27941 research outputs found
Sort by
Choosing Between HLA-Mismatched Unrelated and Haploidentical Donors: Donor Age Considerations
No full textHaploidentical donors and HLA-mismatched unrelated donors (MMUDs) are increasingly utilized for hematopoietic cell transplantation (HCT), with post-transplantation cyclophosphamide (PTCy) emerging as an effective graft-versus-host disease (GVHD) prophylaxis strategy. Despite the growing use of these donor types, comparative data to guide donor selection remain limited. Donor age is a known predictor of HCT outcomes, yet its specific impact when choosing between haploidentical and MMUD donors with PTCy-based prophylaxis has not been thoroughly explored. This study aimed to evaluate the influence of donor age on HCT outcomes in patients receiving haploidentical or MMUD HCT with PTCy-based GVHD prophylaxis, hypothesizing that younger donors (<30 years) would be associated with improved outcomes compared to older donors (≥30 years) regardless of donor type. We conducted a retrospective analysis of 7116 patients with hematologic malignancies from the Center for International Blood and Marrow Transplant Research database, transplanted between 2013 and 2021. Donors were categorized into four groups: younger haploidentical (<30 years), older haploidentical (≥30 years), younger MMUD (<30 years), and older MMUD (≥30 years). The primary outcome was GVHD-free relapse-free survival (GRFS), defined as the absence of grade III to IV acute GVHD, chronic GVHD requiring systemic immunosuppressive therapy (IST), relapse, or death. Secondary outcomes included overall survival, treatment-related mortality (TRM), relapse, grade III to IV acute GVHD, overall chronic GVHD, and chronic GVHD requiring IST. Comparisons were made between (1) younger MMUD versus older haploidentical and (2) younger haploidentical versus older MMUD groups using multivariable Cox proportional hazards models. In multivariable analysis, the older MMUD group exhibited inferior GRFS (hazard ratio [HR] 1.20; 95% confidence interval [CI], 1.06 to 1.36; P = .003), higher TRM (HR 1.49; 95% CI, 1.13 to 1.96; P = .005), and increased grade III to IV acute GVHD (HR 2.88; 95% CI, 1.43 to 5.80; P = .003) compared to the younger haploidentical group. The younger MMUD group had modest GRFS improvement over the older haploidentical group (HR 0.87; 95% CI, 0.78 to 0.98; P = .02) and significantly reduced risks of grade II to IV acute GVHD (HR 0.67; 95% CI, 0.51 to 0.88; P = .003) and chronic GVHD (HR 0.78; 95% CI, 0.65 to 0.94; P = .009). Younger donor age is associated with superior HCT outcomes, emphasizing the importance of prioritizing donors aged <30 years regardless of donor type when feasible.The UMass Center for Clinical and Translational Science (UMCCTS), UL1TR001453, helped fund this study.No embarg
Baseline predictors of poor clinical outcome despite recanalization of distal middle cerebral artery occlusions
No full textObjectiveMechanical thrombectomy (MT) is well-established for the treatment of acute ischemic stroke (AIS) from large vessel occlusion (LVO), with growing data supporting the expansion to distal and medium vessel occlusions (DMVO). Despite successful recanalization in DMVO, certain patients still experience poor long-term clinical outcomes, prompting our study to comprehensively explore pre-MT factors influencing outcome despite excellent recanalization (final modified Thrombolysis in Cerebral Infarction [mTICI] score ≥2c).MethodsWe retrospectively examined data from patients who consecutively underwent MT for a primary middle cerebral artery (MCA) DMVO across 37 centers in North America, Asia, and Europe. We identified baseline clinical and imaging factors associated with poor clinical outcome (defined as a modified Rankin Scale [mRS] score of 3-6) at 3 months, despite excellent recanalization using a multivariable model.ResultsBetween September 2017 and July 2021, 623 patients achieved mTICI > 2b and they were included in our study. Among them, 198 (32%) experienced a poor clinical outcome (mRS 3-6). Predictors of poor clinical outcome included higher age (OR 1.05 [1.03-1.07], 0.001), higher NIHSS at admission (OR 1.12 [1.08-1.15], 0.001), higher baseline mRS (OR 1.77 [0.96-3.26], 0.067), and diabetes (OR 1.59 [1.01-2.48], 0.044). Higher ASPECTS was associated with a decreased risk of poor clinical outcome (OR 0.82 [0.71-0.94], 0.006).ConclusionOlder age, diabetes, higher baseline mRS, and NIHSS were associated with poor clinical outcome in MCA DMVO despite excellent recanalization. Conversely, a higher ASPECTS decreased the probability of such an outcome.No embarg
Factors associated with Early Neurological Deterioration in Minor Distal Medium Vessel Acute Ischemic Stroke: A Multinational Multicenter Study
No full textBackground: Patients with acute ischemic stroke secondary to distal and medium vessel occlusion (AIS-DMVO) and minor strokes present a challenge in determining the most appropriate emergent treatment. Factors leading to early neurological deterioration (END) in this patient population are understudied, but END is known to result in poor functional outcomes. Therefore, we aimed to investigate the factors contributing to END in minor AIS-DMVO cases.
