Deep learning-based whole-body PSMA PET/CT attenuation correction utilizing Pix-2-Pix GAN

PURPOSE

Sequential PET/CT studies oncology patients can undergo during their treatment follow-up course is limited by radiation dosage. We propose an artificial intelligence (AI) tool to produce attenuation-corrected PET (AC-PET) images from non-attenuation-corrected PET (NAC-PET) images to reduce need for low-dose CT scans.

METHODS

A deep learning algorithm based on 2D Pix-2-Pix generative adversarial network (GAN) architecture was developed from paired AC-PET and NAC-PET images. 18F-DCFPyL PSMA PET-CT studies from 302 prostate cancer patients, split into training, validation, and testing cohorts (n = 183, 60, 59, respectively). Models were trained with two normalization strategies: Standard Uptake Value (SUV)-based and SUV-Nyul-based. Scan-level performance was evaluated by normalized mean square error (NMSE), mean absolute error (MAE), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR). Lesion-level analysis was performed in regions-of-interest prospectively from nuclear medicine physicians. SUV metrics were evaluated using intraclass correlation coefficient (ICC), repeatability coefficient (RC), and linear mixed-effects modeling.

RESULTS

Median NMSE, MAE, SSIM, and PSNR were 13.26%, 3.59%, 0.891, and 26.82, respectively, in the independent test cohort. ICC for SUVmax and SUVmean were 0.88 and 0.89, which indicated a high correlation between original and AI-generated quantitative imaging markers. Lesion location, density (Hounsfield units), and lesion uptake were all shown to impact relative error in generated SUV metrics (all p < 0.05).

CONCLUSION

The Pix-2-Pix GAN model for generating AC-PET demonstrates SUV metrics that highly correlate with original images. AI-generated PET images show clinical potential for reducing the need for CT scans for attenuation correction while preserving quantitative markers and image quality.

Evaluation of a Cascaded Deep Learning-based Algorithm for Prostate Lesion Detection at Biparametric MRI

Background Multiparametric MRI (mpMRI) improves prostate cancer (PCa) detection compared with systematic biopsy, but its interpretation is prone to interreader variation, which results in performance inconsistency. Artificial intelligence (AI) models can assist in mpMRI interpretation, but large training data sets and extensive model testing are required. Purpose To evaluate a biparametric MRI AI algorithm for intraprostatic lesion detection and segmentation and to compare its performance with radiologist readings and biopsy results. Materials and Methods This secondary analysis of a prospective registry included consecutive patients with suspected or known PCa who underwent mpMRI, US-guided systematic biopsy, or combined systematic and MRI/US fusion-guided biopsy between April 2019 and September 2022. All lesions were prospectively evaluated using Prostate Imaging Reporting and Data System version 2.1. The lesion- and participant-level performance of a previously developed cascaded deep learning algorithm was compared with histopathologic outcomes and radiologist readings using sensitivity, positive predictive value (PPV), and Dice similarity coefficient (DSC). Results A total of 658 male participants (median age, 67 years [IQR, 61-71 years]) with 1029 MRI-visible lesions were included. At histopathologic analysis, 45% (294 of 658) of participants had lesions of International Society of Urological Pathology (ISUP) grade group (GG) 2 or higher. The algorithm identified 96% (282 of 294; 95% CI: 94%, 98%) of all participants with clinically significant PCa, whereas the radiologist identified 98% (287 of 294; 95% CI: 96%, 99%; P = .23). The algorithm identified 84% (103 of 122), 96% (152 of 159), 96% (47 of 49), 95% (38 of 40), and 98% (45 of 46) of participants with ISUP GG 1, 2, 3, 4, and 5 lesions, respectively. In the lesion-level analysis using radiologist ground truth, the detection sensitivity was 55% (569 of 1029; 95% CI: 52%, 58%), and the PPV was 57% (535 of 934; 95% CI: 54%, 61%). The mean number of false-positive lesions per participant was 0.61 (range, 0-3). The lesion segmentation DSC was 0.29. Conclusion The AI algorithm detected cancer-suspicious lesions on biparametric MRI scans with a performance comparable to that of an experienced radiologist. Moreover, the algorithm reliably predicted clinically significant lesions at histopathologic examination. ClinicalTrials.gov Identifier: NCT03354416 © RSNA, 2024 Supplemental material is available for this article.

Automated Detection and Grading of Extraprostatic Extension of Prostate Cancer at MRI via Cascaded Deep Learning and Random Forest Classification

RATIONALE AND OBJECTIVES

Extraprostatic extension (EPE) is well established as a significant predictor of prostate cancer aggression and recurrence. Accurate EPE assessment prior to radical prostatectomy can impact surgical approach. We aimed to utilize a deep learning-based AI workflow for automated EPE grading from prostate T2W MRI, ADC map, and High B DWI.

MATERIAL AND METHODS

An expert genitourinary radiologist conducted prospective clinical assessments of MRI scans for 634 patients and assigned risk for EPE using a grading technique. The training set and held-out independent test set consisted of 507 patients and 127 patients, respectively. Existing deep-learning AI models for prostate organ and lesion segmentation were leveraged to extract area and distance features for random forest classification models. Model performance was evaluated using balanced accuracy, ROC AUCs for each EPE grade, as well as sensitivity, specificity, and accuracy compared to EPE on histopathology.

RESULTS

A balanced accuracy score of .390 ± 0.078 was achieved using a lesion detection probability threshold of 0.45 and distance features. Using the test set, ROC AUCs for AI-assigned EPE grades 0-3 were 0.70, 0.65, 0.68, and 0.55 respectively. When using EPE≥ 1 as the threshold for positive EPE, the model achieved a sensitivity of 0.67, specificity of 0.73, and accuracy of 0.72 compared to radiologist sensitivity of 0.81, specificity of 0.62, and accuracy of 0.66 using histopathology as the ground truth.

CONCLUSION

Our AI workflow for assigning imaging-based EPE grades achieves an accuracy for predicting histologic EPE approaching that of physicians. This automated workflow has the potential to enhance physician decision-making for assessing the risk of EPE in patients undergoing treatment for prostate cancer due to its consistency and automation.

Reducing False-Positives Due to Urinary Stagnation in the Prostatic Urethra on 18F-DCFPyL PSMA PET/CT With MRI

PURPOSE

Prostate-specific membrane antigen (PSMA)-targeting PET radiotracers reveal physiologic uptake in the urinary system, potentially misrepresenting activity in the prostatic urethra as an intraprostatic lesion. This study examined the correlation between midline 18F-DCFPyL activity in the prostate and hyperintensity on T2-weighted (T2W) MRI as an indication of retained urine in the prostatic urethra.

PATIENTS AND METHODS

Eighty-five patients who underwent both 18F-DCFPyL PSMA PET/CT and prostate MRI between July 2017 and September 2023 were retrospectively analyzed for midline radiotracer activity and retained urine on postvoid T2W MRIs. Fisher's exact tests and unpaired t tests were used to compare residual urine presence and prostatic urethra measurements between patients with and without midline radiotracer activity. The influence of anatomical factors including prostate volume and urethral curvature on urinary stagnation was also explored.

RESULTS

Midline activity on PSMA PET imaging was seen in 14 patients included in the case group, whereas the remaining 71 with no midline activity constituted the control group. A total of 71.4% (10/14) and 29.6% (21/71) of patients in the case and control groups had urethral hyperintensity on T2W MRI, respectively (P < 0.01). Patients in the case group had significantly larger mean urethral dimensions, larger prostate volumes, and higher incidence of severe urethral curvature compared with the controls.

CONCLUSIONS

Stagnated urine within the prostatic urethra is a potential confounding factor on PSMA PET scans. Integrating PET imaging with T2W MRI can mitigate false-positive calls, especially as PSMA PET/CT continues to gain traction in diagnosing localized prostate cancer.

Deep Learning-Based Interpretable AI for Prostate T2W MRI Quality Evaluation

RATIONALE AND OBJECTIVES

Prostate MRI quality is essential in guiding prostate biopsies. However, assessment of MRI quality is subjective with variation. Quality degradation sources exert varying impacts based on the sequence under consideration, such as T2W versus DWI. As a result, employing sequence-specific techniques for quality assessment could yield more advantageous outcomes. This study aims to develop an AI tool that offers a more consistent evaluation of T2W prostate MRI quality, efficiently identifying suboptimal scans while minimizing user bias.

MATERIALS AND METHODS

This retrospective study included 1046 patients from three cohorts (ProstateX [n = 347], All-comer in-house [n = 602], enriched bad-quality MRI in-house [n = 97]) scanned between January 2011 and May 2022. An expert reader assigned T2W MRIs a quality score. A train-validation-test split of 70:15:15 was applied, ensuring equal distribution of MRI scanners and protocols across all partitions. T2W quality AI classification model was based on 3D DenseNet121 architecture using MONAI framework. In addition to multiclassification, binary classification was utilized (Classes 0/1 vs. 2). A score of 0 was given to scans considered non-diagnostic or unusable, a score of 1 was given to those with acceptable diagnostic quality with some usability but with some quality distortions present, and a score of 2 was given to those considered optimal diagnostic quality and usability. Partial occlusion sensitivity maps were generated for anatomical correlation. Three body radiologists assessed reproducibility within a subgroup of 60 test cases using weighted Cohen Kappa.

RESULTS

The best validation multiclass accuracy of 77.1% (121/157) was achieved during training. In the test dataset, multiclassification accuracy was 73.9% (116/157), whereas binary accuracy was 84.7% (133/157). Sub-class sensitivity for binary quality distortion classification for class 0 was 100% (18/18), and sub-class specificity for T2W classification of absence/minimal quality distortions for class 2 was 90.5% (95/105). All three readers showed moderate to substantial agreement with ground truth (R1-R3 κ = 0.588, κ = 0.649, κ = 0.487, respectively), moderate to substantial agreement with each other (R1-R2 κ = 0.599, R1-R3 κ = 0.612, R2-R3 κ = 0.685), fair to moderate agreement with AI (R1-R3 κ = 0.445, κ = 0.410, κ = 0.292, respectively). AI showed substantial agreement with ground truth (κ = 0.704). 3D quality heatmap evaluation revealed that the most critical non-diagnostic quality imaging features from an AI perspective related to obscuration of the rectoprostatic space (94.4%, 17/18).

CONCLUSION

The 3D AI model can assess T2W prostate MRI quality with moderate accuracy and translate whole sequence-level classification labels into 3D voxel-level quality heatmaps for interpretation. Image quality has a significant downstream impact on ruling out clinically significant cancers. AI may be able to help with reproducible identification of MRI sequences requiring re-acquisition with explainability.

Evaluating Diagnostic Accuracy and Inter-reader Agreement of the Prostate Imaging After Focal Ablation Scoring System

Background and objective

Focal therapy (FT) is increasingly recognized as a promising approach for managing localized prostate cancer (PCa), notably reducing treatment-related morbidities. However, post-treatment anatomical changes present significant challenges for surveillance using current imaging techniques. This study aimed to evaluate the inter-reader agreement and efficacy of the Prostate Imaging after Focal Ablation (PI-FAB) scoring system in detecting clinically significant prostate cancer (csPCa) on post-FT multiparametric magnetic resonance imaging (mpMRI).

Methods

A retrospective cohort study was conducted involving patients who underwent primary FT for localized csPCa between 2013 and 2023, followed by post-FT mpMRI and a prostate biopsy. Two expert genitourinary radiologists retrospectively evaluated post-FT mpMRI using PI-FAB. The key measures included inter-reader agreement of PI-FAB scores, assessed by quadratic weighted Cohen's kappa (κ), and the system's efficacy in predicting in-field recurrence of csPCa, with a PI-FAB score cutoff of 3. Additional diagnostic metrics including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy were also evaluated.

Key findings and limitations

Scans from 38 patients were analyzed, revealing a moderate level of agreement in PI-FAB scoring (κ = 0.56). Both radiologists achieved sensitivity of 93% in detecting csPCa, although specificity, PPVs, NPVs, and accuracy varied.

Conclusions and clinical implications

The PI-FAB scoring system exhibited high sensitivity with moderate inter-reader agreement in detecting in-field recurrence of csPCa. Despite promising results, its low specificity and PPV necessitate further refinement. These findings underscore the need for larger studies to validate the clinical utility of PI-FAB, potentially aiding in standardizing post-treatment surveillance.

