Analysis of quality metrics in comprehensive cancer genomic profiling using a dual DNA-RNA panel.

ABSTRACT

Background: The nucleic acid quality from formalin-fixed paraffin-embedded (FFPE) tumor vary among samples, resulting in substantial variability in the quality of comprehensive cancer genomic profiling tests. The objective of the study is to investigate how nucleic acid quality affects sequencing quality. We also examined the variations in nucleic acid quality among different hospitals or cancer types. Methods: Three nucleic acid quality metrics (ddCq, Q-value, and DV200) and five sequencing quality metrics (on-target rate, mean depth, coverage uniformity, target exon coverage, and coverage of the housekeeping gene) were examined using 585 samples from the Todai OncoPanel, a dual DNA-RNA panel. Results: In the DNA panel, ddCq served as an indicator of sequencing depth and Q-value reflected the uniformity of sequencing across different regions. It was essential to have favorable values not only for ddCq but also for Q-value to obtain ideal sequencing results. For the RNA panel, DV200 proved to be a valuable metric for assessing the coverage of the housekeeping genes. Significant inter-hospital differences were observed for DNA quality (ddCq and Q-value), but not for RNA quality (DV200). Differences were also observed among cancer types, with Q-value being the lowest in lung and the highest in cervix, while DV200 was the highest in lung and the lowest in bowel. Conclusions: We demonstrated distinct characteristics and high predictive performances of ddCq, Q-value, and DV200. Variations were observed in the nucleic acid quality across hospitals and cancer types. Further study is warranted on preanalytical factors in comprehensive cancer genomic profiling tests.