Interreader variability in prostate MRI reporting using Prostate Imaging Reporting and Data System version 2.1

ESR Essentials: using the right scoring system in prostate MRI-practice recommendations by ESUR

MRI has gained prominence in the diagnostic workup of prostate cancer (PCa) patients, with the Prostate Imaging Reporting and Data System (PI-RADS) being widely used for cancer detection. Beyond PI-RADS, other MRI-based scoring tools have emerged to address broader aspects within the PCa domain. However, the multitude of available MRI-based grading systems has led to inconsistencies in their application within clinical workflows. The Prostate Cancer Radiological Estimation of Change in Sequential Evaluation (PRECISE) assesses the likelihood of clinically significant radiological changes of PCa during active surveillance, and the Prostate Imaging for Local Recurrence Reporting (PI-RR) scoring system evaluates the risk of local recurrence after whole-gland therapies with curative intent. Underlying any system is the requirement to assess image quality using the Prostate Imaging Quality Scoring System (PI-QUAL). This article offers practicing radiologists a comprehensive overview of currently available scoring systems with clinical evidence supporting their use for managing PCa patients to enhance consistency in interpretation and facilitate effective communication with referring clinicians. KEY POINTS: Assessing image quality is essential for all prostate MRI interpretations and the PI-QUAL score represents  the standardized tool for this purpose. Current urological clinical guidelines for prostate cancer diagnosis and localization recommend adhering to the PI-RADS recommendations. The PRECISE and PI-RR scoring systems can be used for assessing radiological changes of prostate cancer during active surveillance and the likelihood of local recurrence after radical treatments respectively.

Comparison in prostate cancer diagnosis with PSA 4-10 ng/mL: radiomics-based model VS. PI-RADS v2.1

BACKGROUND

To evaluate accuracy of MRI-based radiomics in diagnosing prostate cancer (PCa) in patients with PSA levels between 4 and 10 ng/mL and compare it with the latest Prostate Imaging Reporting and Data System (PI-RADS v2.1) score.

METHODS

221 patients with prostate lesions and PSA levels in 4-10 ng/mL, including 154 and 67 cases in the training and validation groups. Pathological confirmation of all patients was accomplished by the use of MRI-TRUS fusion targeted biopsy or systematic transrectal ultrasound (TRUS) guided biopsy. 851 radiomic features were extracted from each lesion of ADC and T2WI images. The least absolute shrinkage and selection operator (LASSO) regression algorithm and logistic regression were employed to select features and build the ADC and T2WI model. The combined model was obtained based on the ADC and T2WI features. The clinical benefit and diagnostic accuracy of the three radiomics models and PI-RADS v2.1 score were evaluated.

RESULTS

10 radiomic features were ultimately selected from the ADC images, 13 from the T2WI images and 7 from the combined models. The ADC, T2WI and combined models achieved satisfactory diagnostic accuracy in the training [AUC:0.945 (ADC), 0.939 (T2WI), 0.979 (combined)] and validation groups [AUC: 0.942 (ADC), 0.943 (T2WI), 0.959 (combined)], which was significantly higher than those in PI-RADS v2.1 model (0.825 for training cohort and 0.853 for validation cohort). Compared with the PI-RADS v2.1 score, the three radiomics models generated superior PCa diagnostic performance in both the training (p = 0.002, p = 0.005, p < 0.001) and validation groups (p = 0.045, p = 0.035, p = 0.015).

CONCLUSION

Radiomics based on ADC and T2WI images can better identify PCa in patients with PSA 4-10 ng/mL, and MRI-based radiomics significantly outperforms the PI-RADS v2.1 score.

CLINICAL TRIAL NUMBER

Not applicable.

Inter-reader reliability and diagnostic accuracy of PI-RADS scoring between academic and community care networks: How wide is the gap?

IMPORTANCE

The Prostate Imaging Reporting & Data System (PI-RADS) scoring guidelines were developed to address the substantial variation in interpretation and reporting of prostate cancer (PCa) multiparametric MRI (mpMRI) results, and subsequent updates have sought to further improve inter-reader reliability. Nonetheless, the variability of PI-RADS scoring in real-world settings may represent a continuing challenge to the widespread standardization of prostate mpMRI and limit its overall clinical benefit.

OBJECTIVE

To assess variability in mpMRI interpretation and reporting of PCa, we evaluated the discrepancies in PI-RADS scoring between community practices and a tertiary academic care center.

DESIGN, SETTING, AND PARTICIPANTS

We identified 262 mpMRI studies from nonacademic facilities, reinterpreted by radiologists at our institution between January 2016 and July 2022. Results of targeted MRI fusion biopsy were identified for 193 of these patients, totaling 302 lesions. PI-RADS scoring from both community and academic interpreters were recorded in addition to presence of clinically significant PCa (csPCa) on pathological analysis of targeted cores.

MAIN OUTCOME AND MEASURES

The primary outcome was inter-reader reliability via intraclass correlation (ICC) and the kappa statistic. We also assessed the diagnostic accuracy of PI-RADS scoring for detecting csPCa for both cohorts via receiver operator characteristics (ROC) analysis and compared these findings using paired-sample area difference under curve analysis.

RESULTS

Inter-reader agreement and reliability of PI-RADS scoring per lesion was generally poor (absolute agreement ICC = 0.393, 95% CI: 0.288-0.488; consistency ICC = 0.407, 95% CI: 0.308-0.497; kappa = 0.336, 95% CI: 0.267-0.406). Reliability results from studies obtained after the publication of PI-RADSv2.1 were similar to those of the overall analysis. No agreement was observed in the subgroup of lesions scored as PIRADS 3 by community interpreters. No statistically significant difference in diagnostic accuracy was observed between cohorts (ROC area under curve [AUC]: 0.759 vs. 0.785, respectively; P = 0.337). PI-RADS 3 was determined to be the optimal cutoff for detecting clinically significant disease in both cohorts.

CONCLUSIONS AND RELEVANCE

Our results suggest that mpMRI diagnostic accuracy for detecting csPCa is not significantly different between academic and community practices. However, significantly poor reliability of mpMRI was observed between cohorts, suggesting the risk of introducing practice variation for community prostate cancer management. Variability, particularly for PI-RADS 3 lesions, can lead to inconsistent biopsy recommendations, which may result in missed csPCa or unnecessary biopsies.

Perspectives on technology: Prostate Imaging-Reporting and Data System (PI-RADS) interobserver variability

OBJECTIVES

To explore the topic of Prostate Imaging-Reporting and Data System (PI-RADS) interobserver variability, including a discussion of major sources, mitigation approaches, and future directions.

METHODS

A narrative review of PI-RADS interobserver variability.

RESULTS

PI-RADS was developed in 2012 to set technical standards for prostate magnetic resonance imaging (MRI), reduce interobserver variability at interpretation, and improve diagnostic accuracy in the MRI-directed diagnostic pathway for detection of clinically significant prostate cancer. While PI-RADS has been validated in selected research cohorts with prostate cancer imaging experts, subsequent prospective studies in routine clinical practice demonstrate wide variability in diagnostic performance. Radiologist and biopsy operator experience are the most important contributing drivers of high-quality care among multiple interrelated factors including variability in MRI hardware and technique, image quality, and population and patient-specific factors such as prostate cancer disease prevalence. Iterative improvements in PI-RADS have helped flatten the curve for novice readers and reduce variability. Innovations in image quality reporting, administrative and organisational workflows, and artificial intelligence hold promise in improving variability even further.

CONCLUSION

Continued research into PI-RADS is needed to facilitate benchmark creation, reader certification, and independent accreditation, which are systems-level interventions needed to uphold and maintain high-quality prostate MRI across entire populations.

Inter-reader Agreement for Prostate Cancer Detection Using Micro-ultrasound: A Multi-institutional Study

Background and objective

Micro-ultrasound (MUS) uses a high-frequency transducer with superior resolution to conventional ultrasound, which may differentiate prostate cancer from normal tissue and thereby allow targeted biopsy. Preliminary evidence has shown comparable sensitivity to magnetic resonance imaging (MRI), but consistency between users has yet to be described. Our objective was to assess agreement of MUS interpretation across multiple readers.

Methods

After institutional review board approval, we prospectively collected MUS images for 57 patients referred for prostate biopsy after multiparametric MRI from 2022 to 2023. MUS images were interpreted by six urologists at four institutions with varying experience (range 2-6 yr). Readers were blinded to MRI results and clinical data. The primary outcome was reader agreement on the locations of suspicious lesions, measured in terms of Light's κ and positive percent agreement (PPA). Reader sensitivity for identification of grade group (GG) ≥2 prostate cancer was a secondary outcome.

Key findings and limitations

Analysis revealed a κ value of 0.30 (95% confidence interval [CI] 0.21-0.39). PPA was 33% (95% CI 25-42%). The mean patient-level sensitivity for GG ≥2 cancer was 0.66 ± 0.05 overall and 0.87 ± 0.09 when cases with anterior lesions were excluded. Readers were 12 times more likely to detect higher-grade cancers (GG ≥3), with higher levels of agreement for this subgroup (κ 0.41, PPA 45%). Key limitations include the inability to prospectively biopsy reader-delineated targets and the inability of readers to perform live transducer maneuvers.

Conclusions and clinical implications

Inter-reader agreement on the location of suspicious lesions on MUS is lower than rates previously reported for MRI. MUS sensitivity for cancer in the anterior gland is lacking.

Patient summary

The ability to find cancer on imaging scans can vary between doctors. We found that there was frequent disagreement on the location of prostate cancer when doctors were using a new high-resolution scan method called micro-ultrasound. This suggests that the performance of micro-ultrasound is not yet consistent enough to replace MRI (magnetic resonance imaging) for diagnosis of prostate cancer.

Update on PI-RADS Version 2.1 Diagnostic Performance Benchmarks for Prostate MRI: Systematic Review and Meta-Analysis

Background Prostate MRI for the detection of clinically significant prostate cancer (csPCa) is standardized by the Prostate Imaging Reporting and Data System (PI-RADS), currently in version 2.1. A systematic review and meta-analysis infrastructure with a 12-month update cycle was established to evaluate the diagnostic performance of PI-RADS over time. Purpose To provide estimates of diagnostic accuracy and cancer detection rates (CDRs) of PI-RADS version 2.1 categories for prostate MRI, which is required for further evidence-based patient management. Materials and Methods A systematic search of PubMed, Embase, Cochrane Library, and multiple trial registers (English-language studies published from March 1, 2019, to August 30, 2022) was performed. Studies that reported data on diagnostic accuracy or CDRs of PI-RADS version 2.1 with csPCa as the primary outcome were included. For the meta-analysis, pooled estimates for sensitivity, specificity, and CDRs were derived from extracted data at the lesion level and patient level. Sensitivity and specificity for PI-RADS greater than or equal to 3 and PI-RADS greater than or equal to 4 considered as test positive were investigated. In addition to individual PI-RADS categories 1-5, subgroup analyses of subcategories (ie, 2+1, 3+0) were performed. Results A total of 70 studies (11 686 lesions, 13 330 patients) were included. At the patient level, with PI-RADS greater than or equal to 3 considered positive, meta-analysis found a 96% summary sensitivity (95% CI: 95, 98) and 43% specificity (95% CI: 33, 54), with an area under the summary receiver operating characteristic (SROC) curve of 0.86 (95% CI: 0.75, 0.93). For PI-RADS greater than or equal to 4, meta-analysis found an 89% sensitivity (95% CI: 85, 92) and 66% specificity (95% CI: 58, 74), with an area under the SROC curve of 0.89 (95% CI: 0.85, 0.92). CDRs were as follows: PI-RADS 1, 6%; PI-RADS 2, 5%; PI-RADS 3, 19%; PI-RADS 4, 54%; and PI-RADS 5, 84%. The CDR was 12% (95% CI: 7, 19) for transition zone 2+1 lesions and 19% (95% CI: 12, 29) for 3+0 lesions (P = .12). Conclusion Estimates of diagnostic accuracy and CDRs for PI-RADS version 2.1 categories are provided for quality benchmarking and to guide further evidence-based patient management. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Tammisetti and Jacobs in this issue.