Patient summary

Focal therapy has emerged as a promising approach for managing localized prostate cancer, but limitations in current imaging techniques present significant challenges for post-treatment surveillance. The Prostate Imaging after Focal Ablation (PI-FAB) scoring system showed high sensitivity for detecting in-field recurrence of clinically significant prostate cancer. However, its low specificity and positive predictive value necessitate further refinement. Larger, more comprehensive studies are needed to fully validate its clinical utility.

Automated prostate gland segmentation in challenging clinical cases: comparison of three artificial intelligence methods

OBJECTIVE

Automated methods for prostate segmentation on MRI are typically developed under ideal scanning and anatomical conditions. This study evaluates three different prostate segmentation AI algorithms in a challenging population of patients with prior treatments, variable anatomic characteristics, complex clinical history, or atypical MRI acquisition parameters.

MATERIALS AND METHODS

A single institution retrospective database was queried for the following conditions at prostate MRI: prior prostate-specific oncologic treatment, transurethral resection of the prostate (TURP), abdominal perineal resection (APR), hip prosthesis (HP), diversity of prostate volumes (large ≥ 150 cc, small ≤ 25 cc), whole gland tumor burden, magnet strength, noted poor quality, and various scanners (outside/vendors). Final inclusion criteria required availability of axial T2-weighted (T2W) sequence and corresponding prostate organ segmentation from an expert radiologist. Three previously developed algorithms were evaluated: (1) deep learning (DL)-based model, (2) commercially available shape-based model, and (3) federated DL-based model. Dice Similarity Coefficient (DSC) was calculated compared to expert. DSC by model and scan factors were evaluated with Wilcox signed-rank test and linear mixed effects (LMER) model.

RESULTS

683 scans (651 patients) met inclusion criteria (mean prostate volume 60.1 cc [9.05-329 cc]). Overall DSC scores for models 1, 2, and 3 were 0.916 (0.707-0.971), 0.873 (0-0.997), and 0.894 (0.025-0.961), respectively, with DL-based models demonstrating significantly higher performance (p < 0.01). In sub-group analysis by factors, Model 1 outperformed Model 2 (all p < 0.05) and Model 3 (all p < 0.001). Performance of all models was negatively impacted by prostate volume and poor signal quality (p < 0.01). Shape-based factors influenced DL models (p < 0.001) while signal factors influenced all (p < 0.001).

CONCLUSION

Factors affecting anatomical and signal conditions of the prostate gland can adversely impact both DL and non-deep learning-based segmentation models.

Localized high-risk prostate cancer harbors an androgen receptor low subpopulation susceptible to HER2 inhibition

Patients diagnosed with localized high-risk prostate cancer have higher rates of recurrence, and the introduction of neoadjuvant intensive hormonal therapies seeks to treat occult micrometastatic disease by their addition to definitive treatment. Sufficient profiling of baseline disease has remained a challenge in enabling the in-depth assessment of phenotypes associated with exceptional vs. poor pathologic responses after treatment. In this study, we report comprehensive and integrative gene expression profiling of 37 locally advanced prostate tumors prior to six months of androgen deprivation therapy (ADT) plus the androgen receptor (AR) inhibitor enzalutamide prior to radical prostatectomy. A robust transcriptional program associated with HER2 activity was positively associated with poor outcome and opposed AR activity, even after adjusting for common genomic alterations in prostate cancer including PTEN loss and expression of the TMPRSS2:ERG fusion. Patients experiencing exceptional pathologic responses demonstrated lower levels of HER2 and phospho-HER2 by immunohistochemistry of biopsy tissues. The inverse correlation of AR and HER2 activity was found to be a universal feature of all aggressive prostate tumors, validated by transcriptional profiling an external cohort of 121 patients and immunostaining of tumors from 84 additional patients. Importantly, the AR activity-low, HER2 activity-high cells that resist ADT are a pre-existing subset of cells that can be targeted by HER2 inhibition alone or in combination with enzalutamide. In summary, we show that prostate tumors adopt an AR activity-low prior to antiandrogen exposure that can be exploited by treatment with HER2 inhibitors.

Deep Learning-Based Detection and Classification of Bone Lesions on Staging Computed Tomography in Prostate Cancer: A Development Study

RATIONALE AND OBJECTIVES

Efficiently detecting and characterizing metastatic bone lesions on staging CT is crucial for prostate cancer (PCa) care. However, it demands significant expert time and additional imaging such as PET/CT. We aimed to develop an ensemble of two automated deep learning AI models for 1) bone lesion detection and segmentation and 2) benign vs. metastatic lesion classification on staging CTs and to compare its performance with radiologists.

MATERIALS AND METHODS

This retrospective study developed two AI models using 297 staging CT scans (81 metastatic) with 4601 benign and 1911 metastatic lesions in PCa patients. Metastases were validated by follow-up scans, bone biopsy, or PET/CT. Segmentation AI (3DAISeg) was developed using the lesion contours delineated by a radiologist. 3DAISeg performance was evaluated with the Dice similarity coefficient, and classification AI (3DAIClass) performance on AI and radiologist contours was assessed with F1-score and accuracy. Training/validation/testing data partitions of 70:15:15 were used. A multi-reader study was performed with two junior and two senior radiologists within a subset of the testing dataset (n = 36).

RESULTS

In 45 unseen staging CT scans (12 metastatic PCa) with 669 benign and 364 metastatic lesions, 3DAISeg detected 73.1% of metastatic (266/364) and 72.4% of benign lesions (484/669). Each scan averaged 12 extra segmentations (range: 1-31). All metastatic scans had at least one detected metastatic lesion, achieving a 100% patient-level detection. The mean Dice score for 3DAISeg was 0.53 (median: 0.59, range: 0-0.87). The F1 for 3DAIClass was 94.8% (radiologist contours) and 92.4% (3DAISeg contours), with a median false positive of 0 (range: 0-3). Using radiologist contours, 3DAIClass had PPV and NPV rates comparable to junior and senior radiologists: PPV (semi-automated approach AI 40.0% vs. Juniors 32.0% vs. Seniors 50.0%) and NPV (AI 96.2% vs. Juniors 95.7% vs. Seniors 91.9%). When using 3DAISeg, 3DAIClass mimicked junior radiologists in PPV (pure-AI 20.0% vs. Juniors 32.0% vs. Seniors 50.0%) but surpassed seniors in NPV (pure-AI 93.8% vs. Juniors 95.7% vs. Seniors 91.9%).

CONCLUSION

Our lesion detection and classification AI model performs on par with junior and senior radiologists in discerning benign and metastatic lesions on staging CTs obtained for PCa.

Immune determinants of CAR-T cell expansion in solid tumor patients receiving GD2 CAR-T cell therapy

Chimeric antigen receptor T cells (CAR-Ts) have remarkable efficacy in liquid tumors, but limited responses in solid tumors. We conducted a Phase I trial (NCT02107963) of GD2 CAR-Ts (GD2-CAR.OX40.28.z.iC9), demonstrating feasibility and safety of administration in children and young adults with osteosarcoma and neuroblastoma. Since CAR-T efficacy requires adequate CAR-T expansion, patients were grouped into good or poor expanders across dose levels. Patient samples were evaluated by multi-dimensional proteomic, transcriptomic, and epigenetic analyses. T cell assessments identified naive T cells in pre-treatment apheresis associated with good expansion, and exhausted T cells in CAR-T products with poor expansion. Myeloid cell assessment identified CXCR3+ monocytes in pre-treatment apheresis associated with good expansion. Longitudinal analysis of post-treatment samples identified increased CXCR3- classical monocytes in all groups as CAR-T numbers waned. Together, our data uncover mediators of CAR-T biology and correlates of expansion that could be utilized to advance immunotherapies for solid tumor patients.

PI-RADS Version 2.0 Versus Version 2.1: Comparison of Prostate Cancer Gleason Grade Upgrade and Downgrade Rates From MRI-Targeted Biopsy to Radical Prostatectomy

BACKGROUND. Precise risk stratification through MRI/ultrasound (US) fusion-guided targeted biopsy (TBx) can guide optimal prostate cancer (PCa) management. OBJECTIVE. The purpose of this study was to compare PI-RADS version 2.0 (v2.0) and PI-RADS version 2.1 (v2.1) in terms of the rates of International Society of Urological Pathology (ISUP) grade group (GG) upgrade and downgrade from TBx to radical prostatectomy (RP). METHODS. This study entailed a retrospective post hoc analysis of patients who underwent 3-T prostate MRI at a single institution from May 2015 to March 2023 as part of three prospective clinical trials. Trial participants who underwent MRI followed by MRI/US fusion-guided TBx and RP within a 1-year interval were identified. A single genitourinary radiologist performed clinical interpretations of the MRI examinations using PI-RADS v2.0 from May 2015 to March 2019 and PI-RADS v2.1 from April 2019 to March 2023. Upgrade and downgrade rates from TBx to RP were compared using chi-square tests. Clinically significant cancer was defined as ISUP GG2 or greater. RESULTS. The final analysis included 308 patients (median age, 65 years; median PSA density, 0.16 ng/mL2). The v2.0 group (n = 177) and v2.1 group (n = 131) showed no significant difference in terms of upgrade rate (29% vs 22%, respectively; p = .15), downgrade rate (19% vs 21%, p = .76), clinically significant upgrade rate (14% vs 10%, p = .27), or clinically significant downgrade rate (1% vs 1%, p > .99). The upgrade rate and downgrade rate were also not significantly different between the v2.0 and v2.1 groups when stratifying by index lesion PI-RADS category or index lesion zone, as well as when assessed only in patients without a prior PCa diagnosis (all p > .01). Among patients with GG2 or GG3 at RP (n = 121 for v2.0; n = 103 for v2.1), the concordance rate between TBx and RP was not significantly different between the v2.0 and v2.1 groups (53% vs 57%, p = .51). CONCLUSION. Upgrade and downgrade rates from TBx to RP were not significantly different between patients whose MRI examinations were clinically interpreted using v2.0 or v2.1. CLINICAL IMPACT. Implementation of the most recent PI-RADS update did not improve the incongruence in PCa grade assessment between TBx and surgery.

Quality of T2-weighted MRI re-acquisition versus deep learning GAN image reconstruction: A multi-reader study

PURPOSE

To evaluate CycleGAN's ability to enhance T2-weighted image (T2WI) quality.

METHOD

A CycleGAN algorithm was used to enhance T2WI quality. 96 patients (192 scans) were identified from patients who underwent multiple axial T2WI due to poor quality on the first attempt (RAD1) and improved quality on re-acquisition (RAD2). CycleGAN algorithm gave DL classifier scores (0-1) for quality quantification and produced enhanced versions of QI1 and QI2 from RAD1 and RAD2, respectively. A subset (n = 20 patients) was selected for a blinded, multi-reader study, where four radiologists rated T2WI on a scale of 1-4 for quality. The multi-reader study presented readers with 60 image pairs (RAD1 vs RAD2, RAD1 vs QI1, and RAD2 vs QI2), allowing for selecting sequence preferences and quantifying the quality changes.

RESULTS

The DL classifier correctly discerned 71.9 % of quality classes, with 90.6 % (96/106) as poor quality and 48.8 % (42/86) as diagnostic in original sequences (RAD1, RAD2). CycleGAN images (QI1, QI2) demonstrated quantitative improvements, with consistently higher DL classifier scores than original scans (p < 0.001). In the multi-reader analysis, CycleGAN demonstrated no qualitative improvements, with diminished overall quality and motion in QI2 in most patients compared to RAD2, with noise levels remaining similar (8/20). No readers preferred QI2 to RAD2 for diagnosis.

CONCLUSION

Despite quantitative enhancements with CycleGAN, there was no qualitative boost in T2WI diagnostic quality, noise, or motion. Expert radiologists didn't favor CycleGAN images over standard scans, highlighting the divide between quantitative and qualitative metrics.

