Validation of Prostate Imaging-Reporting and Data System Version 2: A Retrospective Analysis

Predictors of Unfavorable Pathology in Patients with Incidental (pT1a-T1b) Prostate Cancer

BACKGROUND

Incidental prostate cancer (IPCa) is encountered in 10% of surgical procedures for benign prostatic obstruction (BPO). Identification of patients with underlying detrimental prostate cancer is paramount for tailored treatment decision-making, but guideline recommendations for this setting are lacking.

OBJECTIVE

To highlight clinical and histological characteristics related to BPO surgery that may predict IPCa with unfavorable pathology.

DESIGN, SETTING, AND PARTICIPANTS

We included men with IPCa who underwent radical prostatectomy (RP) in the short term after IPCa diagnosis. Two cohorts were built according to final pathology for the RP specimen: unfavorable pathology (International Society of Urological Pathology [ISUP] grade group [GG] ≥3 and/or ≥pT3a and/or pN1) versus favorable pathology.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

We performed multivariate regression analysis for the endpoint, which was unfavorable pathology for the RP specimen. Using the model estimates for prostate-specific antigen (PSA), ISUP GG, age, and prostate volume, we established a model for estimating the risk of unfavorable histopathology.

RESULTS AND LIMITATIONS

Overall, 112 patients were included in the final assessment. On multivariate analysis, PSA (odds ratio [OR] 1.083, 95% confidence interval [CI] 1.003-1.170; p = 0.042), ISUP GG for the specimen from BPO surgery (OR 3.090; 95% CI 1.129-8.457; p = 0.028), and age (OR 1.121, 95% CI 1.026-1.225; p = 0.012) were independent predictors for unfavorable histopathology. On receiver operating characteristic analysis, the area under the curve was 0.751. A novel calculator was developed to predict adverse pathology for men with IPCa. The study is limited by its retrospective design.

CONCLUSIONS

For men with IPCa, PSA before surgery for BPO, ISUP GG, and age are independent predictors of unfavorable disease. Our results might improve preoperative risk assessment for patient counseling.

PATIENT SUMMARY

We developed a novel calculator to estimate the risk of underlying detrimental disease in men diagnosed with prostate cancer at surgery for benign prostatic obstruction.

Segmentation of the Prostate Transition Zone and Peripheral Zone on MR Images with Deep Learning

Purpose To develop a deep learning model to delineate the transition zone (TZ) and peripheral zone (PZ) of the prostate on MR images. Materials and Methods This retrospective study was composed of patients who underwent a multiparametric prostate MRI and an MRI/transrectal US fusion biopsy between January 2013 and May 2016. A board-certified abdominal radiologist manually segmented the prostate, TZ, and PZ on the entire data set. Included accessions were split into 60% training, 20% validation, and 20% test data sets for model development. Three convolutional neural networks with a U-Net architecture were trained for automatic recognition of the prostate organ, TZ, and PZ. Model performance for segmentation was assessed using Dice scores and Pearson correlation coefficients. Results A total of 242 patients were included (242 MR images; 6292 total images). Models for prostate organ segmentation, TZ segmentation, and PZ segmentation were trained and validated. Using the test data set, for prostate organ segmentation, the mean Dice score was 0.940 (interquartile range, 0.930-0.961), and the Pearson correlation coefficient for volume was 0.981 (95% CI: 0.966, 0.989). For TZ segmentation, the mean Dice score was 0.910 (interquartile range, 0.894-0.938), and the Pearson correlation coefficient for volume was 0.992 (95% CI: 0.985, 0.995). For PZ segmentation, the mean Dice score was 0.774 (interquartile range, 0.727-0.832), and the Pearson correlation coefficient for volume was 0.927 (95% CI: 0.870, 0.957). Conclusion Deep learning with an architecture composed of three U-Nets can accurately segment the prostate, TZ, and PZ. Keywords: MRI, Genital/Reproductive, Prostate, Neural Networks Supplemental material is available for this article. © RSNA, 2021.