The Impact of Omitting Contralateral Systematic Biopsy on the Surgical Planning of Patients with a Unilateral Suspicious Lesion on Magnetic Resonance Imaging Undergoing Robot-assisted Radical Prostatectomy for Prostate Cancer

Background and objective

A combined approach of magnetic resonance imaging (MRI)-targeted biopsy (TBx) and bilateral systematic biopsy (SBx) is advised in patients who have an increased risk of prostate cancer (PCa). The diagnostic gain of SBx in detecting PCa for treatment planning of patients undergoing robot-assisted radical prostatectomy (RARP) is unknown. This study aims to determine the impact of omitting contralateral SBx on the surgical planning of patients undergoing RARP in terms of nerve-sparing surgery (NSS) and extended pelvic lymph node dissection (ePLND).

Methods

Case files from 80 men with biopsy-proven PCa were studied. All men had a unilateral suspicious lesion on MRI, and underwent TBx and bilateral SBx. Case files were presented to five urologists for the surgical planning of RARP. Each case file was presented randomly using two different sets of information: (1) results of TBx + bilateral SBx, and (2) results of TBx + ipsilateral SBx. The urologists assessed whether they would perform NSS and/or ePLND.

Key findings and limitations

A change in the surgical plan concerning NSS on the contralateral side was observed in 9.0% (95% confidence interval [CI] 6.4-12.2) of cases. Additionally, the indication for ePLND changed in 5.3% (95% CI 3.3-7.9) of cases. Interobserver agreement based on Fleiss' kappa changed from 0.44 to 0.15 for the indication of NSS and from 0.84 to 0.83 for the indication of ePLND.

Conclusions and clinical implications

In our series, the diagnostic information obtained from contralateral SBx has limited impact on the surgical planning of patients with a unilateral suspicious lesion on MRI scheduled to undergo RARP.

Patient summary

In patients with one-sided prostate cancer on magnetic resonance imaging, omitting biopsies on the other side rarely changed the surgical plan with respect to nerve-sparing surgery and the indication to perform extended lymph node dissection.

Variability of mpMRI diagnostic performance according to the upfront individual patient risk of having clinically significant prostate cancer

BACKGROUND

To assess the variation of multiparametric magnetic resonance imaging (mpMRI) positive predictive value (PPV) according to each patient's risk of clinically significant prostate cancer (csPCa) based exclusively on clinical factors.

METHODS

We evaluated 999 patients with positive mpMRI (PI-RADS ≥ 3) receiving targeted (TBx) plus systematic prostate biopsy. We built a multivariable logistic regression analysis (MVA) using clinical risk factors to calculate the individual patients' risk of harboring csPCa at TBx. A second MVA tested the association between individual patients' clinical risk and mpMRI PPV accounting for the PI-RADS score. Finally, we plotted the PPV of each PI-RADS score by the individual patient pretest probability of csPCa using a LOWESS approach.

RESULTS

Overall, TBx found csPCa in 21%, 51%, and 80% of patients with PI-RADS 3, 4, and 5 lesions, respectively. At MVA, age, PSA, digital rectal examination (DRE), and prostate volume were significantly associated with the risk of csPCa at biopsy. DRE yielded the highest odds ratio (OR: 2.88; p < 0.001). The individual patient's clinical risk was significantly associated with mpMRI PPV (OR: 2.49; p < 0.001) using MVA. Plotting the mpMRI PPV according to the predicted clinical risks, we observed that for patients with clinical risk close to 0 versus patients with risk higher than 90%, the mpMRI PPV of PI-RADS 3, 4, and 5 ranged from 0% to 75%, from 0% to 96%, and from 45% to 100%, respectively.

CONCLUSION

mpMRI PPV varies according to the individual pretest patient's risk based on clinical factors. These findings should be considered in the decision-making process for patients with suspect MRI findings referred for a prostate biopsy. Moreover, our data support the need for further studies to create an individualized risk prediction tool.

A Comparative Evaluation of Multiparametric Magnetic Resonance Imaging and Micro-Ultrasound for the Detection of Clinically Significant Prostate Cancer in Patients with Prior Negative Biopsies

BACKGROUND

The diagnostic process for prostate cancer after a negative biopsy is challenging. This study compares the diagnostic accuracy of micro-ultrasound (mUS) with multiparametric magnetic resonance imaging (mpMRI) for such cases.

METHODS

A retrospective cohort study was performed, targeting men with previous negative biopsies and using mUS and mpMRI to detect prostate cancer and clinically significant prostate cancer (csPCa).

RESULTS

In our cohort of 1397 men, 304 had a history of negative biopsies. mUS was more sensitive than mpMRI, with better predictive value for negative results. Importantly, mUS was significantly associated with csPCa detection (adjusted odds ratio [aOR]: 6.58; 95% confidence interval [CI]: 1.15-37.8; p = 0.035).

CONCLUSIONS

mUS may be preferable for diagnosing prostate cancer in previously biopsy-negative patients. However, the retrospective design of this study at a single institution suggests that further research across multiple centers is warranted.

Variability in prostate cancer detection among radiologists and urologists using MRI fusion biopsy

Objectives

The aim of this study is to evaluate the impact of radiologist and urologist variability on detection of prostate cancer (PCa) and clinically significant prostate cancer (csPCa) with magnetic resonance imaging (MRI)-transrectal ultrasound (TRUS) fusion prostate biopsies.

Patients and methods

The Prospective Loyola University MRI (PLUM) Prostate Biopsy Cohort (January 2015 to December 2020) was used to identify men receiving their first MRI and MRI/TRUS fusion biopsy for suspected PCa. Clinical, MRI and biopsy data were stratified by radiologist and urologist to evaluate variation in Prostate Imaging-Reporting and Data System (PI-RADS) grading, lesion number and cancer detection. Multivariable logistic regression (MVR) models and area under the curve (AUC) comparisons assessed the relative impact of individual radiologists and urologists.

Results

A total of 865 patients (469 biopsy-naïve) were included across 5 urologists and 10 radiologists. Radiologists varied with grading 15.4% to 44.8% of patients with MRI lesions as PI-RADS 3. PCa detection varied significantly by radiologist, from 34.5% to 66.7% (p = 0.003) for PCa and 17.2% to 50% (p = 0.001) for csPCa. Urologists' PCa diagnosis rates varied between 29.2% and 55.8% (p = 0.013) and between 24.6% and 39.8% (p = 0.36) for csPCa. After adjustment for case-mix on MVR, a fourfold to fivefold difference in PCa detection was observed between the highest-performing and lowest-performing radiologist (OR 0.22, 95%CI 0.10-0.47, p < 0.001). MVR demonstrated improved AUC for any PCa and csPCa detection when controlling for radiologist variation (p = 0.017 and p = 0.038), but controlling for urologist was not significant (p = 0.22 and p = 0.086). Any PCa detection (OR 1.64, 95%CI 1.06-2.55, p = 0.03) and csPCa detection (OR 1.57, 95%CI 1.00-2.48, p = 0.05) improved over time (2018-2020 vs. 2015-2017).

Conclusions

Variability among radiologists in PI-RADS grading is a key area for quality improvement significantly impacting the detection of PCa and csPCa. Variability for performance of MRI-TRUS fusion prostate biopsies exists by urologist but with less impact on overall detection of csPCa.

Active Surveillance for Prostate Cancer: Expanding the Role of MR Imaging and the Use of PRECISE Criteria

Multiparametric magnetic resonance (MR) imaging has had an expanding role in active surveillance (AS) for prostate cancer. It can improve the accuracy of prostate biopsies, assist in patient selection, and help monitor cancer progression. The PRECISE recommendations standardize reporting of serial MR imaging scans during AS. We summarize the evidence on MR imaging-led AS and provide a clinical primer to help report using the PRECISE criteria. Some limitations to both serial imaging and the PRECISE recommendations must be considered as we move toward a more individualized risk-stratified approach to AS.

Quantification and significance of extraprostatic findings on prostate MRI: a retrospective analysis and three-tier classification

OBJECTIVES

To quantify extraprostatic findings (EPFs) on prostate MRI, estimate the proportion of reported and unreported EPFs, assess their clinical importance, and propose standardized reporting of EPFs.

MATERIALS AND METHODS

Prostate 3-T MRI studies, reports, and clinical data from 623 patients (age 67.9 ± 8.2 years) were retrospectively analyzed and re-evaluated for the presence of EPFs and their clinical significance: E1-no finding or findings that have no clinical significance; E2-potentially significant findings; and E3-significant findings.

RESULTS

Secondary reading identified 1236 EPFs in 593 patients (1.98 ± 1.13 EPFs per patient, no EPFs in 30 patients), from which 468 (37.8%) were mentioned in the original report. The most common findings included diverticulosis (44% of patients), hydrocele (34%), inguinal fat hernia (16%), and bladder wall trabecular hypertrophy (15%). There were 80 (6.5%) E2 EPFs and 30 (2.4%) E3 EPFs. From E3 EPFs, 10 (33%) were not originally reported. A workup was suggested in 35 (52%) of the 67 originally reported E2 and E3 findings with follow-up and performed in 20 (30%). Fourteen (21%) EPFs in 11 patients influenced their management. Four experienced radiologists originally reported 1.8 to 2.5 findings per patient (p < 0.0001).

CONCLUSIONS

EPFs on prostate MRI are frequent, but only 2.4% are clinically significant (E3), and 33% of these are not reported. Only 30% of E2 and E3 findings are further explored, and 21% influence patient management. We suggest that an "E" category should be attached to the PI-RADS system to identify the presence of EPFs that require further workup.

CRITICAL RELEVANCE STATEMENT

Extraprostatic findings on prostate MRI are frequent, but only 2.4% are clinically significant (E3), and 33% of these are not reported. We advocate standardized reporting of extraprostatic findings indicating their clinical significance.

KEY POINTS

• Extraprostatic findings on prostate MRI are frequent with an average of two findings per patient. • 2.4% of extraprostatic findings are significant, and 33% of these are not reported. • There is a significant variability among experienced radiologists in reporting extraprostatic findings.

