Clinical variants paired with phenotype: A rich resource for brain gene curation

Automated extraction of functional biomarkers of verbal and ambulatory ability from multi-institutional clinical notes using large language models

BACKGROUND

Functional biomarkers in neurodevelopmental disorders, such as verbal and ambulatory abilities, are essential for clinical care and research activities. Treatment planning, intervention monitoring, and identifying comorbid conditions in individuals with intellectual and developmental disabilities (IDDs) rely on standardized assessments of these abilities. However, traditional assessments impose a burden on patients and providers, often leading to longitudinal inconsistencies and inequities due to evolving guidelines and associated time-cost. Therefore, this study aimed to develop an automated approach to classify verbal and ambulatory abilities from EHR data of IDD and cerebral palsy (CP) patients. Application of large language models (LLMs) to clinical notes, which are rich in longitudinal data, may provide a low-burden pipeline for extracting functional biomarkers efficiently and accurately.

METHODS

Data from the multi-institutional National Brain Gene Registry (BGR) and a CP clinic cohort were utilized, comprising 3,245 notes from 125 individuals and 5,462 clinical notes from 260 individuals, respectively. Employing three LLMs-GPT-3.5 Turbo, GPT-4 Turbo, and GPT-4 Omni-we provided the models with a clinical note and utilized a detailed conversational format to prompt the models to answer: "Does the individual use any words?" and "Can the individual walk without aid?" These responses were evaluated against ground-truth abilities, which were established using neurobehavioral assessments collected for each dataset.

RESULTS

LLM pipelines demonstrated high accuracy (weighted-F1 scores > .90) in predicting ambulatory ability for both cohorts, likely due to the consistent use of Gross Motor Functional Classification System (GMFCS) as a consistent ground-truth standard. However, verbal ability predictions were more accurate in the BGR cohort, likely due to higher adherence between the prompt and ground-truth assessment questions. While LLMs can be computationally expensive, analysis of our protocol affirmed the cost effectiveness when applied to select notes from the EHR.

CONCLUSIONS

LLMs are effective at extracting functional biomarkers from EHR data and broadly generalizable across variable note-taking practices and institutions. Individual verbal and ambulatory ability were accurately extracted, supporting the method's ability to streamline workflows by offering automated, efficient data extraction for patient care and research. Future studies are needed to extend this methodology to additional populations and to demonstrate more granular functional data classification.

Expansion of the Genotypic and Phenotypic Spectrum of SETD5 Disorder Using Data From the National Brain Gene Registry

Haploinsufficiency of SETD5, which facilitates gene transcription and histone modification, causes intellectual disability. However, the full phenotypic spectrum of SETD5 Disorder is not well characterized. We summarized clinical features of participants with SETD5 Disorder enrolled in the National Brain Gene Registry (BGR). We analyzed data from 13 individuals with a molecular diagnosis of SETD5 Disorder, ages 2-37 years, using data from electronic health records, standardized surveys, and direct assessments. We also determined the number of participants for whom 10 pre-selected clinical features were present based on ICD-10 billing codes in comparison to manual review of clinical documentation. In the cohort, there were 11 unique pathogenic/likely pathogenic variants in 13 individuals from 11 different families. Of these 11 unique variants, 6 were nonsense, 4 were frameshift, and one was splice site. Participants in our cohort had features not previously reported, including brain and musculoskeletal abnormalities. One participant had cerebral palsy. For the 10 pre-selected clinical features, we showed satisfactory representation of these features by ICD-10 billing codes. This report highlights novel genotype and phenotype associations with SETD5 Disorder. Examination of larger patient cohorts with SETD5 Disorder is needed to evaluate the true prevalence of these medical conditions in this disorder.

