Citations

The translation of AI-generated brain metastases (BM) segmentation into clinical practice relies heavily on diverse, high-quality annotated medical imaging datasets. The BraTS-METS 2023 challenge has gained momentum for testing and benchmarking algorithms using rigorously annotated internationally compiled real-world datasets. This study presents the results of the segmentation challenge and characterizes the challenging cases that impacted the performance of the winning algorithms. Untreated brain metastases on standard anatomic MRI sequences (T1, T2, FLAIR, T1PG) from eight contributed international datasets were annotated in stepwise method: published UNET algorithms, student, neuroradiologist, final approver neuroradiologist. Segmentations were ranked based on lesion-wise Dice and Hausdorff distance (HD95) scores. False positives (FP) and false negatives (FN) were rigorously penalized, receiving a score of 0 for Dice and a fixed penalty of 374 for HD95. The mean scores for the teams were calculated. Eight datasets comprising 1303 studies were annotated, with 402 studies (3076 lesions) released on Synapse as publicly available datasets to challenge competitors. Additionally, 31 studies (139 lesions) were held out for validation, and 59 studies (218 lesions) were used for testing. Segmentation accuracy was measured as rank across subjects, with the winning team achieving a LesionWise mean score of 7.9. The Dice score for the winning team was 0.65 ± 0.25. Common errors among the leading teams included false negatives for small lesions and misregistration of masks in space. The Dice scores and lesion detection rates of all algorithms diminished with decreasing tumor size, particularly for tumors smaller than 100 mm3. In conclusion, algorithms for BM segmentation require further refinement to balance high sensitivity in lesion detection with the minimization of false positives and negatives. The BraTS-METS 2023 challenge successfully curated well- annotated, diverse datasets and identified common errors, facilitating the translation of BM segmentation across varied environments and providing the tools for future development of personalized volumetric reports to patients undergoing BM treatment.

Abstract Aims Syncope remains a diagnostic challenge despite advancements in testing and treatment. Cardiac syncope is an independent predictor of mortality and can be difficult to distinguish from other causes of transient loss of consciousness (TLOC). This paper explores whether artificial intelligence (AI) can improve the evaluation and management of patients with syncope. Methods and results We conducted a literature review and incorporated the opinions of experts in the fields of syncope and AI. The cause of TLOC is often unclear, hospitalization criteria are ambiguous, diagnostic tests are frequently non-informative, and assessments are costly. Patients are left with unanswered questions and limited guidance. Artificial intelligence (AI) has the potential to optimize syncope evaluation by processing large data sets, detecting imperceptible patterns, and assisting clinicians. However, AI has limitations, including errors, lack of human empathy, and uncertain clinical utility. Liability issues further complicate its integration. We present three viewpoints: (i) AI is crucial for advancing syncope management; (ii) AI can enhance the patient experience; and (iii) AI in syncope care is inevitable. Conclusion Artificial intelligence may improve syncope diagnosis and management, particularly through machine learning–based test interpretation and wearable device data. However, it has yet to surpass human clinical judgment in complex decision-making. Current challenges include gaps in understanding syncope mechanisms, AI interpretability, generalizability, and clinical integration. Standardized diagnostic approaches, real-world validation, and curated data sets are essential for progress. Artificial intelligence may enhance efficiency and communication but raises concerns regarding confidentiality, bias, inequities, and legal implications.

Abstract Measles virus (MeV) is a highly contagious pathogen capable of infecting immune cells and thereby causing immunosuppression and immune amnesia. A live-attenuated MeV vaccine is highly efficacious and has been safely administered for over 60 years. However, a detailed mechanistic understanding of the mode of its attenuation remains elusive. Here, we determined the molecular mechanisms underlying MeV vaccine attenuation in primary human immune cells. Enhanced infection of innate immune cells resulting in uncontrolled viral RNA synthesis and reduced ability to evade host antiviral responses induced a strong type-I interferon signature that arrested replication of the vaccine strain. In contrast, the wild type virus replicated slowly but steadily, avoiding detection by innate immunity.