Prostate minimally invasive procedures: complications and normal vs. abnormal findings on multiparametric magnetic resonance imaging (mpMRI)

Minimally invasive alternatives to traditional prostate surgery are increasingly utilized to treat benign prostatic hyperplasia and localized prostate cancer in select patients. Advantages of these treatments over prostatectomy include lower risk of complication, shorter length of hospital stay, and a more favorable safety profile. Multiparametric magnetic resonance imaging (mpMRI) has become a widely accepted imaging modality for evaluation of the prostate gland and provides both anatomical and functional information. As prostate mpMRI and minimally invasive prostate procedure volumes increase, it is important for radiologists to be familiar with normal post-procedure imaging findings and potential complications. This paper reviews the indications, procedural concepts, common post-procedure imaging findings, and potential complications of prostatic artery embolization, prostatic urethral lift, irreversible electroporation, photodynamic therapy, high-intensity focused ultrasound, focal cryotherapy, and focal laser ablation.

Interobserver agreement of Prostate Imaging–Reporting and Data System (PI-RADS–v2)

Background

A retrospective study was conducted on 71 consecutive patients with suspected prostate cancer (PCa) with a mean age of 56 years and underwent mp-MRI of the prostate at 3 Tesla MRI. Two readers recognized all prostatic lesions, and each lesion had a score according to Prostate Imaging–Reporting and Data System version 2 (PI-RADS-v2).

Purpose of the study

To evaluate the interobserver agreement of PI-RADS-v2 in characterization of prostatic lesions using multiparametric MRI (mp-MRI) at 3 Tesla MRI.

Results

The overall interobserver agreement of PI-RADS-v2 for both zones was excellent (k = 0.81, percent agreement = 94.9%). In the peripheral zone (PZ) lesions are the interobserver agreement for PI-RADS II (k = 0.78, percent agreement = 83.9%), PI-RADS III (k = 0.66, percent agreement = 91.3 %), PI-RADS IV (k = 0.69, percent agreement = 93.5%), and PI-RADS V (k = 0.91, percent agreement = 95.7 %). In the transitional zone (TZ) lesions are the interobserver agreement for PI-RADS I (k = 0.98, percent of agreement = 96%), PI-RADS II (k = 0.65, percent agreement = 96%), PI-RADS III (k = 0.65, percent agreement = 88%), PI-RADS IV (k = 0.83, percent agreement = 96%), and PI-RADS V (k = 0.82, percent agreement = 92%).

Conclusion

We concluded that PI-RADS-v2 is a reliable and a reproducible imaging modality for the characterization of prostatic lesions and detection of PCa.

Positive Predictive Value of Prostate Imaging Reporting and Data System Version 2 for the Detection of Clinically Significant Prostate Cancer: A Systematic Review and Meta-analysis

CONTEXT

The variability of the positive predictive value (PPV) represents a significant factor affecting the diagnostic performance of multiparametric magnetic resonance imaging (mpMRI).

OBJECTIVE

To analyze published studies reporting mpMRI PPV and the reasons behind the variability of clinically significant prostate cancer (csPCa) detection rates on targeted biopsies (TBx) according to Prostate Imaging Reporting and Data System (PI-RADS) version 2 categories.

EVIDENCE ACQUISITION

A search of PubMed, Cochrane library's Central, EMBASE, MEDLINE, and Scopus databases, from January 2015 to June 2020, was conducted. The primary and secondary outcomes were to evaluate the PPV of PI-RADS version 2 in detecting csPCa and any prostate cancer (PCa), respectively. Individual authors' definitions for csPCa and PI-RADS thresholds for positive mpMRI were accepted. Detection rates, used as a surrogate of PPV, were pooled using random-effect models. Preplanned subgroup analyses tested PPV after stratification for PI-RADS scores, previous biopsy status, TBx technique, and number of sampled cores. PPV variation over cancer prevalence was evaluated.

