Citations

Abstract Immune checkpoint inhibitor–associated immune thrombocytopenia (ICI-ITP) has been described in case reports and small case series, but comprehensive data on its incidence, risk factors, clinical features, treatment, and outcomes are lacking. We reviewed medical records of all adults initiating ICI therapy between 2016 and 2023 at 29 US hospitals across 7 major cancer centers to identify patients with ICI-ITP. Multivariable logistic regression was used to identify risk factors, and Cox modeling was performed to assess the association between ICI-ITP, its severity, and mortality. Among 86 467 patients, ICI-ITP occurred in 214 (0.25%). Independent risk factors included lower baseline platelet count, combination ICI therapy, stage IV cancer, and additional immune-related adverse events. ICI-ITP occurred at a median of 8 weeks (interquartile range [IQR], 4-18) after ICI initiation, with a median nadir platelet count of 41 × 109/L (IQR, 17 × 109/L to 64 × 109/L). Patients were treated with glucocorticoids (n = 106 [49.5%]), immune globulin (n = 39 [18.2%]), and thrombopoietin receptor agonists (n = 29 [13.6%]). Recovery occurred in 161 patients (75.2%) at a median of 2.3 weeks (IQR, 1.0-5.3). Of 76 patients rechallenged with ICIs, 23 (30.3%) developed recurrent ICI-ITP. ICI-ITP and its severity were associated with higher all-cause mortality, with a nearly threefold increase in risk among patients with severe ICI-ITP than those without ICI-ITP (adjusted hazard ratio, 2.96 [95% confidence interval, 2.14-4.08]). These findings establish ICI-ITP as a rare but clinically significant complication of ICI therapy, provide, to our knowledge, the first large-scale description of its risk factors and clinical course, and underscore the importance of timely recognition and management.

OBJECTIVE The aim of this study was to compare 3 electrode designs—a novel 22-contact circular grid, standard 6-contact strip, and high-density (HD) grid between 32 and 64 contacts—for the detection of epileptiform activity during awake craniotomy. METHODS This study included patients who underwent functional brain mapping with and without direct electrical stimulation during awake craniotomy. Demographic, clinical, and electrocorticography (ECoG) data were collected. RESULTS A total of 194 patients were included, with 264 instances of electrode use: 113 circular grid, 96 strip electrode, and 55 HD grid. The HD grid (15%) was used less to record ECoG during direct electrical stimulation than the circular grid (80%, p < 0.001) and strip electrodes (69%, p < 0.001). Sporadic interictal epileptiform activity was better detected with circular (45%, p < 0.001) and HD (40%, p = 0.006) grids compared with strip electrodes (16%). Spontaneous seizures were recorded more frequently with the circular grid (16%, p = 0.018) compared with the strip electrode (9.4%) and HD grid (1.8%). Afterdischarges were more frequently detected by the circular grid (69%, p = 0.031) and with circular and HD grids (combined) compared with strip electrodes (67% vs 48%, p = 0.024), at lower median intensity (3 mA vs 5 mA, p < 0.001). Multivariable analyses demonstrated circular grids (p < 0.001) to be more sensitive in detecting spontaneous epileptiform activity than strip and HD grid electrodes, and grids (circular and HD) to be better in recording stimulus-induced epileptiform activity (p = 0.01). CONCLUSIONS This study demonstrates a higher rate of detection of epileptiform activity with circular and standard grids when compared with strip electrodes. The standard HD grid was used less during electrical stimulation, demonstrating the importance of both the number of contacts and design array when considering optimal conditions for ECoG and functional brain mapping.

