Analytical evaluation of TriVerity, a rapid diagnostic and prognostic host gene expression test performed on the Myrna instrument using RT-LAMP.

We evaluated the analytical performance of the TriVerity test system, a benchtop system for rapid measurement and interpretation of host messenger RNAs (mRNAs) measured quantitatively from PAXgene Blood RNA specimens using the TriVerity cartridge on the Myrna instrument. In approximately 30 minutes, 3 scores are generated from 29 host mRNAs for the likelihood of a bacterial infection, a viral infection, and illness severity (7-day need for ICU-level care), each falling into 1 of 5 interpretation bands (very low, low, moderate, high, and very high). Reproducibility, precision, limit of detection, linearity, interference, and sample and cartridge stability were determined per Clinical Laboratory Standards Institute standards. An operator survey assessed TriVerity's ease of use. Reproducibility demonstrated SDs meeting the acceptance criteria of <5.5 score units. The lowest concentration at which 100% of replicates showed measurable amplification was 1 × 10⁶ cp/mL of in vitro RNA transcripts; the limit of quantification and detection were equivalent to blood containing 500 leukocytes/µL. Interference was not observed for 17 potential interferents. The cartridge was stable at room temperature (RT) for 9 and 12 months using accelerated testing at 37°C. TriVerity results were equivalent using fresh and frozen PAXgene Blood RNA samples. All operators agreed or strongly agreed that the TriVerity test system was easy to use. In summary, the analytical performance presented here, along with diagnostic accuracy established in the SEPSIS-SHIELD trial, demonstrates that the TriVerity system is reliable and user friendly, assisting clinicians in managing patients with suspected acute infections and sepsis in acute care settings.

Importance: The prompt diagnosis of acute infections and sepsis is critical for better patient outcomes. We introduce a groundbreaking messenger RNA-based test, the first of its kind, designed to diagnose the presence of infection and predict illness severity in adult patients with suspected acute infections or sepsis. Several key findings regarding the accuracy and robustness of the test system are presented, which will be relevant to laboratories or acute care settings implementing the test for patient care. Furthermore, these findings may assist clinical researchers in developing analytical trial protocols aimed at the combined evaluation of RNA-based multi-marker tests with both diagnostic and prognostic test characteristics.