Dominic Habgood-Coote (United Kingdom)
Imperial College London Department of Infectious DiseaseAuthor Of 2 Presentations
THE IDENTIFICATION AND SUBSEQUENT CROSS-PLATFORM VALIDATION OF A HOST GENE EXPRESSION SIGNATURE FOR DIFFERENTIATING BETWEEN MIS-C AND OTHER INFECTIOUS AND INFLAMMATORY DISEASES
Abstract
Backgrounds:
Multisystem Inflammatory Syndrome in Children (MIS-C) occurs several weeks after SARS-CoV-2 infection with symptoms including fever, shock and multiorgan failure. Clinical features of MIS-C overlap with Kawasaki Disease (KD), bacterial, and viral infections, making accurate diagnosis challenging. Host genes, measurable through whole blood transcriptomics, are an alternative tool for diagnosing infectious and inflammatory diseases.
Methods
Patients with MIS-C, KD, bacterial, and viral infections were recruited to the EU-funded PERFORM and DIAMONDS studies and the NIH-funded PREVAIL study. Patients were phenotyped using a standardised algorithm. Genome wide RNA sequencing of whole blood was undertaken, and feature selection was performed to identify a diagnostic signature for distinguishing between MIS-C and other infectious and inflammatory conditions. The expression levels of the genes identified were measured using RT-qPCR assays in an independent validation cohort.
Results:
Through feature selection and differential expression analysis, 11 genes with diagnostic potential were identified and taken forward into cross-platform validation using RT-qPCR. With up to 11 genes, it was possible to distinguish between MIS-C vs. KD, bacterial, and viral infections with high accuracy, with an AUC of 92.9% (95% CI: 88.2%-97.6%) in the validation cohort. The diagnostic gene signature retained its high performance when tested within the groups separately in the validation cohort: MIS-C vs. bacterial infections (AUC: 94.6%), vs. viral infections (AUC: 93.1%), and vs. KD (AUC: 89.8%).
Conclusions/Learning Points:
Despite the clinical similarities between MIS-C and other infectious and inflammatory conditions, there are key differences in gene expression profiles that can be used in diagnostic contexts. It will be necessary for the genes reported here to undergo further validation prior to their development into tests with clinical utility.
VALIDATION OF TRANSCRIPTOMIC SIGNATURES FOR FEBRILE CHILDREN USING NANOSTRING TECHNOLOGY AND EXPLORATION OF MULTI-CLASS PREDICTION MODELS
Abstract
Backgrounds:
Many host transcript signatures for paediatric inflammatory and infectious diseases are in development, but require validation in independent cohorts; their translation to clinically useful test platforms lags behind discovery. We used NanoString technology to efficiently validate multiple signatures in parallel and explore the potential for more sophisticated multi-class classification models.
Methods
We validated five transcriptomic diagnostic signatures using prospectively recruited patients from multiple paediatric cohorts. Final phenotypes were assigned using pre-agreed definitions after review of clinical and laboratory data. We quantified 69 transcripts on a custom NanoString nCounter cartridge, normalising expression values using reference genes. Signature performance was assessed using Area Under ROC Curve (AUC) statistics. We explored two approaches to multiclassification diagnostics to develop proof-of-concept methods: a mixed test combining four independent one-vs-all models, and a multinomial model.
Results:
Our cohort of 92 paediatric patients included 23 definite bacterial and 20 definite viral infections, 15 Kawasaki disease, 18 with tuberculosis and 16 healthy controls. The signatures achieved AUCs above 0.82 (Table 1), with confidence intervals overlapping those of the respective discovery studies. However, performance declined in all signatures when tasked with differentiating additional groups. For example, the single-transcript BATF2 had AUC of 0.910 differentiating TB from healthy individuals, reducing to 0.745 when differentiating TB from other febrile diseases. In comparison, the multinomial approach identified a 24-transcript model that correctly classified all 76 non-control patients (0% in-sample error), outperforming the mixed-model (19 transcripts, 19.8% in-sample error).
Conclusions/Learning Points:
The cross-platform, out-of-sample findings validated 5 signatures, but discriminatory power was reduced in patients drawn from outside their remit. An exploratory 24-transcript model had improved accuracy across all diagnostic groups, demonstrating in principle the utility for one-step multi-class diagnosis in patients with broad diagnostic uncertainty.
Poster Author Of 1 e-Poster
PD137 - BRIDGING THE DIAGNOSTIC KAWASAKI DISEASE GENE EXPRESSION PROFILING CLASSIFIER FROM MICROARRAY TO A CLINICALLY APPLICABLE MULTIPLEX QRT-PCR ASSAY (KIDS-GEP) (ID 626)
- Rowan Kuiper (Netherlands)
- Victoria J. Wright (United Kingdom)
- Chisato Shimizu (United States of America)
- Daphne Huigh (United States of America)
- Myrsini Kaforou (United Kingdom)
- Adriana H. Tremoulet (United States of America)
- Daniëlle Van Keulen (Netherlands)
- Jethro Herberg
- Clive J. Hoggart (United Kingdom)
- Dominic Habgood-Coote (United Kingdom)
- Jesus Rodriguez-Manzano (United Kingdom)
- Dennie Tempel (Netherlands)
- Jane C. Burns (United States of America)
- Michael Levin (United Kingdom)