Citations

Abstract Generative artificial intelligence (AI) holds immense potential for medical applications, but the lack of a comprehensive evaluation framework and methodological deficiencies in existing studies hinder its effective implementation. Standardized assessment guidelines are crucial for ensuring reliable and consistent evaluation of generative AI in healthcare. Our objective is to develop robust, standardized guidelines tailored for evaluating generative AI performance in medical contexts. Through a rigorous literature review utilizing the Web of Sciences, Cochrane Library, PubMed, and Google Scholar, we focused on research testing generative AI capabilities in medicine. Our multidisciplinary team of experts conducted discussion sessions to develop a comprehensive 32‐item checklist. This checklist encompasses critical evaluation aspects of generative AI in medical applications, addressing key dimensions such as question collection, querying methodologies, and assessment techniques. The checklist and its broader assessment framework provide a holistic evaluation of AI systems, delineating a clear pathway from question gathering to result assessment. It guides researchers through potential challenges and pitfalls, enhancing research quality and reporting and aiding the evolution of generative AI in medicine and life sciences. Our framework furnishes a standardized, systematic approach for testing generative AI's applicability in medicine. For a concise checklist, please refer to Table S or visit GenAIMed.org .

e13119 Background: Breast cancer metastasis to the central nervous system (CNS) and intracranial area often leads to severe morbidity and mortality. While systemic treatments manage systemic tumors, their impact on CNS/intracranial metastases is limited. Cell based therapies like SV-BR-1-GM may overcome this limitation. SV-BR-1-GM is a GM-CSF secreting, antigen-presenting immortalized breast cancer cell line. We report its efficacy on CNS tumor regression. Methods: Post-hoc analysis of a subset of all advanced MBC patients with CNS/intracranial metastases from previous phase 1, 2, and an ongoing randomized phase 2, trials (n=7 subjects) of SV-BR-1-GM. Bria-IMT monotherapy (intradermal SV-BR-1-GM + cyclophosphamide 300 mg/m2, +interferon-alpha) was administered q3wks to 2 subjects; an immune checkpoint inhibitor (ICI) was administered with each cycle in the combination setting (n=5). Results: There were 7 patients with intracranial metastases at baseline of whom 5 had intracranial tumor responses. Of the 5 responders, median pt age = 66 (range 58-70), median prior lines = 5.5 (3-13), and median time from initial diagnosis = 7.5 years (3-10). 4/5 pts had ≥1 HLA class match with SV-BR-1-GM. 3 were HR + and 1 HER2 +. Pt A002 (ER+/PR+, HER2-) received 6/6 planned cycles, achieving an 83% reduction in breast lesions; after 3 months off therapy, imaging revealed multiple brain metastases and recurrence of right breast lesions. Following 3 additional cycles there was marked regression in both brain and breast lesions. Pt 02-003 (ER-/PR-, HER2 1+ (IHC)) received 5 cycles of SV-BR-1-GM and observed a 60% reduction (5mm -> 2mm) in a left parietal periventricular lesion, and complete resolution (CR) of a left parietal lobe. Pt 06-005 (ER+/PR-, HER2 1+) demonstrated a 56% (9mm -> 4mm) reduction in the thickness of a dural lesion overlying the left anterior temporal lobe after 6 cycles of SV-BR-1-GM + ICI. After 2 additional cycles of SV-BR-1-GM + ICI and 1 cycle of SV-BR-1-GM (ICI skipped on final cycle due to grade 2 AE), patient achieved CR of a left orbital mass observed at baseline (24mm -> 0m). Pt06-007’s (ER-/PR-, HER2 1+) CNS lesions showed a median reduction of 49% compared to baseline after 3 cycles. Pt 11-018 (ER+/PR-, HER2 +) showed standard progression (SD) at initial restaging; 14% increase in sum of diameters of intracranial metastasis. Subsequent imaging, after 3 cycles, showed significant orbital tumor reduction along with reported improvement in eye pain. Conclusions: This post-hoc analysis demonstrated promising efficacy of the Bria-IMT regimen either alone or in combination with ICIs in CNS/intracranial metastasis. The CNS/intracranial regression seen across all breast cancer subtypes among heavily pretreated patients highlights the potential of SV-BR-1-GM in managing CNS metastasis. Ongoing trials will further evaluate the efficacy of the Bria-IMT regimen in patients with CNS/intracranial metastasis. Clinical trial information: NCT03328026 .