Inter-reader variability and reproducibility of the PI-QUAL score in a multicentre setting

PURPOSE

To assess the inter-reader reproducibility of the Prostate Imaging Quality (PI-QUAL) score between readers with varying clinical experience and its reproducibility at assessing imaging quality between different institutions.

METHODS

Following IRB approval, we assessed 60 consecutive prostate MRI scans performed at different academic teaching and non-academic hospitals uploaded to our institutes' PACS for second opinion or discussion in case conferences. Anonymized scans were independently reviewed using the PI-QUAL scoring sheet by three readers - two radiologists (with 1 and 12 years Prostate MRI reporting experience), and an experienced MRI technician with interest in image acquisition and quality. All readers were blinded to the site where scans were acquired.

RESULTS

Agreement coefficients between the 3 readers in paired comparison for each individual PI-QUAL score was moderate. When the scans were clustered into 2 groups according to their ability to rule in or rule out clinically significant prostate cancer [i.e., PI-QUAL score 1-3 vs PI-QUAL score 4-5], the Gwet AC1 coefficients between the three readers in paired comparison was good to very good [Gwet AC 1:0.77, 0.67, 0.836 respectively] with agreement percentage of 88.3%, 83.3% and 91.7% respectively. Agreement coefficient was higher between the experienced radiologist and the experienced MRI technician than between the less experienced trainee radiologist and the other two readers. The mean PI-QUAL score provided by each reader for the scans was significantly higher in the academic hospitals (n = 32) compared to the community hospital (n = 28) [experienced radiologist 4.6 vs 2.9; trainee radiologist 4.5 vs 2.4; experienced technologist 4.4 vs 2.4; p value < 0.001].

CONCLUSION

We observed good to very good reproducibility in the assessment of each MRI sequence and when scans were clustered into two groups [PI-QUAL 1-3 vs PI-QUAL 4-5] between readers with varying clinical experience. However, the reproducibility for each single PI-QUAL score between readers was moderate. Better definitions for each PI-QUAL score criteria may further improve reproducibility between readers. Additionally, the mean PI-QUAL score provided by all three readers was significantly higher for scans performed at academic teaching hospitals compared to community hospital.

MRI Radiomics-Based Machine Learning Models for Ki67 Expression and Gleason Grade Group Prediction in Prostate Cancer

PURPOSE

The Ki67 index and the Gleason grade group (GGG) are vital prognostic indicators of prostate cancer (PCa). This study investigated the value of biparametric magnetic resonance imaging (bpMRI) radiomics feature-based machine learning (ML) models in predicting the Ki67 index and GGG of PCa.

METHODS

A total of 122 patients with pathologically proven PCa who had undergone preoperative MRI were retrospectively included. Radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps. Then, recursive feature elimination (RFE) was applied to remove redundant features. ML models for predicting Ki67 expression and GGG were constructed based on bpMRI and different algorithms, including logistic regression (LR), support vector machine (SVM), random forest (RF), and K-nearest neighbor (KNN). The performances of different models were evaluated with receiver operating characteristic (ROC) analysis. In addition, a joint analysis of Ki67 expression and GGG was performed by assessing their Spearman correlation and calculating the diagnostic accuracy for both indices.

RESULTS

The ML model based on LR and ADC + T2 (LR_ADC + T2, AUC = 0.8882) performed best in predicting Ki67 expression, and ADC_wavelet-LHH_firstorder_Maximum had the highest feature weighting. The SVM_DWI + T2 (AUC = 0.9248) performed best in predicting GGG, and DWI_wavelet HLL_glcm_SumAverage had the highest feature weighting. The Ki67 and GGG exhibited a weak positive correlation (r = 0.382, p < 0.001), and LR_ADC + DWI had the highest diagnostic accuracy in predicting both (0.6230).

CONCLUSION

The proposed ML models are suitable for predicting both Ki67 expression and GGG in PCa. This algorithm could be used to identify indolent or invasive PCa with a noninvasive, repeatable, and accurate diagnostic method.

Role of Perilesional Sampling of Patients Undergoing Fusion Prostate Biopsies

Recently, researchers have proposed perilesional sampling during prostate biopsies to avoid systematic biopsies of patients at risk of prostate cancer. The aim of our study is to evaluate the role of perilesional sampling to avoid systematic biopsies of patients undergoing fusion biopsies. A prospective cohort of patients undergoing transrectal MRI transrectal fusion biopsies were consecutively enrolled. All the patients underwent systematic biopsies (SB), targeted biopsies (TB) and perilesional biopsies within 10 mm from the lesion (PB). The detection rates of different strategies were determined. A total of 262 patients were enrolled. The median age of those enrolled was 70 years. The mean BMI was 27 kg/m2, and the mean and prostate volume was 52 mL. A PIRADS score ≥ 4 was recorded in 163/262 (40%) patients. Overall, the detection rates of cancer were 43.5% (114/262) and 35% (92/262) for csPCa. The use of the target + peri-target strategy resulted in a detection of 32.8% (86/262) of cancer cases and of 29% (76/262) of csPCa cases (Grade Group > 2). Using the target plus peri-target approach resulted in us missing 18/262 (7%) of the csPCa cases, avoiding the diagnosis of 8/262 (3%) of nsPCa cases. A biopsy strategy including lesional and perilesional sampling could avoid unnecessary prostate biopsies. However, the risk of missing significant cancers is present. Future studies should assess the cost-benefit relationship of different strategies.

The role of AI in prostate MRI quality and interpretation: Opportunities and challenges

Prostate MRI plays an important role in imaging the prostate gland and surrounding tissues, particularly in the diagnosis and management of prostate cancer. With the widespread adoption of multiparametric magnetic resonance imaging in recent years, the concerns surrounding the variability of imaging quality have garnered increased attention. Several factors contribute to the inconsistency of image quality, such as acquisition parameters, scanner differences and interobserver variabilities. While efforts have been made to standardize image acquisition and interpretation via the development of systems, such as PI-RADS and PI-QUAL, the scoring systems still depend on the subjective experience and acumen of humans. Artificial intelligence (AI) has been increasingly used in many applications, including medical imaging, due to its ability to automate tasks and lower human error rates. These advantages have the potential to standardize the tasks of image interpretation and quality control of prostate MRI. Despite its potential, thorough validation is required before the implementation of AI in clinical practice. In this article, we explore the opportunities and challenges of AI, with a focus on the interpretation and quality of prostate MRI.

Research progress on deep learning in magnetic resonance imaging-based diagnosis and treatment of prostate cancer: a review on the current status and perspectives

Multiparametric magnetic resonance imaging (mpMRI) has emerged as a first-line screening and diagnostic tool for prostate cancer, aiding in treatment selection and noninvasive radiotherapy guidance. However, the manual interpretation of MRI data is challenging and time-consuming, which may impact sensitivity and specificity. With recent technological advances, artificial intelligence (AI) in the form of computer-aided diagnosis (CAD) based on MRI data has been applied to prostate cancer diagnosis and treatment. Among AI techniques, deep learning involving convolutional neural networks contributes to detection, segmentation, scoring, grading, and prognostic evaluation of prostate cancer. CAD systems have automatic operation, rapid processing, and accuracy, incorporating multiple sequences of multiparametric MRI data of the prostate gland into the deep learning model. Thus, they have become a research direction of great interest, especially in smart healthcare. This review highlights the current progress of deep learning technology in MRI-based diagnosis and treatment of prostate cancer. The key elements of deep learning-based MRI image processing in CAD systems and radiotherapy of prostate cancer are briefly described, making it understandable not only for radiologists but also for general physicians without specialized imaging interpretation training. Deep learning technology enables lesion identification, detection, and segmentation, grading and scoring of prostate cancer, and prediction of postoperative recurrence and prognostic outcomes. The diagnostic accuracy of deep learning can be improved by optimizing models and algorithms, expanding medical database resources, and combining multi-omics data and comprehensive analysis of various morphological data. Deep learning has the potential to become the key diagnostic method in prostate cancer diagnosis and treatment in the future.

PI-RADSAI: introducing a new human-in-the-loop AI model for prostate cancer diagnosis based on MRI

BACKGROUND

This study aims to develop and validate an artificial intelligence (AI)-aided Prostate Imaging Reporting and Data System (PI-RADSAI) for prostate cancer (PCa) diagnosis based on MRI.

METHODS

The deidentified MRI data of 1540 biopsy-naïve patients were collected from four centres. PI-RADSAI is a two-stage, human-in-the-loop AI capable of emulating the diagnostic acumen of subspecialists for PCa on MRI. The first stage uses a UNet-Seg model to detect and segment biopsy-candidate prostate lesions, whereas the second stage leverages UNet-Seg segmentation is trained specifically with subspecialist' knowledge-guided 3D-Resnet to achieve an automatic AI-aided diagnosis for PCa.

RESULTS

In the independent test set, UNet-Seg identified 87.2% (628/720) of target lesions, with a Dice score of 44.9% (range, 22.8-60.2%) in segmenting lesion contours. In the ablation experiment, the model trained with the data from three centres was superior (kappa coefficient, 0.716 vs. 0.531) to that trained with single-centre data. In the internal and external tests, the triple-centre PI-RADSAI model achieved an overall agreement of 58.4% (188/322) and 60.1% (92/153) with a referential subspecialist in scoring target lesions; when one-point margin of error was permissible, the agreement rose to 91.3% (294/322) and 97.3% (149/153), respectively. In the paired test, PI-RADSAI outperformed 5/11 (45.5%) and matched the performance of 3/11 (27.3%) general radiologists in achieving a clinically significant PCa diagnosis (area under the curve, internal test, 0.801 vs. 0.770, p < 0.01; external test, 0.833 vs. 0.867, p = 0.309).

CONCLUSIONS

Our closed-loop PI-RADSAI outperforms or matches the performance of more than 70% of general readers in the MRI assessment of PCa. This system might provide an alternative to radiologists and offer diagnostic benefits to clinical practice, especially where subspecialist expertise is unavailable.

Characterization of high-grade prostate cancer at multiparametric MRI: assessment of PI-RADS version 2.1 and version 2 descriptors across 21 readers with varying experience (MULTI study)

OBJECTIVE

To assess PI-RADSv2.1 and PI-RADSv2 descriptors across readers with varying experience.

METHODS

Twenty-one radiologists (7 experienced (≥ 5 years) seniors, 7 less experienced seniors and 7 juniors) assessed 240 'predefined' lesions from 159 pre-biopsy multiparametric prostate MRIs. They specified their location (peripheral, transition or central zone) and size, and scored them using PI-RADSv2.1 and PI-RADSv2 descriptors. They also described and scored 'additional' lesions if needed. Per-lesion analysis assessed the 'predefined' lesions, using targeted biopsy as reference; per-lobe analysis included 'predefined' and 'additional' lesions, using combined systematic and targeted biopsy as reference. Areas under the curve (AUCs) quantified the performance in diagnosing clinically significant cancer (csPCa; ISUP ≥ 2 cancer). Kappa coefficients (κ) or concordance correlation coefficients (CCC) assessed inter-reader agreement.

