Faculty Bibliography

BACKGROUND/UNASSIGNED:In the DISCOMS (DISCOntinuation of disease-modifying therapies (DMTs) in multiple sclerosis (MS)) randomized clinical trial, we could not demonstrate that discontinuing MS DMTs in older, stable adults was not inferior to continuing DMTs. Relapses were rare in both groups, and most new disease activity was one to two new brain magnetic resonance imaging (MRI) lesions unassociated with clinical changes. OBJECTIVE/AIMS/UNASSIGNED:Describe results of the DISCOMS extension study. METHODS/UNASSIGNED:Among 10/19 of the original sites, we enrolled patients who completed DISCOMS; did not reach the primary endpoint during the original trial; and retained original randomized assignment. Participants completed one study visit and brain MRI at least 30 months after original enrollment in DISCOMS. Primary endpoint was time from entry into the primary study to relapse or new brain MRI activity. RESULTS/UNASSIGNED:= 0.043 from log-rank test). CONCLUSIONS/UNASSIGNED:From entry into DISCOMS extension study, time to new MS activity remained shorter in discontinuers, but relapses were absent and new brain MRI lesions were rare.

BACKGROUND:Underlying intracranial stenosis is the most common cause of failed mechanical thrombectomy in acute ischemic stroke patients with large vessel occlusion. Adjunct emergent stenting is sometimes performed to improve or maintain reperfusion, despite limited data regarding its safety or efficacy. METHODS:We conducted a prospective multicenter observational international cohort study. Patients were enrolled between January 2022 and December 2023 at 25 thrombectomy capable centers in North America, Europe, and Asia. Consecutive patients treated with mechanical thrombectomy were included if they were identified as having underlying intracranial stenosis, defined as 50-99% residual stenosis of the target vessel or intra-procedural re-occlusion. The primary outcome was functional independence, defined as modified Rankin Scale of 0-2 at 90 days. After applying inverse probability of treatment weighting (IPTW) based on propensity scores, we compared outcomes among patients who underwent adjunct emergent intracranial stenting (stenting) versus those who received mechanical thrombectomy alone. RESULTS:A total of 417 patients were included; 218 patients treated with mechanical thrombectomy alone (168 anterior circulation) and 199 with mechanical thrombectomy plus stenting (144 anterior circulation). Patients in the stenting group were less likely to be non-Hispanic White (51.8% vs 62.4%, p=0.03), and less likely to have diabetes (33.2% vs 43.1%, p=0.037) or hyperlipidemia (43.2% vs 56%, p= 0.009). In addition, there was a lower rate of IV thrombolysis use in the stenting group (18.6% vs 27.5%, p=0.03). There was a higher rate of successful reperfusion (modified Treatment In Cerebral Infarction score ≥ 2B) in the stenting versus mechanical thrombectomy alone group (90.9% vs 77.9%, p<0.001) and a higher rate of a 24-hour infarct volume of <30 mL (n=260, 67.9% vs 50.3%, p=0.005). The overall complication rate was higher in the stenting group (12.6% vs 5%, p=0.006), but there was not a significant difference in the rate of symptomatic hemorrhage (9% vs 5.5%, p=0.162). Functional independence at 90 days was significantly higher in the stenting group (42.2% vs. 28.4%, adjusted odds ratio 2.67; 95% CI, 1.66-4.32). CONCLUSIONS:In patients with underlying stenosis who achieved reperfusion with mechanical thrombectomy, adjunct emergent stenting was associated with better functional outcome without a significantly increased risk of symptomatic hemorrhage. REGISTRATION/BACKGROUND:https://clinicaltrials.gov/study/NCT05403593.

The adoption of large language models (LLMs) in healthcare demands a careful analysis of their potential to spread false medical knowledge. Because LLMs ingest massive volumes of data from the open Internet during training, they are potentially exposed to unverified medical knowledge that may include deliberately planted misinformation. Here, we perform a threat assessment that simulates a data-poisoning attack against The Pile, a popular dataset used for LLM development. We find that replacement of just 0.001% of training tokens with medical misinformation results in harmful models more likely to propagate medical errors. Furthermore, we discover that corrupted models match the performance of their corruption-free counterparts on open-source benchmarks routinely used to evaluate medical LLMs. Using biomedical knowledge graphs to screen medical LLM outputs, we propose a harm mitigation strategy that captures 91.9% of harmful content (F1 = 85.7%). Our algorithm provides a unique method to validate stochastically generated LLM outputs against hard-coded relationships in knowledge graphs. In view of current calls for improved data provenance and transparent LLM development, we hope to raise awareness of emergent risks from LLMs trained indiscriminately on web-scraped data, particularly in healthcare where misinformation can potentially compromise patient safety.

T cell receptor (TCR) gene therapy is a potent form of cellular immunotherapy in which patient T cells are genetically engineered to express TCRs with defined tumor reactivity. However, the isolation of therapeutic TCRs is complicated by both the general scarcity of tumor-specific T cells among patient T cell repertoires and the patient-specific nature of T cell epitopes expressed on tumors. Here we describe a high-throughput, personalized TCR discovery pipeline that enables the assembly of complex synthetic TCR libraries in a one-pot reaction, followed by pooled expression in reporter T cells and functional genetic screening against patient-derived tumor or antigen-presenting cells. We applied the method to screen thousands of tumor-infiltrating lymphocyte (TIL)-derived TCRs from multiple patients and identified dozens of CD4⁺ and CD8⁺ T-cell-derived TCRs with potent tumor reactivity, including TCRs that recognized patient-specific neoantigens.