Abstract Lewy body dementia and Alzheimer's disease (AD) are leading causes of cognitive impairment, characterized by distinct but overlapping neuropathological hallmarks. Lewy body disease (LBD) is characterized by α-synuclein aggregates in the form of Lewy bodies as well as the deposition of extracellular amyloid plaques, with many cases also exhibiting neurofibrillary tangle (NFT) pathology. In contrast, AD is characterized by amyloid plaques and neurofibrillary tangles. Both conditions often co-occur with additional neuropathological changes, such as vascular disease and TDP-43 pathology. To elucidate shared and distinct molecular signatures underlying these mixed neuropathologies, we extensively analysed transcriptional changes in the anterior cingulate cortex, a brain region critically involved in cognitive processes. We performed bulk tissue RNA sequencing from the anterior cingulate cortex and determined differentially expressed genes (q-value <0.05) in control (n = 81), LBD (n = 436), AD (n = 53) and pathological amyloid cases consisting of amyloid pathology with minimal or no tau pathology (n = 39). We used gene set enrichment and weighted gene correlation network analysis to understand the pathways associated with each neuropathologically defined group. LBD cases had strong upregulation of inflammatory pathways and downregulation of metabolic pathways. The LBD cases were further subdivided into either high Thal amyloid, Braak NFT, or low pathological burden cohorts. Compared to the control cases, the LBD cohorts consistently showed upregulation for genes involved in protein folding and cytokine immune response, as well as downregulation of fatty acid metabolism. Surprisingly, concomitant tau pathology within the LBD cases resulted in no additional changes. Some core inflammatory pathways were shared between AD and LBD but with numerous disease-specific changes. Direct comparison of LBD cohorts versus AD cases revealed strong enrichment of synaptic signalling, behaviour and neuronal system pathways. Females had a stronger response overall in both LBD and AD, with several sex-specific changes. Overall, the results identify genes commonly and uniquely dysregulated in neuropathologically defined LBD and AD cases, shedding light on shared and distinct molecular pathways. Additionally, the study underscores the importance of considering sex-specific changes in understanding the complex transcriptional landscape of these neurodegenerative diseases.

Abstract BACKGROUND Accurate molecular diagnosis of infiltrating brainstem tumors, including diffuse midline gliomas and IDH mutant astrocytomas, is essential for prognostication and optimal therapy. Surgical biopsy of the brainstem carries risk of permanent neurological deficits and may yield insufficient or non-diagnostic tissue. We hypothesized that minimally invasive “liquid biopsy” analyzing circulating tumor cell-free DNA (cfDNA) in cerebrospinal fluid (CSF) could represent an alternative molecular diagnostic modality. METHODS We performed MSK-IMPACT, a New York State Department of Health authorized hybridization capture-based next-generation panel DNA sequencing clinical assay on 44 CSF samples from 37 unique patients with brainstem tumors. Ages at time of CSF collection ranged from 1–47 years (median: 29 years). For samples in which diagnostic alterations could not be detected, we performed a series of secondary analyses for known hotspot mutations such as H3F3A, IDH1/2, and BRAF, including manual review for mutant reads below the official clinical reporting threshold, as well as droplet digital PCR (ddPCR). RESULTS Among 44 samples included in this study, 10/44 (23%) samples had mutations detected by MSK-IMPACT using standard clinical calling criteria and 14 of the remaining 34 samples (41%) had supporting evidence of a diagnostically relevant mutation based on manual review. Testing by ddPCR was performed on 11 samples (based on DNA availability), uncovering 4 additional mutations. Overall positive diagnostic yield was 64% (28/44). In biopsied patients, CSF confirmed disease-defining mutations in 80% (16/20) of CSF samples, including H3F3A K28M (9/10; 90%) and IDH1/2 (5/6; 83%). Among 24 unbiopsied patients, a disease-defining mutation was identified in 60% (15/25) of CSF samples. CONCLUSIONS Analysis of CSF cfDNA in patients with brainstem tumors has a high diagnostic yield and obviates the need for tissue biopsy in a majority of patients. A tiered testing approach using next-generation sequencing and ddPCR assays helps maximize diagnostic information and sensitivity.