RESULTS

At per-lesion analysis, inter-reader agreement on location and size was moderate-to-good (κ = 0.60-0.73) and excellent (CCC ≥ 0.80), respectively. Agreement on PI-RADSv2.1 scoring was moderate (κ = 0.43-0.47) for seniors and fair (κ = 0.39) for juniors. Using PI-RADSv2.1, juniors obtained a significantly lower AUC (0.74; 95% confidence interval [95%CI]: 0.70-0.79) than experienced seniors (0.80; 95%CI 0.76-0.84; p = 0.008) but not than less experienced seniors (0.74; 95%CI 0.70-0.78; p = 0.75). As compared to PI-RADSv2, PI-RADSv2.1 downgraded 17 lesions/reader (interquartile range [IQR]: 6-29), of which 2 (IQR: 1-3) were csPCa; it upgraded 4 lesions/reader (IQR: 2-7), of which 1 (IQR: 0-2) was csPCa. Per-lobe analysis, which included 60 (IQR: 25-73) 'additional' lesions/reader, yielded similar results.

CONCLUSIONS

Experience significantly impacted lesion characterization using PI-RADSv2.1 descriptors. As compared to PI-RADSv2, PI-RADSv2.1 tended to downgrade non-csPCa lesions, but this effect was small and variable across readers.

Comparison of prostate cancer detection rate at targeted biopsy of hub and spoke centers mpMRI: experience matters

BACKGROUND

Latest changes in European guidelines on prostate cancer determined a widespread of multiparametric magnetic resonance imaging (mpMRI) even in less experienced centers due to an increased demand. This could decrease diagnostic accuracy of targeted biopsy (TB) since image interpretation can be challenging and requires adequate and supervised training. Therefore we aimed to evaluate the prostate cancer (PCa) detection rate on TB according to mpMRI center's volume and experience.

METHODS

We retrospectively analyzed data of 737 patients who underwent mpMRI-TB at our institution. Patients were stratified according to mpMRI center: Hub (high volume >100 exams/year with dedicated radiologists and supervised training) and Spoke center (low volume <100 exams/year without dedicated radiologists and/or supervised training). Detection rate of PCa at TB and possible predictors of clinically significant PCa (csPCa) at TB. Differences in detection rate were explored using Chi-square test. Predictors of csPCa were evaluated through uni and multivariable logistic regression. The adjustment for casemix included: age, PSA, mpMRI center, lesion's location, PSA density, PI-RADS score and index lesion's size.

RESULTS

Four hundred forty-nine (60.9%) and 288 (39.1%) patients underwent mpMRI at a Hub or Spoke center, respectively. Hub group had higher detection rate for both any (60.3% vs. 48.1%) and csPCa (46.9% vs 38.7%; all P≤0.001). After stratifying for PI-RADS score, Hub group had higher detection rate for PI-RADS score 3 (csPCA 25.2% vs. 15.7%; p 0.04) and 4 (csPCa 65.7% vs. 45.7%; P=0.001). At multivariable analyses, receiving an mpMRI scan at a Spoke center was an independent predictor for csPCa on TB (OR 0.65; P=0.04).

CONCLUSIONS

mpMRI performed in Hub centers provided a significantly higher PCa yield on TB. A dedicated team of experienced radiologist, a supervised training for mpMRI and a central revision of mpMRI performed in non-experienced centres are essential to avoid unnecessary and potentially harmful procedures.

Investigating PSMA-PET/CT to resolve prostate MRI PIRADS4-5 and negative biopsy discordance

BACKGROUND

To determine the utility of diagnostic 18F-DCPyL PSMA-PET/CT to aid management of men with highly suspicious multiparametric MRI prostate (PIRAD 4-5 lesions) and discrepant negative prostate biopsy.

METHODS

A multicentre prospective consecutive case series was conducted (2018-2021), recruiting men with prior mpMRI prostate PIRADS 4-5 lesions and negative prostate biopsy. All men had 18F-DCPyL PSMA-PET/CT with subsequent management based on the concordance between MRI and PET: (1) Concordant lesions were biopsied using in-bore MRI targeting; (2) PSMA-PET/CT avidity without MRI correlate were biopsied using cognitive/software targeting with ultrasound guidance and (3) Patients with negative PET/CT were returned to standard of care follow-up.

RESULTS

29 patients were recruited with 48% (n = 14) having concordant MRI/PET abnormalities. MRI targeted biopsy found prostate cancer in six patients, with grade groups GG3 (n = 1), GG2 (n = 1), GG1 (n = 4) found. Of the 20 men who PSMA-PET/CT avidity and biopsy, analysis showed higher SUVmax (20.1 vs 6.8, p = 0.036) predicted prostate cancer. Of patients who had PSMA-PET avidity without MRI correlate, and those with no PSMA-PET avidity, only one patient was subsequently found to have prostate cancer (GG1). The study is limited by small size and short follow-up of 17 months (IQR 12.5-29.9).

CONCLUSIONS

PSMA-PET/CT is useful in this group of men but requires further investigation. Avidity (higher SUVmax) that correlates to the mpMRI prostate lesion should be considered for targeted biopsy.

The diagnostic performance in clinically significant prostate cancer with PI-RADS version 2.1: simplified bpMRI versus standard mpMRI

OBJECTIVES

To compare the diagnostic performance for the detection of clinically significant prostate cancer (csPCa) between bpMRI with only axial T2WI (simplified bpMRI) and standard-multiparametric MRI (mpMRI).

METHODS

A total of 569 patients who underwent mpMRI followed by biopsy or prostatectomy were enrolled in this retrospective study. According to PI-RADS v2.1, three radiologists (A, B, C) from three centers blinded to clinical variables were assigned scores on lesions with simplified bpMRI and then with mpMRI 2 weeks later. Diagnostic performance of simplified bpMRI was compared with mpMRI using histopathology as reference standard.

RESULTS

For all the three radiologists, the diagnostic sensitivity was significantly higher with mpMRI than with simplified bpMRI (P < 0.001 to P = 0.035); and although specificity was also higher with mpMRI than with simplified bpMRI for radiologist B and radiologist C, it was statistically significant only for radiologist B (P = 0.011, P = 0.359, respectively). On the contrary, for radiologist A, specificity was higher with simplified bpMRI than with mpMRI (P = 0.001). The area under the receiver operating characteristic curve (AUC) was significantly higher for mpMRI than for simplified bpMRI except for radiologist A (radiologist A: 0.903 vs 0.913, P = 0.1542; radiologist B: 0.861 vs 0.834 P = 0.0013; and radiologist C: 0.884 vs 0.848, P = 0.0003). Interobserver reliability of PI-RADS v2.1 showed good agreement for both simplified bpMRI (kappa = 0.665) and mpMRI (kappa = 0.739).

CONCLUSION

Although the detection of csPCa with simplified bpMRI was comparatively lower than that with mpMRI, the diagnostic performance was still high in simplified bpMRI. Our data justify using mpMRI outperforms simplified bpMRI for prostate cancer screening and imply simplified bpMRI as a potential screening tool.

Automated prostate multi-regional segmentation in magnetic resonance using fully convolutional neural networks

OBJECTIVE

Automatic MR imaging segmentation of the prostate provides relevant clinical benefits for prostate cancer evaluation such as calculation of automated PSA density and other critical imaging biomarkers. Further, automated T2-weighted image segmentation of central-transition zone (CZ-TZ), peripheral zone (PZ), and seminal vesicle (SV) can help to evaluate clinically significant cancer following the PI-RADS v2.1 guidelines. Therefore, the main objective of this work was to develop a robust and reproducible CNN-based automatic prostate multi-regional segmentation model using an intercontinental cohort of prostate MRI.

METHODS

A heterogeneous database of 243 T2-weighted prostate studies from 7 countries and 10 machines of 3 different vendors, with the CZ-TZ, PZ, and SV regions manually delineated by two experienced radiologists (ground truth), was used to train (n = 123) and test (n = 120) a U-Net-based model with deep supervision using a cyclical learning rate. The performance of the model was evaluated by means of dice similarity coefficient (DSC), among others. Segmentation results with a DSC above 0.7 were considered accurate.

RESULTS

The proposed method obtained a DSC of 0.88 ± 0.01, 0.85 ± 0.02, 0.72 ± 0.02, and 0.72 ± 0.02 for the prostate gland, CZ-TZ, PZ, and SV respectively in the 120 studies of the test set when comparing the predicted segmentations with the ground truth. No statistically significant differences were found in the results obtained between manufacturers or continents.

CONCLUSION

Prostate multi-regional T2-weighted MR images automatic segmentation can be accurately achieved by U-Net like CNN, generalizable in a highly variable clinical environment with different equipment, acquisition configurations, and population.

KEY POINTS

• Deep learning techniques allows the accurate segmentation of the prostate in three different regions on MR T2w images. • Multi-centric database proved the generalization of the CNN model on different institutions across different continents. • CNN models can be used to aid on the diagnosis and follow-up of patients with prostate cancer.

Radiologists' Contribution to Variation in Detecting Clinically Significant Prostate Cancer in Men With Prostate MRI

OBJECTIVE

Assess radiologists' contribution to variation in clinically significant prostate cancer (csPCa) detection in patients with elevated prostate-specific antigen (PSA) and multiparametric MRI (mpMRI).

METHODS

This institutional review board-approved, retrospective cohort study was performed at a tertiary, academic, National Cancer Institute-designated Comprehensive Cancer Center with a multidisciplinary prostate cancer program. Men undergoing mpMRI examinations from January 1, 2015, to December 31, 2019, with elevated PSA (≥4 ng/mL) and biopsy within 6 months pre- or post-MRI or prostatectomy within 6 months post-mpMRI were included. Univariate and multivariable hierarchical logistic regression assessed impact of patient, provider, mpMRI examination, mpMRI report, and pathology factors on the diagnosis of Grade Group ≥ 2 csPCa.

RESULTS

Study cohort included 960 MRIs in 928 men, mean age 64.0 years (SD ± 7.4), and 59.8% (555 of 928) had csPCa. Interpreting radiologist was not significant individually (P > .999) or combined with mpMRI ordering physician and physician performing biopsy or prostatectomy (P = .41). Prostate Imaging Reporting and Data System (PI-RADS) category 2 (odds ratio [OR] 0.18, P = .04), PI-RADS category 4 (OR 2.52, P < .001), and PI-RADS category 5 (OR 4.99, P < .001) assessment compared with no focal lesion; PSA density of 0.1 to 0.15 ng/mL/cc (OR 2.46, P < .001), 0.15 to 0.2 ng/mL/cc (OR 2.77, P < .001), or ≥0.2 ng/mL/cc (OR 4.52, P < .001); private insurance (reference = Medicare, OR 0.52, P = .001), and unambiguous extraprostatic extension on mpMRI (OR 2.94, P = .01) were independently associated with csPCa. PI-RADS 3 assessment (OR 1.18, P = .56), age (OR 0.99, P = .39), and African American race (OR 0.90, P = .75) were not.

DISCUSSION

Although there is known in-practice variation in radiologists' interpretation of mpMRI, in our multidisciplinary prostate cancer program we found no significant radiologist-attributable variation in csPCa detection.