ASSESSMENT OF CELL SURFACE TARGETS IN METASTATIC PROSTATE CANCER: EXPRESSION LANDSCAPE AND MOLECULAR CORRELATES

Therapeutic approaches targeting proteins on the surface of cancer cells have emerged as an important strategy for precision oncology. To fully capitalize on the potential impact of drugs targeting surface proteins, detailed knowledge about the expression patterns of the target proteins in tumor tissues is required. In castration-resistant prostate cancer (CRPC), agents targeting prostate-specific membrane antigen (PSMA) have demonstrated clinical activity. However, PSMA expression is lost in a significant number of CRPC tumors, and the identification of additional cell surface targets is necessary in order to develop new therapeutic approaches. Here, we performed a comprehensive analysis of the expression and co-expression patterns of trophoblast cell-surface antigen 2 (TROP2), delta-like ligand 3 (DLL3), and carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAM5) in CRPC samples from a rapid autopsy cohort. We show that DLL3 and CEACAM5 exhibit the highest expression in neuroendocrine prostate cancer (NEPC), while TROP2 is expressed across different CRPC molecular subtypes, except for NEPC. We observed variable intra-tumoral and inter-tumoral heterogeneity and no dominant metastatic site predilections for TROP2, DLL3, and CEACAM5. We further show that AR amplifications were associated with higher expression of PSMA and TROP2 but lower DLL3 and CEACAM5 levels. Conversely, PSMA and TROP2 expression was lower in RB1-altered tumors. In addition to genomic alterations, we demonstrate a tight correlation between epigenetic states, particularly histone H3 lysine 27 methylation (H3K27me3) at the transcriptional start site and gene body of TACSTD2 (encoding TROP2), DLL3, and CEACAM5, and their respective protein expression in CRPC patient-derived xenografts. Collectively, these findings provide novel insights into the patterns and determinants of expression of TROP2, DLL3, and CEACAM5 with important implications for the clinical development of cell surface targeting agents in CRPC.

Comparison of MRI-Based Staging and Pathologic Staging for Predicting Biochemical Recurrence of Prostate Cancer After Radical Prostatectomy

BACKGROUND. Currently most clinical models for predicting biochemical recurrence (BCR) of prostate cancer (PCa) after radical prostatectomy (RP) incorporate staging information from RP specimens, creating a gap in preoperative risk assessment. OBJECTIVE. The purpose of our study was to compare the utility of presurgical staging information from MRI and postsurgical staging information from RP pathology in predicting BCR in patients with PCa. METHODS. This retrospective study included 604 patients (median age, 60 years) with PCa who underwent prostate MRI before RP from June 2007 to December 2018. A single genitourinary radiologist assessed MRI examinations for extraprostatic extension (EPE) and seminal vesicle invasion (SVI) during clinical interpretations. The utility of EPE and SVI on MRI and RP pathology for BCR prediction was assessed through Kaplan-Meier and Cox proportional hazards analyses. Established clinical BCR prediction models, including the University of California San Francisco Cancer of the Prostate Risk Assessment (UCSF-CAPRA) model and the Cancer of the Prostate Risk Assessment Postsurgical (CAPRA-S) model, were evaluated in a subset of 374 patients with available Gleason grade groups from biopsy and RP pathology; two CAPRA-MRI models (CAPRA-S model with modifications to replace RP pathologic staging features with MRI staging features) were also assessed. RESULTS. Univariable predictors of BCR included EPE on MRI (HR = 3.6), SVI on MRI (HR = 4.4), EPE on RP pathology (HR = 5.0), and SVI on RP pathology (HR = 4.6) (all p < .001). Three-year BCR-free survival (RFS) rates for patients without versus with EPE were 84% versus 59% for MRI and 89% versus 58% for RP pathology, and 3-year RFS rates for patients without versus with SVI were 82% versus 50% for MRI and 83% versus 54% for RP histology (all p < .001). For patients with T3 disease on RP pathology, 3-year RFS rates were 67% and 41% for patients without and with T3 disease on MRI. AUCs of CAPRA models, including CAPRA-MRI models, ranged from 0.743 to 0.778. AUCs were not significantly different between CAPRA-S and CAPRA-MRI models (p > .05). RFS rates were significantly different between low- and intermediate-risk groups for only CAPRA-MRI models (80% vs 51% and 74% vs 44%; both p < .001). CONCLUSION. Presurgical MRI-based staging features perform comparably to postsurgical pathologic staging features for predicting BCR. CLINICAL IMPACT. MRI staging can preoperatively identify patients at high BCR risk, helping to inform early clinical decision-making. TRIAL REGISTRATION. ClinicalTrials.gov NCT00026884 and NCT02594202.

Is prostatic adenocarcinoma with cribriform architecture more difficult to detect on prostate MRI?

BACKGROUND

Cribriform (CBFM) pattern on prostate biopsy has been implicated as a predictor for high-risk features, potentially leading to adverse outcomes after definitive treatment. This study aims to investigate whether the CBFM pattern containing prostate cancers (PCa) were associated with false negative magnetic resonance imaging (MRI) and determine the association between MRI and histopathological disease burden.

METHODS

Patients who underwent multiparametric magnetic resonance imaging (mpMRI), combined 12-core transrectal ultrasound (TRUS) guided systematic (SB) and MRI/US fusion-guided biopsy were retrospectively queried for the presence of CBFM pattern at biopsy. Biopsy cores and lesions were categorized as follows: C0 = benign, C1 = PCa with no CBFM pattern, C2 = PCa with CBFM pattern. Correlation between cancer core length (CCL) and measured MRI lesion dimension were assessed using a modified Pearson correlation test for clustered data. Differences between the biopsy core groups were assessed with the Wilcoxon-signed rank test with clustering.

RESULTS

Between 2015 and 2022, a total of 131 consecutive patients with CBFM pattern on prostate biopsy and pre-biopsy mpMRI were included. Clinical feature analysis included 1572 systematic biopsy cores (1149 C0, 272 C1, 151 C2) and 736 MRI-targeted biopsy cores (253 C0, 272 C1, 211 C2). Of the 131 patients with confirmed CBFM pathology, targeted biopsy (TBx) alone identified CBFM in 76.3% (100/131) of patients and detected PCa in 97.7% (128/131) patients. SBx biopsy alone detected CBFM in 61.1% (80/131) of patients and PCa in 90.8% (119/131) patients. TBx and SBx had equivalent detection in patients with smaller prostates (p = 0.045). For both PCa lesion groups there was a positive and significant correlation between maximum MRI lesion dimension and CCL (C1 lesions: p < 0.01, C2 lesions: p < 0.001). There was a significant difference in CCL between C1 and C2 lesions for T2 scores of 3 and 5 (p ≤ 0.01, p ≤ 0.01, respectively) and PI-RADS 5 lesions (p ≤ 0.01), with C2 lesions having larger CCL, despite no significant difference in MRI lesion dimension.

CONCLUSIONS

The extent of disease for CBFM-containing tumors is difficult to capture on mpMRI. When comparing MRI lesions of similar dimensions and PIRADS scores, CBFM-containing tumors appear to have larger cancer yield on biopsy. Proper staging and planning of therapeutic interventions is reliant on accurate mpMRI estimation. Special considerations should be taken for patients with CBFM pattern on prostate biopsy.

Evaluation of a Deep Learning-based Algorithm for Post-Radiotherapy Prostate Cancer Local Recurrence Detection Using Biparametric MRI

OBJECTIVE

To evaluate a biparametric MRI (bpMRI)-based artificial intelligence (AI) model for the detection of local prostate cancer (PCa) recurrence in patients with radiotherapy history.

MATERIALS AND METHODS

This study included post-radiotherapy patients undergoing multiparametric MRI and subsequent MRI/US fusion-guided and/or systematic biopsy. Histopathology results were used as ground truth. The recurrent cancer detection sensitivity of a bpMRI-based AI model, which was developed on a large dataset to primarily identify lesions in treatment-naïve patients, was compared to a prospective radiologist assessment using the Wald test. Subanalysis was conducted on patients stratified by the treatment modality (external beam radiation treatment [EBRT] and brachytherapy) and the prostate volume quartiles.

RESULTS

Of the 62 patients included (median age = 70 years; median PSA = 3.51 ng/ml; median prostate volume = 27.55 ml), 56 recurrent PCa foci were identified within 46 patients. The AI model detected 40 lesions in 35 patients. The AI model performance was lower than the prospective radiology interpretation (Rad) on a patient-(AI: 76.1% vs. Rad: 91.3%, p = 0.02) and lesion-level (AI: 71.4% vs. Rad: 87.5%, p = 0.01). The mean number of false positives per patient was 0.35 (range: 0-2). The AI model performance was higher in EBRT group both on patient-level (EBRT: 81.5% [22/27] vs. brachytherapy: 68.4% [13/19]) and lesion-level (EBRT: 79.4% [27/34] vs. brachytherapy: 59.1% [13/22]). In patients with gland volumes >34 ml (n = 25), detection sensitivities were 100% (11/11) and 94.1% (16/17) on patient- and lesion-level, respectively.

CONCLUSION

The reported bpMRI-based AI model detected the majority of locally recurrent prostate cancer after radiotherapy. Further testing including external validation of this model is warranted prior to clinical implementation.

Deep Learning-Based T2-weighted MR Image Quality Assessment and Its Impact on Prostate Cancer Detection Rates

BACKGROUND

Image quality evaluation of prostate MRI is important for successful implementation of MRI into localized prostate cancer diagnosis.

PURPOSE

To examine the impact of image quality on prostate cancer detection using an in-house previously developed artificial intelligence (AI) algorithm.

STUDY TYPE

Retrospective.

SUBJECTS

615 consecutive patients (median age 67 [interquartile range [IQR]: 61-71] years) with elevated serum PSA (median PSA 6.6 [IQR: 4.6-9.8] ng/mL) prior to prostate biopsy.

FIELD STRENGTH/SEQUENCE

3.0T/T2-weighted turbo-spin-echo MRI, high b-value echo-planar diffusion-weighted imaging, and gradient recalled echo dynamic contrast-enhanced.

ASSESSMENTS

Scans were prospectively evaluated during clinical readout using PI-RADSv2.1 by one genitourinary radiologist with 17 years of experience. For each patient, T2-weighted images (T2WIs) were classified as high-quality or low-quality based on evaluation of both general distortions (eg, motion, distortion, noise, and aliasing) and perceptual distortions (eg, obscured delineation of prostatic capsule, prostatic zones, and excess rectal gas) by a previously developed in-house AI algorithm. Patients with PI-RADS category 1 underwent 12-core ultrasound-guided systematic biopsy while those with PI-RADS category 2-5 underwent combined systematic and targeted biopsies. Patient-level cancer detection rates (CDRs) were calculated for clinically significant prostate cancer (csPCa, International Society of Urological Pathology Grade Group ≥2) by each biopsy method and compared between high- and low-quality images in each PI-RADS category.

STATISTICAL TESTS

Fisher's exact test. Bootstrap 95% confidence intervals (CI). A P value <0.05 was considered statistically significant.

RESULTS

385 (63%) T2WIs were classified as high-quality and 230 (37%) as low-quality by AI. Targeted biopsy with high-quality T2WIs resulted in significantly higher clinically significant CDR than low-quality images for PI-RADS category 4 lesions (52% [95% CI: 43-61] vs. 32% [95% CI: 22-42]). For combined biopsy, there was no significant difference in patient-level CDRs for PI-RADS 4 between high- and low-quality T2WIs (56% [95% CI: 47-64] vs. 44% [95% CI: 34-55]; P = 0.09).

DATA CONCLUSION

Higher quality T2WIs were associated with better targeted biopsy clinically significant cancer detection performance for PI-RADS 4 lesions. Combined biopsy might be needed when T2WI is lower quality.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 1.

Radiogenomic profiling of prostate tumors prior to external beam radiotherapy converges on a transcriptomic signature of TGF-β activity driving tumor recurrence

Background

Patients with localized prostate cancer have historically been assigned to clinical risk groups based on local disease extent, serum prostate specific antigen (PSA), and tumor grade. Clinical risk grouping is used to determine the intensity of treatment with external beam radiotherapy (EBRT) and androgen deprivation therapy (ADT), yet a substantial proportion of patients with intermediate and high risk localized prostate cancer will develop biochemical recurrence (BCR) and require salvage therapy. Prospective identification of patients destined to experience BCR would allow treatment intensification or selection of alternative therapeutic strategies.

Methods

Twenty-nine individuals with intermediate or high risk prostate cancer were prospectively recruited to a clinical trial designed to profile the molecular and imaging features of prostate cancer in patients undergoing EBRT and ADT. Whole transcriptome cDNA microarray and whole exome sequencing were performed on pretreatment targeted biopsy of prostate tumors (n=60). All patients underwent pretreatment and 6-month post EBRT multiparametric MRI (mpMRI), and were followed with serial PSA to assess presence or absence of BCR. Genes differentially expressed in the tumor of patients with and without BCR were investigated using pathways analysis tools and were similarly explored in alternative datasets. Differential gene expression and predicted pathway activation were evaluated in relation to tumor response on mpMRI and tumor genomic profile. A novel TGF-β gene signature was developed in the discovery dataset and applied to a validation dataset.