The expanding clinical and genetic spectrum of DYNC1H1-related disorders

Intracellular trafficking involves an intricate machinery of motor complexes, including the dynein complex, to shuttle cargo for autophagolysosomal degradation. Deficiency in dynein axonemal chains, as well as cytoplasmic light and intermediate chains, have been linked with ciliary dyskinesia and skeletal dysplasia. The cytoplasmic dynein 1 heavy chain protein (DYNC1H1) serves as a core complex for retrograde trafficking in neuronal axons. Dominant pathogenic variants in DYNC1H1 have been previously implicated in peripheral neuromuscular disorders (NMD) and neurodevelopmental disorders (NDD). As heavy-chain dynein is ubiquitously expressed, the apparent selectivity of heavy chain dyneinopathy for motor neuronal phenotypes remains currently unaccounted for. Here, we aimed to evaluate the full DYNC1H1-related clinical, molecular and imaging spectrum, including multisystem features and novel phenotypes presenting throughout life. We identified 47 cases from 43 families with pathogenic heterozygous variants in DYNC1H1 (aged 0-59 years) and collected phenotypic data via a comprehensive standardized survey and clinical follow-up appointments. Most patients presented with divergent and previously unrecognized neurological and multisystem features, leading to significant delays in genetic testing and establishing the correct diagnosis. Neurological phenotypes include novel autonomic features, previously rarely described behavioral disorders, movement disorders and periventricular lesions. Sensory neuropathy was identified in nine patients (median age of onset 10.6 years), of which five were only diagnosed after the second decade of life, and three had a progressive age-dependent sensory neuropathy. Novel multisystem features included primary immunodeficiency, bilateral sensorineural hearing loss, organ anomalies and skeletal manifestations, resembling the phenotypic spectrum of other dyneinopathies. We also identified an age-dependent biphasic disease course with developmental regression in the first decade and, following a period of stability, neurodegenerative progression after the second decade of life. Of note, we observed several cases in whom neurodegeneration appeared to be prompted by intercurrent systemic infections with double-stranded DNA viruses (Herpesviridae) or single-stranded RNA viruses (Ross River fever, SARS-CoV-2). Moreover, the disease course appeared to be exacerbated by viral infections regardless of age and/or severity of neurodevelopmental disorder manifestations, indicating a role of dynein in anti-viral immunity and neuronal health. In summary, our findings expand the clinical, imaging and molecular spectrum of pathogenic DYNC1H1 variants beyond motor neuropathy disorders and suggest a life-long continuum and age-related progression due to deficient intracellular trafficking. This study will facilitate early diagnosis and improve counselling and health surveillance of affected patients.

The Brain Gene Registry: a data snapshot

Monogenic disorders account for a large proportion of population-attributable risk for neurodevelopmental disabilities. However, the data necessary to infer a causal relationship between a given genetic variant and a particular neurodevelopmental disorder is often lacking. Recognizing this scientific roadblock, 13 Intellectual and Developmental Disabilities Research Centers (IDDRCs) formed a consortium to create the Brain Gene Registry (BGR), a repository pairing clinical genetic data with phenotypic data from participants with variants in putative brain genes. Phenotypic profiles are assembled from the electronic health record (EHR) and a battery of remotely administered standardized assessments collectively referred to as the Rapid Neurobehavioral Assessment Protocol (RNAP), which include cognitive, neurologic, and neuropsychiatric assessments, as well as assessments for attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD). Co-enrollment of BGR participants in the Clinical Genome Resource's (ClinGen's) GenomeConnect enables display of variant information in ClinVar. The BGR currently contains data on 479 participants who are 55% male, 6% Asian, 6% Black or African American, 76% white, and 12% Hispanic/Latine. Over 200 genes are represented in the BGR, with 12 or more participants harboring variants in each of these genes: CACNA1A, DNMT3A, SLC6A1, SETD5, and MYT1L. More than 30% of variants are de novo and 43% are classified as variants of uncertain significance (VUSs). Mean standard scores on cognitive or developmental screens are below average for the BGR cohort. EHR data reveal developmental delay as the earliest and most common diagnosis in this sample, followed by speech and language disorders, ASD, and ADHD. BGR data has already been used to accelerate gene-disease validity curation of 36 genes evaluated by ClinGen's BGR Intellectual Disability (ID)-Autism (ASD) Gene Curation Expert Panel. In summary, the BGR is a resource for use by stakeholders interested in advancing translational research for brain genes and continues to recruit participants with clinically reported variants to establish a rich and well-characterized national resource to promote research on neurodevelopmental disorders.