Methods: We included patients with AIS-DMVO and minor strokes from 37 sites across North America, Asia, and Europe. Minor stroke was defined as a baseline National Institutes of Health Stroke Scale (NIHSS) score of ≤ 5. The primary outcome measure, END, was defined as a shift of ≥ 4 points in the NIHSS score at day one after treatment compared to baseline. Univariable and multivariable logistic regression analyses were performed to identify factors associated with END.
Results: Among 559 consecutive patients with DMVO and minor strokes, END was reported in 68 patients. In multivariable analysis, mechanical thrombectomy (MT) was independently associated with higher odds of END (aOR 2.37; 95% CI: 1.12-5.02; p = 0.02), while intravenous thrombolysis (IVT) was associated with lower odds of END (aOR 0.46; 95% CI: 0.26-0.81; p = 0.008). However, the association between MT and END was no longer statistically significant in the IPTW-adjusted analysis (OR 1.65; 95% CI: 0.69-3.98; p = 0.26). Hypertension and antiplatelet use at baseline were also independently associated with END. Among MT-treated patients, successful and excellent recanalization and first-pass effect were protective against END.
Conclusion: MT was associated with END in patients with minor AIS-DMVO, although this association was not significant after IPTW adjustment. IVT was independently associated with reduced risk of END. These findings support careful patient selection and further study in randomized trials.No embarg
Transforming Informed Consent Generation Using Large Language Models: Mixed Methods Study
Get PDFBackground: Informed consent forms (ICFs) for clinical trials have become increasingly complex, often hindering participant comprehension and engagement due to legal jargon and lengthy content. The recent advances in large language models (LLMs) present an opportunity to streamline the ICF creation process while improving readability, understandability, and actionability.
objectives: This study aims to evaluate the performance of the Mistral 8x22B LLM in generating ICFs with improved readability, understandability, and actionability. Specifically, we evaluate the model's effectiveness in generating ICFs that are readable, understandable, and actionable while maintaining the accuracy and completeness.
Methods: We processed 4 clinical trial protocols from the institutional review board of UMass Chan Medical School using the Mistral 8x22B model to generate key information sections of ICFs. A multidisciplinary team of 8 evaluators, including clinical researchers and health informaticians, assessed the generated ICFs against human-generated counterparts for completeness, accuracy, readability, understandability, and actionability. Readability, Understandability, and Actionability of Key Information indicators, which include 18 binary-scored items, were used to evaluate these aspects, with higher scores indicating greater accessibility, comprehensibility, and actionability of the information. Statistical analysis, including Wilcoxon rank sum tests and intraclass correlation coefficient calculations, was used to compare outputs.
Results: LLM-generated ICFs demonstrated comparable performance to human-generated versions across key sections, with no significant differences in accuracy and completeness (P>.10). The LLM outperformed human-generated ICFs in readability (Readability, Understandability, and Actionability of Key Information score of 76.39% vs 66.67%; Flesch-Kincaid grade level of 7.95 vs 8.38) and understandability (90.63% vs 67.19%; P=.02). The LLM-generated content achieved a perfect score in actionability compared with the human-generated version (100% vs 0%; P<.001). Intraclass correlation coefficient for evaluator consistency was high at 0.83 (95% CI 0.64-1.03), indicating good reliability across assessments.