Abstract Background The purpose of this study was to develop a series of benchmarks specifically for large language model (LLM) performance in the identification of drug-drug interactions (DDIs). Methods We evaluated six LLMs using a clinician-designed dataset of 750 DDI scenarios. Tasks included: (1) a pointwise two-drug classification task (2) a pairwise three-drug discrimination task and (3) a listwise 4–6 drug selection task. Zero-shot prompting was applied and performance was assessed using precision, recall, F1 score, accuracy, and self-consistency. Results Model performance varied by task structure and interaction severity. GPT-5-Chat demonstrated the highest recall and F1 score in the pointwise task (80.7% and 80.2%), whereas DrugGPT achieved superior accuracy in the pairwise (93.7%) and listwise task (92.6%). All models had improved performance with more severe compared to less severe DDIs (accuracy 72.0-99.2% vs. 58.3-86.7%). All models tended to have higher self-consistency in the pairwise and listwise experiments compared to the pointwise experiment (68.8-93.6% vs. 31.0-74.2%). No model exhibited uniformly high reliability across all tasks. Conclusions Current LLMs show inconsistent capabilities in identifying DDIs. Performance degrades as reasoning space expands with limited stability. These findings emphasize the need for multi-format evaluation frameworks and reliability-aware assessment when considering LLMs for medication-safety applications.

BACKGROUND: Metachronous cancer risk varies by mismatch repair mutation in patients with hereditary nonpolyposis colorectal cancer. Total colectomy offers decreased risk of metachronous colorectal cancers but decreased quality of life compared with segmental colectomy. OBJECTIVE: Compare segmental versus total colectomy for colon cancer via a mutation-specific Markov decision model. DESIGN: Simulated survival and colorectal metachronous occurrence among patients with MLH1, MSH2 , or MSH6 mutations undergoing either segmental or total colectomy. SETTINGS: National databases, literature review. PATIENTS: Simulated patients with hereditary nonpolyposis colorectal cancer. MAIN OUTCOME MEASURES: Overall survival, quality-adjusted life years based on procedure-specific utility states (segmental colectomy 0.99, total colectomy 0.95). RESULTS: Among patients aged 25 years, total colectomy offered less than 1 year of increased mean survival compared with segmental resection across all mutations ( MLH1: 0.8 years, MSH2: 0.9, MSH6: 0.3). This difference in survival decreased with increasing age. Segmental resection offered a small increase in quality-adjusted life years compared with total colectomy across all ages. This difference was highest for MSH6 (0.9 years vs MLH1 : 0.6 and MSH2 : 0.4 at age 25), particularly at younger ages of diagnosis. In a sensitivity analysis, lower total colectomy utility states resulted in greater increases in quality-adjusted life years with segmental colectomy across all mutations. LIMITATIONS: Metachronous cancer risk data do not distinguish between colon and rectal cancers and is extrapolated beyond age 70 years. CONCLUSIONS: Segmental and total colectomy offer similar survival and quality-adjusted life year outcomes. With decreased risk of metachronous cancers, patients with MSH6 mutations across all ages may prefer segmental resection owing to higher quality-adjusted life years and lower survival benefits from total colectomy compared with MLH1 and MSH2 patients. Operative strategy in patients with hereditary nonpolyposis colorectal cancer should be individualized, and this study shows that mutation-specific differences in survival and quality-adjusted life years are small. See Video Abstract . MODELO DE DECISIÓN ESPECÍFICO PARA LA MUTACIÓN EN LA COLECTOMÍA ABDOMINAL SEGMENTARIA