Rapid Paediatric Sequencing (RaPS): comprehensive real-life workflow for rapid diagnosis of critically ill children

Abstract

Background: Rare genetic conditions are frequent risk factors for, or direct causes of, paediatric intensive care unit (PICU) admission. Such conditions are frequently suspected but unidentified at PICU admission. Compassionate and effective care is greatly assisted by definitive diagnostic information. There is therefore a need to provide a rapid genetic diagnosis to inform clinical management.To date, whole genome sequencing (WGS) approaches have proved successful in diagnosing a proportion of children with rare diseases, but results may take months to report. Our aim was to develop an end-to-end workflow for the use of rapid WGS for diagnosis in critically ill children in a UK National Health Service (NHS) diagnostic setting.

Methods: We sought to establish a multidisciplinary Rapid Paediatric Sequencing team for case selection, trio WGS, rapid bioinformatics sequence analysis and a phased analysis and reporting system to prioritise genes with a high likelihood of being causal.

Results: Trio WGS in 24 critically ill children led to a molecular diagnosis in 10 (42%) through the identification of causative genetic variants. In 3 of these 10 individuals (30%), the diagnostic result had an immediate impact on the individual's clinical management. For the last 14 trios, the shortest time taken to reach a provisional diagnosis was 4 days (median 8.5 days).

Conclusion: Rapid WGS can be used to diagnose and inform management of critically ill children within the constraints of an NHS clinical diagnostic setting. We provide a robust workflow that will inform and facilitate the rollout of rapid genome sequencing in the NHS and other healthcare systems globally.

Keywords: genomics; paediatric intensive care unit; rapid diagnosis; rare disease; whole genome sequencing.

Figure 1

Description of RaPS workflow. A flow diagram representing the three stages of the RaPS workflow showing how trio samples progress from the stage of patient referral to the issuing of a diagnostic report. This diagram provides brief details of the variant filtering steps applied to samples and the phased analysis strategy. Detailed methods are provided in online supplementary material 2.

Figure 2

Number of trios referred and diagnoses made per clinical speciality. Graph showing the number of patients referred from specialist clinical teams and whether that patient received a molecular diagnosis.

Figure 3

RaPS Time Frame for the last 14 cases. (Note, the first 10 cases were used for proof of principle and establishment of the workflow). (A) The 5 calendar days time frame was achieved as indicated on the left panel. *To provide a comparison with the time frames published by previous studies, we have calculated the median time frame of the last 14 cases from the time library preparation was initialised. Note that the timeframe to ascertain patients was variable and depended on a number of factors such as availability of parents for consenting. (B) Histogram of time frame of genomic sequencing calculated from library preparation to return of clinical findings. Weekends, holidays and delays due to reagents failure or unavailability of sequencers are not excluded from the time frame to reflect real-life working conditions. (C) Table shows the quartile distribution of time frame (calculated from library preparation to return of clinical findings).