68Ga-PSMA-11 PET/MRI versus multiparametric MRI in men referred for prostate biopsy: primary tumour localization and interreader agreement

Background

Magnetic resonance imaging (MRI) is recommended by the European Urology Association guidelines as the standard modality for imaging-guided biopsy. Recently positron emission tomography with prostate-specific membrane antigen (PSMA PET) has shown promising results as a tool for this purpose. The aim of this study was to compare the accuracy of positron emission tomography with prostate-specific membrane antigen/magnetic resonance imaging (PET/MRI) using the gallium-labeled prostate-specific membrane antigen (⁶⁸Ga-PSMA-11) and multiparametric MRI (mpMRI) for pre-biopsy tumour localization and interreader agreement for visual and semiquantitative analysis. Semiquantitative parameters included apparent diffusion coefficient (ADC) and maximum lesion diameter for mpMRI and standardized uptake value (SUV_max) and PSMA-positive volume (PSMA_vol) for PSMA PET/MRI.

Results

Sensitivity and specificity were 61.4% and 92.9% for mpMRI and 66.7% and 92.9% for PSMA PET/MRI for reader one, respectively. RPE was available in 23 patients and 41 of 47 quadrants with discrepant findings. Based on RPE results, the specificity for both imaging modalities increased to 98% and 99%, and the sensitivity improved to 63.9% and 72.1% for mpMRI and PSMA PET/MRI, respectively. Both modalities yielded a substantial interreader agreement for primary tumour localization (mpMRI kappa = 0.65 (0.52–0.79), PSMA PET/MRI kappa = 0.73 (0.61–0.84)). ICC for SUV_max, PSMA_vol and lesion diameter were almost perfect (≥ 0.90) while for ADC it was only moderate (ICC = 0.54 (0.04–0.78)). ADC and lesion diameter did not correlate significantly with Gleason score (ρ = 0.26 and ρ = 0.16) while SUV_max and PSMA_vol did (ρ =  − 0.474 and ρ =  − 0.468).

Conclusions

PSMA PET/MRI has similar accuracy and reliability to mpMRI regarding primary prostate cancer (PCa) localization. In our cohort, semiquantitative parameters from PSMA PET/MRI correlated with tumour grade and were more reliable than the ones from mpMRI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41824-022-00135-4.

Inter-reader agreement of the prostate imaging reporting and data system version v2.1 for detection of prostate cancer: A systematic review and meta-analysis

Objectives

We aimed to systematically assess the inter-reader agreement of the Prostate Imaging Reporting and Data System Version (PI-RADS) v2.1 for the detection of prostate cancer (PCa).

Methods

We included studies reporting inter-reader agreement of different radiologists that applied PI-RADS v2.1 for the detection of PCa. Quality assessment of the included studies was performed with the Guidelines for Reporting Reliability and Agreement Studies. The summary estimates of the inter-reader agreement were pooled with the random-effect model and categorized (from slight to almost perfect) according to the kappa (κ) value. Multiple subgroup analyses and meta-regression were performed to explore various clinical settings.

Results

A total of 12 studies comprising 2475 patients were included. The pooled inter-reader agreement for whole gland was κ=0.65 (95% CI 0.56-0.73), and for transitional zone (TZ) lesions was κ=0.62 (95% CI 0.51-0.72). There was substantial heterogeneity presented throughout the studies (I 2= 95.6%), and meta-regression analyses revealed that only readers' experience (<5 years vs. ≥5 years) was the significant factor associated with heterogeneity (P<0.01). In studies providing head-to-head comparison, there was no significant difference in inter-reader agreement between PI-RADS v2.1 and v2.0 for both the whole gland (0.64 vs. 0.57, p=0.37), and TZ (0.61 vs. 0.59, p=0.81).

Conclusions

PI-RADS v2.1 demonstrated substantial inter-reader agreement among radiologists for whole gland and TZ lesions. However, the difference in agreement between PI-RADS v2.0 and v2.1 was not significant for the whole gland or the TZ.

Inter-Reader Variability Using PI-RADS v2 Versus PI-RADS v2.1: Most New Disagreement Stems from Scores 1 and 2

PURPOSE

To analyze possible differences in the inter-reader variability between PI-RADS version 2 (v2) and version 2.1 (v2.1) for the classification of prostate lesions using multiparametric MRI (mpMRI) of the prostate.

METHODS

In this retrospective and randomized study, 239 annotated and histopathologically correlated prostate lesions (104 positive and 135 negative for prostate cancer) were rated twice by three experienced uroradiologists using PI-RADS v2 and v2.1 with an interval of at least two months between readings. Results were tabulated across readers and reading timepoints and inter-reader variability was determined using Fleiss' kappa (κ). Thereafter, an additional analysis of the data was performed in which PI-RADS scores 1 and 2 were combined, as they have the same clinical consequences.

RESULTS

PI-PI-RADS v2.1 showed better inter-reader agreement in the peripheral zone (PZ), but poorer inter-reader agreement in the transition zone (TZ) (PZ: κ = 0.63 vs. κ = 0.58; TZ: κ = 0.47 vs. κ = 0.57). When PI-RADS scores 1 and 2 were combined, the use of PI-RADS v2.1 resulted in almost perfect inter-reader agreement in the PZ and substantial agreement in the TZ (PZ: κ = 0.81; TZ: κ = 0.80).

CONCLUSION

PI-RADS v2.1 improves inter-reader agreement in the PZ. New differences in inter-reader agreement were mainly the result of the assignment of PI-RADS v2.1 scores 1 and 2 to lesions in the TZ. Combining scores 1 and 2 improved inter-reader agreement both in the TZ and in the PZ, indicating that refined definitions may be warranted for these PI-RADS scores.

KEY POINTS

· PI-RADSv2.1 improves inter-reader agreement in the PZ but not in the TZ.. · New differences derived from PI-RADSv2.1 scores 1 and 2 in the TZ.. · Combined PI-RADSv2.1 scores of 1 and 2 yielded better inter-reader agreement.. · PI-RADSv2.1 appears to provide more precise description of lesions in the PZ.. · Improved inter-reader agreement in the PZ stresses the importance of appropriate lexicon description..

CITATION FORMAT

· Beetz N, Haas M, Baur A et al. Inter-Reader Variability Using PI-RADS v2 Versus PI-RADS v2.1: Most New Disagreement Stems from Scores 1 and 2. Fortschr Röntgenstr 2022; 194: 852 - 861.

Promoting the use of the PRECISE score for prostate MRI during active surveillance: results from the ESOR Nicholas Gourtsoyiannis teaching fellowship

Objectives

The PRECISE criteria for serial multiparametric magnetic resonance imaging (MRI) of the prostate during active surveillance recommend the use of a dedicated scoring system (PRECISE score) to assess the likelihood of clinically significant radiological change. This pilot study assesses the effect of an interactive teaching course on prostate MRI during active surveillance in assessing radiological change in serial imaging.

Methods

Eleven radiology fellows and registrars with different experience in prostate MRI reading participated in a dedicated teaching course where they initially evaluated radiological change (based on their previous training in prostate MRI reading) independently in fifteen patients on active surveillance (baseline and follow-up scan), and then attended a lecture on the PRECISE score. The initial scans were reviewed for teaching purposes and afterwards the participants re-assessed the degree of radiological change in a new set of images (from fifteen different patients) applying the PRECISE score. Receiver operating characteristic analysis was performed. Confirmatory biopsies and PRECISE scores given in consensus by two radiologists (involved in the original draft of the PRECISE score) were the reference standard.

Results

There was a significant improvement in the average area under the curve (AUC) for the assessment of radiological change from baseline (AUC: 0.60 [Confidence Intervals: 0.51–0.69] to post-teaching (AUC: 0.77 [0.70–0.84]). This was an improvement of 0.17 [0.016–0.28] (p = 0.004).

Conclusions

A dedicated teaching course on the use of the PRECISE score improves the accuracy in the assessment of radiological change in serial MRI of the prostate.

Radiomics-Based Machine Learning Models for Predicting P504s/P63 Immunohistochemical Expression: A Noninvasive Diagnostic Tool for Prostate Cancer

Objective

To develop and validate a noninvasive radiomic-based machine learning (ML) model to identify P504s/P63 status and further achieve the diagnosis of prostate cancer (PCa).

Methods

A retrospective dataset of patients with preoperative prostate MRI examination and P504s/P63 pathological immunohistochemical results between June 2016 and February 2021 was conducted. As indicated by P504s/P63 expression, the patients were divided into label 0 (atypical prostatic hyperplasia), label 1 (benign prostatic hyperplasia, BPH) and label 2 (PCa) groups. This study employed T2WI, DWI and ADC sequences to assess prostate diseases and manually segmented regions of interest (ROIs) with Artificial Intelligence Kit software for radiomics feature acquisition. Feature dimensionality reduction and selection were performed by using a mutual information algorithm. Based on screened features, P504s/P63 prediction models were established by random forest (RF), gradient boosting decision tree (GBDT), logistic regression (LR), adaptive boosting (AdaBoost) and k-nearest neighbor (KNN) algorithms. The performance was evaluated by the area under the ROC curve (AUC) and accuracy.

Results

A total of 315 patients were enrolled. Among the 851 radiomic features, the 32 top features were derived from T2WI, in which the gray-level run length matrix (GLRLM) and gray-level cooccurrence matrix (GLCM) features accounted for the largest proportion. Among the five models, the RF algorithm performed best in general evaluations (microaverage AUC=0.920, macroaverage AUC=0.870) and provided the most accurate result in further sublabel prediction (the accuracies of label 0, 1, and 2 were 0.831, 0.831, and 0.932, respectively). In comparative sequence analyses, T2WI was the best single-sequence candidate (microaverage AUC=0.94 and macroaverage AUC=0.78). The merged datasets of T2WI, DWI, and ADC yielded optimal AUCs (microaverage AUC=0.930 and macroaverage AUC=0.900).

Conclusions

The radiomic-based RF classifier has the potential to be used to evaluate the presurgical P504s/P63 status and further diagnose PCa noninvasively and accurately.

Predictors of clinically significant prostate cancer in biopsy-naïve and prior negative biopsy men with a negative prostate MRI: improving MRI-based screening with a novel risk calculator

Purpose:

A pre-biopsy decision aid is needed to counsel men with a clinical suspicion for clinically significant prostate cancer (csPCa), despite normal prostate magnetic resonance imaging (MRI).

Methods:

A risk calculator (RC) for csPCa (International Society of Urological Pathology grade group (ISUP) ⩾ 2) presence in men with a negative-MRI (Prostate Imaging–Reporting and Data System (PI-RADS) ⩽ 2) was developed, and its performance was compared with RCs of the European Randomized Study of Screening for Prostate Cancer (ERSPC), Prostate Biopsy Collaborative Group (PBCG), and Prospective Loyola University mpMRI (PLUM). All biopsy-naïve and prior negative biopsy men with a negative-MRI followed by systematic prostate biopsy were included from October 2015 to September 2021. The RC was developed using multivariable logistic regression with the following parameters: age (years), family history of PCa (first- or second-degree family member), ancestry (African Caribbean/other), digital rectal exam (benign/malignant), MRI field strength (1.5/3.0 Tesla), prior negative biopsy status, and prostate-specific antigen (PSA) density (ng/ml/cc). Performance of RCs was compared using receiver operating characteristic (ROC) curve analysis.

