Improving RNA Fusion Call Confidence and Reliability in Molecular Diagnostic Testing
Mariusz Shrestha1, Sasha Blay2, Sydney Liang3
1Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada.
The Journal of Molecular Diagnostics : JMD
|March 23, 2023
Summary
Next-generation sequencing (NGS) reliably detects RNA fusions, but poor sample quality can cause false negatives. A novel proxy quality control (pQC) metric using 15 genes improves RNA quality assessment and fusion call confidence in clinical settings.
Area of Science:
- Molecular Biology
- Genomics
- Cancer Research
Background:
- Next-generation sequencing (NGS) offers superior RNA fusion detection compared to FISH and RT-PCR.
- Poor RNA quality in formalin-fixed, paraffin-embedded (FFPE) tissues can compromise NGS accuracy, leading to false negatives.
- Current quality control metrics may not adequately reflect sample integrity for reliable fusion calling.
Purpose of the Study:
- To develop a robust quality control (QC) metric for NGS to improve confidence in RNA fusion detection.
- To establish a proxy quality control (pQC) gene set that reflects internal sample quality.
- To minimize false-negative calls in clinical NGS assays.
Main Methods:
- Evaluated gene expression across 361 patient tumor samples to identify 15 robust genes for a pQC metric.
- Assessed normalized expression of the 15 pQC genes using NGS data.
- Tested a revised library preparation method to improve the pass rate of low-quality samples.
Main Results:
- A pQC metric using 11 out of 15 genes demonstrated a 4.71% fail rate, deemed acceptable for clinical stringency.
- A revised library preparation method, increasing cDNA input, enabled 75% of previously failed samples to pass pQC.
- The pQC metric serves as a surrogate for housekeeping genes, enhancing fusion call reliability.
Conclusions:
- The developed pQC metric provides a reliable assessment of RNA quality for NGS.
- Implementing the pQC tool enhances confidence in RNA fusion detection, reducing false negatives.
- Optimized library preparation protocols can improve the success rate of analyzing low-quality FFPE samples with NGS.
