Virtual Library

Start Your Search

Patrizia Viola



Author of

  • +

    EP1.09 - Pathology (ID 199)

    • Event: WCLC 2019
    • Type: E-Poster Viewing in the Exhibit Hall
    • Track: Pathology
    • Presentations: 1
    • Now Available
    • Moderators:
    • Coordinates: 9/08/2019, 08:00 - 18:00, Exhibit Hall
    • +

      EP1.09-20 - Somatic Variants Interpretation and Classification in Lung Adenocarcinoma Patients from a Targeted Gene Panel in a Molecular Diagnostic Setting (Now Available) (ID 110)

      08:00 - 18:00  |  Author(s): Patrizia Viola

      • Abstract
      • Slides

      Background

      The key driver mutations of lung adenocarcinoma are well described, and several targeted therapies have been developed. Personalized medicine with the availability of rapid and actionable clinical NGS-based genomic profiling has become a reality. But, as the molecular data continue to expand, we face a challenge to reliably estimate the pathogenicity of the identified variants.

      Thus, the aim of the present study is to interpret and classify somatic variants generated from next-generation sequencing of lung adenocarcinoma patients’ samples in routine molecular diagnostic laboratory.

      Method

      Variant call format files generated from next-generation sequencing of 150 lung adenocarcinoma patients’ samples, using Ion torrent hotspot panel of 50 oncogenes covering 207 amplicons, during a three year period (2014-2017) at Department of Molecular Pathology, Hammersmith Hospital, Imperial College Healthcare NHS Trust, were retreived from the database. The variant classification and interpretation was done by an in-house generated bioinformatics pipeline based on the clinical and biological relevance of variants, levels of evidence available in the literature and in-silico pathogenicity prediction tools score.

      Result

      A total of 125 VCF files were available from the 150 lung adenocarcinoma patients' samples. Variant annotation of the these files for the 50 cancer panel genes resulted in altogether 26,9180 variants. Application of variant filtering parameters to all the variants resulted in 117 unique variants. We classified these 117 variants into 3 categories: variants of strong clinical significance in lung cancer (55, 47%), varaints of potential clinical significance (19, 16%) and variants of unknown significance (43, 36.7%) which were further categorized into possible pathogenic (37) and unlikely pathogenic (6) based on their in-silico predicted pathogenicity scores and their effect on protein function.

      Conclusion

      By using this classification system, we could classify 47% of variants as of strong clinical evidence and 16% variants of potential clinical significance in our cohort of lung adenocarcinoma cases. The VUS were further characterized into possibly pathogenic (31.6%) and unlikely pathogenic (5%), thus providing a structure and value to the interface between the oncogenic variants and non-oncogenic variants.The classification system proposed in the study is simple, robust and can be applied to a range of cancer panels in a clinical laboratory and will help in reducing the turn-around time to analyse and report variants. The variants identified in the study can be used to build an in-house database of VUS and the potential clinical significance of these variants could be assessed by correlating with patients' histopathological and clinical data.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.