AbstractPurposeSinonasal malignancies (SNMs) adversely impact patients’ quality of life (QOL) and are frequently identified at an advanced stage. Because these tumors are rare, there are few studies that examine the specific QOL areas that are impacted. This knowledge would help improve the care of these patients.MethodsIn this prospective, multi‐institutional study, 273 patients with SNMs who underwent definitive treatment with curative intent were evaluated. We used the University of Washington Quality of Life (UWQOL) instrument over 5 years from diagnosis to identify demographic, treatment, and disease‐related factors that influence each of the 12 UWQOL subdomains from baseline to 5 ‐years post‐treatment.ResultsMultivariate models found endoscopic resection predicted improved pain (vs. nonsurgical treatment CI 2.4, 19.4, p = 0.01) and appearance versus open (CI 27.0, 35.0, p < 0.001) or combined (CI 10.4, 17.1, p < 0.001). Pterygopalatine fossa involvement predicted worse swallow (CI −10.8, −2.4, p = 0.01) and pain (CI −17.0, −4.0, p < 0.001). Neck dissection predicted worse swallow (CI −14.8, −2.8, p < 0.001), taste (CI −31.7, −1.5, p = 0.02), and salivary symptoms (CI −28.4, −8.6, p < 0.001). Maxillary involvement predicted worse chewing (CI 9.8, 33.2; p < 0.001) and speech (CI −21.8, −5.4, p < 0.001) relative to other sites. Advanced T stage predicted worse anxiety (CI −13.0, −2.0, p = 0.03).ConclusionsSurgical approach, management of cervical disease, tumor extent, and site of involvement impacted variable UWQOL symptom areas. Endoscopic resection predicted better pain, appearance, and chewing compared with open. These results may aid in counseling patients regarding potential QOL expectations in their SNM treatment and recovery course.

Colorectal cancer (CRC) is the fourth most frequently diagnosed cancer and the second leading cause of cancer death in the United States. Management of disseminated metastatic CRC involves various active drugs, either in combination or as single agents. The choice of therapy is based on consideration of the goals of therapy, the type and timing of prior therapy, the mutational profile of the tumor, and the differing toxicity profiles of the constituent drugs. This manuscript summarizes the data supporting the systemic therapy options recommended for metastatic CRC in the NCCN Guidelines for Colon Cancer.

Importance Neurological manifestations during acute SARS-CoV-2–related multisystem inflammatory syndrome in children (MIS-C) are common in hospitalized patients younger than 18 years and may increase risk of new neurocognitive or functional morbidity. Objective To assess the association of severe neurological manifestations during a SARS-CoV-2–related hospital admission with new neurocognitive or functional morbidities at discharge. Design, Setting, and Participants This prospective cohort study from 46 centers in 10 countries included patients younger than 18 years who were hospitalized for acute SARS-CoV-2 or MIS-C between January 2, 2020, and July 31, 2021. Exposure Severe neurological manifestations, which included acute encephalopathy, seizures or status epilepticus, meningitis or encephalitis, sympathetic storming or dysautonomia, cardiac arrest, coma, delirium, and stroke. Main Outcomes and Measures The primary outcome was new neurocognitive (based on the Pediatric Cerebral Performance Category scale) and/or functional (based on the Functional Status Scale) morbidity at hospital discharge. Multivariable logistic regression analyses were performed to examine the association of severe neurological manifestations with new morbidity in each SARS-CoV-2–related condition. Results Overall, 3568 patients younger than 18 years (median age, 8 years [IQR, 1-14 years]; 54.3% male) were included in this study. Most (2980 [83.5%]) had acute SARS-CoV-2; the remainder (588 [16.5%]) had MIS-C. Among the patients with acute SARS-CoV-2, 536 (18.0%) had a severe neurological manifestation during hospitalization, as did 146 patients with MIS-C (24.8%). Among survivors with acute SARS-CoV-2, those with severe neurological manifestations were more likely to have new neurocognitive or functional morbidity at hospital discharge compared with those without severe neurological manifestations (27.7% [n = 142] vs 14.6% [n = 356]; P &amp;lt; .001). For survivors with MIS-C, 28.0% (n = 39) with severe neurological manifestations had new neurocognitive and/or functional morbidity at hospital discharge compared with 15.5% (n = 68) of those without severe neurological manifestations ( P = .002). When adjusting for risk factors in those with severe neurological manifestations, both patients with acute SARS-CoV-2 (odds ratio, 1.85 [95% CI, 1.27-2.70]; P = .001) and those with MIS-C (odds ratio, 2.18 [95% CI, 1.22-3.89]; P = .009) had higher odds of having new neurocognitive and/or functional morbidity at hospital discharge. Conclusions and Relevance The results of this study suggest that children and adolescents with acute SARS-CoV-2 or MIS-C and severe neurological manifestations may be at high risk for long-term impairment and may benefit from screening and early intervention to assist recovery.

