Abstract 2936: Nanopore adaptive sampling detects nucleotide variants and improves large scale rearrangement characterization for diagnosis of cancer predisposition
Sandy Chevrier,Corentin Richard,Marie Mille,Denis Bertrand,Michael Blum,Nicolas Philippe,Romain Boidot
DOI: https://doi.org/10.1158/1538-7445.am2024-2936
IF: 11.2
2024-03-22
Cancer Research
Abstract:Abstract Molecular diagnosis has become highly significant for patient management. Facing new challenges to generate more comprehensive data in reduced time and cost, we tested Nanopore (ONT) adaptive sampling to analyze a gene panel for germline predisposition to cancer. We tested 30 clinical germline samples carrying known alterations on BRCA1, BRCA2, PALB2 and MLH1 genes to detect large scale rearrangements (LSR) and single nucleotide variants (SNV). We used adaptive sampling to enrich the full sequence (from promoters to 3’UTR) of 152 cancer predisposition genes on ONT R10.4.1 MinION flowcells and the Q20 chemistry. The analysis was performed using the long-read bioinformatics pipeline developed by SeqOne Genomics. We observed a mean enrichment of 7.85 (3.91-9.85) and a mean coverage on target genes of 14.73X (4.71-35). In our cohort, 11 samples had LSR that were all detected with ONT sequencing. In addition to perfectly detect the locus of the LSR (starts and stops always in intronic regions), we discovered that a known MLPA amplification of exon 13 in the BRCA1 gene was in fact an amplification of both exons 12 and 13. This 2-exon duplication was also confirmed on a patient’s relative, originally diagnosed with an exon 13 duplication as well. Another LSR was a total deletion of the BRCA1 gene. ONT sequencing highlighted this complete deletion of BRCA1 was the consequence of a large deletion of almost 140 kbp carrying 5 different genes (BRCA1, IFI35, NBR2, RND2, and VAT1). ONT sequencing was also able to detect all class-5 SNVs (point mutations and small indel) present in 16 patient samples at low coverage (between 6 and 27X). We faced bioinformatics detection issues for 3 mutations located in homopolymer regions using standard pipelines. These were all confirmed correctly when visualizing the raw data. As we analyzed complete genes and more genes than with short read sequencing, we detected novel unknown variants. To check the reliability of those novel variant by ONT, we wondered whether we could confirm results obtained at a low coverage (10X) with average quality (Phread score >/= 15). For that we randomly selected 8 new variants we analyzed by Sanger sequencing, and all of these were confirmed, suggesting that variants detected with ONT, even at a 10X coverage and a Phread score higher than 15, could be considered as real variants. In conclusion, we showed that ONT adaptive sampling sequencing is suitable for the analysis of germline alterations. It improves the characterization of LSR and is able to detect SNV even at low coverage. Nevertheless, an improvement of bioinformatics detection of variants located in homopolymer regions should be considered. A larger comparison study between ONT adaptive sampling and short read sequencings should be performed on hundreds of samples to adapt international guidelines to third generation sequencing for germline diagnosis. Citation Format: Sandy Chevrier, Corentin Richard, Marie Mille, Denis Bertrand, Michael Blum, Nicolas Philippe, Romain Boidot. Nanopore adaptive sampling detects nucleotide variants and improves large scale rearrangement characterization for diagnosis of cancer predisposition [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2936.
oncology