Findings

Baseline MRI lesion volume and PTEN/TP53 status in prostate tumor biopsies correlated with the activation state of TGF-β signaling measured using pathway analysis. All three measures correlated with the risk of BCR after definitive RT. A prostate cancer-specific TGF-β signature discriminated between patients that experienced BCR vs. those that did not. The signature retained prognostic utility in an independent cohort.

Interpretation

TGF-β activity is a dominant feature of intermediate-to-unfavorable risk prostate tumors prone to biochemical failure after EBRT with ADT. TGF-β activity may serve as a prognostic biomarker independent of existing risk factors and clinical decision-making criteria.

Funding

This research was supported by the Prostate Cancer Foundation, the Department of Defense Congressionally Directed Medical Research Program, National Cancer Institute, and the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research.

Prospective Evaluation of PI-RADS Version 2.1 for Prostate Cancer Detection and Investigation of Multiparametric MRI-derived Markers

Background Data regarding the prospective performance of Prostate Imaging Reporting and Data System (PI-RADS) version 2.1 alone and in combination with quantitative MRI features for prostate cancer detection is limited. Purpose To assess lesion-based clinically significant prostate cancer (csPCa) rates in different PI-RADS version 2.1 categories and to identify MRI features that could improve csPCa detection. Materials and Methods This single-center prospective study included men with suspected or known prostate cancer who underwent multiparametric MRI and MRI/US-guided biopsy from April 2019 to December 2021. MRI scans were prospectively evaluated using PI-RADS version 2.1. Atypical transition zone (TZ) nodules were upgraded to category 3 if marked diffusion restriction was present. Lesions with an International Society of Urological Pathology (ISUP) grade of 2 or higher (range, 1-5) were considered csPCa. MRI features, including three-dimensional diameter, relative lesion volume (lesion volume divided by prostate volume), sphericity, and surface to volume ratio (SVR), were obtained from lesion contours delineated by the radiologist. Univariable and multivariable analyses were conducted at the lesion and participant levels to determine features associated with csPCa. Results In total, 454 men (median age, 67 years [IQR, 62-73 years]) with 838 lesions were included. The csPCa rates for lesions categorized as PI-RADS 1 (n = 3), 2 (n = 170), 3 (n = 197), 4 (n = 319), and 5 (n = 149) were 0%, 9%, 14%, 37%, and 77%, respectively. csPCa rates of PI-RADS 4 lesions were lower than PI-RADS 5 lesions (P < .001) but higher than PI-RADS 3 lesions (P < .001). Upgraded PI-RADS 3 TZ lesions were less likely to harbor csPCa compared with their nonupgraded counterparts (4% [one of 26] vs 20% [20 of 99], P = .02). Predictors of csPCa included relative lesion volume (odds ratio [OR], 1.6; P < .001), SVR (OR, 6.2; P = .02), and extraprostatic extension (EPE) scores of 2 (OR, 9.3; P < .001) and 3 (OR, 4.1; P = .02). Conclusion The rates of csPCa differed between consecutive PI-RADS categories of 3 and higher. MRI features, including lesion volume, shape, and EPE scores of 2 and 3, predicted csPCa. Upgrading of PI-RADS category 3 TZ lesions may result in unnecessary biopsies. ClinicalTrials.gov registration no. NCT03354416 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Goh in this issue.

Predicting Outcomes of Indeterminate Bone Lesions on 18F-DCFPyL PSMA PET/CT Scans in the Setting of High-Risk Primary or Recurrent Prostate Cancer

Indeterminate bone lesions (IBLs) on prostate-specific membrane antigen (PSMA) PET/CT are common. This study aimed to define variables that predict whether such lesions are likely malignant or benign using features on PSMA PET/CT. Methods: ¹⁸F-DCFPyL PET/CT imaging was performed on 243 consecutive patients with high-risk primary or biochemically recurrent prostate cancer. IBLs identified on PSMA PET/CT could not definitively be interpreted as benign or malignant. Medical records of patients with IBLs were reviewed to determine the ultimate status of each lesion. IBLs were deemed malignant or benign on the basis of evidence of progression or stability at follow-up, respectively, or by biopsy results; IBLs were deemed equivocal when insufficient or unclear evidence existed. Post hoc patient, lesion, and scan variables accounting for clustered data were evaluated using Wilcoxon rank-sum and χ² tests to determine features that favored benign or malignant interpretation. Results: Overall, 98 IBLs within 267 bone lesions (36.7%) were identified in 48 of 243 patients (19.8%). Thirty-seven of 98 IBLs were deemed benign, and 42 were deemed malignant, of which 8 had histologic verification; 19 remained equivocal. Location and SUV_max categorical variables were predictive of IBL interpretation (P = 0.0201 and P = 0.0230, respectively). For IBLs with new interpretations, 34 of 37 (91.9%) considered benign showed an SUV_max of less than 5 or exhibited focal uptake without coexisting bone metastases; 37 of 42 (88.1%) deemed malignant demonstrated an SUV_max of at least 5 or were present with coexisting bone metastases. Logistic regression predicted IBLs with a high SUV_max (univariable: odds ratio [OR], 9.29 [P = 0.0016]; multivariable: OR, 13.87 [P = 0.0089]) or present with other bone metastases (univariable: OR, 9.87 [P = 0.0112]; multivariable: OR, 11.35 [P = 0.003]) to be malignant. Conclusion: IBLs on PSMA PET/CT are concerning; however, characterizing their location, SUV, and additional scan findings can aid interpretation. IBLs displaying an SUV_max of at least 5 or present with other bone metastases favor malignancy. IBLs without accompanying bone metastases that exhibit an SUV_max of less than 5 and are observed only in atypical locations favor benign processes. These guidelines may assist in the interpretation of IBLs on PSMA PET/CT.

Ipsilateral hemigland prostate biopsy may underestimate cancer burden in patients with unilateral mpMRI-visible lesions

PURPOSE

To evaluate the cancer detection rates of reduced-core biopsy schemes in patients with unilateral mpMRI-visible intraprostatic lesions and to analyze the contribution of systematic biopsy cores in clinically significant prostate cancer (csPCa) detection.

METHODS

212 patients with mpMRI-visible unilateral intraprostatic lesions undergoing MRI/TRUS fusion-guided targeted biopsy (TBx) and systematic biopsy (SBx) were included. Cancer detection rates of TBx + SBx, as determined by highest Gleason Grade Group (GG), were compared to 3 reduced-core biopsy schemes: TBx alone, TBx + ipsilateral systematic biopsy (IBx; MRI-positive hemigland), and TBx + contralateral systematic biopsy (CBx; MRI-negative hemigland). Patient-level and biopsy core-level data were analyzed using descriptive statistics with confidence intervals. Univariable and multivariable logistic regression analysis was conducted to identify predictors of csPCa (≥ GG2) detected in MRI-negative hemiglands at p < 0.05.

RESULTS

Overall, 43.4% (92/212) of patients had csPCa and 66.0% (140/212) of patients had any PCa detected by TBx + SBx. Of patients with csPCa, 81.5% had exclusively ipsilateral involvement (MRI-positive), 7.6% had only contralateral involvement (MRI-negative), and 10.9% had bilateral involvement. The csPCa detection rates of reduced-core biopsy schemes were 35.4% (75/212), 40.1% (85/212), and 39.6% (84/212) for TBx alone, TBx + IBx, and TBx + CBx, respectively, with detection sensitivities of 81.5%, 92.4%, and 91.3% compared to TBx + SBx.

CONCLUSION

Reduced-core prostate biopsy strategies confined to the ipsilateral hemigland underestimate csPCa burden by at least 8% in patients with unilateral mpMRI-visible intraprostatic lesions. The combined TBx + SBx strategy maximizes csPCa detection.

The Low Rate of Adherence to Checklist for Artificial Intelligence in Medical Imaging Criteria Among Published Prostate MRI Artificial Intelligence Algorithms

OBJECTIVE

To determine the rigor, generalizability, and reproducibility of published classification and detection artificial intelligence (AI) models for prostate cancer (PCa) on MRI using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) guidelines, a 42-item checklist that is considered a measure of best practice for presenting and reviewing medical imaging AI research.

MATERIALS AND METHODS

This review searched English literature for studies proposing PCa AI detection and classification models on MRI. Each study was evaluated with the CLAIM checklist. The additional outcomes for which data were sought included measures of AI model performance (eg, area under the curve [AUC], sensitivity, specificity, free-response operating characteristic curves), training and validation and testing group sample size, AI approach, detection versus classification AI, public data set utilization, MRI sequences used, and definition of gold standard for ground truth. The percentage of CLAIM checklist fulfillment was used to stratify studies into quartiles. Wilcoxon's rank-sum test was used for pair-wise comparisons.

RESULTS

In all, 75 studies were identified, and 53 studies qualified for analysis. The original CLAIM items that most studies did not fulfill includes item 12 (77% no): de-identification methods; item 13 (68% no): handling missing data; item 15 (47% no): rationale for choosing ground truth reference standard; item 18 (55% no): measurements of inter- and intrareader variability; item 31 (60% no): inclusion of validated interpretability maps; item 37 (92% no): inclusion of failure analysis to elucidate AI model weaknesses. An AUC score versus percentage CLAIM fulfillment quartile revealed a significant difference of the mean AUC scores between quartile 1 versus quartile 2 (0.78 versus 0.86, P = .034) and quartile 1 versus quartile 4 (0.78 versus 0.89, P = .003) scores. Based on additional information and outcome metrics gathered in this study, additional measures of best practice are defined. These new items include disclosure of public dataset usage, ground truth definition in comparison to other referenced works in the defined task, and sample size power calculation.

CONCLUSION

A large proportion of AI studies do not fulfill key items in CLAIM guidelines within their methods and results sections. The percentage of CLAIM checklist fulfillment is weakly associated with improved AI model performance. Additions or supplementations to CLAIM are recommended to improve publishing standards and aid reviewers in determining study rigor.

Development of a 3D CNN-based AI Model for Automated Segmentation of the Prostatic Urethra

RATIONALE AND OBJECTIVE

The combined use of prostate cancer radiotherapy and MRI planning is increasingly being used in the treatment of clinically significant prostate cancers. The radiotherapy dosage quantity is limited by toxicity in organs with de-novo genitourinary toxicity occurrence remaining unperturbed. Estimation of the urethral radiation dose via anatomical contouring may improve our understanding of genitourinary toxicity and its related symptoms. Yet, urethral delineation remains an expert-dependent and time-consuming procedure. In this study, we aim to develop a fully automated segmentation tool for the prostatic urethra.

MATERIALS AND METHODS

This study incorporated 939 patients' T2-weighted MRI scans (train/validation/test/excluded: 657/141/140/1 patients), including in-house and public PROSTATE-x datasets, and their corresponding ground truth urethral contours from an expert genitourinary radiologist. The AI model was developed using MONAI framework and was based on a 3D-UNet. AI model performance was determined by Dice score (volume-based) and the Centerline Distance (CLD) between the prediction and ground truth centers (slice-based). All predictions were compared to ground truth in a systematic failure analysis to elucidate the model's strengths and weaknesses. The Wilcoxon-rank sum test was used for pair-wise comparison of group differences.

RESULTS

The overall organ-adjusted Dice score for this model was 0.61 and overall CLD was 2.56 mm. When comparing prostates with symmetrical (n = 117) and asymmetrical (n = 23) benign prostate hyperplasia (BPH), the AI model performed better on symmetrical prostates compared to asymmetrical in both Dice score (0.64 vs. 0.51 respectively, p < 0.05) and mean CLD (2.3 mm vs. 3.8 mm respectively, p < 0.05). When calculating location-specific performance, the performance was highest at the apex and lowest at the base location of the prostate for Dice and CLD. Dice location dependence: symmetrical (Apex, Mid, Base: 0.69 vs. 0.67 vs. 0.54 respectively, p < 0.05) and asymmetrical (Apex, Mid, Base: 0.68 vs. 0.52 vs. 0.39 respectively, p < 0.05). CLD location dependence: symmetrical (Apex, Mid, Base: 1.43 mm vs. 2.15 mm vs. 3.28 mm, p < 0.05) and asymmetrical (Apex, Mid, Base: 1.83 mm vs. 3.1 mm vs. 6.24 mm, p < 0.05).

CONCLUSION

We developed a fully automated prostatic urethra segmentation AI tool yielding its best performance in prostate glands with symmetric BPH features. This system can potentially be used to assist treatment planning in patients who can undergo whole gland radiation therapy or ablative focal therapy.

