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.

Image-Guided Transbronchial Pulmonary Cryoablation with a Flexible Cryoprobe in Swine: Performance and Radiology-Pathology Correlation.

PURPOSE: To evaluate the performance of a prototype flexible transbronchial cryoprobe compared with that of percutaneous transthoracic cryoablation and to define cone-beam computed tomography (CT) imaging and pathology cryolesion features in an in vivo swine model. MATERIALS AND METHODS: Transbronchial cryoablation was performed with a prototype flexible cryoprobe (3 central and 3 peripheral lung ablations in 3 swine) and compared with transthoracic cryoablation performed with a commercially available rigid cryoprobe (2 peripheral lung ablations in 1 swine). Procedural time and cryoablation success rates for endobronchial navigation and cryoneedle deployment were measured. Intraoperative cone-beam CT imaging features of cryolesions were characterized and correlated with gross pathology and hematoxylin and eosin-stained sections of the explanted cryolesions. RESULTS: The flexible cryoprobe was successfully navigated and delivered to each target through a steerable guiding sheath (6/6). At 4 minutes after ablation, 5 of 6 transbronchial and 2 of 2 transthoracic cryolesions were visible on cone-beam CT. The volumes on cone-beam CT images were 55.5 cm3 (SE ± 8.0) for central transbronchial ablations (n = 2), 72.5 cm3 (SE ± 8.1) for peripheral transbronchial ablations (n = 3), and 79.5 cm3 (SE ±11.6) for peripheral transthoracic ablations (n = 2). Pneumothorax developed in 1 animal after transbronchial ablation and during ablation in the transthoracic cryoablation. Images of cryoablation zones on cone-beam CT correlated well with the matched gross pathology and histopathology sections of the cryolesions. CONCLUSIONS: Transbronchial cryoablation with a flexible cryoprobe, delivered through a steerable guiding sheath, is feasible. Transbronchial cryoablation zones are imageable with cone-beam CT, with gross pathology and histopathology similar to those of transthoracic cryoablation.

A Review of Ultrasound-Mediated Checkpoint Inhibitor Immunotherapy.

Over the past decade, immunotherapy has emerged as a major modality in cancer medicine. However, despite its unprecedented success, immunotherapy currently benefits only a subgroup of patients, may induce responses of limited duration and is associated with potentially treatment-limiting side effects. In addition, responses to immunotherapeutics are sometimes diminished by the emergence of a complex array of resistance mechanisms. The efficacy of immunotherapy depends on dynamic interactions between tumour cells and the immune landscape in the tumour microenvironment. Ultrasound, especially in conjunction with cavitation-promoting agents such as microbubbles, can assist in the uptake and/or local release of immunotherapeutic agents at specific target sites, thereby increasing treatment efficacy and reducing systemic toxicity. There is also increasing evidence that ultrasound and/or cavitation may themselves directly stimulate a beneficial immune response. In this review, we summarize the latest developments in the use of ultrasound and cavitation agents to promote checkpoint inhibitor immunotherapy.

C-DARL: Contrastive diffusion adversarial representation learning for label-free blood vessel segmentation.

Blood vessel segmentation in medical imaging is one of the essential steps for vascular disease diagnosis and interventional planning in a broad spectrum of clinical scenarios in image-based medicine and interventional medicine. Unfortunately, manual annotation of the vessel masks is challenging and resource-intensive due to subtle branches and complex structures. To overcome this issue, this paper presents a self-supervised vessel segmentation method, dubbed the contrastive diffusion adversarial representation learning (C-DARL) model. Our model is composed of a diffusion module and a generation module that learns the distribution of multi-domain blood vessel data by generating synthetic vessel images from diffusion latent. Moreover, we employ contrastive learning through a mask-based contrastive loss so that the model can learn more realistic vessel representations. To validate the efficacy, C-DARL is trained using various vessel datasets, including coronary angiograms, abdominal digital subtraction angiograms, and retinal imaging. Experimental results confirm that our model achieves performance improvement over baseline methods with noise robustness, suggesting the effectiveness of C-DARL for vessel segmentation.Our source code is available at https://github.com/boahK/MEDIA_CDARL.2.

Innovations in Image-Guided Procedures: Unraveling Robot-Assisted Non-Hepatic Percutaneous Ablation.