Results:

A total of 232 men were included for analysis, of which 18.1% had csPCa. Parameters associated with csPCa were family history of PCa (p < 0.0001), African Caribbean ancestry (p = 0.005), PSA density (p = 0.002), prior negative biopsy (p = 0.06), and age at biopsy (p = 0.157). The area under the curve (AUC) of the developed RC was 0.76 (95% CI 0.68–0.85). This was significantly better than the RCs of the ERSPC (AUC: 0.59; p = 0.001) and PBCG (AUC: 0.60; p = 0.002), yet similar to PLUM (AUC: 0.69; p = 0.09).

Conclusion:

The developed RC (Prostate Biopsy Cohort Amsterdam (‘PROBA’ RC), integrated predictors for csPCa at prostate biopsy in negative-MRI men and outperformed other widely used RCs. These findings require external validation before introduction in daily practice.

Comparison of prostate imaging reporting and data system v2.1 and 2 in transition and peripheral zones: evaluation of interreader agreement and diagnostic performance in detecting clinically significant prostate cancer

OBJECTIVE

To evaluate the interreader agreement and diagnostic performance of the Prostate Imaging Reporting and Data System (PI-RADS) v. 2.1, in comparison with v. 2.

METHODS

Institutional review board approval was obtained for this retrospective study. 77 consecutive patients who underwent a prostate multiparametric magnetic resonance imaging at 3.0 T before radical prostatectomy were included. Four radiologists (two experienced uroradiologists and two inexperienced radiologists) independently scored eight regions [six peripheral zones (PZ) and two transition zones (TZ)] using v. 2.1 and v. 2. Interreader agreement was assessed using κ statistics. To evaluate diagnostic performance for clinically significant prostate cancer (csPC), area under the curve (AUC) was estimated.

RESULTS

228 regions were pathologically diagnosed as positive for csPC. With a cut-off ≥3, the agreement among all readers was better with v. 2.1 than v. 2 in TZ, PZ, or both zones combined (κ-value: TZ, 0.509 vs 0.414; PZ, 0.686 vs 0.568; both zones combined, 0.644 vs 0.531). With a cut-off ≥4, the agreement among all readers was also better with v. 2.1 than v. 2 in the PZ or both zones combined (κ-value: PZ, 0.761 vs 0.701; both zones combined, 0.756 vs 0.709). For all readers, AUC with v. 2.1 was higher than with v. 2 (TZ, 0.826-0.907 vs 0.788-0.856; PZ, 0.857-0.919 vs 0.853-0.902).

CONCLUSION

Our study suggests that the PI-RADS v. 2.1 could improve the interreader agreement and might contribute to improved diagnostic performance compared with v. 2.

ADVANCES IN KNOWLEDGE

PI-RADS v. 2.1 has a potential to improve interreader variability and diagnostic performance among radiologists with different levels of expertise.

Comparison of diagnostic performance and inter-reader agreement between PI-RADS v2.1 and PI-RADS v2: systematic review and meta-analysis

OBJECTIVES

To perform a systematic review and meta-analysis comparing diagnostic performance and inter reader agreement between PI-RADS v. 2.1 and PI-RADS v. 2 in the detection of clinically significant prostate cancer (csPCa).

METHODS

A systematic review was performed, searching the major biomedical databases (Medline, Embase, Scopus), using the keywords "PIRADS 2.1" or "PI RADS 2.1" or "PI-RADS 2.1". Studies reporting on head-to-head diagnostic comparison between PI-RADS v. 2.1 and v. 2 were included. Pooled sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were compared between PI-RADS v. 2.1 and v. 2. Summary receiver operator characteristic graphs were plotted. Analysis was performed for whole gland, and pre-planned subgroup analysis was performed by tumour location (whole gland vs transition zone (TZ)), high b-value DWI (b-value ≥1400 s/mm2), and reader experience (<5 years vs ≥5 years with prostate MRI interpretation). Inter-reader agreement and pooled rates of csPCa for PI-RADS 1-3 lesions were compared between PI-RADS v. 2.1 and v. 2. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool v. 2 (QUADAS-2).

RESULTS

Eight studies (1836 patients, 1921 lesions) were included. Pooled specificity for PI-RADS v. 2.1 was significantly lower than PI-RADS v. 2 for whole gland (0.62 vs 0.66, p = 0.02). Pooled sensitivities, PPVs and NPVs were not significantly different (p = 0.17, 0.31, 0.41). Pooled specificity for PI-RADS v. 2.1 was significantly lower than PI-RADS v. 2 for TZ only (0.67 vs 0.72, p = 0.01). Pooled sensitivities, PPVs and NPVs were not significantly different (p = 0.06, 0.36, 0.17). Amongst studies utilising diffusion-weighted imaging with highest b-value of ≥1400 s/mm2, pooled sensitivities, specificities, PPVs and NPVs were not significantly different (p = 0.52, 0.4, 0.5, 0.47). There were no significant differences in pooled sensitivities, specificities, PPVs and NPVs between PI-RADS v. 2.1 and PI-RADS v. 2 for less-experienced readers (p = 0.65, 0.37, 0.65, 0.81) and for more experienced readers (p = 0.57, 0.90, 0.91, 0.65). For PI-RADS v. 2.1 alone, there were no significant differences in pooled sensitivity, specificity, PPV and NPV between less and more experienced readers (p = 0.38, 0.70, 1, 0.48). Inter-reader agreement was moderate to substantial for both PI-RADS v. 2.1 and v. 2. There were no significant differences between pooled csPCa rates between PI-RADS v. 2.1 and v. 2 for PI-RADS 1-2 lesions (6.6% vs  7.3%, p = 0.53), or PI-RADS 3 lesions (24.1% vs  26.8%, p = 0.28).

CONCLUSIONS

Diagnostic performance and inter-reader agreement for PI-RADS v. 2.1 is comparable to PI-RADS v. 2, however the significantly lower specificity of PI-RADS v. 2.1 may result in increased number of unnecessary biopsies.

ADVANCES IN KNOWLEDGE

1. Compared to PI-RADS v. 2, PI-RADS v. 2.1 has a non-significantly higher sensitivity but a significantly lower specificity for detection of clinically significant prostate cancer.2. PI-RADS v. 2.1 could potentially result in considerable increase in number of negative targeted biopsy rates for PI-RADS 3 lesions, which could have been potentially avoided.

MRI as a screening tool for prostate cancer: current evidence and future challenges

Purpose

Prostate cancer (PCa) screening, which relies on prostate-specific antigen (PSA) testing, is a contentious topic that received negative attention due to the low sensitivity and specificity of PSA to detect clinically significant PCa. In this context, due to the higher sensitivity and specificity of magnetic resonance imaging (MRI), several trials investigate the feasibility of “MRI-only” screening approaches, and question if PSA testing may be replaced within prostate cancer screening programs.

Methods

This narrative review discusses the current literature and the outlook on the potential of MRI-based PCa screening.

Results

Several prospective randomized population-based trials are ongoing. Preliminary study results appear to favor the “MRI-only” approach. However, MRI-based PCa screening programs face a variety of obstacles that have yet to be fully addressed. These include the increased cost of MRI, lack of broad availability, differences in MRI acquisition and interpretation protocols, and lack of long-term impact on cancer-specific mortality. Partly, these issues are being addressed by shorter and simpler MRI approaches (5–20 min bi-parametric MRI), novel quality indicators (PI-QUAL) and the implementation of radiomics (deep learning, machine learning).

Conclusion

Although promising preliminary results were reported, MRI-based PCa screening still lack long-term data on crucial endpoints such as the impact of MRI screening on mortality. Furthermore, the issues of availability, cost-effectiveness, and differences in MRI acquisition and interpretation still need to be addressed.

Interobserver Agreement and Accuracy in Interpreting mpMRI of the Prostate: a Systematic Review

PURPOSE OF REVIEW

To present the latest evidence related to interobserver agreement and accuracy; evaluate the strengths, weaknesses, and implications of use; and outline opportunities for improvement and future development of the Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) for detection of prostate cancer (PCa) on multiparametric magnetic resonance imaging (mpMRI).

RECENT FINDINGS

Our review of currently available evidence suggests that recent improvements to the PI-RADS system with PI-RADS v2.1 slightly improved interobserver agreement, with generally high sensitivity and moderate specificity for the detection of clinically significant PCa. Recent evidence additionally demonstrates substantial improvement in diagnostic specificity with PI-RADS v2.1 compared with PI-RADS v2. However, results of studies examining the comparative performance of v2.1 are limited by small sample sizes and retrospective cohorts, potentially introducing selection bias. Some studies suggest a substantial improvement between v2.1 and v2, while others report no statistically significant difference. Additionally, in PI-RADS v2.1, the interpretation and reporting of certain findings remain subjective, particularly for category 2 lesions, and reader experience continues to vary significantly. These factors further contribute to a remaining degree of interobserver variability and findings of improved performance among more experienced readers. PI-RADS v2.1 appears to show at least minimal improvement in interobserver agreement, diagnostic performance, and both sensitivity and specificity, with greater improvements seen among more experienced readers. However, given the decrescent nature of these improvements and the limited power of all studies examined, the clinical impact of this progress may be marginal. Despite improvements in PI-RADS v2.1, practitioner experience in interpreting mpMRI of the prostate remains the most important factor in prostate cancer detection.

Second-opinion reads in prostate MRI: added value of subspecialty interpretation and review at multidisciplinary rounds

Purpose

This study evaluates the impact of second-opinion review of multiparametric prostate MRI for cancer detection by a multidisciplinary review board at a tertiary care center when compared with the initial community radiologist interpretation.

Methods

Cases were collected retrospectively from multidisciplinary prostate MRI rounds from 2017 to 2020 at a single tertiary care center. Patients with suspected prostate cancer or on active surveillance were referred for consideration of TRUS/MRI-fusion biopsy based on community-read prostate MRIs. All MRIs were re-read by subspecialized abdominal radiologists and a PI-RADS score assigned. Targeted fusion and 8–12 core systematic biopsy was performed in patients with PIRADS ≥ 3 lesions. Cohen kappa values were used to quantify interobserver agreement. Positive predictive value (PPV) was used to determine accuracy of PI-RADS score for detection of clinically significant prostate cancer (csPCa) (ISUP Grade Group ≥ 2).

Results

Three hundred and thirty-two lesions in 303 patients were reviewed and 252 lesions in 198 patients biopsied. The PI-RADS score was concordant in 60.5% of lesions, downgraded in 17.8%, and upgraded in 7.8%. Agreement between community and tertiary center interpretation was fair (κ = 0.354), with greater agreement for PI-RADS ≥ 4 (κ = 0.523) than PI-RADS ≥ 3 (κ = 0.456), and peripheral zone (κ = 0.419) than transition zone lesions (κ = 0.251). Prevalence of csPCa in biopsied lesions was 40.9%.

