Citations

Background/Objectives: Diagnostic evaluation and management of nontraumatic osteonecrosis of the femoral head (ONFH) vary substantially. This systematic review was conducted to inform development of the Association Research Circulation Osseous (ARCO) clinical practice guideline for diagnosis and treatment of ARCO stages I to III ONFH. Methods: We searched MEDLINE, EMBASE, Web of Science, SCOPUS, Global Index Medicus, and the Cochrane Library for studies evaluating imaging modalities and treatments for adult ONFH. We assessed risk of bias using the QUADAS-2, the ROB-2, and the ROBINS-I tools; conducted meta-analyses using random-effects regression; and evaluated certainty of evidence using GRADE methodology. Results: Among 36 included studies, 18 addressed diagnostic test accuracy and 18 addressed comparative effectiveness of treatments. Magnetic resonance imaging (MRI) demonstrated the highest pooled sensitivity (0.91; 95% confidence interval (CI), 0.87 to 0.94) and specificity (0.96; 95% CI, 0.87 to 0.99) for ONFH diagnosis. Bone marrow edema and grade 2+ joint effusion on MRI differentiated symptomatic versus asymptomatic disease. Computed tomography and MRI better detected subchondral fractures than plain radiography. Very low-grade evidence suggested lower rates of femoral head collapse with core decompression plus bone marrow concentrate compared with core decompression alone (pooled relative risk [RR], 0.55; 95% CI, 0.36 to 0.83), and with vascularized versus non-vascularized bone grafting (RR, 0.35; 95% CI, 0.14 to 0.84) over a ≤5-year follow-up. Based on three non-comparative case series, osteotomies might have a lower risk of collapse over a 10- to 20-year follow-up, but this needs to be evaluated in future comparative research. Inconsistent outcome reporting hindered treatment outcome pooling. There were no comparative studies that evaluated observation only versus intervention in asymptomatic disease or strategies for monitoring treatment response. Conclusions: Evidence supporting optimal imaging modalities and early joint-preserving interventions remains limited and predominantly observational, underscoring the need for high-quality comparative studies with consistently defined core outcomes to guide clinical decision-making.

Abstract Background Young age is a known risk factor for breast cancer recurrence and mortality, particularly in hormone receptor-positive disease. However, no specific evidence exists on whether age at diagnosis influences clinical behavior and outcomes among carriers of germline BRCA pathogenic or likely pathogenic variants (PVs). Methods The BRCA BCY Collaboration (NCT03673306) is an international, multicenter, hospital-based, retrospective cohort study including women with germline BRCA1 and/or BRCA2 PVs and diagnosed with stage I-III breast cancer at the age of 40 years or younger between January 2000 and December 2020. The present analysis aimed to evaluate clinical features and outcomes of breast cancer in young BRCA carriers according to age at diagnosis, by comparing patients aged ≤30 years with those aged 31-35 and 36-40 years at breast cancer diagnosis. The primary endpoint was disease-free survival (DFS). Overall survival (OS) was a secondary endpoint. Multivariable Cox proportional hazards models were used to estimate adjusted hazard ratios (aHRs), adjusting for risk-reducing mastectomy (RRM) and risk-reducing salpingo-oophorectomy (RRSO) (time-dependent), and stratifying by country, year of diagnosis, tumor size, nodal involvement, tumor grade and hormone receptor status. Results From 109 centers in 5 continents, 5350 young BRCA carriers were included, of whom 1106 (20.7%) were ≤30 years of age, 1945 (36.4%) were 31-35 years and 2299 (43.0%) were 36-40 years at the time of breast cancer diagnosis. Compared to patients aged 31-35 and 36-40 years, those aged ≤30 years were more likely to harbor BRCA1 PVs (71.3%, 62.4%, 61.1%, respectively), to present with high-grade (G3 in 71.3%, 66.6%, 65.0%, respectively), triple-negative breast cancer (56.3%, 48.1%, 46.6%, respectively), and less frequently smaller tumor size (T1: 31.7%, 36.6% and 39.5%, respectively). Consequently, chemotherapy was administered more frequently among patients aged ≤30 years (94.4%, 92.6%, 89.7%, respectively). At a median follow-up of 8.2 years (IQR, 4.7-12.8), 8-year DFS rate was 65.0% among patients aged ≤30 years, 64.7% in the 31-35 years group, and 65.8% in the 36-40 years group. No significant association between age at diagnosis and DFS outcome was observed (31-35 vs. ≤30 years: aHR 1.03, 95% CI 0.89-1.19; 36-40 vs. ≤30 years aHR 1.00, 95% CI 0.87-1.16). Similarly, no significant association between age at diagnosis and OS was observed (31-35 vs. ≤30 years: aHR 1.10, 95% CI 0.86-1.41; 36-40 vs. ≤30 years: aHR 1.11, 95% CI 0.87-1.43). Subgroup analyses by specific BRCA gene, tumor subtype, tumor size, nodal status, and chemotherapy use revealed no statistically significant interaction with age at diagnosis in terms of DFS or OS (all p for interaction > 0.05). Specifically, no significant differences between age groups were observed among patients with luminal-like tumors for both DFS and OS. Conclusions In this unique international cohort of BRCA carriers diagnosed with breast cancer at a young age (≤40 years), we found no significant differences in DFS nor OS across different age groups. Notably, very young age (≤30 years) at diagnosis was not associated with worse outcomes compared to 31-35 years or 36-40 years. Age per se should not be considered a negative prognostic factor in BRCA carriers when appropriate treatment is provided. Citation Format: E. Blondeaux, A. Meacci, M. Fragío Gil, E. Klocker, F. Coussy, H. Kim, R. Bernstein Molho, F. Hilbers, A. Di Meglio, A. Kwong, K. Pogoda, J. Bajpai, H. Wildiers, E. Agostinetto, J. Balmana, H. C. Moore, K. Phillips, A. H. Partridge, C. Rousset-Jablonski, F. A. Peccatori, A. Toss, T. Renaud, C. Anghelone, A. Benitez Cruz, S. Paluch-Shimon, J. Lee, R. Fruscio, W. Cui, S. M. Wong, M. Lee, M. Rozenblit, C. Vernieri, C. Villarreal-Garza, A. Matikas, M. Dieci, L. De Marchis, J. K. Plichta, K. J. Ruddy, D. Can Guven, F. Puglisi, Z. Kemp, P. A. Meireles, L. Del Mastro, V. Delucchi, M. Lambertini. Association between age at diagnosis and survival among young BRCA carriers with breast cancer: results from an international multicenter hospital-based cohort study [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PD4-07.

