Long-Term Storage of Formalin-Fixed Paraffin-Embedded Tissues Negatively Impacts Next-Generation Sequencing: Successful Restoration with a DNA Repair Enzyme

BACKGROUND: High-quality formalin-fixed paraffin-embedded tissue (FFPET) is needed for next-generation sequencing (NGS) to identify clinically actionable tumor mutations. Long-term storage of FFPET may increase sequencing artifacts confounding interpretation. This study assesses the effects of long-term storage on NGS quality control (QC) metrics, variant calling, and tumor mutation burden (TMB), and tests the hypothesis that a DNA repair enzyme may aid in the analysis of stored FFPET. METHODS: Sequencing was performed on 281 FFPET (0-17 years old) using a large hybrid-capture NGS panel, and QC metrics were evaluated. Paired sequencing was performed, with and without DNA enzyme repair, on an additional 10 FFPET at time 0 (T0) and after storage for 2 years or 4 years. Sequencing quality metrics, variant calling, and TMB scores were compared. RESULTS: Multiple QC metrics, including sequencing read depth, declined with prolonged storage, particularly for samples stored ≥5 years. Even 2-4 years of storage introduced sequencing artifacts totaling 191 false-positive variants, as compared with 154 true-positive variants. Artifact variant types included deamination single-nucleotide variants (SNVs), nondeamination SNVs, and insertion/deletion variants, all of which were detected above 5% variant allele frequency. Four samples were TMB-low at baseline, but TMB was artifactually elevated in the stored sample. The storage-associated artifacts, including TMB, were largely restored by pretreatment with a DNA repair enzyme. CONCLUSIONS: Long-term storage of FFPET resulted in reduced DNA quality and sequencing quality metrics, as well as many artifactual variants. FFPET stored for longer than 2 years may benefit from pretreatment with a DNA repair enzyme.