Interventional oncology is routinely tasked with the feat of tumor characterization or destruction, via image-guided biopsy and tumor ablation, which may pose difficulties due to challenging-to-reach structures, target complexity, and proximity to critical structures. Such procedures carry a risk-to-benefit ratio along with measurable radiation exposure. To streamline the complexity and inherent variability of these interventions, various systems, including table-, floor-, gantry-, and patient-mounted (semi-) automatic robotic aiming devices, have been developed to decrease human error and interoperator and intraoperator outcome variability. Their implementation in clinical practice holds promise for enhancing lesion targeting, increasing accuracy and technical success rates, reducing procedure duration and radiation exposure, enhancing standardization of the field, and ultimately improving patient outcomes. This narrative review collates evidence regarding robotic tools and their implementation in interventional oncology, focusing on clinical efficacy and safety for nonhepatic malignancies.

VB-111 (ofranergene obadenovec) in combination with nivolumab in patients with microsatellite stable colorectal liver metastases: a single center, single arm, phase II trial.

BACKGROUND: Microsatellite stable colorectal liver metastases (MSS CLM) maintain an immunosuppressive tumor microenvironment (TME). Historically, immune-based approaches have been ineffective. VB-111 (ofranergene obadenovec) is a genetically-modified adenoviral vector targeting the TME; its unique dual mechanism induces an immune response and disrupts neovascularization. Checkpoint inhibition may synergize the immune response induced by viral-mediated anti-angiogenic gene therapy. We aimed to examine the safety and antitumor activity of VB-111 and nivolumab in patients with refractory MSS CLM and to characterize immunological treatment-response. METHODS: This was a phase II study of adult patients with histologically-confirmed MSS CLM who progressed on prior therapy. A priming dose of VB-111 1×1013 viral particles was given intravenously 2 weeks prior to starting biweekly nivolumab 240 mg and continued every 6 weeks. The combination continued until disease progression or unacceptable toxicity. The primary objectives were overall response rate and safety/tolerability. Secondary objectives included median overall survival and progression-free survival. Correlative studies were performed on paired tumor biopsies and blood. RESULTS: Between August 2020 and December 2021, 14 patients were enrolled with median age 50.5 years (40-75), and 14% were women. Median follow-up was 5.5 months. Of the 10 evaluable patients, the combination of VB-111 and nivolumab failed to demonstrate radiographic responses; at best, 2 patients had stable disease. Median overall survival was 5.5 months (95% CI: 2.3 to 10.8), and median progression-free survival was 1.8 months (95% CI: 1.4 to 1.9). The most common grade 3-4 treatment-related adverse events were fever/chills, influenza-like symptoms, and lymphopenia. No treatment-related deaths were reported. Qualitative analysis of immunohistochemical staining of paired tumor biopsies did not demonstrate significant immune infiltration after treatment, except for one patient who had exceptional survival (26.0 months). Immune analysis of peripheral blood mononuclear cells showed an increase of PD-1highKi67highCD8+ T cells and HLA-DRhigh T cells after VB-111 priming dose. Plasma cytokines interleukin-10 and tumor necrosis factor-α increased after treatment with both drugs. CONCLUSION: In patients with MSS CLM, VB-111 and nivolumab did not improve overall response rate or survival but were tolerated with minimal toxicities. While challenging to distinguish between antiviral or antitumor, correlative studies demonstrated an immune response with activation and proliferation of CD8+ T cells systemically that was poorly sustained. TRIAL REGISTRATION NUMBER: NCT04166383.

Smart goggles augmented reality CT-US fusion compared to conventional fusion navigation for percutaneous needle insertion.