Conclusions: The Mistral 8x22B LLM showed promising capabilities in enhancing the readability, understandability, and actionability of ICFs without sacrificing accuracy or completeness. LLMs present a scalable, efficient solution for ICF generation, potentially enhancing participant comprehension and consent in clinical trials.No embarg
Expanding Structural Knowledge of EV68-3C Protease through Protein Crystallography: A Structural Basis of Specificity, Potential Inhibitor Resistance, and Foundation for Designing Resilient 3C Direct Acting Antivirals.
No full textEnterovirus-D68 has emerged as a global health concern over the last decade with severe symptomatic infections resulting in long-lasting neurological deficits and death. Despite previous attempts, there are currently no FDA-approved antiviral drugs for EV68 or any other non-polio enterovirus. One particularly attractive class of potential drugs are small molecules inhibitors, which can target the conserved active site of EV68-3C protease. However, the structural characterization of EV68- 3C protease bound to its substrates, and therefore the substrate specificity has been lacking. For other viral proteases, we previously demonstrated that the emergence of drug resistance can be minimized by designing inhibitors that leverage the evolutionary constraints of substrate specificity. A better understanding of this molecular basis of drug resistance is therefore central to developing robust protease inhibitors with a high barrier to drug resistant mutations.
As a result of this need, my research characterizes the structural basis of specificity of EV68-3C protease and explores the resistance profiles of current inhibitors with plans for developing robust protease inhibitors. Chapter II explores the structural characterization EV68-3C Protease, both with computational modeling and experimental protein crystallography, to define the common volume shared by EV68-3C protease viral substrates – the substrate envelope. We utilize the substrate envelope to assess potential sites of resistance as well as uncover protein conformational changes during ligand-binding. Chapter III assesses additional 3C and 3C-like inhibitors through protein crystallography to further understand possible resistance mutations where inhibitors protrude from the substrate envelope. We also analyzed structural patterns and motifs from ligand-bound protein co-crystals that can aid in the design of robust EV68-3C inhibitors. I hope this works lays down a strong foundation for other scientists to expand upon when developing robust therapeutics against this debilitating pathogen as well as applying this knowledge to similar viruses.MD/PhD2 years2027-01-3
A Case for Automated Segmentation of MRI Data in Milder Neurodegenerative Diseases [preprint]
Get PDFThis article is a preprint. Preprints are preliminary reports of work that have not been certified by peer review.Background: Volumetric analysis and segmentation of magnetic resonance imaging (MRI) data is an important tool for evaluating neurological disease progression and neurodevelopment. Fully automated segmentation pipelines offer faster and more reproducible results. However, since these analysis pipelines were trained on or run based on atlases consisting of neurotypical controls, it is important to evaluate how accurate these methods are for neurodegenerative diseases. In this study, we compared 5 fully automated segmentation pipelines including FSL, Freesurfer, volBrain, SPM12, and SimNIBS with a manual segmentation process in GM1 gangliosidosis patients and neurotypical controls.
Methods: We analyzed 45 MRI scans from 16 juvenile GM1 gangliosidosis patients, 11 MRI scans from 8 late-infantile GM1 gangliosidosis patients, and 19 MRI scans from 11 neurotypical controls. We compared results for 7 brain structures including volumes of the total brain, bilateral thalamus, ventricles, bilateral caudate nucleus, bilateral lentiform nucleus, corpus callosum, and cerebellum.
Results: We found volBrain's vol2Brain pipeline to have the strongest correlations with the manual segmentation process for the whole brain, ventricles, and thalamus. We also found Freesurfer's recon-all pipeline to have the strongest correlations with the manual segmentation process for the caudate nucleus. For the cerebellum, we found a combination of volBrain's vol2Brain and SimNIBS' headreco to have the strongest correlations depending on the cohort. For the lentiform nucleus, we found a combination of recon-all and FSL's FIRST to give the strongest correlations depending on the cohort. Lastly, we found segmentation of the corpus callosum to be highly variable.