Abstract Chimeric antigen receptor T (CART) cell therapy has shown clinical successes in B cell hematological malignancies. Two B cell maturation antigen (BCMA)-directed CART cell therapies treating multiple myeloma (MM) were approved by FDA but most patients relapse within 1-2 years. Ex vivo-expanded antigen specific T cells as cancer treatment rely on TCR-mediated killing of tumors. We hypothesized that arming ex vivo enriched tumor-specific T cells with CARs targeting a different antigen improves T cell functions. First, we expanded mucin 1 (MUC1)-activated T cells from patients with MM through ex vivo stimulation by MUC1 peptides-loaded autologous PBMCs for 19 days. Then, we engineered MUC1-activated T cells with 4-1BB-co-stimulated BCMA CAR to generate MUC1 TCR+BCMA CAR+ T cells. Baseline T cells without MUC1 stimulation were engineered with BCMA CAR as control. Baseline T cells, MUC1 T cells, baseline BCMA CART cells, or MUC1 BCMA CART cells were cocultured with luc+ RPMI-8226, a MUC1+BCMA+ MM cell line, for 3 days. MUC1 BCMA CART cells showed stronger tumor killing than BCMA CART or MUC1 T cells in all 3 patients (p<0.0001). MUC1 BCMA CART showed improved proliferation than BCMA CART or MUC1 T cells (p=0.0765, p<0.0001, respectively) after a 5-day stimulation by RPMI-8226. To exclude possibility of metabolic alterations by 19-day PBMC incubation leading to better T cell functions in MUC1 BCMA CART, baseline and MUC1-stimulated T cells from 3 healthy donors were processed for the same assays. Day-19 BCMA CART cells showed inferior tumor killing compared to baseline BCMA CART in all 3 donors (p<0.0001). Lower T cell proliferation was also observed in day-19 BCMA CART (p=0.0838). Our data indicate that improved T cell functions in MUC1 BCMA CART were unlikely due to metabolic changes and required stimulation by tumor antigens prior to CAR engineering likely through MUC1 TCR. Additionally, MUC1 T cells, BCMA CART, or MUC1 BCMA CART cells from tested patients were stimulated by RPMI-8226 for bulk RNA sequencing. Both BCMA CART and MUC1 T cells had distinct transcriptomic features from MUC1 BCMA CART cells. Differential gene expression analysis showed upregulation of T cell function-associated genes (RYR2, ITGA1, GNLY, GZMA, etc.) in MUC1 BCMA CART vs BCMA CART. Enrichment of T cell function-associated pathways was also observed in MUC1 BCMA CART. Analysis between MUC1 T cell and MUC1 BCMA CART showed upregulation of T cell function-associated genes (CSF2, CRTAM, IFNG, GZMB, etc.) and enrichment of immune response pathways in MUC1 BCMA CART. Our findings suggest that equipping ex vivo-enriched tumor-specific T cells with CARs targeting a different antigen improves T cell functions in vitro. Importantly, this provides a rationale for a therapeutic regimen where patients receive ex vivo-enriched tumor-induced T cells followed by CAR engineering. Persistence of the engineered T cells is being investigated Citation Format: Kun Yun, R. Leo Sakemura, Ismail Can, Omar Gutierrez Ruiz, Ekene J. Ogbodo, Truc N. Huynh, Claudia Manriquez Roman, Olivia Sirpilla, Carli M. Stewart, Jennifer M. Feigin, Grace E. DeFranco, Sophia Y. Goldberg, Dominic A. Skeele, Mariam Stein, Dara Missan, Latha Pathangey, Long Mai, Hong Xia, Brooke Kimball, Lionel A. Kankeu Fonkoua, Mehrdad Hefazi, Michael W. Ruff, Elizabeth L. Siegler, Leif Bergsagel, Michael Gustafson, Sandra Gendler, Saad S. Kenderian. BCMA CAR engineering arms T cells with superior functions post ex vivo stimulation by MUC1 peptide-loaded autologous PBMCs [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6103.