FRENTE A LA COLECTOMÍA ABDOMINAL TOTAL EN EL CÁNCER COLORRECTAL HEREDITARIO NO POLIPOSO ANTECEDENTES: El riesgo de cáncer metacrónico varía según la mutación de reparación de desajustes en pacientes con cáncer colorrectal hereditario no poliposo. La colectomía total ofrece un menor riesgo de cánceres colorrectales metacrónicos, pero una menor calidad de vida en comparación con la colectomía segmentaria. OBJETIVO: Comparar la colectomía segmentaria con la colectomía total para el cáncer de colon mediante un modelo de decisión de Markov específico para mutaciones. DISEÑO: Simulación de la supervivencia y la aparición metacrónica de cáncer colorrectal en pacientes con mutaciones MLH1, MSH2 o MSH6 sometidos a colectomía segmentaria o total. ENTORNO: Bases de datos nacionales, revisión de la literatura. PACIENTES: Pacientes simulados con cáncer colorrectal hereditario no poliposo. PRINCIPALES MEDIDAS DE RESULTADO: Supervivencia global, años de vida ajustados por calidad basados en estados de utilidad específicos del procedimiento (colectomía segmentaria 0,99, colectomía total 0,95). RESULTADOS: Entre los pacientes de 25 años, la colectomía total ofreció menos de un año de aumento de la supervivencia media en comparación con la resección segmentaria en todas las mutaciones ( MLH1: 0,8 años, MSH2: 0,9, MSH6: 0,3). Esta diferencia en la supervivencia disminuyó con el aumento de la edad. La resección segmentaria ofreció un pequeño aumento en los años de vida ajustados por calidad en comparación con la colectomía total en todas las edades. Esta diferencia fue mayor para MSH6 (0,9 años frente a MLH1 : 0,6 y MSH2 : 0,4 a los 25 años), especialmente en las edades más jóvenes de diagnóstico. En un análisis de sensibilidad, los estados de utilidad más bajos de la colectomía total dieron lugar a mayores aumentos en los años de vida ajustados por calidad con la colectomía segmentaria en todas las mutaciones. LIMITACIONES: Los datos sobre el riesgo de cáncer metacrónico no distinguen entre cánceres de colon y recto y se extrapolan más allá de los 70 años. CONCLUSIONES: La colectomía segmentaria y la colectomía total ofrecen resultados similares en cuanto a la supervivencia y los años de vida ajustados por calidad. Con la disminución del riesgo de cánceres metacrónicos, los pacientes con mutaciones MSH6 de todas las edades pueden preferir la resección segmentaria debido a los mayores años de vida ajustados por calidad y los menores beneficios de supervivencia de la colectomía total en comparación con los pacientes MLH1 y MSH2 . La estrategia quirúrgica en pacientes con cáncer colorrectal hereditario no poliposo debe individualizarse, y este estudio muestra que las diferencias específicas de la mutación en la supervivencia y los años de vida ajustados por calidad son pequeñas. ( AI-generated translation )

BACKGROUND: End colostomy reversal with colorectal anastomosis (Hartmann’s reversal) is a technically challenging procedure associated with significant morbidity due to adhesions and distorted anatomy after the index surgery. OBJECTIVE: To compare perioperative outcomes of open, laparoscopic, and robotic approaches for Hartmann’s reversal. DESIGN: Retrospective analysis of prospectively maintained data. SETTINGS: High-volume tertiary referral center with specialized colorectal surgery expertise. PATIENTS: A total of 218 patients who underwent Hartmann’s reversal between May 2018 and April 2023. MAIN OUTCOME MEASURES: Primary outcomes included hospital length of stay and rate of conversion. Secondary outcomes included overall complication rates and time to return of bowel function. RESULTS: Of 218 patients, 139 (63.8%) underwent open surgery, 48 (22.0%) underwent laparoscopic surgery, and 31 (14.2%) underwent robotic surgery. Robotic surgery had the lowest estimated blood loss (median 77.5 vs 100 mL laparoscopic and 150 mL open, p < 0.0001). Conversion to open surgery occurred in 16.1% of robotic cases compared with 35.4% of laparoscopic cases ( p = 0.0618). Minimally invasive approaches were associated with shorter hospital length of stay (median 3.0 vs 4.0 days, p < 0.0001) and faster return of bowel function (median 2.0 vs 3.0 days, p = 0.0095) compared to open surgery. Overall 30-day morbidity was 23.4%, with no significant difference among approaches ( p = 0.30). The temporal trend showed increasing adoption of robotic techniques, from 0% in 2018 to 53.8% in 2023, and a decrease in the proportion of cases approached by open surgery, from 76.9% in 2018 to 23.1% in 2023. LIMITATIONS: Retrospective design and single-institution experience with limited follow-up duration. CONCLUSIONS: Minimally invasive approaches to Hartmann’s reversal are associated with improved short-term outcomes compared to open surgery. The robotic approach shows promise in reducing conversion rates, potentially extending the benefits of minimally invasive surgery to more patients undergoing this challenging procedure. See Video Abstract . COMPARACIÓN DE LOS ENFOQUES ABIERTO, LAPAROSCÓPICO Y ROBÓTICO PARA LA REVERSIÓN DE HARTMANN: LA EXPERIENCIA DE LA CLÍNICA MAYO ANTECEDENTES: La reversión de la colostomía terminal con anastomosis colorrectal (reversión de Hartmann) es un procedimiento técnicamente complejo asociado a una morbilidad significativa debido a las adherencias y la distorsión anatómica tras la cirugía inicial. OBJETIVO: Comparar los resultados perioperatorios de los abordajes abiertos, laparoscópicos y robóticos para la reversión de Hartmann. DISEÑO: Análisis retrospectivo de datos mantenidos prospectivamente. ENTORNO: Centro terciario de referencia de gran volumen con experiencia especializada en cirugía colorrectal. PACIENTES: Un total de 218 se sometieron a la reversión de Hartmann entre mayo de 2018 y abril de 2023. PRINCIPALES MEDIDAS DE RESULTADOS: Los resultados primarios incluyeron la duración de la estancia hospitalaria y la tasa de conversión. Los resultados secundarios incluyeron las tasas generales de complicaciones y el tiempo de recuperación de la función intestinal. RESULTADOS: De los 218 pacientes, 139 (63,8 %) se sometieron a cirugía abierta, 48 (22,0 %) a cirugía laparoscópica y 31 (14,2 %) a cirugía robótica. La cirugía robótica tuvo la menor pérdida de sangre estimada (mediana, 77,5 ml frente a 100 ml en laparoscopia y 150 ml en cirugía abierta, p < 0,0001). La conversión a cirugía abierta se produjo en el 16,1 % de los casos robóticos frente al 35,4 % de los casos laparoscópicos ( p = 0,0618). Los abordajes mínimamente invasivos se asociaron con una estancia hospitalaria más corta (mediana de 3,0 días frente a 4,0 días, p < 0,0001) y un retorno más rápido de la función intestinal (mediana de 2,0 días frente a 3,0 días, p = 0,0095) en comparación con la cirugía abierta. La morbilidad global a los 30 días fue del 23,4 %, sin diferencias significativas entre los distintos enfoques ( p = 0,30). La tendencia temporal mostró un aumento en la adopción de técnicas robóticas, del 0 % en 2018 al 53,8 % en 2023, y una disminución en la proporción de casos abordados mediante cirugía abierta, del 76,9 % en 2018 al 23,1 % en 2023. LIMITACIONES: Diseño retrospectivo y experiencia de una sola institución con una duración de seguimiento limitada. CONCLUSIONES: Los enfoques mínimamente invasivos para la reversión de Hartmann se asocian con mejores resultados a corto plazo en comparación con la cirugía abierta. El enfoque robótico es prometedor para reducir las tasas de conversión, lo que podría ampliar los beneficios de la cirugía mínimamente invasiva a más pacientes que se someten a este procedimiento tan complejo. ( AI-generated translation ).