EVIDENCE SYNTHESIS

Fifty-six studies, with a total of 16 537 participants, were included in the quantitative synthesis. The PPV of suspicious mpMRI for csPCa was 40% (95% confidence interval 36-43%), with large heterogeneity between studies (I² 94%, p < 0.01). PPV increased according to PCa prevalence. In subgroup analyses, PPVs for csPCa were 13%, 40%, and 69% for, respectively, PI-RADS 3, 4, and 5 (p < 0.001). TBx missed 6%, 6%, and 5% of csPCa in PI-RADS 3, 4, and 5 lesions, respectively. In biopsy-naïve and prior negative biopsy groups, PPVs for csPCa were 42% and 32%, respectively (p = 0.005). Study design, TBx technique, and number of sampled cores did not affect PPV.

CONCLUSIONS

Our meta-analysis underlines that the PPV of mpMRI is strongly dependent on the disease prevalence, and that the main factors affecting PPV are PI-RADS version 2 scores and prior biopsy status. A substantially low PPV for PI-RADS 3 lesions was reported, while it was still suboptimal in PI-RADS 4 and 5 lesions. Lastly, even if the added value of a systematic biopsy for csPCa is relatively low, this rate can improve patient risk assessment and staging.

PATIENT SUMMARY

Targeted biopsy of Prostate Imaging Reporting and Data System 3 lesions should be considered carefully in light of additional individual risk assessment corroborating the presence of clinically significant prostate cancer. On the contrary, the positive predictive value of highly suspicious lesions is not high enough to omit systematic prostate sampling.

A 3D-2D Hybrid U-Net Convolutional Neural Network Approach to Prostate Organ Segmentation of Multiparametric MRI

OBJECTIVE

Prostate cancer is the most commonly diagnosed cancer in men in the United States with more than 200,000 new cases in 2018. Multiparametric MRI (mpMRI) is increasingly used for prostate cancer evaluation. Prostate organ segmentation is an essential step of surgical planning for prostate fusion biopsies. Deep learning convolutional neural networks (CNNs) are the predominant method of machine learning for medical image recognition. In this study, we describe a deep learning approach, a subset of artificial intelligence, for automatic localization and segmentation of prostates from mpMRI.

MATERIALS AND METHODS

This retrospective study included patients who underwent prostate MRI and ultrasound-MRI fusion transrectal biopsy between September 2014 and December 2016. Axial T2-weighted images were manually segmented by two abdominal radiologists, which served as ground truth. These manually segmented images were used for training on a customized hybrid 3D-2D U-Net CNN architecture in a fivefold cross-validation paradigm for neural network training and validation. The Dice score, a measure of overlap between manually segmented and automatically derived segmentations, and Pearson linear correlation coefficient of prostate volume were used for statistical evaluation.

RESULTS

The CNN was trained on 299 MRI examinations (total number of MR images = 7774) of 287 patients. The customized hybrid 3D-2D U-Net had a mean Dice score of 0.898 (range, 0.890-0.908) and a Pearson correlation coefficient for prostate volume of 0.974.

CONCLUSION

A deep learning CNN can automatically segment the prostate organ from clinical MR images. Further studies should examine developing pattern recognition for lesion localization and quantification.

Applications of Artificial Intelligence to Prostate Multiparametric MRI (mpMRI): Current and Emerging Trends

Prostate carcinoma is one of the most prevalent cancers worldwide. Multiparametric magnetic resonance imaging (mpMRI) is a non-invasive tool that can improve prostate lesion detection, classification, and volume quantification. Machine learning (ML), a branch of artificial intelligence, can rapidly and accurately analyze mpMRI images. ML could provide better standardization and consistency in identifying prostate lesions and enhance prostate carcinoma management. This review summarizes ML applications to prostate mpMRI and focuses on prostate organ segmentation, lesion detection and segmentation, and lesion characterization. A literature search was conducted to find studies that have applied ML methods to prostate mpMRI. To date, prostate organ segmentation and volume approximation have been well executed using various ML techniques. Prostate lesion detection and segmentation are much more challenging tasks for ML and were attempted in several studies. They largely remain unsolved problems due to data scarcity and the limitations of current ML algorithms. By contrast, prostate lesion characterization has been successfully completed in several studies because of better data availability. Overall, ML is well situated to become a tool that enhances radiologists' accuracy and speed.