We describe the design and results from the BraTS 2023 Intracranial Meningioma Segmentation Challenge. The BraTS Meningioma Challenge differed from prior BraTS Glioma challenges in that it focused on meningiomas, which are typically benign extra-axial tumors with diverse radiologic and anatomical presentation and a propensity for multiplicity. Nine participating teams each developed deep-learning automated segmentation models using image data from the largest multi-institutional systematically expert annotated multilabel multi-sequence meningioma MRI dataset to date, which included 1000 training set cases, 141 validation set cases, and 283 hidden test set cases. Each case included T2, FLAIR, T1, and T1Gd brain MRI sequences with associated tumor compartment labels delineating enhancing tumor, non-enhancing tumor, and surrounding non-enhancing FLAIR hyperintensity. Participant automated segmentation models were evaluated and ranked based on a scoring system evaluating lesion-wise metrics including dice similarity coefficient (DSC) and 95% Hausdorff Distance. The top ranked team had a lesion-wise median dice similarity coefficient (DSC) of 0.976, 0.976, and 0.964 for enhancing tumor, tumor core, and whole tumor, respectively and a corresponding average DSC of 0.899, 0.904, and 0.871, respectively. These results serve as state-of-the-art benchmarks for future pre-operative meningioma automated segmentation algorithms. Additionally, we found that 1286 of 1424 cases (90.3%) had at least 1 compartment voxel abutting the edge of the skull-stripped image edge, which requires further investigation into optimal pre-processing face anonymization steps.

773 Background: BCG-unresponsive High-risk non-muscle invasive bladder cancer (HRNMIBC) tumours have an ominous prognosis and require effective, tolerable treatments. Uptake remains low for FDA approved agents (including systemic pembrolizumab, nogapendekin alfa inbakicept-pmln or N-803) due to toxicity concerns. We hypothesize direct intravesical administration of a PDL1 inhibitor could be effective with less systemic toxicity. However, it is unknown whether antibodies delivered via this route can reach tumor vasculature. INVEST is a phase Ib window of opportunity study investigating the safety and preliminary activity of passive instillation and direct injection of intravesical atezolizumab into the tumour/bladder wall. Methods: Eligible participants (ECOG performance status 0-2) are awaiting radical cystectomy (RC) for any stage urothelial cell carcinoma. Participants with muscle invasive bladder cancer, must be ineligible for/ refuse cisplatin based neo-adjuvant chemotherapy. Atezolizumab (600mg or 1200mg) is administered via direct injection into the tumour/bladder or by instillation into the bladder. The 3+3 design is utilised in the dose confirmation stages where participants receive either Single or Multiple (between 3 and 6) dose(s) of treatment before RC. Efficacy signals are derived from pathological complete response at RC and progression-free survival at 2 years. Primary endpoint, Dose confirmation stage: The number of dose-limiting toxicities (DLTs) observed from first dose of trial treatment to RC. Results: Enrolment began in May 2023. 13 participants were recruited to the single dose cohorts: 11 male, and 2 female. Age (Range; 43-86); median 70y, 9 NMIBC, 4 MIBC. 7 for direct injection (4 at 600mg and 3 at 1200mg dose) and 6 for passive instillation (3 at 600mg and 3 at 1200mg). Due to technical issues (syringe size) the first direct injection participant did not receive the full 600mg and were replaced. 12 participants received trial treatment as planned. There were no DLTs; 5 participants experienced non-trial treatment related SAEs (4 infections, 2 ileus, 2 renal impairment; 7/8 events occurred post-RC and all SAEs resolved). All participants underwent planned RC within the protocol stipulated timeline. Median time from last treatment to RC was18 days. 5 of 13 participants had downstaging of tumour in cystectomy specimen. Following independent Safety Review Committee review, recruitment is now open for the dose confirmation phase (Multiple 1200mg dose) in both treatment routes. Conclusions: Single dose atezolizumab treatment at a dose of 600mg and 1200mg in passive instillation and direct injection cohorts was well tolerated. Early efficacy data is encouraging. Multiple weekly doses of 1200mg in each treatment route are now being investigated within the INVEST trial. Clinical trial information: ISRCTN15842444 .

