Validation of a Comprehensive Genomic Profiling Next-Generation Sequencing Assay for Diagnostic Use
Kylätie, Riina (2023)
Kylätie, Riina
2023
Bioteknologian ja biolääketieteen tekniikan maisteriohjelma - Master's Programme in Biotechnology and Biomedical Engineering
Lääketieteen ja terveysteknologian tiedekunta - Faculty of Medicine and Health Technology
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-05-10
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202304274794
https://urn.fi/URN:NBN:fi:tuni-202304274794
Tiivistelmä
Cancer is caused by myriad of different kinds of damaging mutations in genes regulating cell growth and division. Traditionally, these genetic changes have been diagnosed with multiple targeted small-scale test methods, which identify only to a limited number of specific genetic mutations. In recent years, the development of comprehensive genomic profiling (CGP) methods has enabled the evolution of targeted and personalized cancer therapies by providing fast determination of numerous cancer biomarkers in one assay, thus saving time, money and sample material.
The aim of this work was to validate a commercial pan-cancer CGP assay TruSight Oncology 500 (TSO500) by Illumina for diagnostic use in clinical laboratory to reduce the need for multiple diagnostic tests and to enable more personalized treatment for cancer patients. A total of 68 individual tumor samples were analysed in this study, including one DNA control and one RNA control sample. The patient specimens were provided by Fimlab Pathology unit, and they were selected to contribute as a representative sample of different cancer and variant types with differing DNA and RNA input amounts, variant allele frequencies, tumor cell percentages and sample qualities. Tissue specimens consisted mostly of FFPE samples, in addition a few freshly frozen tissue samples, bone marrow samples, RNAlater fixed samples and one EDTA-coagulated whole blood were included as well. From this set of samples, total of 144 variants from 54 gene with previous results from reference methods were studied from the patient samples, 16 gene fusions and two splice site variants from RNA control, and 12 genes with 16 different variant types from DNA control.
The results were compared to the results from the reference methods, as well as the validation results from the manufacturer. Reference methods used were immunohistochemistry staining, fluorescence in situ hybridisation, and smaller NGS cancer panels already in clinical use. Total of 86 out of 89 small DNA variants (single nucleotide variants and small insertions and deletions) were detected (96.6 %) with TSO500 assay. From 21 analysed copy number variants (CNVs), TSO500 detected 16 (76.1 %). Six out of fourteen detected CNVs from FFPE samples had a fold change under the detection limit given by the manufacturer (2.2x). All five undetected CNVs had been classified as “possible but not unambiguous” by the reference method. The results suggest that the method can reliably detect small DNA variants when the variant allele frequency is > 5 %, CNVs with fold change of ≥ 2.2x, and small variants and CNVs when the DNA input amount is 100 ng and tumor cell percentage (TCP) > 20 %. Gene fusions were detected from samples with TCP of ≥ 20% and RNA input amount of 85 ng, when the RNA quality score DV200 was ≥ 40 %.
The results from TSO500 assay proved to be comparable with the reference methods when sample quality, input amount and tumor cell percentage requirements were met. According to the results, the analytical sensitivity of the assay meets the threshold given by the manufacturer as well. In conclusion, in accordance with the results of this validation study, the TSO500 assay meets the requirements for a clinical assay and can be implemented to diagnostic use.
The aim of this work was to validate a commercial pan-cancer CGP assay TruSight Oncology 500 (TSO500) by Illumina for diagnostic use in clinical laboratory to reduce the need for multiple diagnostic tests and to enable more personalized treatment for cancer patients. A total of 68 individual tumor samples were analysed in this study, including one DNA control and one RNA control sample. The patient specimens were provided by Fimlab Pathology unit, and they were selected to contribute as a representative sample of different cancer and variant types with differing DNA and RNA input amounts, variant allele frequencies, tumor cell percentages and sample qualities. Tissue specimens consisted mostly of FFPE samples, in addition a few freshly frozen tissue samples, bone marrow samples, RNAlater fixed samples and one EDTA-coagulated whole blood were included as well. From this set of samples, total of 144 variants from 54 gene with previous results from reference methods were studied from the patient samples, 16 gene fusions and two splice site variants from RNA control, and 12 genes with 16 different variant types from DNA control.
The results were compared to the results from the reference methods, as well as the validation results from the manufacturer. Reference methods used were immunohistochemistry staining, fluorescence in situ hybridisation, and smaller NGS cancer panels already in clinical use. Total of 86 out of 89 small DNA variants (single nucleotide variants and small insertions and deletions) were detected (96.6 %) with TSO500 assay. From 21 analysed copy number variants (CNVs), TSO500 detected 16 (76.1 %). Six out of fourteen detected CNVs from FFPE samples had a fold change under the detection limit given by the manufacturer (2.2x). All five undetected CNVs had been classified as “possible but not unambiguous” by the reference method. The results suggest that the method can reliably detect small DNA variants when the variant allele frequency is > 5 %, CNVs with fold change of ≥ 2.2x, and small variants and CNVs when the DNA input amount is 100 ng and tumor cell percentage (TCP) > 20 %. Gene fusions were detected from samples with TCP of ≥ 20% and RNA input amount of 85 ng, when the RNA quality score DV200 was ≥ 40 %.
The results from TSO500 assay proved to be comparable with the reference methods when sample quality, input amount and tumor cell percentage requirements were met. According to the results, the analytical sensitivity of the assay meets the threshold given by the manufacturer as well. In conclusion, in accordance with the results of this validation study, the TSO500 assay meets the requirements for a clinical assay and can be implemented to diagnostic use.