response: Thank you for your valuable comment. We apologize for not providing detailed description of our biopsy protocol. The 12-core transperineal systematic prostate biopsy is strongly recommended as a standard procedure in the 2022 EAU guidelines for prostate cancer (EAU Guidelines. Edn. presented at the EAU Annual Congress Amsterdam 2022. ISBN 978-94-92671-16-5.). In our study, the 12 biopsy cores distribute bilaterally to cover anterior TZ, posterior TZ, anterior horn of PZ, anterior lateral PZ, posterior lateral PZ, and posterior medial PZ, comprising the areas of high CSPCa incidence. Lacking previous studies in patients with TURP history, whether targeted biopsy can replace or provide additional value to systematic biopsy remains inconclusive. On the other hand, targeted biopsy is recommended to be combined with systematic biopsy in cases with positive mpMRI findings, while our patient cohort may have negative findings in certain prostatic zones. Therefore, we conducted our well-practiced 12-core systematic prostate biopsy, during which we made slight adjustments to cover all suspected lesions found on imaging. We have revised the description of our biopsy protocol to provide more details and added a supplementary figure of our biopsy template for illustration.

Line 90:

All patients underwent 12-core transrectal ultrasound (TRUS)-guided transperineal prostate biopsy. The well-designed biopsy template covers bilateral anterior TZ, posterior TZ, anterior horn of PZ, anterior lateral PZ, posterior lateral PZ, and posterior medial PZ. The biopsy template is shown in Figure S1. Core No. 1-8 are targeted to the PZ of the prostate while Core No. 9-12 are targeted to the TZ, slight adjustments were made to adapt to the tissue defect caused by TURP and to cover all suspected lesions found on imaging.

2. Why did authors perform only 8 cores of the PZ and 4 cores of the previously removed TZ?

Response: Thank you very much for your valuable comment. As stated above, the systematic 12-core TRUS-guided transperineal biopsy procedure used in our study follows a well-designed and widely-used protocol, which is strongly recommended as a standard procedure in the 2022 EAU guidelines for prostate cancer. The 12 biopsy cores distribute bilaterally to cover anterior TZ, posterior TZ, anterior horn of PZ, anterior lateral PZ, posterior lateral PZ, and posterior medial PZ, comprising the areas of high CSPCa incidence. Long-term outcomes of TURP revealed a considerable proportion of patients undergo adenoma regrowth in the TZ of the prostate (PMID: 20184573), which is commonly seen in our cohort of patients with elevated PSA and suspectable prostatic examinations. Therefore, we consider it appropriate to perform the standard biopsy protocol in our patient cohort. We apologize for not providing detailed description of our biopsy protocol, and we have revised the manuscript to further illustrate the biopsy procedure.

Line 90:

All patients underwent 12-core transrectal ultrasound (TRUS)-guided transperineal prostate biopsy. The well-designed biopsy template covers bilateral anterior TZ, posterior TZ, anterior horn of PZ, anterior lateral PZ, posterior lateral PZ, and posterior medial PZ. The biopsy template is shown in Figure S1. Core No. 1-8 are targeted to the PZ of the prostate while Core No. 9-12 are targeted to the TZ, slight adjustments were made to adapt to the tissue defect caused by TURP and to cover all suspected lesions found on imaging.

3. Subgroup analyses according to PI-RADS score are needed (also dichotomizing PIRADS>=3 vs PIRADS<3)

Response: Thank you very much for your comment. We have added the subgroup analyses of PIRADS >=3 vs. <3 in Supplementary Table 2. Consistent with our previous findings, PIRADS >=3 in PZ is an independent predictor for CSPCa (OR=6.333, 95%CI 2.000-20.052, P=0.002), while PIRADS >=3 in TZ showed limited significance for predicting CSPCa (OR=0.778, 95%CI 0.335-1.803, P=0.558). We have revised our manuscript to display the results of the analyses in Line 164 as below:

In univariate logistic regression analysis (Table S2), compared to PI-RADS PZ score 2, the odds ratio was 7.200 (95%CI 2.140-24.230) for patients with a PI-RADS PZ score of 4, and 30.000 (95%CI 4.714-190.939) for patients with a PI-RADS PZ score of 5. PI-RADS PZ score ≥3 was an independent predictor for CSPCa (OR=6.333, 95%CI 2.000-20.052, P=0.002). As for the transitional zone, PI-RADS TZ scores of 2, 3, 4, and 5 showed no significant odds ratio compared to no suspected lesions by PI-RADS v2.1. PI-RADS TZ score ≥3 showed limited significance for predicting CSPCa (OR=0.778, 95%CI 0.335-1.803, P=0.558).

4. Could authors evaluate PSA-D?

Response: Thank you very much for your comment. PSA-D is a strong predictor in CSPCa risk calculators, and combinations of PSA-D and PI-RADS have been reported to provide guidance for biopsy decisions. Hence, the addition of PSA-D analysis would be valuable for our study design. However, patients in our study cohort demonstrate impaired anatomy in the TZ as a result of the surgical removal of TZ tissue and adenoma regrowth post-operatively. The anatomy of the TZ showed drastic heterogeneity within our cohort, which created difficulties in standardizing prostate volume measurement. Therefore, to maintain the integrity of the current study, PSA-D was not calculated and analyzed in our cohort. We have revised our manuscript to address this shortcoming of our study in Line 251 as below:

PSA-Density is another effective predictor for CSPCa in the whole patient population [30]. However, impaired anatomy at the previous TURP site created difficulties for prostate volume measurement.