PURPOSE: Targeting accuracy determines outcomes for percutaneous needle interventions. Augmented reality (AR) in IR may improve procedural guidance and facilitate access to complex locations. This study aimed to evaluate percutaneous needle placement accuracy using a goggle-based AR system compared to an ultrasound (US)-based fusion navigation system. METHODS: Six interventional radiologists performed 24 independent needle placements in an anthropomorphic phantom (CIRS 057A) in four needle guidance cohorts (n = 6 each): (1) US-based fusion, (2) goggle-based AR with stereoscopically projected anatomy (AR-overlay), (3) goggle AR without the projection (AR-plain), and (4) CT-guided freehand. US-based fusion included US/CT registration with electromagnetic (EM) needle, transducer, and patient tracking. For AR-overlay, US, EM-tracked needle, stereoscopic anatomical structures and targets were superimposed over the phantom. Needle placement accuracy (distance from needle tip to target center), placement time (from skin puncture to final position), and procedure time (time to completion) were measured. RESULTS: Mean needle placement accuracy using US-based fusion, AR-overlay, AR-plain, and freehand was 4.5 ± 1.7 mm, 7.0 ± 4.7 mm, 4.7 ± 1.7 mm, and 9.2 ± 5.8 mm, respectively. AR-plain demonstrated comparable accuracy to US-based fusion (p = 0.7) and AR-overlay (p = 0.06). Excluding two outliers, AR-overlay accuracy became 5.9 ± 2.6 mm. US-based fusion had the highest mean placement time (44.3 ± 27.7 s) compared to all navigation cohorts (p < 0.001). Longest procedure times were recorded with AR-overlay (34 ± 10.2 min) compared to AR-plain (22.7 ± 8.6 min, p = 0.09), US-based fusion (19.5 ± 5.6 min, p = 0.02), and freehand (14.8 ± 1.6 min, p = 0.002). CONCLUSION: Goggle-based AR showed no difference in needle placement accuracy compared to the commercially available US-based fusion navigation platform. Differences in accuracy and procedure times were apparent with different display modes (with/without stereoscopic projections). The AR-based projection of the US and needle trajectory over the body may be a helpful tool to enhance visuospatial orientation. Thus, this study refines the potential role of AR for needle placements, which may serve as a catalyst for informed implementation of AR techniques in IR.

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.

Reducing False-Positives Due to Urinary Stagnation in the Prostatic Urethra on 18 F-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 18 F-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 18 F-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.

Timing, incidence and management of delayed bleeding after partial nephrectomy in patients at risk for recurrent, bilateral, multifocal renal tumors.

INTRODUCTION: Delayed bleeding is a potentially serious complication after partial nephrectomy (PN), with reported rates of 1%-2%. Patients with multiple renal tumors, including those with hereditary forms of kidney cancer, are often managed with resection of multiple tumors in a single kidney which may increase the risk of delayed bleeding, though outcomes have not previously been reported specifically in this population. The objective of this study was to evaluate the incidence and timing of delayed bleeding as well as the impact of intervention on renal functional outcomes in a cohort primarily made up of patients at risk for bilateral, multifocal renal tumors. METHODS: A retrospective review of a prospectively maintained database of patients with known or suspected predisposition to bilateral, multifocal renal tumors who underwent PN from 2003 to 2023 was conducted. Patients who presented with delayed bleeding were identified. Patients with delayed bleeding were compared to those without. Comparative statistics and univariate logistic regression were used to determine potential risk factors for delayed bleeding. RESULTS: A total of 1256 PN were performed during the study period. Angiographic evidence of pseudoaneurysm, AV fistula and/or extravasation occurred in 24 cases (1.9%). Of these, 21 were symptomatic presenting with gross hematuria in 13 (54.2%), decreasing hemoglobin in 4(16.7%), flank pain in 2(8.3%), and mental status change in 2 (8.3%), while 3 patients were asymptomatic. Median number of resected tumors was 5 (IQR 2-8). All patients underwent angiogram with super-selective embolization. Median time to bleed event was 13.5 days (IQR 7-22). Factors associated with delayed bleeding included open approach (OR 2.2, IQR(1.06-5.46), P = 0.04 and left-sided surgery (OR 4.93, IQR(1.67-14.5), P = 0.004. Selective embolization had little impact on ultimate renal functional outcomes, with a median change of 11% from the baseline eGFR after partial nephrectomy and embolization. One patient required total nephrectomy for refractory bleeding after embolization. CONCLUSIONS: Delayed bleeding after PN in a cohort of patients with multifocal tumors is an infrequent event, with similar rates to single tumor series. Patients should be counseled regarding timing and symptoms of delayed bleeding and multidisciplinary management with interventional radiology is critical for timely diagnosis and treatment.

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.

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 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.