PURPOSE Telisotuzumab vedotin (Teliso-V) is a c-Met–directed antibody-drug conjugate with a monomethyl auristatin E cytotoxic payload. The phase II LUMINOSITY trial (ClinicalTrials.gov identifier: NCT03539536 ) aimed to identify the optimal c-Met protein–overexpressing non–small cell lung cancer (NSCLC) population for treatment with Teliso-V (stage I) and expand the selected group for efficacy evaluation (stage II). Stage II enrolled patients with nonsquamous epidermal growth factor receptor ( EGFR )-wildtype NSCLC. METHODS Eligible patients had locally advanced/metastatic c-Met protein–overexpressing NSCLC and ≤2 previous lines of therapy (including ≤1 line of systemic chemotherapy). c-Met protein overexpression in nonsquamous EGFR -wildtype NSCLC was defined as ≥25% tumor cells with 3+ staining (high [≥50% 3+]; intermediate [≥25%–<50%]). Teliso-V was administered at 1.9 mg/kg once every 2 weeks. The primary end point was overall response rate (ORR) by independent central review. RESULTS In total, 172 patients with nonsquamous EGFR -wildtype NSCLC received Teliso-V in stages I and II. ORR was 28.6% (95% CI, 21.7 to 36.2; c-Met high, 34.6% [95% CI, 24.2 to 46.2]; c-Met intermediate, 22.9% [95% CI, 14.4 to 33.4]). The median duration of response was 8.3 months (95% CI, 5.6 to 11.3; c-Met high, 9.0 [95% CI, 4.2 to 13.0]; c-Met intermediate: 7.2 [95% CI, 5.3 to 11.5]). The median overall survival was 14.5 months (95% CI, 9.9 to 16.6; c-Met high, 14.6 [95% CI, 9.2 to 25.6]; c-Met intermediate, 14.2 [95% CI, 9.6 to 16.6]). The median progression-free survival was 5.7 months (95% CI, 4.6 to 6.9; c-Met high, 5.5 [95% CI, 4.1 to 8.3]; c-Met intermediate: 6.0 [95% CI, 4.5 to 8.1]). Most common any-grade treatment-related adverse events (AEs) were peripheral sensory neuropathy (30%), peripheral edema (16%), and fatigue (14%); the most common grade ≥3 AE was peripheral sensory neuropathy (7%). CONCLUSION Teliso-V was associated with durable responses in c-Met protein–overexpressing nonsquamous EGFR -wildtype NSCLC, especially in those with high c-Met. AEs were generally manageable.

AbstractNucleic acid amplification testing (NAAT) remains one of the most reliable methods for pathogen identification. However, conventional bulk NAATs may not be sufficiently fast or sensitive enough for the detection of clinically‐relevant pathogens in point‐of‐care testing. Here, we have developed a digital droplet RT‐LAMP (ddRT‐LAMP) assay that rapidly and quantitatively detects the SARS‐CoV‐2 viral E gene in microfluidic drops. Droplet partitioning using ddRT‐LAMP significantly accelerates detection times across a wide range of template concentrations compared to bulk RT‐LAMP assays. We discover that a reduction in droplet diameter decreases assay times up to a certain size, upon which surface adsorption of the RT‐LAMP polymerase reduces reaction efficiency. Optimization of drop size and polymerase concentration enables rapid, sensitive, and quantitative detection of the SARS‐CoV‐2 E gene in only 8 min. These results highlight the potential of ddRT‐LAMP assays as an excellent platform for quantitative point‐of‐care testing.