ImportanceMyeloperoxidase is one of the most abundant peroxidase enzymes in activated myeloid cells. Myeloperoxidase inhibitors may have a clinical benefit in amyotrophic lateral sclerosis (ALS) by slowing neurodegeneration via reduced neuroinflammation and oxidative stress.ObjectiveTo determine the safety, tolerability, and efficacy of verdiperstat, a selective myeloperoxidase inhibitor, in ALS.Design, Settings, and ParticipantsVerdiperstat was tested as a regimen of the HEALEY ALS Platform Trial, a multicenter, double-blind, perpetual platform design, randomized clinical trial, with sharing of trial infrastructure and placebo data across multiple regimens. The study was conducted at 54 ALS referral centers across the US from July 2020 to April 2022. Adult participants with a diagnosis of clinically possible, probable, laboratory-supported probable, or definite ALS defined by the revised El Escorial criteria were randomized to verdiperstat or regimen-specific placebo. An additional group of participants concurrently randomized to placebo from other regimens was included in the analyses.InterventionsEligible participants were randomized in a 3:1 ratio to receive oral verdiperstat, 600 mg, twice daily or matching placebo for a planned placebo-controlled duration of 24 weeks.Main Outcomes and MeasuresThe primary efficacy outcome was change from baseline through week 24 in disease severity, as measured by a joint model of ALS Functional Rating Scale–Revised and survival, with the treatment effect quantified by the disease rate ratio (DRR), with DRR less than 1 indicating a slowing in disease progression of verdiperstat relative to placebo.ResultsA total of 167 participants (mean [SD] age, 58.5 [11.4] years; 59 [35.3%] female; 108 [64.6%] male) were randomized to either verdiperstat (126 [75.4%]) or to placebo (41 [25.6%]). Among the participants randomized to the verdiperstat regimen, 130 (78%) completed the trial. The estimated DRR was 0.98 (95% credible interval, 0.77-1.24; posterior probability = 0.57 for slowing of disease progression [DRR <1]). Verdiperstat was estimated to slow progression by 2% vs placebo (95% credible interval, −23% to 24%; posterior probability 0.57). Verdiperstat was overall safe and well tolerated. Common adverse events in the verdiperstat group were nausea, insomnia, and elevated thyrotropin levels.Conclusions and RelevanceResults demonstrate that treatment with verdiperstat was unlikely to alter disease progression in ALS.Trial RegistrationClinical Trial Identifiers: NCT04297683 and NCT04436510

ImportanceAmyotrophic lateral sclerosis (ALS) is a fatal disease. The sigma-1 (σ1) receptor emerged as a target for intervention.ObjectiveTo determine the effects of pridopidine, a σ1-receptor agonist, in ALS.Design, Settings, and ParticipantsPridopidine was tested as a regimen of the HEALEY ALS Platform Trial, a phase 2/3, multicenter, randomized, double-blind, platform trial. The study was conducted at 54 sites in the US from January 2021 to July 2022 (final follow-up, July 14, 2022). A total of 163 participants with ALS were randomized to receive pridopidine or placebo. An additional 122 concurrently randomized participants were assigned to receive placebo in other regimens and included in the analyses.InterventionsEligible participants were randomized 3:1 to receive oral pridopidine 45 mg twice daily (n = 121) or matching oral placebo (n = 42) for a planned duration of 24 weeks.Main Outcomes and MeasuresThe primary efficacy outcome was change from baseline through week 24 in ALS disease severity, analyzed using a bayesian shared parameter model, which has components for function (Revised Amyotrophic Lateral Sclerosis Functional Rating Scale [ALSFRS-R]) and survival that were linked through an integrated estimate of treatment-dependent disease slowing across these 2 components. This was denoted as the disease rate ratio (DRR), with DRR less than 1 indicating a slowing in disease progression on pridopidine relative to placebo. There were 5 key secondary end points: time to 2-point or greater reduction in ALSFRS-R total score among participants with bulbar dysfunction at baseline, rate of decline in slow vital capacity among participants with bulbar dysfunction at baseline, percentage of participants with no worsening in the ALSFRS-R bulbar domain score, time to 1-point or greater change in the ALSFRS-R bulbar domain score, and time to death or permanent assisted ventilation.ResultsAmong 162 patients (mean age, 57.5 years; 35% female) who were randomized to receive the pridopidine regimen and included in the primary efficacy analysis, 136 (84%) completed the trial. In the primary analysis comparing pridopidine vs the combined placebo groups, there was no significant difference between pridopidine and placebo in the primary end point (DRR, 0.99 [95% credible interval, 0.80-1.21]; probability of DRR <1, 0.55) and no differences were seen in the components of ALSFRS-R or survival. There was no benefit of pridopidine on the secondary end points. In the safety dataset (pridopidine, n = 121; placebo, n = 163), the most common adverse events were falls (28.1% vs 29.3%, respectively) and muscular weakness (24.0% vs 31.7%, respectively).Conclusions and RelevanceIn this 24-week study, pridopidine did not impact the progression of ALS.Trial RegistrationClinicalTrials.gov Identifiers: NCT04297683, NCT04615923