Validation of an Interferon Type I Score Test on the Nanostring nCounter Platform

Type I interferons have wide-ranging impact on the immune system in response to viruses, bacteria, and parasites. However, excessive and uncontrolled levels could lead to various autoimmunity and autoinflammatory conditions, a subset of which have been linked to monogenetic mutations in a growing number of genes and termed type I interferonopathies (monogenic autoinflammatory diseases). The gold standard for diagnosis remains genetically determining a disease-causing mutation, but the IFN signature has emerged as a useful and relatively rapid diagnostic screening tool. Traditionally performed by RT-PCR, more recent RNA-hybridization technologies promise to significantly improve patient diagnosis. The Nanostring nCounter platform was selected as it can be multiplexed and directly quantifies the number of RNA transcripts without the need for prior reverse transcription and amplification, avoiding bias. Mirroring previous NIH studies, 29 genes associated with the IFN I pathway were picked. For clinical accuracy, we reanalyzed samples previously tested on a similar platform at NIH and recruited known affected patients locally that were freshly collected into PAXgene tubes. On the other hand, target oligos were synthesized for the entire panel and pooled for analytical validation. Scores obtained from 16 patients and 9 controls through NIH highly correlated between centers. Reference ranges were estimated from 82 individuals aged 1-69 years old. We found scores do not trend with either age or sex, thus not requiring brackets. An ROC comparing heathy controls against 31 known patients pooled from both NIH and local sources returned a 99% accuracy for type I interferonopathies (Figure 1). Using the pooled oligos, precision studies were preformed, including repeatability, reproducibility, interlot, instrument linearity, and inter-operator. Additionally, whole blood samples were used to test for sample stability and processed sample stability, all giving results within acceptable bias or variation. Figure 1. Type I interferon scores of patients with known elevated or normal levels performed at Cincinnati Children’s. Positive samples from NIH and Cincinnati were pooled (n = 31) and compared against all controls (n = 98), as shown as blue open circles in an ROC curve (right plot). Red horizontal lines represent median and interquartile range. Dotted black horizontal line represents the 97.5th percentile of the reference cohort. We report data on both analytical and clinical validation of an interferon score test using the Nanostring nCounter platform performed at the Cincinnati Children’s Hospital Diagnostic Immunology Laboratory. This shall soon be the first clinically available test in North America for the diagnosis and treatment monitoring of various interferon-mediated diseases.