73. Optimization and clinical validation of the Ion Torrent Oncomine comprehensive assay plus

Aly Abdelkareem

Jennifer Crow

Dr. Jennifer Crow received her PhD in Pathology from the University of Kansas School of Medicine in 2016. Her post-doctoral research centered on elucidating the nucleic acid content of small extracellular vesicles in pediatric sarcomas, including Ewing’s Sarcoma through next generation sequencing. Her work in this field was supported by grants from Hyundai Hope on Wheels, Braden’s Hope for Childhood Cancer, and Noah’s Bandage Project. In addition, she was the recipient of the NIH Loan Repayment Program for Pediatric Research. In the spring of 2021 Dr. Crow transitioned from academia to the Clinical Molecular Oncology Laboratory of the University of Kansas Health System where she currently oversees new assay development and validation. 


Jennifer Crowa, Stephen Hytera, Sarah Schmitta, Andrew Godwinb

aThe University of Kansas Health System, Kansas City, KS, United States; bThe University of Kansas Medical Center, Kansas City, KS, United States

The Oncomine Comprehensive Assay Plus v3, (OCAPlus, ThermoFisher) provides an in-house comprehensive genomic profiling solution appropriate for formalin-fixed, paraffin-embedded (FFPE) tissues. The assay uses matched DNA and RNA (cDNA) to detect single nucleotide variants (SNVs) and small deletions and insertions (DELINS) in over 500 genes, copy number variants (CNVs) in over 300 genes, gene fusions (SVs) in 49 genes, tumor mutational burden (TMB), and microsatellite stability (MSI). Validation consisted of 100 clinical samples representing 8 different cancer types, 6 synthetic controls, and 2 cell lines. Library construction and templating was automated on Ion Chef instruments (ThermoFisher) using Ampliseq DL8 and OCAPlus chemistries. Sequencing was completed on the Gene Studio S5 (ThermoFisher). Signal processing and alignment was conducted with Torrent Suite 5.18.1 and data analysis was completed on a modified OCAPlus protocol using Ion Reporter. Preliminary analysis revealed several weaknesses in the OCAplus pipeline. Modifications were made to the OCAPlus protocol to adjust MSI calling parameters, CNV deletion thresholds, and tumor cellularity calling. The hotspot BED file was amended to increase sensitivity for TERT c.-124C>T and c.-146C>T variant calling. Lastly, we created an additional filter to capture 98% of variants of uncertain significance that were originally filtered out during preliminary analysis. Performance analysis under these modified parameters resulted in > 90% PPA and PPV for SNV, SV, CNV, MSI, and TMB calling and >85% PPA and PPV for DELINS. Overall, with these modifications and additional filter, the OCAPlus is a robust platform for in-house comprehensive genetic profiling.