A deep learning–based algorithm automatically detected cancerous lesions on prostate biparametric MRI scans with reasonable performance and reliably predicted clinically significant prostate cancer at biopsy.

Radiologists play a crucial role in pretreatment selection and posttreatment surveillance of patients undergoing focal therapy for primary prostate cancer.

Purpose To evaluate the performance of an artificial intelligence (AI) model in detecting overall and clinically significant prostate cancer (csPCa)-positive lesions on paired external and in-house biparametric MRI (bpMRI) scans and assess performance differences between each dataset. Materials and Methods This single-center retrospective study included patients who underwent prostate MRI at an external institution and were rescanned at the authors' institution between May 2015 and May 2022. A genitourinary radiologist performed prospective readouts on in-house MRI scans following the Prostate Imaging Reporting and Data System (PI-RADS) version 2.0 or 2.1 and retrospective image quality assessments for all scans. A subgroup of patients underwent an MRI/US fusion-guided biopsy. A bpMRI-based lesion detection AI model previously developed using a completely separate dataset was tested on both MRI datasets. Detection rates were compared between external and in-house datasets with use of the paired comparison permutation tests. Factors associated with AI detection performance were assessed using multivariable generalized mixed-effects models, incorporating features selected through forward stepwise regression based on the Akaike information criterion. Results The study included 201 male patients (median age, 66 years [IQR, 62-70 years]; prostate-specific antigen density, 0.14 ng/mL² [IQR, 0.10-0.22 ng/mL²]) with a median interval between external and in-house MRI scans of 182 days (IQR, 97-383 days). For intraprostatic lesions, AI detected 39.7% (149 of 375) on external and 56.0% (210 of 375) on in-house MRI scans (P < .001). For csPCa-positive lesions, AI detected 61% (54 of 89) on external and 79% (70 of 89) on in-house MRI scans (P < .001). On external MRI scans, better overall lesion detection was associated with a higher PI-RADS score (odds ratio [OR] = 1.57; P = .005), larger lesion diameter (OR = 3.96; P < .001), better diffusion-weighted MRI quality (OR = 1.53; P = .02), and fewer lesions at MRI (OR = 0.78; P = .045). Better csPCa detection was associated with a shorter MRI interval between external and in-house scans (OR = 0.58; P = .03) and larger lesion size (OR = 10.19; P < .001). Conclusion The AI model exhibited modest performance in identifying both overall and csPCa-positive lesions on external bpMRI scans. Keywords: MR Imaging, Urinary, Prostate Supplemental material is available for this article. © RSNA, 2024.

Widespread adoption of mpMRI has led to a decrease in the number of patients requiring prostate biopsies. ⁶⁸Ga-PSMA-11 PET/CT has demonstrated added benefits in identifying csPCa. Integrating the use of these imaging techniques may hold promise for predicting the presence of csPCa without invasive biopsy. A retrospective analysis of 42 consecutive patients who underwent mpMRI, ⁶⁸Ga-PSMA-11 PET/CT, prostatic biopsy, and radical prostatectomy (RP) was carried out. A lesion-based model (n = 122) using prostatectomy histopathology as reference standard was used to analyze the accuracy of ⁶⁸Ga-PSMA-11 PET/CT, mpMRI alone, and both in combination to identify ISUP-grade group ≥ 2 lesions. ⁶⁸Ga-PSMA-11 PET/CT demonstrated greater specificity and positive predictive value (PPV), with values of 73.3% (vs. 40.0%) and 90.1% (vs. 82.2%), while the mpMRI Prostate Imaging Reporting and Data System (PI-RADS) 4–5 had better sensitivity and negative predictive value (NPV): 90.2% (vs. 78.5%) and 57.1% (vs. 52.4%), respectively. When used in combination, the sensitivity, specificity, PPV, and NPV were 74.2%, 83.3%, 93.2%, and 51.0%, respectively. Subgroup analysis of PI-RADS 3, 4, and 5 lesions was carried out. For PI-RADS 3 lesions, ⁶⁸Ga-PSMA-11 PET/CT demonstrated a NPV of 77.8%. For PI-RADS 4–5 lesions, ⁶⁸Ga-PSMA-11 PET/CT achieved PPV values of 82.1% and 100%, respectively, with an NPV of 100% in PI-RADS 5 lesions. A combination of ⁶⁸Ga-PSMA-11 PET/CT and mpMRI improved the radiological diagnosis of csPCa. This suggests that avoidance of prostate biopsy prior to RP may represent a valid option in a selected subgroup of high-risk patients with a high suspicion of csPCa on mpMRI and ⁶⁸Ga-PSMA-11 PET/CT.

Immunoglobulin G4-related disease (IgG4-RD) is a systemic fibroinflammatory disease characterized by raised serum IgG4 levels and tumefactive inflammation affecting multiple organ systems, typically involving the pancreas and biliary tree. Though rare, prostatic involvement has been reported in a few cases and is suspected to be an underreported entity. Our patient is a 63-year-old gentleman who has presented with an incidental “PI-RADS 5” (Prostate Imaging Reporting & Data System) prostate lesion and perivascular soft tissue cuffing of the superior rectal vessels on MRI rectum performed for surveillance of rectal neuroendocrine tumor. He had a history of lacrimal gland IgG4-RD. The lentiform prostate lesion subtly indents the prostate capsule, reminiscent of a periprostatic rather than an intraprostatic lesion. Perivascular cuffing of superior rectal vessels suggest inflammatory vasculitis of IgG4-RD. Differential diagnosis of periprostatic inflammatory IgG4-RD was considered, subsequently proven on MRI-ultrasound fusion targeted biopsy. Reported radiological findings of prostate IgG4-RD typically show diffuse chronic inflammation of the prostate, with a minority of the reports describing focal involvement, often mimicking focal prostate adenocarcinoma. Focal periprostatic involvement of IgG4-RD is an unusual manifestation which should be considered in patients with IgG4-RD who present with a periprostatic pseudotumor. IgG4-RD of the prostate usually responds well to steroid treatment without the need for surgery.

Focal therapy (FT) for localized prostate cancer induces architectural changes that complicate post-treatment multiparametric magnetic resonance imaging (mpMRI) surveillance. The Prostate Imaging after Focal Ablation (PI-FAB) system was developed to standardize the evaluation of in-field recurrence on mpMRI. This study assessed the diagnostic performance and inter-reader agreement of PI-FAB following focal cryoablation for localized prostate cancer. In this retrospective study (October 2019 to January 2024), 85 patients (140 lesion sites) underwent post-FT mpMRI and biopsy. Two radiologists (11 and 4 years of experience) independently scored mpMRIs using the three-point PI-FAB scale. Metrics included sensitivity, specificity, PPV, NPV, and accuracy. Inter-reader agreement was measured via quadratic weighted Cohen’s kappa (κ). Reader 1 (more experienced) demonstrated 83.9% sensitivity, 84.4% specificity, 60.5% PPV, and 94.8% NPV. Reader 2 demonstrated 71.4% sensitivity, 87.5% specificity, 58.8% PPV, and 92.5% NPV. Both achieved 84.3% accuracy. Inter-reader agreement was moderate (κ = 0.60). PI-FAB provides good diagnostic performance for detecting in-field recurrence after cryoablation. While reader experience may impact sensitivity, moderate agreement across experience levels supports PI-FAB’s validity and potential clinical utility in standardizing post-FT surveillance.