Conclusion

There is variability in community and tertiary care center interpretation of prostate MRI in cancer detection, with higher concordance rates for higher grade and peripheral zone lesions. These differences demonstrate the added value of multidisciplinary round review and highlight the need for ongoing education and feedback.

Graphical abstract

Machine Learning in Prostate MRI for Prostate Cancer: Current Status and Future Opportunities

Advances in our understanding of the role of magnetic resonance imaging (MRI) for the detection of prostate cancer have enabled its integration into clinical routines in the past two decades. The Prostate Imaging Reporting and Data System (PI-RADS) is an established imaging-based scoring system that scores the probability of clinically significant prostate cancer on MRI to guide management. Image fusion technology allows one to combine the superior soft tissue contrast resolution of MRI, with real-time anatomical depiction using ultrasound or computed tomography. This allows the accurate mapping of prostate cancer for targeted biopsy and treatment. Machine learning provides vast opportunities for automated organ and lesion depiction that could increase the reproducibility of PI-RADS categorisation, and improve co-registration across imaging modalities to enhance diagnostic and treatment methods that can then be individualised based on clinical risk of malignancy. In this article, we provide a comprehensive and contemporary review of advancements, and share insights into new opportunities in this field.

Diagnostic Accuracy of Abbreviated Bi-Parametric MRI (a-bpMRI) for Prostate Cancer Detection and Screening: A Multi-Reader Study

(1) Background: There is currently limited evidence on the diagnostic accuracy of abbreviated biparametric MRI (a-bpMRI) protocols for prostate cancer (PCa) detection and screening. In the present study, we aim to investigate the performance of a-bpMRI among multiple readers and its potential application to an imaging-based screening setting. (2) Methods: A total of 151 men who underwent 3T multiparametric MRI (mpMRI) of the prostate and transperineal template prostate mapping biopsies were retrospectively selected. Corresponding bpMRI (multiplanar T2WI, DWI, ADC maps) and a-bpMRI (axial T2WI and b 2000 s/mm² DWI only) dataset were derived from mpMRI. Three experienced radiologists scored a-bpMRI, standard biparametric MRI (bpMRI) and mpMRI in separate sessions. Diagnostic accuracy and interreader agreement of a-bpMRI was tested for different positivity thresholds and compared to bpMRI and mpMRI. Predictive values of a-bpMRI were computed for lower levels of PCa prevalence to simulate a screening setting. The primary definition of clinically significant PCa (csPCa) was Gleason ≥ 4 + 3, or cancer core length ≥ 6 mm. (3) Results: The median age was 62 years, the median PSA was 6.8 ng/mL, and the csPCa prevalence was 40%. Using a cut off of MRI score ≥ 3, the sensitivity and specificity of a-bpMRI were 92% and 48%, respectively. There was no significant difference in sensitivity compared to bpMRI and mpMRI. Interreader agreement of a-bpMRI was moderate (AC1 0.58). For a low prevalence of csPCa (e.g., <10%), higher cut offs (MRI score ≥ 4) yield a more favourable balance between the predictive values and positivity rate of MRI. (4) Conclusion: Abbreviated bpMRI protocols could match the diagnostic accuracy of bpMRI and mpMRI for the detection of csPCa. If a-bpMRI is used in low-prevalence settings, higher cut-offs for MRI positivity should be prioritised.

AutoProstate: Towards Automated Reporting of Prostate MRI for Prostate Cancer Assessment Using Deep Learning

Multiparametric magnetic resonance imaging (mpMRI) of the prostate is used by radiologists to identify, score, and stage abnormalities that may correspond to clinically significant prostate cancer (CSPCa). Automatic assessment of prostate mpMRI using artificial intelligence algorithms may facilitate a reduction in missed cancers and unnecessary biopsies, an increase in inter-observer agreement between radiologists, and an improvement in reporting quality. In this work, we introduce AutoProstate, a deep learning-powered framework for automatic MRI-based prostate cancer assessment. AutoProstate comprises of three modules: Zone-Segmenter, CSPCa-Segmenter, and Report-Generator. Zone-Segmenter segments the prostatic zones on T2-weighted imaging, CSPCa-Segmenter detects and segments CSPCa lesions using biparametric MRI, and Report-Generator generates an automatic web-based report containing four sections: Patient Details, Prostate Size and PSA Density, Clinically Significant Lesion Candidates, and Findings Summary. In our experiment, AutoProstate was trained using the publicly available PROSTATEx dataset, and externally validated using the PICTURE dataset. Moreover, the performance of AutoProstate was compared to the performance of an experienced radiologist who prospectively read PICTURE dataset cases. In comparison to the radiologist, AutoProstate showed statistically significant improvements in prostate volume and prostate-specific antigen density estimation. Furthermore, AutoProstate matched the CSPCa lesion detection sensitivity of the radiologist, which is paramount, but produced more false positive detections.

The utility of prostate MRI within active surveillance: description of the evidence

Purpose

We present an overview of the literature regarding the use of MRI in active surveillance of prostate cancer.

Methods

Both MEDLINE^® and Cochrane Library were queried up to May 2020 for studies of men on active surveillance with MRI and later confirmatory biopsy. The terms studied were ‘prostate cancer’ as the anchor followed by two of the following: active surveillance, surveillance, active monitoring, MRI, NMR, magnetic resonance imaging,  MRI, and multiparametric MRI. Studies were excluded if pathologic reclassification (GG1 →  ≥ GG2) and PI-RADS or equivalent was not reported.

Results

Within active surveillance, baseline MRI is effective for identifying clinically significant prostate cancer and thus associated with fewer reclassification events. A positive initial MRI (≥ PI-RADS 3) with GG1 identified at biopsy has a positive predictive value (PPV) of 35–40% for reclassification by 3 years. MRI possessed a stronger negative predictive value, with a negative MRI (≤ PI-RADS 2) yielding a negative predictive value of up to 85% at 3 years. Surveillance MRI, obtained after initial biopsy, yielded a PPV of 11–65% and NPV of 85–95% for reclassification.

Conclusion

MRI is useful for initial risk stratification of prostate cancer in men on active surveillance, especially if MRI is negative when imaging is obtained during surveillance. While useful, MRI cannot replace biopsy and further research is necessary to fully integrate MRI into active surveillance.

The Role of PSA Density among PI-RADS v2.1 Categories to Avoid an Unnecessary Transition Zone Biopsy in Patients with PSA 4-20 ng/mL

Objective

To evaluate the role of prostate-specific antigen density (PSAD) in different Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) categories to avoid an unnecessary biopsy in transition zone (TZ) patients with PSA ranging from 4 to 20 ng/mL.

Materials and Methods

In this retrospective and single-center study, 333 biopsy-naïve patients with TZ lesions who underwent biparametric magnetic resonance imaging (bp-MRI) were analyzed from January 2016 to March 2020. Multivariate logistic regression analyses were performed to determine independent predictors of clinically significant prostate cancer (cs-PCa). The receiver operating characteristic (ROC) curve was used to compare diagnostic performance.

Results

PI-RADS v2.1 and PSAD were the independent predictors for TZ cs-PCa in patients with PSA 4-20 ng/mL. 0.9% (2/213), 10.0% (7/70), and 48.0% (24/50) of PI-RADS v2.1 score 1-2, 3, and 4-5 had TZ cs-PCa. However, for patients with PI-RADS v2.1 score 1-2, there were no obvious changes in the detection of TZ cs-PCa (0.8% (1/129), 1.3% (1/75), and 0.0% (0/9)) combining with different PSAD stratification (PSAD < 0.15, 0.15-0.29, and ≥0.30 ng/mL/mL). For patients with PI-RADS v2.1 score ≥ 3, the TZ cs-PCa detection rate significantly varied according to different PSAD stratification. A PI-RADS v2.1 score 3 and PSAD < 0.15 and 0.15-0.29 ng/mL/mL had 8.6% (3/35) and 3.7% (1/27) of TZ cs-PCa, while a PI-RADS v2.1 score 3 and PSAD ≥ 0.30 ng/mL/mL had a higher TZ cs-PCa detection rate (37.5% (3/8)). A PI-RADS v2.1 score 4-5 and PSAD <0.15 ng/mL/mL had no cs-PCa (0.0% (0/9)). In contrast, a PI-RADS v2.1 score 4-5 and PSAD 0.15-0.29 and ≥0.30 ng/mL/mL had the highest cs-PCa detection rate (50.0% (10/20), 66.7% (14/21)). It showed the highest AUC in the combination of PI-RADS v2.1 and PSAD (0.910), which was significantly higher than PI-RADS v2.1 (0.889, P = 0.039) or PSAD (0.803, P < 0.001).

Conclusions

For TZ patients with PSA 4-20 ng/mL, PI-RADS v2.1 score ≤ 2 can avoid an unnecessary biopsy regardless of PSAD. PI-RADS v2.1 score ≥ 3 may avoid an unnecessary biopsy after combining with PSAD. PI-RADS v2.1 combined with PSAD could significantly improve diagnostic performance.

MRI and Targeted Biopsy Essential Tools for an Accurate Diagnosis and Treatment Decision Making in Prostate Cancer

Prostate cancer (PCa) is one of the most frequent causes of cancer death worldwide. Historically, diagnosis was based on physical examination, transrectal (TRUS) images, and TRUS biopsy resulting in overdiagnosis and overtreatment. Recently magnetic resonance imaging (MRI) has been identified as an evolving tool in terms of diagnosis, staging, treatment decision, and follow-up. In this review we provide the key studies and concepts of MRI as a promising tool in the diagnosis and management of prostate cancer in the general population and in challenging scenarios, such as anteriorly located lesions, enlarged prostates determining extracapsular extension and seminal vesicle invasion, and prior negative biopsy and the future role of MRI in association with artificial intelligence (AI).

Inter-reader agreement of the PI-QUAL score for prostate MRI quality in the NeuroSAFE PROOF trial

Objectives

The Prostate Imaging Quality (PI-QUAL) score assesses the quality of multiparametric MRI (mpMRI). A score of 1 means all sequences are below the minimum standard of diagnostic quality, 3 implies that the scan is of sufficient diagnostic quality, and 5 means that all three sequences are of optimal diagnostic quality. We investigated the inter-reader reproducibility of the PI-QUAL score in patients enrolled in the NeuroSAFE PROOF trial.

Methods

We analysed the scans of 103 patients on different MR systems and vendors from 12 different hospitals. Two dedicated radiologists highly experienced in prostate mpMRI independently assessed the PI-QUAL score for each scan. Interobserver agreement was assessed using Cohen’s kappa with standard quadratic weighting (κw) and percent agreement.

Results

The agreement for each single PI-QUAL score was strong (κw = 0.85 and percent agreement = 84%). A similar agreement (κw = 0.82 and percent agreement = 84%) was observed when the scans were clustered into three groups (PI-QUAL 1–2 vs PI-QUAL 3 vs PI-QUAL 4–5). The agreement in terms of diagnostic quality for each single sequence was highest for T2-weighted imaging (92/103 scans; 89%), followed by dynamic contrast-enhanced sequences (91/103; 88%) and diffusion-weighted imaging (80/103; 78%).