A Cascaded Deep Learning-Based Artificial Intelligence Algorithm for Automated Lesion Detection and Classification on Biparametric Prostate Magnetic Resonance Imaging

RATIONALE AND OBJECTIVES

Prostate MRI improves detection of clinically significant prostate cancer; however, its diagnostic performance has wide variation. Artificial intelligence (AI) has the potential to assist radiologists in the detection and classification of prostatic lesions. Herein, we aimed to develop and test a cascaded deep learning detection and classification system trained on biparametric prostate MRI using PI-RADS for assisting radiologists during prostate MRI read out.

MATERIALS AND METHODS

T2-weighted, diffusion-weighted (ADC maps, high b value DWI) MRI scans obtained at 3 Tesla from two institutions (n = 1043 in-house and n = 347 Prostate-X, respectively) acquired between 2015 to 2019 were used for model training, validation, testing. All scans were retrospectively reevaluated by one radiologist. Suspicious lesions were contoured and assigned a PI-RADS category. A 3D U-Net-based deep neural network was used to train an algorithm for automated detection and segmentation of prostate MRI lesions. Two 3D residual neural network were used for a 4-class classification task to predict PI-RADS categories 2 to 5 and BPH. Training and validation used 89% (n = 1290 scans) of the data using 5 fold cross-validation, the remaining 11% (n = 150 scans) were used for independent testing. Algorithm performance at lesion level was assessed using sensitivities, positive predictive values (PPV), false discovery rates (FDR), classification accuracy, Dice similarity coefficient (DSC). Additional analysis was conducted to compare AI algorithm's lesion detection performance with targeted biopsy results.

RESULTS

Median age was 66 years (IQR = 60-71), PSA 6.7 ng/ml (IQR = 4.7-9.9) from in-house cohort. In the independent test set, algorithm correctly detected 111 of 198 lesions leading to 56.1% (49.3%-62.6%) sensitivity. PPV was 62.7% (95% CI 54.7%-70.7%) with FDR of 37.3% (95% CI 29.3%-45.3%). Of 79 true positive lesions, 82.3% were tumor positive at targeted biopsy, whereas of 57 false negative lesions, 50.9% were benign at targeted biopsy. Median DSC for lesion segmentation was 0.359. Overall PI-RADS classification accuracy was 30.8% (95% CI 24.6%-37.8%).

CONCLUSION

Our cascaded U-Net, residual network architecture can detect, classify cancer suspicious lesions at prostate MRI with good detection, reasonable classification performance metrics.

Deep learning-based artificial intelligence for prostate cancer detection at biparametric MRI

PURPOSE

To present fully automated DL-based prostate cancer detection system for prostate MRI.

METHODS

MRI scans from two institutions, were used for algorithm training, validation, testing. MRI-visible lesions were contoured by an experienced radiologist. All lesions were biopsied using MRI-TRUS-guidance. Lesions masks, histopathological results were used as ground truth labels to train UNet, AH-Net architectures for prostate cancer lesion detection, segmentation. Algorithm was trained to detect any prostate cancer ≥ ISUP1. Detection sensitivity, positive predictive values, mean number of false positive lesions per patient were used as performance metrics.

RESULTS

525 patients were included for training, validation, testing of the algorithm. Dataset was split into training (n = 368, 70%), validation (n = 79, 15%), test (n = 78, 15%) cohorts. Dice coefficients in training, validation sets were 0.403, 0.307, respectively, for AHNet model compared to 0.372, 0.287, respectively, for UNet model. In validation set, detection sensitivity was 70.9%, PPV was 35.5%, mean number of false positive lesions/patient was 1.41 (range 0-6) for UNet model compared to 74.4% detection sensitivity, 47.8% PPV, mean number of false positive lesions/patient was 0.87 (range 0-5) for AHNet model. In test set, detection sensitivity for UNet was 72.8% compared to 63.0% for AHNet, mean number of false positive lesions/patient was 1.90 (range 0-7), 1.40 (range 0-6) in UNet, AHNet models, respectively.

CONCLUSION

We developed a DL-based AI approach which predicts prostate cancer lesions at biparametric MRI with reasonable performance metrics. While false positive lesion calls remain as a challenge of AI-assisted detection algorithms, this system can be utilized as an adjunct tool by radiologists.

Clinical Application of Artificial Intelligence in Positron Emission Tomography: Imaging of Prostate Cancer

PET imaging with targeted novel tracers has been commonly used in the clinical management of prostate cancer. The use of artificial intelligence (AI) in PET imaging is a relatively new approach and in this review article, we will review the current trends and categorize the currently available research into the quantification of tumor burden within the organ, evaluation of metastatic disease, and translational/supplemental research which aims to improve other AI research efforts.

Multiresolution Application of Artificial Intelligence in Digital Pathology for Prediction of Positive Lymph Nodes From Primary Tumors in Bladder Cancer

PURPOSE

To develop an artificial intelligence (AI)-based model for identifying patients with lymph node (LN) metastasis based on digital evaluation of primary tumors and train the model using cystectomy specimens available from The Cancer Genome Atlas (TCGA) Project; patients from our institution were included for validation of the leave-out test cohort.

METHODS

In all, 307 patients were identified for inclusion in the study (TCGA, n = 294; in-house, n = 13). Deep learning models were trained from image patches at 2.5×, 5×, 10×, and 20× magnifications, and spatially resolved prediction maps were combined with microenvironment (lymphocyte infiltration) features to derive a final patient-level AI score (probability of LN metastasis). Training and validation included 219 patients (training, n = 146; validation, n = 73); 89 patients (TCGA, n = 75; in-house, n = 13) were reserved as an independent testing set. Multivariable logistic regression models for predicting LN status based on clinicopathologic features alone and a combined model with AI score were fit to training and validation sets.

RESULTS

Several patients were determined to have positive LN metastasis in TCGA (n = 105; 35.7%) and in-house (n = 3; 23.1%) cohorts. A clinicopathologic model that considered using factors such as age, T stage, and lymphovascular invasion demonstrated an area under the curve (AUC) of 0.755 (95% CI, 0.680 to 0.831) in the training and validation cohorts compared with the cross validation of the AI score (likelihood of positive LNs), which achieved an AUC of 0.866 (95% CI, 0.812 to 0.920; P = .021). Performance in the test cohort was similar, with a clinicopathologic model AUC of 0.678 (95% CI, 0.554 to 0.802) and an AI score of 0.784 (95% CI, 0.702 to 0.896; P = .21). In addition, the AI score remained significant after adjusting for clinicopathologic variables (P = 1.08 × 10^-9), and the combined model significantly outperformed clinicopathologic features alone in the test cohort with an AUC of 0.807 (95% CI, 0.702 to 0.912; P = .047).

CONCLUSION

Patients who are at higher risk of having positive LNs during cystectomy can be identified on primary tumor samples using novel AI-based methodologies applied to digital hematoxylin and eosin-stained slides.

Neoadjuvant PROSTVAC prior to radical prostatectomy enhances T-cell infiltration into the tumor immune microenvironment in men with prostate cancer

Background

Clinical trials have shown the ability of therapeutic vaccines to generate immune responses to tumor-associated antigens (TAAs). What is relatively less known is if this translates into immune-cell (IC) infiltration into the tumor microenvironment. This study examined whether neoadjuvant prostate-specific antigen (PSA)-targeted vaccination with PROSTVAC could induce T-cell immunity, particularly at the tumor site.

Methods

An open-label, phase II study of neoadjuvant PROSTVAC vaccine enrolled 27 patients with localized prostate cancer awaiting radical prostatectomy (RP). We evaluated increases in CD4 and CD8 T-cell infiltrates (RP tissue vs baseline biopsies) using a six-color multiplex immunofluorescence Opal method. Antigen-specific responses were assessed by intracellular cytokine staining after in vitro stimulation of peripheral blood mononuclear cells with overlapping 15-mer peptide pools encoding the TAAs PSA, brachyury and MUC-1.

Results

Of 27 vaccinated patients, 26 had matched prevaccination (biopsy) and postvaccination (RP) prostate samples available for non-compartmentalized analysis (NCA) and compartmentalized analysis (CA). Tumor CD4 T-cell infiltrates were significantly increased in postvaccination RP specimens compared with baseline biopsies by NCA (median 176/mm² vs 152/mm²; IQR 136–317/mm² vs 69–284/mm²; p=0.0249; median ratio 1.20; IQR 0.64–2.25). By CA, an increase in both CD4 T-cell infiltrates at the tumor infiltrative margin (median 198/mm² vs 151/mm²; IQR 123–500/mm² vs 85–256/mm²; p=0.042; median ratio 1.44; IQR 0.59–4.17) and in CD8 T-cell infiltrates at the tumor core (median 140/mm² vs 105/mm²; IQR 91–175/mm² vs 83–163/mm²; p=0.036; median ratio 1.25; IQR 0.88–2.09) were noted in postvaccination RP specimens compared with baseline biopsies. A total of 13/25 patients (52%) developed peripheral T-cell responses to any of the three tested TAAs (non-neoantigens); five of these had responses to more than one antigen of the three evaluated.

Conclusion

Neoadjuvant PROSTVAC can induce both tumor immune response and peripheral immune response.

Trial registration number

NCT02153918.

Deep Learning Based Staging of Bone Lesions From Computed Tomography Scans

In this study, we formulated an efficient deep learning-based classification strategy for characterizing metastatic bone lesions using computed tomography scans (CTs) of prostate cancer patients. For this purpose, 2,880 annotated bone lesions from CT scans of 114 patients diagnosed with prostate cancer were used for training, validation, and final evaluation. These annotations were in the form of lesion full segmentation, lesion type and labels of either benign or malignant. In this work, we present our approach in developing the state-of-the-art model to classify bone lesions as benign or malignant, where (1) we introduce a valuable dataset to address a clinically important problem, (2) we increase the reliability of our model by patient-level stratification of our dataset following lesion-aware distribution at each of the training, validation, and test splits, (3) we explore the impact of lesion texture, morphology, size, location, and volumetric information on the classification performance, (4) we investigate the functionality of lesion classification using different algorithms including lesion-based average 2D ResNet-50, lesion-based average 2D ResNeXt-50, 3D ResNet-18, 3D ResNet-50, as well as the ensemble of 2D ResNet-50 and 3D ResNet-18. For this purpose, we employed a train/validation/test split equal to 75%/12%/13% with several data augmentation methods applied to the training dataset to avoid overfitting and to increase reliability. We achieved an accuracy of 92.2% for correct classification of benign vs. malignant bone lesions in the test set using an ensemble of lesion-based average 2D ResNet-50 and 3D ResNet-18 with texture, volumetric information, and morphology having the greatest discriminative power respectively. To the best of our knowledge, this is the highest ever achieved lesion-level accuracy having a very comprehensive data set for such a clinically important problem. This level of classification performance in the early stages of metastasis development bodes well for clinical translation of this strategy.

Prognostic Features of Biochemical Recurrence of Prostate Cancer Following Radical Prostatectomy Based on Multiparametric MRI and Immunohistochemistry Analysis of MRI-guided Biopsy Specimens

Background Although prostate MRI is routinely used for the detection and staging of localized prostate cancer, imaging-based assessment and targeted molecular sampling for risk stratification are an active area of research. Purpose To evaluate features of preoperative MRI and MRI-guided biopsy immunohistochemistry (IHC) findings associated with biochemical recurrence (BCR) of prostate cancer after surgery. Materials and Methods In this retrospective case-control study, patients underwent multiparametric MRI before MRI-guided biopsy followed by radical prostatectomy between 2008 and 2016. Lesions were retrospectively scored with the Prostate Imaging Reporting and Data System (PI-RADS) (version 2) by radiologists who were blinded to the clinical-pathologic results. The IHC staining, including stains for the ETS-related gene, phosphatase and tensin homolog, androgen receptor, prostate specific antigen, and p53, was performed with targeted biopsy specimens of the index lesion (highest suspicion at MRI and pathologic grade) and scored by pathologists who were blinded to clinical-pathologic outcomes. Cox proportional hazards regression analysis was used to evaluate associations with recurrence-free survival (RFS). Results The median RFS was 31.7 months (range, 1-101 months) for 39 patients (median age, 62 years; age range, 47-76 years) without BCR and 14.6 months (range, 1-61 months) for 40 patients (median age, 59 years; age range, 47-73 years) with BCR. MRI features that showed a significant relationship with the RFS interval included an index lesion with a PI-RADS score of 5 (hazard ratio [HR], 2.10; 95% CI: 1.05, 4.21; P = .04); index lesion burden, defined as ratio of index lesion volume to prostate volume (HR, 1.55; 95% CI: 1.2, 2.1; P = .003); and suspicion of extraprostatic extension (EPE) (HR, 2.18; 95% CI: 1.1, 4.2; P = .02). Presurgical multivariable analysis indicated that suspicion of EPE at MRI (adjusted HR, 2.19; 95% CI: 1.1, 4.3; P = .02) and p53 stain intensity (adjusted HR, 2.22; 95% CI: 1.0, 4.7; P = .04) were significantly associated with RFS. Conclusion MRI features, including Prostate Imaging Reporting and Data System score, index lesion burden, extraprostatic extension, and preoperative guided biopsy p53 immunohistochemistry stain intensity are associated with biochemical relapse of prostate cancer after surgery. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Costa in this issue.

