Reference Samples to Compare Next-Generation Sequencing Test Performance for Oncology Therapeutics and Diagnostics.

Area of Science:

• Molecular diagnostics
• Genomics
• Oncology

Background:

• Laboratory-developed tests (LDTs) using next-generation sequencing (NGS) are crucial for selecting targeted therapies.
• Concerns exist regarding the accuracy and standardization of these NGS LDTs across clinical laboratories.

Purpose of the Study:

• To assess the performance of NGS LDTs in clinical laboratories.
• To evaluate the accuracy of variant detection for targeted therapy selection.
• To introduce traceable reference samples for NGS LDT validation.

Main Methods:

• Engineered human cell lines (wet samples) and in silico mutagenized sequence files (dry samples) were used as reference materials.
• CRISPR/Cas9 technology was employed for cell line engineering.
• Single and multinucleotide variants in KRAS and NRAS genes were tested at varying allele fractions.

Main Results:

• Twenty-one laboratories participated, testing both wet and dry samples.
• Only 37% of laboratories correctly reported all variants.
• The majority of errors were false negatives, indicating underdetection of genetic variants.

Conclusions:

• Genetically engineered cell lines and in silico data serve as valuable complementary reference samples for NGS LDT evaluation.
• Variable accuracy in variant detection among NGS LDTs can impact patient stratification for targeted therapies.
• Standardized performance evaluation is essential for reliable NGS LDTs in clinical practice.