Background Patients with anti-leucine-rich glioma-inactivated 1 (anti-LGI1) autoimmune encephalitis (AIE) frequently present with faciobrachial dystonic seizures (FBDS), which are often under-detected due to their subtle motor features and lack of consistent EEG correlates. Conventional in-hospital monitoring and patient-reported seizure diaries are limited, particularly during nocturnal periods. In this pilot feasibility study, we evaluated the use of a wearable device (Empatica E4) to objectively monitor FBDS-related motor and autonomic activity in seven patients with anti-LGI1 AIE and four control subjects. Methods Wearable recordings included accelerometry (ACC), electrodermal activity (EDA), photoplethysmography (PPG) and skin temperature. A support vector machine-based algorithm was developed to identify disorder-related events using ACC and EDA features. Analyses were restricted to sleep periods to minimise confounding from voluntary movements. Results Pretreatment recordings from two patients (P1 and P2) showed significantly elevated motor event frequency, ACC magnitude and EDA activity compared to controls. Post-treatment recordings demonstrated marked reductions in these metrics, consistent with patient-reported clinical improvement. Other patients showed variable results depending on treatment timing and signal quality. Conclusion Our pilot study demonstrates the feasibility of using wearable technology for objective, real-world monitoring and assessment of patient status in anti-LGI1 AIE. Nocturnal monitoring offers a low-noise baseline for detecting seizure-related activity and may support earlier diagnosis, more accurate evaluation of treatment response and reduced reliance on resource-intensive inpatient evaluations. Future studies should expand monitoring to daytime periods and validate findings against conventional modalities such as video-EEG and electromyography polygraphy.

Abstract Objectives To comprehensively evaluate the validity of International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes for both prevalent diagnoses and less common diseases, and to assess the performance of a large language model (LLM)-based system in validating these codes. Materials and Methods This retrospective study analyzed hospital admissions from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. We developed a validated LLM-based system using GPT-4o, refined through iterative prompt engineering, to assess ICD-10-CM code validity. We measured the positive predictive value (PPV) of ICD-10-CM codes, PPV of principal and secondary diagnoses, and the performance of an LLM-based system in code validation. Results Among 865 079 assigned codes, the PPV was 84.6% (95% CI, 84.5%-84.6%). Principal diagnoses had a PPV of 93.9% (95% CI, 93.7%-94.1%), while secondary diagnoses had a PPV of 83.8% (95% CI, 83.7%-83.9%). The LLM system demonstrated high performance in validating ICD codes, achieving 93.6% accuracy, 95.4% sensitivity, and 85.2% specificity. Among correctly assigned secondary diagnoses, the majority (67.9%) represented historical or baseline conditions, while 32.1% reflected active conditions that deviated from baseline status; 22.3% of these emerged after hospital admission. PPV decreases with later diagnosis positions, with the largest decline occurring between principal and secondary diagnoses. Discussion and Conclusion In this large-scale evaluation, ICD-10-CM codes exhibited generally high accuracy, though variability existed by position and condition type. A validated LLM system performed comparably to physician review and offers a scalable means to improve coding accuracy. These findings support the potential for integrating LLM-based auditing into routine workflows to strengthen the quality of administrative and research data.