Nascent Prostate Cancer Heterogeneity Drives Evolution and Resistance to Intense Hormonal Therapy

BACKGROUND

Patients diagnosed with high risk localized prostate cancer have variable outcomes following surgery. Trials of intense neoadjuvant androgen deprivation therapy (NADT) have shown lower rates of recurrence among patients with minimal residual disease after treatment. The molecular features that distinguish exceptional responders from poor responders are not known.

OBJECTIVE

To identify genomic and histologic features associated with treatment resistance at baseline.

DESIGN, SETTING, AND PARTICIPANTS

Targeted biopsies were obtained from 37 men with intermediate- to high-risk prostate cancer before receiving 6 mo of ADT plus enzalutamide. Biopsy tissues were used for whole-exome sequencing and immunohistochemistry (IHC).

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

We assessed the relationship of molecular features with final pathologic response using a cutpoint of 0.05 cm³ for residual cancer burden to compare exceptional responders to incomplete and nonresponders. We assessed intratumoral heterogeneity at the tissue and genomic level, and compared the volume of residual disease to the Shannon diversity index for each tumor. We generated multivariate models of resistance based on three molecular features and one histologic feature, with and without multiparametric magnetic resonance imaging estimates of baseline tumor volume.

RESULTS AND LIMITATIONS

Loss of chromosome 10q (containing PTEN) and alterations to TP53 were predictive of poor response, as were the expression of nuclear ERG on IHC and the presence of intraductal carcinoma of the prostate. Patients with incompletely and nonresponding tumors harbored greater tumor diversity as estimated via phylogenetic tree reconstruction from DNA sequencing and analysis of IHC staining. Our four-factor binary model (area under the receiver operating characteristic curve [AUC] 0.89) to predict poor response correlated with greater diversity in our cohort and a validation cohort of 57 Gleason score 8-10 prostate cancers from The Cancer Genome Atlas. When baseline tumor volume was added to the model, it distinguished poor response to NADT with an AUC of 0.98. Prospective use of this model requires further retrospective validation with biopsies from additional trials.

CONCLUSIONS

A subset of prostate cancers exhibit greater histologic and genomic diversity at the time of diagnosis, and these localized tumors have greater fitness to resist therapy.

PATIENT SUMMARY

Some prostate cancer tumors do not respond well to a hormonal treatment called androgen deprivation therapy (ADT). We used tumor volume and four other parameters to develop a model to identify tumors that will not respond well to ADT. Treatments other than ADT should be considered for these patients.

Quality of Prostate MRI: Is the PI-RADS Standard Sufficient?

RATIONALE AND OBJECTIVE

The Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) published a set of minimum technical standards (MTS) to improve image quality and reduce variability in multiparametric prostate MRI. The effect of PIRADSv2 MTS on image quality has not been validated. We aimed to determine whether adherence to PI-RADSv2 MTS improves study adequacy and perceived quality.

MATERIALS AND METHODS

Sixty-two prostate MRI examinations including T2 weighted (T2W) and diffusion weighted image (DWI) consecutively referred to our center from 62 different institutions within a 12-month period (September 2017 to September 2018) were included. Six readers assessed images as adequate or inadequate for use in PCa detection and a numerical image quality ranking was given using a 1-5 scale. The PI-RADSv2 MTS were synthesized into sets of seven and 10 rules for T2W and DWI, respectively. Image adherence was assessed using Digital Imaging and Communications in Medicine (DICOM) metadata. Statistical analysis of survey results and image adherence was performed based on reader quality scoring (Kendall Rank tau-b) and reader adequate scoring (Wilcoxon test for association) for T2 and DWI quality assessment.

RESULTS

Out of 62 images, 52 (83%) T2W and 38 (61%) DWIs were rated to be adequate by a majority of readers. Reader adequacy scores showed no significant association with adherence to PI-RADSv2. There was a weak (tau-b = 0.22) but significant (p value = 0.01) correlation between adherence to PIRADSv2 MTS and image quality for T2W. Studies following all PI-RADSv2 T2W rules achieved a higher median average quality score (3.58 for 7/7 vs. 3.0 for <7/7, p = 0.012). No statistical relationship with PI-RADSv2 MTS adherence and DWI quality was found.

CONCLUSION

Among 62 sites performing prostate MRI, few were considered of high quality, but the majority were considered adequate. DWI showed considerably lower rates of adequate studies in the sample. Adherence to PI-RADSv2 MTS did not increase the likelihood of having a qualitatively adequate T2W or DWI.

Quick guide on radiology image pre-processing for deep learning applications in prostate cancer research

Purpose: Deep learning has achieved major breakthroughs during the past decade in almost every field. There are plenty of publicly available algorithms, each designed to address a different task of computer vision in general. However, most of these algorithms cannot be directly applied to images in the medical domain. Herein, we are focused on the required preprocessing steps that should be applied to medical images prior to deep neural networks. Approach: To be able to employ the publicly available algorithms for clinical purposes, we must make a meaningful pixel/voxel representation from medical images which facilitates the learning process. Based on the ultimate goal expected from an algorithm (classification, detection, or segmentation), one may infer the required pre-processing steps that can ideally improve the performance of that algorithm. Required pre-processing steps for computed tomography (CT) and magnetic resonance (MR) images in their correct order are discussed in detail. We further supported our discussion by relevant experiments to investigate the efficiency of the listed preprocessing steps. Results: Our experiments confirmed how using appropriate image pre-processing in the right order can improve the performance of deep neural networks in terms of better classification and segmentation. Conclusions: This work investigates the appropriate pre-processing steps for CT and MR images of prostate cancer patients, supported by several experiments that can be useful for educating those new to the field (https://github.com/NIH-MIP/Radiology_Image_Preprocessing_for_Deep_Learning).

Changes in Magnetic Resonance Imaging Using the Prostate Cancer Radiologic Estimation of Change in Sequential Evaluation Criteria to Detect Prostate Cancer Progression for Men on Active Surveillance

BACKGROUND

The ability of serial magnetic resonance imaging (MRI) to capture pathologic progression during active surveillance (AS) remains in question.

OBJECTIVE

To determine whether changes in MRI are associated with pathologic progression for patients on AS.

DESIGN, SETTING, AND PARTICIPANTS

From July 2007 through January 2020, we identified all patients evaluated for AS at our institution. Following confirmatory biopsy, a total of 391 patients who underwent surveillance MRI and biopsy at least once were identified (median follow-up of 35.6 mo, interquartile range 19.7-60.6).

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

All MRI intervals were scored using the "Prostate Cancer Radiologic Estimation of Change in Sequential Evaluation" (PRECISE) criteria, with PRECISE scores =4 considered a positive change in MRI. A generalized estimating equation-based logistic regression analysis was conducted for all intervals with a PRECISE score of <4 to determine the predictors of Gleason grade group (GG) progression despite stable MRI.

RESULTS AND LIMITATIONS

A total of 621 MRI intervals were scored by PRECISE and validated by biopsy. The negative predictive value of stable MRI (PRECISE score <4) was greatest for detecting GG1 to?=?GG3 disease (0.94 [0.91-0.97]). If 2-yr surveillance biopsy were performed exclusively for a positive change in MRI, 3.7% (4/109) of avoided biopsies would have resulted in missed progression from GG1 to?=?GG3 disease. Prostate-specific antigen (PSA) density (odds ratio 1.95 [1.17-3.25], p?=? 0.01) was a risk factor for progression from GG1 to =GG3 disease despite stable MRI.

CONCLUSIONS

In patients with GG1 disease and stable MRI (PRECISE score <4) on surveillance, grade progression to?=?GG3 disease is not common. In patients with grade progression detected on biopsy despite stable MRI, elevated PSA density appeared to be a risk factor for progression to?=?GG3 disease.

PATIENT SUMMARY

For patients with low-risk prostate cancer on active surveillance, the risk of progressing to grade group 3 disease is low with a stable magnetic resonance image (MRI) after 2?yr. Having higher prostate-specific antigen density increases the risk of progression, despite having a stable MRI.

Sequential Prostate Magnetic Resonance Imaging in Newly Diagnosed High-risk Prostate Cancer Treated with Neoadjuvant Enzalutamide is Predictive of Therapeutic Response

PURPOSE

For high-risk prostate cancer, standard treatment options include radical prostatectomy (RP) or radiotherapy plus androgen deprivation therapy (ADT). Despite definitive therapy, many patients will have disease recurrence. Imaging has the potential to better define characteristics of response and resistance. In this study, we evaluated prostate multiparametric MRI (mpMRI) before and after neoadjuvant enzalutamide plus ADT.

PATIENTS AND METHODS

Men with localized intermediate- or high-risk prostate cancer underwent a baseline mpMRI and mpMRI-targeted biopsy followed by a second mpMRI after 6 months of enzalutamide and ADT prior to RP. Specimens were sectioned in the same plane as mpMRI using patient-specific 3D-printed molds to permit mpMRI-targeted biopsies to be compared with the same lesion from the RP. Specimens were analyzed for imaging and histologic correlates of response.

RESULTS

Of 39 patients enrolled, 36 completed imaging and RP. Most patients (92%) had high-risk disease. Fifty-eight lesions were detected on baseline mpMRI, of which 40 (69%) remained measurable at 6-month follow-up imaging. Fifty-five of 59 lesions (93%) demonstrated >50% volume reduction on posttreatment mpMRI. Three of 59 lesions (5%) demonstrated growth in size at follow-up imaging, with two lesions increasing more than 3-fold in volume. On whole-mount pathology, 15 patients demonstrated minimal residual disease (MRD) of <0.05 cc or pathologic complete response. Low initial mpMRI relative tumor burden was most predictive of MRD on final pathology.

CONCLUSIONS

Low relative lesion volume at baseline mpMRI was predictive of pathologic response. A subset of patients had limited response. Selection of patients based on these metrics may improve outcomes in high-risk disease.

Apical periurethral transition zone lesions: MRI and histology findings

PURPOSE

Apical periurethral transition zone (TZ) cancers can pose unique problems for surgery and radiation therapy. Here, we describe the appearance of such cancers on multiparametric MRI (mpMRI) and correlate this with histopathology derived from MRI-targeted biopsy.

MATERIALS AND METHODS

Between May 2011 and January 2019, a total of 4381 consecutive patients underwent 3 T mpMRI. Of these, 53 patients with 58 apical periurethral TZ lesions underwent TRUS/MRI fusion-guided biopsy and 12-core systematic TRUS-guided biopsy. Correlation was made with patient age, PSA, PSA density, whole prostate volume, and Gleason scores.

RESULTS

A total 53 men (median age 68 years, median PSA 7.94 ng/ml) were identified as having at least one apical periurethral TZ lesion on mpMRI and 5 (9%) patients had more than one apical periurethral lesion. Thus, 58 lesions were identified in 53 patients. Of these 37/53 patients (69%) and 40/58 lesions were positive at biopsy for prostate cancer. Seven were diagnosed by 12-core systematic TRUS-guided biopsy and 34 were diagnosed by TRUS/MRI fusion-guided biopsy. Gleason score was ≥ 3 + 4 in 34/58 (58%) lesions.

CONCLUSION

Identification of apical periurethral TZ prostate cancers is important to help guide surgical and radiation therapy as these tumors are adjacent to critical structures. Because of the tendency to undersample the periurethral zone during TRUS biopsy, MRI-guided biopsy is particularly helpful for detecting apical periurethral TZ prostate cancers many of which prove to be clinically significant.

Artificial intelligence for the detection of COVID-19 pneumonia on chest CT using multinational datasets

Chest CT is emerging as a valuable diagnostic tool for clinical management of COVID-19 associated lung disease. Artificial intelligence (AI) has the potential to aid in rapid evaluation of CT scans for differentiation of COVID-19 findings from other clinical entities. Here we show that a series of deep learning algorithms, trained in a diverse multinational cohort of 1280 patients to localize parietal pleura/lung parenchyma followed by classification of COVID-19 pneumonia, can achieve up to 90.8% accuracy, with 84% sensitivity and 93% specificity, as evaluated in an independent test set (not included in training and validation) of 1337 patients. Normal controls included chest CTs from oncology, emergency, and pneumonia-related indications. The false positive rate in 140 patients with laboratory confirmed other (non COVID-19) pneumonias was 10%. AI-based algorithms can readily identify CT scans with COVID-19 associated pneumonia, as well as distinguish non-COVID related pneumonias with high specificity in diverse patient populations.

