A novel ORF-1a based SARS-CoV-2 RT-PCR assay to resolve inconclusive samples.

ABSTRACT Background India bears the second largest burden of SARS-CoV-2 infection. A multitude of RT-PCR detection assays with disparate gene targets including automated high throughput platforms are available. Varying concordance and interpretation of diagnostic results in this setting can result in significant reporting delays leading to suboptimal disease management. Here, we report the development of a novel ORF-1a based SARS-CoV-2 RT-PCR assay, Viroselect, showing high concordance with conventional assays and the ability to resolve inconclusive results generated during the peak of the epidemic in Mumbai, India. Methods We identified a unique target region within SARS-CoV-2 ORF1a, non-structural protein (nsp3), that was used to design and develop our assay. This hypervariable region (1923-3956) between SARS-CoV-2, SARS-CoV, and MERS-CoV was utilized to design our primers and probe for RT-PCR assay. We further evaluated concordance of our assay with commonly used EUA (USFDA) manual kits as well as an automated high throughput testing platform. Further, a retrospective analysis using Viroselect on samples reported as ‘inconclusive’ during April-October 2020 was carried out. Results A total of 701 samples were tested. Concordance analysis of 477 samples demonstrated high overall agreement of Viroselect assay with both manual (87.6%; 95% CI) as well as automated (84.7%; 95% CI) testing assays. Also, in the retrospective analysis of 224 additional samples reported as ‘inconclusive’, Viroselect was able to resolve 100% (19/19) and 93.7% (192/205) samples which were termed inconclusive by manual and automated high throughput platform respectively. Conclusion We show that Viroselect had high concordance with conventional assays, both manual and automated, as well as highlight its potential in resolving inconclusive samples.