BACKGROUND:This study primarily aimed to explore the etiology of the phenotypic relationships between personality traits and executive functions. We assessed common genetic and environmental sources of variance in these phenomena by examining different dimensions of the Five Factor Model (FFM) and executive function factors. An additional research question focused on whether specific facets within the FFM share a greater genetic overlap with executive functions compared to broader personality dimensions. METHODS:The NEO-PI-R, the Wisconsin Card Sorting Test, the Trail Making Test, and verbal fluency tests were applied to a sample of 468 twins from the Serbian Twin Advanced Registry (153 monozygotic twin pairs). RESULTS:Several facets from all domains had significant genetic associations with executive functions, with Ideas and Values showing the highest positive correlations, and Order and Vulnerability showing the highest negative genetic correlations. Most phenotypic associations between the two domains of individual differences were almost entirely explained by shared genetic influences. CONCLUSIONS:These findings underscore the importance of examining personality traits at the facet level, as facet-level analysis provides more detailed insights into the relationship between personality and executive functions than broader personality dimensions. The results emphasize the role of shared genetic influences in shaping both personality and executive abilities, suggesting that future research should investigate the underlying neural mechanisms driving these associations.

BACKGROUND:Arteriovenous malformations (AVMs) are uncommon cerebral lesions that can cause significant neurological complications. Surgical resection is the gold standard for treatment, but endovascular embolization and stereotactic radiosurgery (SRS) are viable alternatives. OBJECTIVE:To compare the outcomes of endovascular embolization versus SRS in the treatment of AVMs with Spetzler-Martin grades I-III. METHODS:This study combined retrospective data from 10 academic institutions in North America and Europe. Patients aged 1 to 90 years who underwent endovascular embolization or SRS for AVMs with Spetzler-Martin grades I-III between January 2010 and December 2023 were included. RESULTS:The study included 244 patients, including 84 who had endovascular embolization and 160 who had SRS. Before propensity score matching (PSM), complete obliteration at the last follow-up was achieved in 74.5% of the SRS group compared with 57.8% of the embolization group (OR=0.47; 95% CI 0.26 to 0.48; P=0.01). After propensity score matching, SRS still achieved significantly higher occlusion rates at last follow-up (78.9% vs 55.3%; OR=0.32; 95% CI 0.12 to 0.90; P=0.03).Hemorrhagic complications were higher in the embolization group than in the SRS group, although this difference did not reach statistical significance after PSM (13.2% vs 2.6%; OR=5.6; 95% CI 0.62 to 50.47; P=0.12). Similarly, re-treatment rate was higher in the embolization group (10.5% vs 5.3%; OR=2.11; 95% CI 0.36 to 12.31; P=0.40) compared with the SRS group. CONCLUSION/CONCLUSIONS:Our findings indicate that SRS has a significantly higher obliteration rate at last follow-up compared with endovascular embolization. Also, SRS has a higher tendency for fewer hemorrhagic complications and lower re-treatment rate. Further prospective studies are needed.

BACKGROUND:With transradial access (TRA) being more progressively used in neuroendovascular procedures, we compared TRA with transfemoral access (TFA) in middle meningeal artery embolization (MMAE) for chronic subdural hematoma (cSDH). METHODS:Consecutive patients undergoing MMAE for cSDH at 14 North American centers (2018-23) were included. TRA and TFA groups were compared using propensity score matching (PSM) controlling for: age, sex, concurrent surgery, previous surgery, hematoma thickness and side, midline shift, and pretreatment antithrombotics. The primary outcome was access site and overall complications, and procedure duration; secondary endpoints were surgical rescue, radiographic improvement, and technical success and length of stay. RESULTS:872 patients (median age 73 years, 72.9% men) underwent 1070 MMAE procedures (54% TFA vs 46% TRA). Access site hematoma occurred in three TFA cases (0.5%; none required operative intervention) versus 0% in TRA (P=0.23), and radial-to-femoral conversion occurred in 1% of TRA cases. TRA was more used in right sided cSDH (58.4% vs 44.8%; P<0.001). Particle embolics were significantly higher in TFA while Onyx was higher in TRA (P<0.001). Following PSM, 150 matched pairs were generated. Particles were more utilized in the TFA group (53% vs 29.7%) and Onyx was more utilized in the TRA group (56.1% vs 31.5%) (P=0.001). Procedural duration was longer in the TRA group (median 68.5 min (IQR 43.1-95) vs 59 (42-84); P=0.038), and radiographic success was higher in the TFA group (87.3% vs 77.4%; P=0.036). No differences were noted in surgical rescue (8.4% vs 10.1%, P=0.35) or technical failures (2.4% vs 2%; P=0.67) between TFA and TRA. Sensitivity analysis in the standalone MMAE retained all associations but differences in procedural duration. CONCLUSIONS:In this study, TRA offered comparable outcomes to TFA in MMAE for cSDH in terms of access related and overall complications, technical feasibility, and functional outcomes. Procedural duration was slightly longer in the TRA group, and radiographic success was higher in the TFA group, with no differences in surgical rescue rates.