Evaluating Biochemically Recurrent Prostate Cancer: Histologic Validation of 18F-DCFPyL PET/CT with Comparison to Multiparametric MRI

Background Prostate cancer recurrence is found in up to 40% of men with prior definitive (total prostatectomy or whole-prostate radiation) treatment. Prostate-specific membrane antigen PET agents such as 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine 3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid (¹⁸F-DCFPyL) may improve detection of recurrence compared with multiparametric MRI; however, histopathologic validation is lacking. Purpose To determine the sensitivity, specificity, and positive predictive value (PPV) of ¹⁸F-DCFPyL PET/CT based on histologic analysis and to compare with pelvic multiparametric MRI in men with biochemically recurrent prostate cancer. Materials and Methods Men were prospectively recruited after prostatectomy and/or radiation therapy with rising prostate-specific antigen level (median, 2.27 ng/mL; range, 0.2-27.45 ng/mL) and a negative result at conventional imaging (bone scan and/or CT). Participants underwent ¹⁸F-DCFPyL PET/CT imaging and 3.0-T pelvic multiparametric MRI. Statistical analysis included Wald and modified χ² tests. Results A total of 323 lesions were visualized in 77 men by using ¹⁸F-DCFPyL or multiparametric MRI, with imaging detection concordance of 25% (82 of 323) when including all lesions in the MRI field of view and 53% (52 of 99) when only assessing prostate bed lesions. ¹⁸F-DCFPyL depicted more pelvic lymph nodes than did MRI (128 vs 23 nodes). Histologic validation was obtained in 80 locations with sensitivity, specificity, and PPV of 69% (25 of 36; 95% confidence interval [CI]: 51%, 88%), 91% (40 of 44; 95% CI: 74%, 98%), and 86% (25 of 29; 95% CI: 73%, 97%) for ¹⁸F-DCFPyL and 69% (24 of 35; 95% CI: 50%, 86%), 74% (31 of 42; 95% CI: 42%, 89%), and 69% (24 of 35; 95% CI: 50%, 88%) for multiparametric MRI (P = .95, P = .14, and P = .07, respectively). In the prostate bed, sensitivity, specificity, and PPV were 57% (13 of 23; 95% CI: 32%, 81%), 86% (18 of 21; 95% CI: 73%, 100%), and 81% (13 of 16; 95% CI: 59%, 100%) for ¹⁸F-DCFPyL and 83% (19 of 23; 95% CI: 59%, 100%), 52% (11 of 21; 95% CI: 29%, 74%), and 66% (19 of 29; 95% CI: 44%, 86%) for multiparametric MRI (P = .19, P = .02, and P = .17, respectively). The addition of ¹⁸F-DCFPyL to multiparametric MRI improved PPV by 38% overall (P = .02) and by 30% (P = .09) in the prostate bed. Conclusion Findings with 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine 3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid (¹⁸F-DCFPyL) were histologically validated and demonstrated high specificity and positive predictive value. In the pelvis, ¹⁸F-DCFPyL depicted more lymph nodes and improved positive predictive value and specificity when added to multiparametric MRI. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Zukotynski and Rowe in this issue.

Adipose Tissue Segmentation in Unlabeled Abdomen MRI using Cross Modality Domain Adaptation

Abdominal fat quantification is critical since multiple vital organs are located within this region. Although computed tomography (CT) is a highly sensitive modality to segment body fat, it involves ionizing radiations which makes magnetic resonance imaging (MRI) a preferable alternative for this purpose. Additionally, the superior soft tissue contrast in MRI could lead to more accurate results. Yet, it is highly labor intensive to segment fat in MRI scans. In this study, we propose an algorithm based on deep learning technique(s) to automatically quantify fat tissue from MR images through a cross modality adaptation. Our method does not require supervised labeling of MR scans, instead, we utilize a cycle generative adversarial network (C-GAN) to construct a pipeline that transforms the existing MR scans into their equivalent synthetic CT (s-CT) images where fat segmentation is relatively easier due to the descriptive nature of HU (hounsfield unit) in CT images. The fat segmentation results for MRI scans were evaluated by expert radiologist. Qualitative evaluation of our segmentation results shows average success score of 3.80/5 and 4.54/5 for visceral and subcutaneous fat segmentation in MR images*.

Deep-Learning-Based Artificial Intelligence for PI-RADS Classification to Assist Multiparametric Prostate MRI Interpretation: A Development Study

BACKGROUND

The Prostate Imaging Reporting and Data System (PI-RADS) provides guidelines for risk stratification of lesions detected on multiparametric MRI (mpMRI) of the prostate but suffers from high intra/interreader variability.

PURPOSE

To develop an artificial intelligence (AI) solution for PI-RADS classification and compare its performance with an expert radiologist using targeted biopsy results.

STUDY TYPE

Retrospective study including data from our institution and the publicly available ProstateX dataset.

POPULATION

In all, 687 patients who underwent mpMRI of the prostate and had one or more detectable lesions (PI-RADS score >1) according to PI-RADSv2.

FIELD STRENGTH/SEQUENCE

T₂ -weighted, diffusion-weighted imaging (DWI; five evenly spaced b values between b = 0-750 s/mm² ) for apparent diffusion coefficient (ADC) mapping, high b-value DWI (b = 1500 or 2000 s/mm² ), and dynamic contrast-enhanced T₁ -weighted series were obtained at 3.0T.

ASSESSMENT

PI-RADS lesions were segmented by a radiologist. Bounding boxes around the T₂ /ADC/high-b value segmentations were stacked and saved as JPEGs. These images were used to train a convolutional neural network (CNN). The PI-RADS scores obtained by the CNN were compared with radiologist scores. The cancer detection rate was measured from a subset of patients who underwent biopsy.

STATISTICAL TESTS

Agreement between the AI and the radiologist-driven PI-RADS scores was assessed using a kappa score, and differences between categorical variables were assessed with a Wald test.

RESULTS

For the 1034 detection lesions, the kappa score for the AI system vs. the expert radiologist was moderate, at 0.40. However, there was no significant difference in the rates of detection of clinically significant cancer for any PI-RADS score in 86 patients undergoing targeted biopsy (P = 0.4-0.6).

DATA CONCLUSION

We developed an AI system for assignment of a PI-RADS score on segmented lesions on mpMRI with moderate agreement with an expert radiologist and a similar ability to detect clinically significant cancer.

LEVEL OF EVIDENCE

4 TECHNICAL EFFICACY STAGE: 2.

A case report of multiple primary prostate tumors with differential drug sensitivity

Localized prostate cancers are genetically variable and frequently multifocal, comprising spatially distinct regions with multiple independently-evolving clones. To date there is no understanding of whether this variability can influence management decisions for patients with prostate tumors. Here, we present a single case from a clinical trial of neoadjuvant intense androgen deprivation therapy. A patient was diagnosed with a large semi-contiguous tumor by imaging, histologically composed of a large Gleason score 9 tumor with an adjacent Gleason score 7 nodule. DNA sequencing demonstrates these are two independent tumors, as only the Gleason 9 tumor harbors single-copy losses of PTEN and TP53. The PTEN/TP53-deficient tumor demonstrates treatment resistance, selecting for subclones with mutations to the remaining copies of PTEN and TP53, while the Gleason 7 PTEN-intact tumor is almost entirely ablated. These findings indicate that spatiogenetic variability is a major confounder for personalized treatment of patients with prostate cancer.

Spatial density and diversity of architectural histology in prostate cancer: influence on diffusion weighted magnetic resonance imaging

Background

To assess the influence of specific histopathologic patterns on MRI diffusion characteristics by performing rigorous whole-mount/imaging registration and correlating histologic architectures observed in prostate cancer with diffusion characteristics in prostate MRIs.

Methods

Fifty-two whole-mount pathology blocks from 15 patients who underwent multiparametric MRI (mpMRI) at a single institution prior to radical prostatectomy were retrospectively analyzed. Regions containing individual morphologic patterns (N=21 patterns, including variations of cribriforming, expansile sheets, single cells, patterns of early intraluminal complexity, and mucin rupture patterns) were digitally annotated by an expert genitourinary pathologist. Distinct tumor foci on each slide were also assigned a Gleason grade and scored as having any high-risk histologic pattern. Digital sections were aligned to MRI using a patient-specific mold and registered using local mean weighted piecewise transformation based on anatomic control points. Density and presence of morphological patterns was correlated to apparent diffusion coefficient (ADC) signal intensity using mixed effects model accounting for nested intra-foci, intra-patient correlation. Influence of intra-tumoral heterogeneity was assessed by affinity propagation clustering (APC) of morphology features and correlated to foci- and cluster-level ADC metrics.

Results

One hundred eleven distinct tumor foci were evaluated. Beta diversity, reflecting average morphology representation across inter- and intra-foci areas, demonstrated higher intra-tumor diversity within high-risk foci (P<0.05). ADC signal demonstrated an inverse correlation with foci-level Gleason grade (P>0.05), which was strengthened in cluster-level analysis for intra-foci regions containing high-risk morphologies (P=0.017). In voxel-based analysis, dense regions demonstrate lower ADC, but the presence and density for each morphology influenced ADC independently (ANOVA P<0.001).

Conclusions

Architectural features influence ADC characteristics of MRI, with more complex tumors having lower ADC values regulated by presence and density of specific morphologies.

18F-DCFPyL PET/CT Imaging in Patients with Biochemically Recurrent Prostate Cancer After Primary Local Therapy

Our objective was to investigate the lesion detection rate of ¹⁸F-DCFPyL PET/CT, a prostate-specific membrane antigen (PSMA)-targeted PET agent, in patients with biochemically relapsed prostate cancer after primary local therapy. Methods: This was a prospective institutional review board-approved study of 90 patients with documented biochemical recurrence (median prostate-specific antigen [PSA], 2.5 ng/mL; range, 0.21-35.5 ng/mL) and negative results on conventional imaging after primary local therapies, including radical prostatectomy (n = 38), radiation (n = 27), or a combination of the two (n = 25). Patients on androgen deprivation therapy were excluded. Patients underwent whole-body ¹⁸F-DCFPyL PET/CT (299.9 ± 15.5 MBq) at 2 h after injection. The PSMA PET lesion detection rate was correlated with PSA, PSA kinetics, and original primary tumor grade. Results: Seventy patients (77.8%) showed positive PSMA PET results, with a total of 287 lesions identified: 37 prostate bed foci, 208 lesions in lymph nodes, and 42 in distant sites in bones or organs, Eleven patients had negative results, and 9 patients showed indeterminate lesions, which were considered negative in this study. The detection rates were 47.6% (n = 10/21), 50% (n = 5/10), 88.9% (n = 8/9), and 94% (n = 47/50) for PSA levels of >0.2 to <0.5, 0.5 to <1.0, 1 to <2.0, and ≥2.0 ng/mL, respectively. In postsurgical patients, PSA, PSA doubling time, and PSA velocity correlated with PET results, but the same was not true for postradiation patients. These parameters also correlated with the extent of disease on PET (intrapelvic vs. extrapelvic). There was no significant difference in the rate of positive scans between patients with higher-grade and lower-grade primary tumors (Gleason score of ≥4 + 3 vs. <3 + 4). Tumor recurrence was histology-confirmed in 40% (28/70) of patients. On a per-patient basis, positive predictive value was 93.3% (95% confidence interval, 77.6%-99.2%) by histopathologic validation and 96.2% (95% confidence interval, 86.3%-99.7%) by the combination of histology and imaging/clinical follow-up. Conclusion: ¹⁸F-DCFPyL PET/CT imaging offers high detection rates in biochemically recurrent prostate cancer patients and is positive in about 50% of patients with a PSA level of less than 0.5 ng/mL, which could substantially impact clinical management. In postsurgical patients, ¹⁸F-DCFPyL PET/CT correlates with PSA, PSA doubling time, and PSA velocity, suggesting it may have prognostic value. ¹⁸F-DCFPyL PET/CT is highly promising for localizing sites of recurrent prostate cancer.

