Open Collections

Germline variant calling in formalin-fixed paraffin-embedded tumours

University of British Columbia

Germline alterations can have clinical implications for both cancer patients and their families. Because the tumour genome may contain both germline and somatic variants, the increasingly common practice of clinical tumour sequencing presents an opportunity to also pre-screen for germline variants. This framework is time- and cost-effective because only patients with potential germline variants are referred to downstream confirmatory testing. However, a key challenge is that tumour specimens are commonly formalin-fixed and paraffin-embedded (FFPE), which induces DNA damage that may interfere with molecular testing. Another challenge is distinguishing between germline and somatic variants in the tumour in order to accurately select candidates for follow-up screening. In order to leverage tumour sequencing for identifying germline variants, these challenges must be addressed. To this end, we retrospectively analyzed clinical amplicon sequencing data from 213 patients with a range of tumours, for whom matched-normal samples were available. We assessed formalin-induced DNA damage by comparing amplicon enrichment and sequencing results of FFPE DNA to the matched-normal DNA isolated from peripheral blood mononuclear cells, a gold standard for germline testing. Although formalin-induced DNA fragmentation and cytosine deamination were detectable, we determined that the discrepancies were minor and could be mitigated by using shorter amplicons and enriching for longer DNA templates. We also found that 98.0% of germline alterations identified in the blood were retained in the tumours, suggesting that FFPE tumour DNA can be a reliable source for germline variant calling. Finally, we applied variant allele frequency (VAF) thresholds to delineate germline and somatic variants in tumour-only analyses. We reported that a VAF cut-off of 15% would correctly identify 99% of germline alterations in FFPE tumours, but erroneously submit 14% of somatic mutations (false positives) for follow-up germline testing. This underscores the high sensitivity and positive predictive value of using VAF to discriminate between germline and somatic variants. Collectively, our results demonstrate that clinical tumour amplicon sequencing could also be used to provide cost-efficient first-line germline testing.

Text

2018-01-05

Attribution-NonCommercial-ShareAlike 4.0 International

http://hdl.handle.net/2429/64281

Experimental Medicine

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-sa/4.0/