ABSTRACT Objective To comprehensively evaluate the validity of ICD-10-CM codes for both prevalent diagnoses and less common diseases, and to assess the performance of a large language model (LLM)-based system in validating these codes. Materials and Methods This retrospective study analyzed hospital admissions from the Medical Information Mart for Intensive Care (MIMIC-IV) database. We developed a validated LLM-based system using GPT-4o, refined through iterative prompt engineering, to assess ICD-10-CM code validity. We measured the PPV of ICD-10-CM codes, PPV of principal and secondary diagnoses, and the performance of an LLM-based system in code validation. Results Among 865,079 assigned codes, the PPV was 84.6% (95% CI, 84.5%-84.6%). Principal diagnoses had a PPV of 93.9% (95% CI, 93.7%-94.1%), while secondary diagnoses had a PPV of 83.8% (95% CI, 83.7%-83.9%). The LLM system demonstrated high performance in validating ICD codes, achieving 93.6% accuracy, 95.4% sensitivity and 85.2% specificity. Among correctly assigned secondary diagnoses, the majority (67.9%) represented historical or baseline conditions, while 32.1% reflected active conditions that deviated from baseline status; 22.3% of these emerged after hospital admission. PPV decreases with later diagnosis positions, with the largest decline occurring between principal and secondary diagnoses. Discussion and Conclusion In this large-scale evaluation, ICD-10-CM codes exhibited generally high accuracy, though variability existed by position and condition type. A validated LLM system performed comparably to physician review and offers a scalable means to improve coding accuracy. These findings support the potential for integrating LLM-based auditing into routine workflows to strengthen the quality of administrative and research data.

Abstract Background Dynamically predicting patients at risk of Alzheimer's disease (AD) is crucial for timely treatment and care in precision medicine. Method We present a novel architecture, a Kolmogorov‐Arnold networks (KANs)‐based joint prediction model of longitudinal and survival data (JM‐KAN). Real‐life data of patients with Mild Cognitive Impairment (MCI) with no prior AD diagnosis from National Alzheimer's Coordinating Center (NACC) dataset was utilized for individualized predictions. We built joint models with landmark times every 1 year up to 4 years after the MCI diagnosis, and predicted survival outcomes one year after each landmark (i.e. 1, 2, 3, and 4 years). Integrated AUC (iAUC), and integrated Brier score (iBS) were measured for existing prediction models and our novel architecture. Result Out of 2711 unique patients with MCI, 821 patients progressed to develop AD. Mean time‐to‐event in years was 3.19 (standard deviation (SD): 2.29). The number of visits prior to AD diagnosis was 4.03 (SD: 2.41). With iAUC of 0.789, iBS of 0.118, JM‐KAN has both the highest iAUC and lowest iBS among joint models, which indicates superiority in discrimination and overall performance. For individualized predictions, the model successfully predicted both longitudinal risk factors and survival probability, as new information came in every year. Conclusion Using neural network‐based Artificial Intelligence (AI) model, we demonstrated that this architecture improves upon prior methods in dynamically updating predictions of AD as new longitudinal data becomes available.

Abstract Background Chronic traumatic encephalopathy (CTE) is a neurodegenerative tauopathy associated with repetitive head impacts (RHI). Autopsy studies show that RHI influences risk for other later‐life neuropathologies, particularly white matter and vascular injury. We have observed a unique pattern of white matter hyperintensities (WMH) on FLAIR MRI in people exposed to RHI (RHI‐WMH) characterized by small, discrete lesions near the depths of sulci (Figure 1a). The objective of this study was to characterize this pattern empirically in former American football players and to investigate neuropathology co‐localizing with the WMH. Method The sample included 164 male former football players ( n = 111 professional, n = 53 college) and 55 asymptomatic males without RHI from the DIAGNOSE CTE Project (Table 1). Visual ratings of WMH blind to the participant's RHI and clinical status were conducted to count lesions that were a.) within 1.0cm of the gray/white matter boundary, b.) as bright as the cortex, c.) spherical and d.) <1.0cm in size. ANCOVA compared groups on total number of WMH adjusted for age and vascular risk factors. We selected one individual (former professional player) with the typical RHI‐WMH pattern and who donated their brain for neuropathological examination. Guided by in vivo MRI, ex vivo MRI was done to locate the WMH near the depths of the sulci in the tissue. Tissue sections were tested for phosphorylated tau immunoreactivity and reactive astrocytes and microglia in areas surrounding the lesion seen on MRI. Result Number of RHI‐WMH was greater in football players compared to controls ( p = 0.007) (Figure 1b). Number of RHI‐WMH was greater in former professional players compared to controls ( p = 0.001) but not compared to the college players ( p = 0.983). In the neuropathology case study, we located the punctate RHI‐WMH on ex vivo MRI. Immunostaining revealed that the RHI‐WMH was adjacent to a sulcus with the pathognomonic CTE p ‐tau lesion and was characterized by myelin pallor, dense GFAP‐positive glial scarring, and increased total and activated microglia (Figure 2). Conclusion We provide empirical support for exposure to RHI as potentially leading to a unique pattern of WMH. Importantly, histopathological evidence showed that the RHI‐WMH co‐localized with RHI and CTE‐related neuropathologies at the depths of the sulci.