Conclusion: Previous studies have considered automated segmentation techniques to be unreliable, particularly in neurodegenerative diseases. However, in our study we produced results comparable to those obtained with a manual segmentation process. While manual segmentation processes conducted by neuroradiologists remain the gold standard, we present evidence to the capabilities and advantages of using an automated process including the ability to segment white matter throughout the brain or analyze large datasets, which pose feasibility issues to fully manual processes. Future investigations should consider the use of artificial intelligence-based segmentation pipelines to determine their accuracy in GM1 gangliosidosis, lysosomal storage disorders, and other neurodegenerative diseases.No embarg
Medical decision-making experiences of persons with dementia and their carepartners: a qualitative study
Get PDFBackground: Persons with dementia (PWD) and their carepartners must often make complex medical decisions, weighing the benefits of medical (surgical and non-surgical) interventions with uncertainty regarding outcomes, both dementia- and non-dementia related, in the short-term and long-term. This study informs gaps in clinical guidance for patient-centered decision-making about medical and surgical interventions for PWD and advancecare planning.
Methods: We conducted a qualitative study using thematic analysis based on semi-structured interviews with PWD and carepartners.
Results: We interviewed 30 participants (9 PWD, 21 carepartners). Four themes were identified (with related subthemes): 1) PWD and carepartners varied in using decision-making approaches for medical interventions for PWD (a) variations in views about decision-making load; (b) Progressive involvement of carepartners in ACP decision-making as cognition erodes; 2) medical intervention decisions were an inflection point to evaluate values for dyads and involved tradeoffs with implications for end-of-life care and quality of life 3) lack of discussion with clinical team about impact of medical interventions on dementia burdened dyads; 4) decisional quality was facilitated by: (a) a trusting relationship with clinicians; and (b) a multidisciplinary team approach.
Conclusion: Most patients with mild-to-moderate dementia and carepartners approach medical intervention decision-making guided by their understanding of the dementia prognosis, but the risks of medical interventions are often unaddressed in discussions with the clinical team, sometimes burdening dyads with undesirable consequences to their quality-of-life. Clinicians should provide dementia-related risks regarding medical intervention outcomes to best facilitate decision-making conversations and advance care planning.No embarg
Endovascular Therapy for Patients With Low NIHSS Scores and Large Vessel Occlusion in the 6- to 24-Hour Window: Analysis of the CLEAR Study
No full textBackground and objectives: There is uncertainty about whether patients with an anterior circulation large vessel occlusion (LVO) and a low NIH Stroke Scale (NIHSS) score (≤5) benefit from endovascular therapy (EVT) in the late time window (6-24 hours). We compared the clinical outcomes of these patients receiving EVT with those receiving medical management (MM).
Methods: The CT for Late Endovascular Reperfusion multinational cohort study was conducted at 66 sites across 10 countries from January 2014 to May 2022. This subanalysis included consecutive patients with late-window stroke due to an anterior circulation LVO, defined as occlusion of the internal carotid artery or proximal middle cerebral artery (M1/M2 segments), and a baseline NIHSS score ≤5 who received EVT or MM alone. The primary end point was a 90-day ordinal shift in the modified Rankin Scale (mRS) score. Secondary outcomes were 90-day excellent outcome (defined as mRS scores 0-1 or return to baseline mRS score in patients with a prestroke mRS score >1) and favorable outcome (defined as mRS scores 0-2 or return to baseline mRS score in patients with prestroke mRS score >2). Safety outcomes were symptomatic intracranial hemorrhage and 90-day mortality. We used ordinal and binary logistic regression models to test for outcome differences.
Results: Among 5,098 patients, 318 patients were included (median [interquartile range] age 67 [56-76] years; 149 [46.9%] were female; baseline NIHSS score was 4 [2-5]). A total of 202 patients (63.5%) received EVT and 116 MM (36.5%). There was no difference in favorable 90-day ordinal mRS score shift (adjusted common odds ratio [OR] 0.77, 95% CI 0.45-1.32), excellent outcome (adjusted OR 0.86, 95% CI 0.49-1.50), or favorable outcome (adjusted OR 0.72, 95% CI 0.35-1.50) in the EVT group compared with MM. Symptomatic intracranial hemorrhage risk (adjusted OR 3.40, 95% CI 0.84-13.73) and mortality at 90 days (adjusted OR 2.44, 95% CI 0.60-10.02) were not statistically different between treatment groups.
Discussion: In patients with an anterior LVO and low NIHSS score in the 6-24-hour time window, there was no statistical difference in disability outcomes or intracranial bleeding risk between patients treated with EVT compared with MM. The retrospective and observational design limits our findings. Ongoing randomized controlled trials will provide further insight.