Abstract Systems-based hematology (SBH) is an emerging discipline focused on optimizing care delivery for patients with hematologic disorders within complex health care systems. Despite its growing importance, the field lacks a formal definition, standardized curriculum, and clear career pathway, hindering widespread adoption and training of future SBH experts. We sought to define core principles, competencies, challenges, and future directions of SBH to identify foundational components for a potential educational curriculum. Semistructured interviews were conducted with 8 SBH experts. Expert participants were identified through their committee involvement with the American Society of Hematology. Five overarching themes emerged: (1) The accidental expert: SBH careers have developed organically and reactively in response to pressing institutional needs, not through formal training; (2) Defining the unseen: experts define the field pragmatically by its function and roles; (3) The SBH toolkit: success requires a hybrid skillset combining clinical hematology expertise with health systems science (HSS) principles; (4) The architect's dilemma: the field's growth is hindered by interdependent challenges of building a curriculum, awareness, and securing resources; and (5) Charting the future: the discipline is poised for growth, driven by increasing health care complexity and specialization. SBH represents the nexus of HSS principles and hematology care. Formalizing SBH requires a multipronged approach, including developing a standardized curriculum that equips learners with a hybrid skillset, creating clear value for health systems and patients, and fostering a community of practice through professional societies. These findings provide a foundational framework for developing educational resources and institutional support necessary to continue advancing SBH.

BACKGROUND AND OBJECTIVES: Preventive treatment of unruptured intracranial aneurysms (UIAs) requires assessment of treatment risks vs expected benefit. Although established scores exist to estimate rupture and growth risk, currently, no externally validated tools exist to estimate the risks of microsurgical treatment of UIAs. Clinical prediction models based on machine learning enable generation of personalized risk estimates for each individual patient based on their specific patient and aneurysm characteristics. METHODS: Using data from 20 international centers from the prediction of adverse events after microsurgery for intracranial unruptured aneurysms study on patients treated microsurgically for UIAs, we developed and externally validated clinical prediction models for 3 outcomes measured at hospital discharge: poor neurological outcome (modified Rankin Score ≥3), new sensorimotor neurological deficits, and all-cause adverse events (Clavien-Dindo Grade ≥1). RESULTS: A total of 3705 patients were included. Data from 13 centers (2881, 78%) were used for model development. Fully trained models were evaluated on 824 patients (22%) from 7 additional centers. Average age was 56 ± 12 years, and 1049 (28%) were male. At discharge, poor neurological outcome was seen in 514 patients (14%). New sensorimotor deficits were observed in 534 patients (14%), and 894 patients (24%) experienced adverse events until discharge. At external validation, prediction of poor neurological outcome was achieved with good calibration and an area under the curve (AUC) = 0.70 (95% CI: 0.63-0.75). Similarly, new neurological deficits were predicted with good calibration and with an AUC = 0.69 (95% CI: 0.63-0.74). Prediction of all-cause adverse events only achieved an AUC = 0.59 (95% CI: 0.55-0.64) with fair calibration. The prediction model was integrated into a web application accessible at https://neurosurgery.shinyapps.io/PRAEMIUM/. CONCLUSION: The developed models for prediction of poor neurological outcome and new sensorimotor neurological deficits at discharge exhibit good calibration and fair discrimination based on a multinational external validation, indicating that the predicted probabilities correspond well to real-world risks and may thus be clinically useful in more objectively estimating the risk of microsurgical treatment.

Abstract Wiskott-Aldrich syndrome (WAS), an X-linked disorder characterized by immunodeficiency, thrombocytopenia, autoimmunity, and malignancy, can be effectively treated with allogeneic hematopoietic cell transplantation (HCT). Older age at HCT and mismatched donors are known to affect overall survival (OS). However, the influence of specific clinical manifestations or WAS variant class on OS and factors associated with event-free survival (EFS) remain incompletely defined. We analyzed outcomes of 308 patients with WAS who underwent HCT at 37 institutions of the Primary Immune Deficiency Treatment Consortium from 1990 to 2018. With a median follow-up of 5.3 years, the 5-year OS and EFS were 87.2% and 79.7%, respectively. Age ≥5 years, donor type, and a pre-HCT history of severe infection had a negative impact on OS and EFS, whereas pre-HCT autoimmunity had no impact. Reduced-intensity regimens were associated with lower T-cell and myeloid donor chimerism, particularly when non–busulfan-based regimens were used. Low myeloid donor chimerism was associated with lower platelet counts. Mixed chimerism was not consistently associated with post-HCT autoimmunity. Patients with class I (exon 1-2 missense and intron 5 hot spot variants) and class II variants (all others) had similar pre-HCT clinical symptom severity and no difference in OS, EFS, or platelet recovery post-HCT. In conclusion, our study showed excellent long-term OS and EFS after HCT for WAS, highlighting the importance of early HCT, before the development of severe infections. We confirmed that HCT using busulfan-based conditioning was associated with improved donor chimerism and platelet recovery. This trial was registered at www.clinicaltrials.gov as NCT02064933.