Conclusion

We observed strong reproducibility in the assessment of PI-QUAL between two radiologists with high expertise in prostate mpMRI. At present, PI-QUAL offers clinicians the only available tool for evaluating and reporting the quality of prostate mpMRI in a systematic manner but further refinements of this scoring system are warranted.

Key Points

• Inter-reader agreement for each single Prostate Imaging Quality (PI-QUAL) score (i.e., PI-QUAL 1 to PI-QUAL 5) was strong, with weighted kappa = 0.85 (95% confidence intervals: 0.51 – 1) and percent agreement = 84%.

• Interobserver agreement was strong when the scans were clustered into three groups according to the ability (or not) to rule in and to rule out clinically significant prostate cancer (i.e., PI-QUAL 1-2 vs PI-QUAL 3 vs PI-QUAL 4–5), with weighted kappa = 0.82 (95% confidence intervals: 0.68 – 0.96) and percent agreement = 84%.

• T2-weighted acquisitions were the most compliant with the Prostate Imaging Reporting and Data System (PI-RADS) v. 2.0 technical recommendations and were the sequences of highest diagnostic quality for both readers in 95/103 (92%) scans, followed by dynamic contrast enhanced acquisition with 81/103 (79%) scans and lastly by diffusion-weighted imaging with 79/103 (77%) scans.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-08169-1.

Prostate MRI quality: a critical review of the last 5 years and the role of the PI-QUAL score

There is increasing interest in the use of multiparametric magnetic resonance imaging (mpMRI) in the prostate cancer pathway. The European Association of Urology (EAU) and the British Association of Urological Surgeons (BAUS) now advise mpMRI prior to biopsy, and the Prostate Imaging Reporting and Data System (PI-RADS) recommendations set out the minimal technical requirements for the acquisition of mpMRI of the prostate.The widespread and swift adoption of this technique has led to variability in image quality. Suboptimal image acquisition reduces the sensitivity and specificity of mpMRI for the detection and staging of clinically significant prostate cancer.This critical review outlines the studies aimed at improving prostate MR quality that have been published over the last 5 years. These span from the use of specific MR sequences, magnets and coils to patient preparation. The rates of adherence of prostate mpMRI to technical standards in different cohorts across the world are also discussed.Finally, we discuss the first standardised scoring system (i.e., Prostate Imaging Quality, PI-QUAL) that has been created to evaluate image quality, although further iterations of this score are expected in the future.

Artificial Intelligence Compared to Radiologists for the Initial Diagnosis of Prostate Cancer on Magnetic Resonance Imaging: A Systematic Review and Recommendations for Future Studies

Computer-aided diagnosis (CAD) of prostate cancer on multiparametric magnetic resonance imaging (mpMRI), using artificial intelligence (AI), may reduce missed cancers and unnecessary biopsies, increase inter-observer agreement between radiologists, and alleviate pressures caused by rising case incidence and a shortage of specialist radiologists to read prostate mpMRI. However, well-designed evaluation studies are required to prove efficacy above current clinical practice. A systematic search of the MEDLINE, EMBASE, and arXiv electronic databases was conducted for studies that compared CAD for prostate cancer detection or classification on MRI against radiologist interpretation and a histopathological reference standard, in treatment-naïve men with a clinical suspicion of prostate cancer. Twenty-seven studies were included in the final analysis. Due to substantial heterogeneities in the included studies, a narrative synthesis is presented. Several studies reported superior diagnostic accuracy for CAD over radiologist interpretation on small, internal patient datasets, though this was not observed in the few studies that performed evaluation using external patient data. Our review found insufficient evidence to suggest the clinical deployment of artificial intelligence algorithms at present. Further work is needed to develop and enforce methodological standards, promote access to large diverse datasets, and conduct prospective evaluations before clinical adoption can be considered.

How does PI-RADS v2.1 impact patient classification? A head-to-head comparison between PI-RADS v2.0 and v2.1

BACKGROUND

PI-RADS classification has recently been updated, with the magnitude of changes implemented currently unknown.

PURPOSE

To quantify the categorization shifts between PI-RADS v2.0 and v2.1.

MATERIAL AND METHODS

Retrospective review of 535 consecutive diagnostic magnetic resonance imaging (MRI) studies performed over 18 months, assigning to each case a PI-RADS category in the peripheral zone (PZ), the transition zone (TZ), and the whole gland using both PI-RADS v2.0 and v2.1. Significance of changes in category assignments and of differences in the number of positive or negative MRIs were evaluated using the McNemar test.

RESULTS

Comparing v2.0 to v2.1 for the whole gland, 11.2% of PI-RADS 2 categories shifted to PI-RADS 1 (6.9% in the PZ, 56.8% in the TZ), 16.1% of PI-RADS 3 categories shifted to PI-RADS 2 (15.0% in the PZ, 20.0% in the TZ), and 2.1% of PI-RADS 2 categories shifted to PI-RADS 3 (0.3% in the PZ, 1.9% in the TZ). The proportion of PI-RADS 1 significantly increased from 0.6% to 7.3%, PI-RADS 2 significantly decreased from 60.0% to 53.8%, and PI-RADS 3 non-significantly decreased from 11.6% to 11.0%. The total number of positive exams (PI-RADS 3-5) did not change significantly (39.4% versus 38.8%).

CONCLUSION

The most prominent change between v2.0 and v2.1 was observed in the TZ with the downgrading of typical benign prostatic hyperplasia nodules from category 2 into category 1. Overall, there were no significant changes in the number of positive and negative MRI results, with an expected low influence in clinical management.

Prostate Imaging Reporting and Data System (PI-RADS): What the radiologists need to know?

We aim to review the new modifications in MR imaging technique, image interpretation, lexicon, and scoring system of the last version of Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) in a simple and practical way. This last version of PI-RADS v2.1 describes the new technical modifications in the protocol of Multiparametric MRI (MpMRI) including T2, diffusion-weighted imaging (DWI), and dynamic contrast enhancement (DCE) parameters. It includes also; new guidelines in the image interpretation specifications in new locations (lesions located in the central zone and anterior fibromuscular stroma), clarification of T2 scoring of lesions of the transition zone, the distinction between DWI score 2 and 3 lesions in the transition zone and peripheral zone, as well as between positive and negative enhancement in DCE. Biparametric MRI (BpMRI) along with simplified PI-RADS is gaining more acceptances in the assessment of clinically significant prostatic cancer.

Diagnostic Accuracy and Interobserver Agreement of PI-RADS Version 2 and Version 2.1 for the Detection of Transition Zone Prostate Cancers

BACKGROUND. PI-RADS version 2.1 (v2.1) introduced a number of key changes to the assessment of transition zone (TZ) lesions. OBJECTIVE. The purpose of this study was to evaluate interobserver agreement and diagnostic accuracy for detecting TZ prostate cancer (PCa) and clinically significant PCa (csPCa) by use of PI-RADS v2 and PI-RADS v2.1 among radiologists with different levels of experience. METHODS. This retrospective study included 355 biopsy-naïve patients who from January 2017 to March 2020 underwent prostate MRI that showed a TZ lesion and underwent subsequent biopsy. PCa was diagnosed in 93 patients (International Society of Urological Pathology [ISUP] grade group 1, n = 34; ISUP grade group ≥ 2, n = 59) and non-cancerous lesions in 262 patients. Five radiologists with varying experience in prostate MRI scored lesions using PI-RADS v2 and PI-RADS v2.1 in sessions separated by at least 4 weeks. Interobserver agreement was evaluated with kappa and Kendall W statistics. ROC curve analysis was used to evaluate performance in detection of TZ PCa and csPCa. RESULTS. Interobserver agreement among all readers was higher for PI-RADS v2.1 than for PI-RADS v2 (mean weighted κ = 0.700 vs 0.622; Kendall W = 0.805 vs 0.728; p = .03). The pooled AUC values for detecting TZ PCa and csPCa were higher among all readers using PI-RADS v2.1 (0.866 vs 0.827 for TZ PCa; 0.929 vs 0.899 for TZ csPCa; p < .001). For detecting TZ PCa, the pooled sensitivity, specificity, and accuracy were 86.9%, 79.4%, and 75.4% among all readers for PI-RADS v2.1 compared with 79.4%, 71.8%, and 73.8% for PI-RADS v2. For detecting TZ csPCa, the pooled sensitivity, specificity, and accuracy were 84.8%, 90.9%, and 89.9% among all readers for PI-RADS v2.1 compared with 81.4%, 89.9%, and 88.5% for PI-RADS v2. Reader 1, who had the least experience, had the lowest sensitivity, specificity, and accuracy (78.0%, 89.2%, and 87.3%). Reader 5, who had the most experience, had the highest sensitivity, specificity, and accuracy (88.1%, 92.9%, and 92.1%) in detecting csPCa. CONCLUSION. PI-RADS v2.1 had better interobserver agreement and diagnostic accuracy than PI-RADS v2 for evaluating TZ lesions. Reader experience continues to affect the performance of prostate MRI interpretation with PI-RADS v2.1. CLINICAL IMPACT. PI-RADS v2.1 is more accurate and reproducible than PI-RADS v2 for the diagnosis of TZ PCa.

Comparison between biparametric and multiparametric MR imaging of Prostate Imaging Reporting and Data System Version 2.1 in detection of prostate cancer

Background

Prostate cancer (PCa) is considered to be the commonest cancer among males. Early and precise diagnosis of PCa is essential for adequate treatment. Multiparametric MR imaging (mpMRI) is actually the most precise imaging technique used for early diagnosis of PCa. The aim of this work was to assess the diagnostic capability of biparametric MRI (bpMRI) and multiparametric MRI (mpMRI) of PI-RADS V2.1 in detection of prostate cancer (PCa). This prospective study was carried on 60 male patients with high PSA. bpMRI and mpMRI were performed for all patients using a 3-T MRI scanner. The diagnostic performance of bpMRI of PI-RADS V2.1 was compared to that of mpMRI of PI-RADS V 2.1. The diagnosis of Pca was confirmed by transrectal ultrasound-guided biopsy and the results of open prostatectomy specimens.

Results

When considering PI-RADS categories 1, 2, and 3 as benign and categories 4 and 5 as malignant, mpMRI had higher sensitivity and diagnostic accuracy when compared with bpMRI (sensitivity was 88.6% for mpMRI versus 60% for bpMRI and diagnostic accuracy was 91.7% for mpMRI versus 75% for bpMRI). When considering PI-RADS categories 1 and 2 as benign and PI-RADS categories 3.4 and 5 as malignant, the sensitivity and diagnostic accuracy of bpMRI and mpMRI were comparable (sensitivity was 94.3% for both bpMRI and mpMRI and diagnostic accuracy was 86.7% for both bpMRI and mpMRI).

Conclusion

Considering PI-RADS scores 4 and 5 as malignant, mpMRI had higher sensitivity and diagnostic accuracy when compared with bpMRI; however, when considering PI-RADS scores 3, 4, and 5 as malignant, both bpMRI and mpMRI had similar diagnostic accuracy.