# Scaling genomic reanalysis to unlock diagnoses and transform rare disease care

## TL;DR

A new system at Boston Children’s Hospital enables ongoing genomic reanalysis for rare diseases, improving diagnoses as patient data and knowledge evolve.

## Contribution

A centralized, semi-automated, clinician-facing genomic reanalysis workflow that enables iterative and scalable rare disease diagnosis.

## Key findings

- Proactive genomic reanalysis identified 33 high-suspicion causative variants in 42 patients.
- Three additional candidate variants of uncertain significance were identified and reported.
- The system demonstrates feasibility for iterative genomic reanalysis integrated into clinical care.

## Abstract

Genomic reanalysis can identify causative variants for rare diseases as patient phenotypes evolve and gene-disease knowledge expands. Despite its diagnostic value, routine reanalysis is limited by clinician capacity, lack of patient follow-up, data silos, cost, and lack of availability of clinical data to testing laboratories that are not obligated to conduct reanalysis. The Children’s Rare Disease Collaborative at Boston Children’s Hospital (BCH) has integrated genomic and phenotypic data from over 15,500 patients into a clinician-facing platform. Leveraging this infrastructure, we developed a Proactive Genomic Reanalysis (PGR) workflow at BCH for clinical sequencing data that is centralized, semi-automated, and clinically integrated. Here, we report initial results and outline required resources and transferable insights applicable to other healthcare settings. Initial pilot implementation, applied to a subset of clinical sequencing patients’ data, revealed practical challenges, notably clinician turnover and patient recontact difficulties. Of 42 patients’ candidate variants discovered by the PGR bioinformatics pipeline and returned to treating clinicians, 33 were determined to have a high suspicion of disease causality and an additional 3 were determined to be candidate variant of uncertain significance. A process to generate reports and return results to patients was initiated when applicable. Although the initial pilot implementation was limited, the PGR bioinformatics pipeline is designed to be utilized iteratively, making reanalysis a continuing process. This work highlights the feasibility and impact of centralized PGR processes and the potential for healthcare institutions to scale genomic reanalysis.

Centralized, semi-automated reanalysis at Boston Children’s Hospital enables ongoing rare disease diagnoses. This proactive, clinician-driven model transforms reanalysis from an occasional event into a dynamic, iterative process integrated with clinical care.