Abstract Introduction Bruton tyrosine kinase inhibitors (BTKis) have changed the treatment algorithm for patients with high-risk relapsed/refractory (R/R) chronic lymphocytic leukemia (CLL). Efficacy was demonstrated for zanubrutinib and ibrutinib in high-risk R/R CLL patients in the ALPINE (NCT03734016) trial, for ibrutinib and acalabrutinib in the ELEVATE-RR (NCT02477696) trial, and for acalabrutinib in the ASCEND (NCT00135226) trial. While there is limited head-to-head comparative trial data of all covalent BTKis in the treatment of high-risk R/R CLL, a previously published network meta-analysis (NMA) used data from the ALPINE, ELEVATE-RR, and ASCEND trials, and reported that zanubrutinib demonstrated significantly improved relative efficacy compared to ibrutinib and acalabrutinib in high-risk R/R CLL (Shadman 2025). Within this NMA, high-risk R/R CLL populations were defined based on the prespecified definitions within each trial, including patients with del(17p) and/or TP53 mutations in the ALPINE and ASCEND trials, and del(17p)/del(11q) in the ELEVATE-RR trial. This study aimed to compare zanubrutinib versus ibrutinib and acalabrutinib in high-risk R/R CLL by calculating the number needed to treat (NNT) to avoid one progression or death. Methods A model was developed to evaluate the NNT among high-risk R/R CLL patients to avoid a progression or death. Progression free survival (PFS) data for zanubrutinib in high-risk patients were extracted from the ALPINE trial. PFS values for ibrutinib and acalabrutinib were derived from the previously published NMA study. PFS for ibrutinib and acalabrutinib were calculated by applying PFS hazard ratio from the NMA results to the zanubrutinib PFS value for the high-risk population in the ALPINE trial. 24-month PFS (72.6% for zanubrutinib, 52.0% for ibrutinib, and 55.9% for acalabrutinib) were used for the base case analysis of the model. Sensitivity analyses were conducted to examine the impact of alternative PFS inputs for ibrutinib, derived directly from the high-risk populations in ALPINE trial (unadjusted). Results The base case results from the NNT model indicate that for the treatment of patients with high-risk R/R CLL over 24-months, every five patients treated with zanubrutinib instead of ibrutinib results in the avoidance of one disease progression or death and every six patients treated with zanubrutinib instead of acalabrutinib results in the avoidance of one disease progression or death. Applying the model result to a hypothetical scenario of 100 patients with high-risk R/R CLL treated with zanubrutinib versus ibrutinib for two years finds approximately 20 patients will avoid disease progression events or deaths. Treating the same population with zanubrutinib versus acalabrutinib for two years will result in the avoidance of 17 disease progression events or death. Sensitivity analysis showed consistent NNT results.Conclusions The NNT analysis suggests that among high-risk R/R CLL patients, zanubrutinib is associated with more favorable clinical outcomes compared to ibrutinib and acalabrutinib. Over 24 months, zanubrutinib demonstrated a NNT of five to prevent one disease progression or death compared to ibrutinib, and an NNT of six compared to acalabrutinib, in high-risk R/R CLL patients. Study findings should be interpreted based on modeling assumptions and potential patient population differences across trials.