A Magnetic Resonance Imaging-Based Prediction Model for Prostate Biopsy Risk Stratification

Importance

Multiparametric magnetic resonance imaging (MRI) in conjunction with MRI-transrectal ultrasound (TRUS) fusion-guided biopsies have improved the detection of prostate cancer. It is unclear whether MRI itself adds additional value to multivariable prediction models based on clinical parameters.

Objective

To determine whether an MRI-based prediction model can reduce unnecessary biopsies in patients with suspected prostate cancer.

Design, Setting, and Participants

Patients underwent MRI, MRI-TRUS fusion-guided biopsy, and 12-core systematic biopsy in 1 session. The development cohort used to derive the prediction model consisted of 400 patients from 1 institution enrolled between May 14, 2015, and August 31, 2016, and the validation cohort included 251 patients from 2 independent institutions who underwent biopsies between April 1, 2013, and June 30, 2016, at 1 institution and between July 1, 2015, and October 31, 2016, at the other institution. The MRI model included MRI-derived parameters in addition to clinical variables. Area under the curve of receiver operating characteristic curves and decision curve analysis were performed.

Main Outcomes and Measures

Risk of clinically significant prostate cancer on biopsy, defined as a Gleason score of 3 + 4 or higher in at least 1 biopsy core.

Results

Overall, 193 (48.3%) of the 400 patients in the development cohort (mean [SD] age at biopsy, 64.3 [7.1] years) and 96 (38.2%) of the 251 patients in the validation cohort (mean [SD] age at biopsy, 64.9 [7.2] years) had clinically significant prostate cancer, defined as a Gleason score greater than or equal to 3 + 4. By applying the model to the external validation cohort, the area under the curve increased from 64% to 84% compared with the baseline model (P < .001). At a risk threshold of 20%, the MRI model had a lower false-positive rate than the baseline model (46% [95% CI, 32%-66%] vs 92% [95% CI, 70%-100%]), with only a small reduction in the true-positive rate (89% [95% CI, 85%-96%] vs 99% [95% CI, 89%-100%]). Eighteen of 100 fewer biopsies could have been performed, with no increase in the number of patients with missed clinically significant prostate cancers.

Conclusions and Relevance

The inclusion of MRI-derived parameters in a risk model could reduce the number of unnecessary biopsies while maintaining a high rate of diagnosis of clinically significant prostate cancers.

Impact of Anatomic Location of Bone Metastases on Prognosis in Metastatic Castration-Resistant Prostate Cancer

BACKGROUND

Whole-body assessments of ¹⁸F-NaF positron emission tomography (PET)/computed tomography (CT) provide promising quantitative imaging biomarkers of metastatic castration-resistant prostate cancer (mCRPC). This study investigated whether the distribution of metastases across anatomic regions is prognostic of progression-free survival.

PATIENTS AND METHODS

Fifty-four mCRPC patients with osseous metastases received baseline NaF PET/CT. Patients received chemotherapy (n = 16) or androgen receptor pathway inhibitors (n = 38). Semiautomated analysis using Quantitative Total Bone Imaging software extracted imaging metrics for the whole, axial, and appendicular skeleton as well as 11 skeletal regions. Five PET metrics were extracted for each region: number of lesions (N_L), standardized maximum uptake value (SUV_max), average uptake (SUV_mean), sum of uptake (SUV_total), and diseased fraction of the skeleton (volume fraction). Progression included that discovered by clinical, biochemical, or radiographic means. Univariate and multivariate Cox proportional hazard regression analyses were performed between imaging metrics and progression-free survival, and were assessed according to their hazard ratios (HR) and concordance (C)-indices.

RESULTS

The strongest univariate models of progression-free survival were pelvic N_L and SUV_max with HR = 1.80 (N_L: false discovery rate adjusted P = .001, SUV_max: adjusted P = .001). Three other region-specific metrics (axial N_L: HR = 1.59, adjusted P = .02, axial SUV_max: HR = 1.61, adjusted P = .02, and skull SUV_max: HR = 1.58, adjusted P = .04) were found to be stronger prognosticators relative to their whole-body counterparts. Multivariate model including region-specific metrics (C-index = 0.727) outperformed that of whole-body metrics (C-index = 0.705). The best performance was obtained when region-specific and whole-body metrics were included (C-index = 0.742).

CONCLUSION

Quantitative characterization of metastatic spread by anatomic location on NaF PET/CT enhances potential prognostication. Further study is warranted to optimize the prognostic and predictive value of NaF PET/CT in mCRPC patients.

Intra- and interreader reproducibility of PI-RADSv2: A multireader study

BACKGROUND

The Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) has been in use since 2015; while interreader reproducibility has been studied, there has been a paucity of studies investigating the intrareader reproducibility of PI-RADSv2.

PURPOSE

To evaluate both intra- and interreader reproducibility of PI-RADSv2 in the assessment of intraprostatic lesions using multiparametric magnetic resonance imaging (mpMRI).

STUDY TYPE

Retrospective.

POPULATION/SUBJECTS

In all, 102 consecutive biopsy-naïve patients who underwent prostate MRI and subsequent MR/transrectal ultrasonography (MR/TRUS)-guided biopsy.

FIELD STRENGTH/SEQUENCES

Prostate mpMRI at 3T using endorectal with phased array surface coils (TW MRI, DW MRI with ADC maps and b2000 DW MRI, DCE MRI).

ASSESSMENT

Previously detected and biopsied lesions were scored by four readers from four different institutions using PI-RADSv2. Readers scored lesions during two readout rounds with a 4-week washout period.

STATISTICAL TESTS

Kappa (κ) statistics and specific agreement (Po ) were calculated to quantify intra- and interreader reproducibility of PI-RADSv2 scoring. Lesion measurement agreement was calculated using the intraclass correlation coefficient (ICC).

RESULTS

Overall intrareader reproducibility was moderate to substantial (κ = 0.43-0.67, Po  = 0.60-0.77), while overall interreader reproducibility was poor to moderate (κ = 0.24, Po  = 46). Readers with more experience showed greater interreader reproducibility than readers with intermediate experience in the whole prostate (P = 0.026) and peripheral zone (P = 0.002). Sequence-specific interreader agreement for all readers was similar to the overall PI-RADSv2 score, with κ = 0.24, 0.24, and 0.23 and Po  = 0.47, 0.44, and 0.54 in T2 -weighted, diffusion-weighted imaging (DWI), and dynamic contrast-enhanced (DCE), respectively. Overall intrareader and interreader ICC for lesion measurement was 0.82 and 0.71, respectively.

DATA CONCLUSION

PI-RADSv2 provides moderate intrareader reproducibility, poor interreader reproducibility, and moderate interreader lesion measurement reproducibility. These findings suggest a need for more standardized reader training in prostate MRI.

LEVEL OF EVIDENCE

2 Technical Efficacy: Stage 2.

Artificial intelligence at the intersection of pathology and radiology in prostate cancer

Pathologic grading plays a key role in prostate cancer risk stratification and treatment selection, traditionally assessed from systemic core needle biopsies sampled throughout the prostate gland. Multiparametric magnetic resonance imaging (mpMRI) has become a well-established clinical tool for detecting and localizing prostate cancer. However, both pathologic and radiologic assessment suffer from poor reproducibility among readers. Artificial intelligence (AI) methods show promise in aiding the detection and assessment of imaging-based tasks, dependent on the curation of high-quality training sets. This review provides an overview of recent advances in AI applied to mpMRI and digital pathology in prostate cancer which enable advanced characterization of disease through combined radiology-pathology assessment.

A comparison of prostate cancer bone metastases on 18F-Sodium Fluoride and Prostate Specific Membrane Antigen (18F-PSMA) PET/CT: Discordant uptake in the same lesion

Purpose

Prostate-Specific Membrane Antigen (PSMA) PET/CT has been introduced as a sensitive method for characterizing metastatic prostate cancer. The purpose of this study is to compare the spatial concordance of ¹⁸F-NaF PET/CT and ¹⁸F-PSMA-targeted PET/CT within prostate cancer bone metastases.

Methods

Prostate cancer patients with known bone metastases underwent PSMA-targeted PET/CT (¹⁸F-DCFBC or ¹⁸F-DCFPyL) and ¹⁸F-NaF PET/CT. In pelvic and spinal lesions detected by both radiotracers, regions-of-interest (ROIs) derived by various thresholds of uptake intensity were compared for spatial colocalization. Overlap volume was correlated with uptake characteristics and disease status.

Results

The study included 149 lesions in 19 patients. Qualitatively, lesions exhibited a heterogeneous range of spatial concordance between PSMA and NaF uptake from completely matched to completely discordant. Quantitatively, overlap volume decreased as a function of tracer intensity. and disease status, where lesions from patients with castration-sensitive disease showed higher spatial concordance while lesions from patients with castration-resistant disease demonstrated more frequent spatial discordance.

Conclusion

As metastatic prostate cancer progresses from castration-sensitive to castration-resistant, greater discordance is observed between NaF PET and PSMA PET uptake. This may indicate a possible phenotypic shift to tumor growth that is more independent of bone remodeling via osteoblastic formation.

Evaluating the size criterion for PI-RADSv2 category 5 upgrade: is 15 mm the best threshold?

PURPOSE

The purpose of the study was to determine if the ≥ 15 mm threshold currently used to define PIRADS 5 lesions is the optimal size threshold for predicting high likelihood of clinically significant (CS) cancers.

MATERIALS

Three hundred and fifty-eight lesions that may be changed from category 4 to 5 or vice versa on the basis of the size criterion (category 4: n = 288, category 5: n = 70) from 255 patients were evaluated. Kendall's tau-b statistic accounting for inter-lesion correlation, generalized estimation equation logistic regression, and receiver operating curve analysis evaluated two lesion size-metrics (lesion diameter and relative lesion diameter-defined as lesion diameter/prostate volume) for ability to identify CS (Gleason grade ≥ 3 + 4) cancer at targeted biopsy. Optimal cut-points were identified using the Youden index. Analyses were performed for the whole prostate (WP) and zone-specific sub-cohorts of lesions in the peripheral and transition zones (PZ and TZ).

RESULTS

Lesion diameter showed a modest correlation with Gleason grade (WP: τB = 0.21, p < 0.0001; PZ: τB = 0.13, p = 0.02; TZ: τB = 0.32, p = 0.001), and association with CS cancer detection (WP: AUC = 0.63, PZ: AUC = 0.59, TZ: AUC = 0.74). Empirically derived thresholds (WP: 14 mm, PZ: 13 mm, TZ: 16 mm) performed similarly to the current ≥ 15 mm standard. Lesion relative lesion diameter improved identification of CS cancers compared to lesion diameter alone (WP: τB = 0.30, PZ: τB = 0.24, TZ: τB = 0.42, all p < 0.0001). AUC also improved for WP and PZ lesions (WP: AUC = 0.70, PZ: AUC = 0.68, and TZ: AUC = 0.74).

CONCLUSIONS

The current ≥ 15 mm diameter threshold is a reasonable delineator of PI-RADS category 4 and category 5 lesions in the absence of extraprostatic extension to predict CS cancers. Additionally, relative lesion diameter can improve identification of CS cancers and may serve as another option for distinguishing category 4 and 5 lesions.

Can computer-aided diagnosis assist in the identification of prostate cancer on prostate MRI? a multi-center, multi-reader investigation

For prostate cancer detection on prostate multiparametric MRI (mpMRI), the Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) and computer-aided diagnosis (CAD) systems aim to widely improve standardization across radiologists and centers. Our goal was to evaluate CAD assistance in prostate cancer detection compared with conventional mpMRI interpretation in a diverse dataset acquired from five institutions tested by nine readers of varying experience levels, in total representing 14 globally spread institutions. Index lesion sensitivities of mpMRI-alone were 79% (whole prostate (WP)), 84% (peripheral zone (PZ)), 71% (transition zone (TZ)), similar to CAD at 76% (WP, p=0.39), 77% (PZ, p=0.07), 79% (TZ, p=0.15). Greatest CAD benefit was in TZ for moderately-experienced readers at PI-RADSv2 <3 (84% vs mpMRI-alone 67%, p=0.055). Detection agreement was unchanged but CAD-assisted read times improved (4.6 vs 3.4 minutes, p<0.001). At PI-RADSv2 ≥ 3, CAD improved patient-level specificity (72%) compared to mpMRI-alone (45%, p<0.001). PI-RADSv2 and CAD-assisted mpMRI interpretations have similar sensitivities across multiple sites and readers while CAD has potential to improve specificity and moderately-experienced radiologists' detection of more difficult tumors in the center of the gland. The multi-institutional evidence provided is essential to future prostate MRI and CAD development.