Combined immune checkpoint inhibition with durvalumab and tremelimumab with and without radiofrequency ablation in patients with advanced biliary tract carcinoma.

BACKGROUND: Current standard of care for advanced biliary tract cancer (BTC) is gemcitabine, cisplatin plus anti-PD1/PD-L1, but response rates are modest. The purpose of this study was to explore the efficacy and safety of durvalumab (anti-PD-L1) and tremelimumab (anti-CTLA-4), with and without an interventional radiology (IR) procedure in advanced BTC. METHODS: Eligible patients with advanced BTC who had received or refused at least one prior line of systemic therapy were treated with tremelimumab and durvalumab for four combined doses followed by monthly durvalumab alone with and without an IR procedure until the progression of disease or unacceptable toxicity. Objective response was assessed through CT or MRI by Response Evaluation Criteria in Solid Tumors (RECIST, version 1.1) every 8 weeks. Adverse events (AEs) were recorded and managed. The primary endpoint was 6-month progression-free survival (PFS). RESULTS: Twenty-three patients with advanced BTC were enrolled; 17 patients were assigned to treatment with durvalumab and tremelimumab (Durva/Treme); and 6 patients were treated with the combination of durvalumab, tremelimumab plus IR procedure (Durva/Treme + IR). The best clinical responses in the Durva/Treme arm were partial response (n = 1), stable disease (n = 5), progressive disease (n = 5), and in the Durva/Treme + IR arm: partial response (n = 0), stable disease (n = 3), progressive disease (n = 3). The median PFS was 2.2 months (95% CI: 1.3-3.1 months) in the Durva/Treme arm and 2.9 months (95% CI: 1.9-4.7 months) in the Durva/Treme + IR arm (p = 0.27). The median OS was 5.1 months (95% CI: 2.5-6.9 months) in the Durva/Treme arm and 5.8 months (95% CI: 2.9-40.1 months) in the Durva/Treme + IR arm (p = 0.31). The majority of AEs were grades 1-2. CONCLUSION: Durva/Treme and Durva/Treme + IR showed similar efficacy. With a manageable safety profile. Larger studies are needed to fully characterize the efficacy of Durva/Treme ± IR in advanced BTC.

X-ray imageable, drug-loaded hydrogel that forms at body temperature for image-guided, needle-based locoregional drug delivery.

Liver cancer ranks as the fifth leading cause of cancer-related death globally. Direct intratumoral injections of anti-cancer therapeutics may improve therapeutic efficacy and mitigate adverse effects compared to intravenous injections. Some challenges of intratumoral injections are that the liquid drug formulation may not remain localized and have unpredictable volumetric distribution. Thus, drug delivery varies widely, highly-dependent upon technique. An X-ray imageable poloxamer 407 (POL)-based drug delivery gel was developed and characterized, enabling real-time feedback. Utilizing three needle devices, POL or a control iodinated contrast solution were injected into an ex vivo bovine liver. The 3D distribution was assessed with cone beam computed tomography (CBCT). The 3D distribution of POL gels demonstrated localized spherical morphologies regardless of the injection rate. In addition, the gel 3D conformal distribution could be intentionally altered, depending on the injection technique. When doxorubicin (DOX) was loaded into the POL and injected, DOX distribution on optical imaging matched iodine distribution on CBCT suggesting spatial alignment of DOX and iodine localization in tissue. The controllability and localized deposition of this formulation may ultimately reduce the dependence on operator technique, reduce systemic side effects, and facilitate reproducibility across treatments, through more predictable standardized delivery.

Deep learning-based image quality assessment: impact on detection accuracy of prostate cancer extraprostatic extension on MRI.