5. Sensitivity, specificity, PPV and NPV are required

Response: Thank you for your valuable comment. We have revised our manuscript, and PPV and NPV values are added in our analysis in Line 127:

Setting the threshold at 23.81 ng/ml, PSA showed a best Youden’s Index of 0.338, with a sensitivity, specificity, PPV, and NPV of 37.2%,96.6%, 88.9%, and 67.9%, respectively, in differentiating CSPCa from this cohort. At a cutoff value of 10 ng/ml, PSA showed a sensitivity of 76.7% but a specificity of only 45.8%, and a PPV and NPV of 76.7% and 73.7% in differentiating CSPCa from this cohort. The best cutoff value for F/T PSA was obtained at 0.135. At this level, the Youden’s index, sensitivity, specificity, PPV, and NPV for F/T PSA were 0.467, 72.1%, 74.6%, 67.4%, and 78.6%. For PI-RADS v2.1, when the cutoff value is set as ≥4, the best Youden’s index, sensitivity, specificity, PPV, and NPV are 0.460, 88.4%, 57.6%, 60.3%, and 87.2%, respectively. If the cutoff value is set as ≥3, the sensitivity, specificity, PPV, and NPV were 97.7%, 25.4%, 48.8%, and 93.8%, respectively.

And Line 157:

At a cutoff value of ≥3, PI-RADS v2.1 PZ had a sensitivity, specificity, PPV, and NPV of 90.5%, 40.0%, 51.3%, and 85.7%.

6. Clinical implications of these findings are warranted? Possible implication for Active Surveillance (reference PMID: 34687344)?

Response: Thank you for your valuable comment. As stated in our manuscript, the patient cohort in our study is commonly excluded from clinical studies of diagnostic tools and treatment strategies, which strengthens the importance of our study. However, the patients we focus on are those who are suspectable of developing CSPCa in long-term follow-ups after TURP, instead of incidental PCa or missed PCa in the initial procedure. PMID: 34687344 concludes that AS represents a safe management strategy for incidental PCa, which is PCa discovered in pathological examinations of the removed tissue in the initial TURP procedure. Therefore, the findings may not apply in our cohort. In our cohort, as CSPCa is possibly developed independent of the TURP procedure, it may be reasonable to treat them through risk stratifications the same as the whole patient population. At all events, clinical implications of these findings require further follow-up and well-designed clinical studies, and we have revised our manuscript to address this point in Line 253 as below:

Clinical implications of our findings remain inconclusive. Further large-sample prospective studies are needed to conclude the characteristics of such a cohort and explore optimal diagnostic procedures and treatment strategies.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear Authors, 

thanks for answering to the requested revisions.

There is only one point which needs to be revised: as stated in the first revision, you refer to lesion volume as determinant for csPCa in gleason 6 category. I agree with You on that, but the PIRADS v2 criteria DO NOT consider lesion volume but lesion diameter. So you have to correct this statement. It will be nice to add this on the discussion section, as lesion volume is a more relevant predictor of csPCa than lesion diameter, citing the two relevant argument on this theme by Martorana E et al :

1-Martorana E, Pirola GM, Scialpi M et al. Lesion volume predicts prostate cancer risk and aggressiveness: validation of its value alone and matched with prostate imaging reporting and data system score. BJU Int. 2017 Jul;120(1):92-103. doi: 10.1111/bju.13649.  

2-Martorana E, Aisa MC, Grisanti R et Al. Lesion Volume in a Bi- or Multivariate Prediction Model for the Management of PI-RADS v2.1 Score 3 Category Lesions. Turk J Urol. 2022 Jul;48(4):268-277. doi: 10.5152/tud.2022.22038

Author Response

Thank you for your valuable comments. The suggestion to take lesion volume into consideration has been greatly inspirational and useful for us. However, after literature review, we have to insist that PIRADS V2 has defined clinically significant prostate cancer considering prostate lesion volume≥0.5cc instead of lesion diameter. We have attached a PDF version of PI-RADS Prostate Imaging– Reporting and Data System: 2015, Version 2 after this revision note, and the definition of CSPCa is located at 4. Section III: Assessment and Reporting cited as: “Based on the current uses and capabilities of mpMRI and MRI-targeted procedures, for PI-RADS™ v2 clinically significant cancer is defined on pathology/histology as Gleason score ≥7 (including 3+4 with prominent but not predominant Gleason 4 component), and/or volume ≥0.5cc, and/or extra prostatic extension (EPE).” Therefore, we did not revise the definition of CSPCa in our manuscript. The two relevant arguments on the theme of lesion volume have focused on the predictive value of MRI-estimated lesion volume in addition to PI-RADS V2 on detecting CSPCa and have concluded positive results. MRI-estimated lesion volume may be a valuable variable to be included in our diagnostic tests. However, MRI-estimated lesion volumes are best used in studies applying targeted biopsy techniques to link specific lesion volumes and biopsy results. In addition, as explained in the manuscript, impaired anatomy at the previous TURP site created difficulties for lesion volume measurement. Hence, it was not included to our study.

We have revised the manuscript and cited the two relevant studies to address this issue in the discussion section as a limitation of the current study in Line 251:

PSA-Density, as well as MRI-estimated lesion volume, has been reported to provide additional predictive value to the detection of CSPCa in the whole patient population [30-32]. However, impaired anatomy at the previous TURP site created difficulties for prostate and lesion volume measurement.

Author Response File: Author Response.pdf

Reviewer 2 Report

No more comments

Author Response

Thank you very much for your previous suggestions and approval.

Round 3

Reviewer 1 Report

In this revised version, the discussion has been implemented according to reviewers' concerns. The paper is now suitable for publication.