Classification of evidence: This study provides Class III evidence that in adult patients with anterior circulation LVO and low NIHSS score (≤5) presenting in the late time window (6-24 hours), EVT does not improve clinical outcome vs MM.
Trial registration: This study was registered at clinicaltrials.gov under NCT04096248.No embarg
Patient Perceptions on Opioid Therapy for Acute Pain Management in the Emergency Department and Post-Surgical Settings: A Qualitative Study
Get PDFPurpose: Opioids are highly effective in treating acute pain but have various detrimental, non-therapeutic effects that may lead to the development of opioid use disorder (OUD) in certain individuals. This creates a complex landscape of opioid prescribing and use. The present manuscript uses a qualitative research lens to understand the patient experience during opioid therapy for acute pain, including perceptions, behaviors, and factors that drive both during the third wave of the opioid epidemic.
Patients and methods: Patients receiving opioids for acute pain were recruited from the emergency department and general surgery clinic as part of a larger study to perform continuous physiologic monitoring during opioid use. Semi-structured interviews were conducted within 10 days of hospital discharge. Interviews were recorded, transcribed verbatim and subject to applied thematic analysis by two independent coders.
Results: Of the 60 subjects that completed the parent study, 44 completed the semi-structured interviews (mean age 47 years, 50% female). Three main themes emerged from the interviews: the direct effects of opioid analgesia, internal factors and external factors that influenced therapy. Symptoms of opioid tolerance, withdrawal and dependence were described in a subset of participants, even with a relatively short duration of therapy. Participants expressed a desire for more education around opioid therapy, and more engagement with their plan of care.
Conclusion: Opioid therapy, even in short courses for acute pain, can be a complicated experience for patients to navigate. Prescribers should consider how patients' unique experiences and perspectives shape their responses when providing counseling to patients and monitoring of response to opioid therapy.No embarg
GrantCheck-an AI Solution for Guiding Grant Language to New Policy Requirements: Development Study
Get PDFBACKGROUND: Academic institutions face increasing challenges in grant writing due to evolving federal and state policies that restrict the use of specific language. Manual review processes are labor-intensive and may delay submissions, highlighting the need for scalable, secure solutions that ensure compliance without compromising scientific integrity.
OBJECTIVE: This study aimed to develop a secure, artificial intelligence-powered tool that assists researchers in writing grants consistent with evolving state and federal policy requirements.
METHODS: GrantCheck (University of Massachusetts Chan Medical School) was built on a private Amazon Web Services virtual private cloud, integrating a rule-based natural language processing engine with large language models accessed via Amazon Bedrock. A hybrid pipeline detects flagged terms and generates alternative phrasing, with validation steps to prevent hallucinations. A secure web-based front end enables document upload and report retrieval. Usability was assessed using the System Usability Scale.
RESULTS: GrantCheck achieved high performance in detecting and recommending alternatives for sensitive terms, with a precision of 1.00, recall of 0.73, and an F-score of 0.84-outperforming general-purpose models including GPT-4o (OpenAI; F=0.43), Deepseek R1 (High-Flyer; F=0.40), Llama 3.1 (Meta AI; F=0.27), Gemini 2.5 Flash (Google; F=0.58), and even Gemini 2.5 Pro (Google; F=0.72). Usability testing among 25 faculty and staff yielded a mean System Usability Scale score of 85.9 (SD 13.4), indicating high user satisfaction and strong workflow integration.
CONCLUSIONS: GrantCheck demonstrates the feasibility of deploying institutionally hosted, artificial intelligence-driven systems to support compliant and researcher-friendly grant writing. Beyond administrative efficiency, such systems can indirectly safeguard public health research continuity by minimizing grant delays and funding losses caused by language-related policy changes. By maintaining compliance without suppressing scientific rigor or inclusivity, GrantCheck helps protect the pipeline of research that advances biomedical discovery, health equity, and patient outcomes. This capability is particularly relevant for proposals in sensitive domains-such as social determinants of health, behavioral medicine, and community-based research-that are most vulnerable to evolving policy restrictions. As a proof-of-concept development study, our implementation is tailored to one institution's policy environment and security infrastructure, and findings should be interpreted as preliminary rather than universally generalizable.No embarg