OBJECTIVE: To assess impact of image quality on prostate cancer extraprostatic extension (EPE) detection on MRI using a deep learning-based AI algorithm. MATERIALS AND METHODS: This retrospective, single institution study included patients who were imaged with mpMRI and subsequently underwent radical prostatectomy from June 2007 to August 2022. One genitourinary radiologist prospectively evaluated each patient using the NCI EPE grading system. Each T2WI was classified as low- or high-quality by a previously developed AI algorithm. Fisher's exact tests were performed to compare EPE detection metrics between low- and high-quality images. Univariable and multivariable analyses were conducted to assess the predictive value of image quality for pathological EPE. RESULTS: A total of 773 consecutive patients (median age 61 [IQR 56-67] years) were evaluated. At radical prostatectomy, 23% (180/773) of patients had EPE at pathology, and 41% (131/318) of positive EPE calls on mpMRI were confirmed to have EPE. The AI algorithm classified 36% (280/773) of T2WIs as low-quality and 64% (493/773) as high-quality. For EPE grade ≥ 1, high-quality T2WI significantly improved specificity for EPE detection (72% [95% CI 67-76%] vs. 63% [95% CI 56-69%], P = 0.03), but did not significantly affect sensitivity (72% [95% CI 62-80%] vs. 75% [95% CI 63-85%]), positive predictive value (44% [95% CI 39-49%] vs. 38% [95% CI 32-43%]), or negative predictive value (89% [95% CI 86-92%] vs. 89% [95% CI 85-93%]). Sensitivity, specificity, PPV, and NPV for EPE grades ≥ 2 and ≥ 3 did not show significant differences attributable to imaging quality. For NCI EPE grade 1, high-quality images (OR 3.05, 95% CI 1.54-5.86; P < 0.001) demonstrated a stronger association with pathologic EPE than low-quality images (OR 1.76, 95% CI 0.63-4.24; P = 0.24). CONCLUSION: Our study successfully employed a deep learning-based AI algorithm to classify image quality of prostate MRI and demonstrated that better quality T2WI was associated with more accurate prediction of EPE at final pathology.

X-ray imageable, drug-loaded hydrogel that forms at body temperature for image-guided, needle- based locoregional drug delivery.

Liver cancer ranks as the fifth leading cause of cancer-related death globally. Direct intratumoral injections of anti-cancer therapeutics may improve therapeutic efficacy and mitigate adverse effects compared to intravenous injections. Some challenges of intratumoral injections are that the liquid drug formulation may not remain localized and have unpredictable volumetric distribution. Thus, drug delivery varies widely, highly-dependent upon technique. An x-ray imageable poloxamer 407 (POL)-based drug delivery gel was developed and characterized, enabling real-time feedback. Utilizing three needle devices, POL or a control iodinated contrast solution were injected into an ex vivo bovine liver. The 3D distribution was assessed with cone beam computed tomography (CBCT). The 3D distribution of POL gels demonstrated localized spherical morphologies regardless of the injection rate. In addition, the gel 3D conformal distribution could be intentionally altered, depending on the injection technique. When doxorubicin (DOX) was loaded into the POL and injected, DOX distribution on optical imaging matched iodine distribution on CBCT suggesting spatial alignment of DOX and iodine localization in tissue. The controllability and localized deposition of this formulation may ultimately reduce the dependence on operator technique, reduce systemic side effects, and facilitate reproducibility across treatments, through more predictable standardized delivery.

Ultrasound and x-ray imageable poloxamer-based hydrogel for loco-regional therapy delivery in the liver.

Intratumoral injections have the potential for enhanced cancer treatment efficacy while reducing costs and systemic exposure. However, intratumoral drug injections can result in substantial off-target leakage and are invisible under standard imaging modalities like ultrasound (US) and x-ray. A thermosensitive poloxamer-based gel for drug delivery was developed that is visible using x-ray imaging (computed tomography (CT), cone beam CT, fluoroscopy), as well as using US by means of integrating perfluorobutane-filled microbubbles (MBs). MBs content was optimized using tissue mimicking phantoms and ex vivo bovine livers. Gel formulations less than 1% MBs provided gel depositions that were clearly identifiable on US and distinguishable from tissue background and with minimal acoustic artifacts. The cross-sectional areas of gel depositions obtained with US and CT imaging were similar in studies using ex vivo bovine liver and postmortem in situ swine liver. The gel formulation enhanced multimodal image-guided navigation, enabling fusion of ultrasound and x-ray/CT imaging, which may enhance targeting, definition of spatial delivery, and overlap of tumor and gel. Although speculative, such a paradigm for intratumoral drug delivery might streamline clinical workflows, reduce radiation exposure by reliance on US, and boost the precision and accuracy of drug delivery targeting during procedures. Imageable gels may also provide enhanced temporal and spatial control of intratumoral conformal drug delivery.

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.

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.