Documenting Trofinetide Treatment Phenotypes in Rett Syndrome with a Dual-LLM Consensus Pipeline
Abstract
Objective: Trofinetide, approved in 2023 as the first Rett syndrome treatment, lacks scalable real-world monitoring because chart review is labor-intensive. We developed a dual large language model (LLM) consensus pipeline to extract 18 treatment-response phenotypes, comprising 4 adverse effects, 13 Rett clinical features, and medication refusal, from clinical notes. Materials and Methods: We analyzed 26,763 notes from 108 patients with confirmed sustained trofinetide exposure at Texas Children's Hospital. The pipeline filters notes by keyword, classifies each through two LLMs requiring consensus, and aggregates a patient-by-timepoint matrix. Architecture and prompt were selected on a diarrhea development set, then frozen and applied to the other 17 phenotypes with the phenotype definition substituted, as an out-of-sample test. The reported prompt comparison scores 183 expert-labeled cells. Results: A prompt requiring diarrhea to be active and severe performed best (84.2 ± 0.0% accuracy; F1 = 0.674), 47 percentage points above regular-expression matching (p < 0.0001). Expert reviewers judged six of the pipeline's calls per phenotype, three positive and three negative, and agreed with 90.2% (92/102). The 13 Rett clinical features are manifestations of the disorder itself rather than adverse effects of trofinetide. Gross and fine motor impairment, language/communication, and stereotypy were documented present in 92-100% of patients at every visit. Discussion: In-context LLM extraction captured multi-domain treatment responses. The outcome is documented presence at each visit, not a direction-of-change judgment. Conclusion: Coupling in-context LLM reasoning with a consensus architecture offers a scalable alternative to keyword and regular-expression extraction for real-world evidence, adapting to new phenotypes and medications through configuration alone.
Related
- Progress in mast cell activation syndrome: the global consensus-2 diagnostic criteria at six years
- Novel Compound Heterozygous RECQL4 Pathogenic Variants and Associated Phenotypes in a Chinese Patient With Rothmund-Thomson Syndrome: Case Report and Literature Review.
- Sex-associated territory-specific vascular phenotypes in genetically confirmed vascular Ehlers-Danlos syndrome
- Differential Phenotypes in Type I and Type II Pediatric Stickler Syndrome
- Effectiveness and drug survival of adalimumab across clinical phenotypes of Behçet’s syndrome: an exploratory real-world cohort study
BibTeX
@misc{palacios2026documenting,
title = {{Documenting Trofinetide Treatment Phenotypes in Rett Syndrome with a Dual-LLM Consensus Pipeline}},
author = {Daniel Palacios and Guantong Qi and Mikael Guzman Karlsson and Shawntel K. David and Mohammed Sameeruddin and Bernhard Suter and Zhandong Liu and Davut Pehlivan and Hu Chen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.22062},
eprint = {2610.22062},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.22062},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Documenting Trofinetide Treatment Phenotypes in Rett Syndrome with a Dual-LLM Consensus Pipeline
AU - Daniel Palacios
AU - Guantong Qi
AU - Mikael Guzman Karlsson
AU - Shawntel K. David
AU - Mohammed Sameeruddin
AU - Bernhard Suter
AU - Zhandong Liu
AU - Davut Pehlivan
AU - Hu Chen
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.22062
UR - https://arxiv.org/abs/2610.22062
AB - Objective: Trofinetide, approved in 2023 as the first Rett syndrome treatment, lacks scalable real-world monitoring because chart review is labor-intensive. We developed a dual large language model (LLM) consensus pipeline to extract 18 treatment-response phenotypes, comprising 4 adverse effects, 13 Rett clinical features, and medication refusal, from clinical notes. Materials and Methods: We analyzed 26,763 notes from 108 patients with confirmed sustained trofinetide exposure at Texas Children's Hospital. The pipeline filters notes by keyword, classifies each through two LLMs requiring consensus, and aggregates a patient-by-timepoint matrix. Architecture and prompt were selected on a diarrhea development set, then frozen and applied to the other 17 phenotypes with the phenotype definition substituted, as an out-of-sample test. The reported prompt comparison scores 183 expert-labeled cells. Results: A prompt requiring diarrhea to be active and severe performed best (84.2 ± 0.0% accuracy; F1 = 0.674), 47 percentage points above regular-expression matching (p < 0.0001). Expert reviewers judged six of the pipeline's calls per phenotype, three positive and three negative, and agreed with 90.2% (92/102). The 13 Rett clinical features are manifestations of the disorder itself rather than adverse effects of trofinetide. Gross and fine motor impairment, language/communication, and stereotypy were documented present in 92-100% of patients at every visit. Discussion: In-context LLM extraction captured multi-domain treatment responses. The outcome is documented presence at each visit, not a direction-of-change judgment. Conclusion: Coupling in-context LLM reasoning with a consensus architecture offers a scalable alternative to keyword and regular-expression extraction for real-world evidence, adapting to new phenotypes and medications through configuration alone.
ER -