Tissue-based genomic profiling of 300,000 tumors highlights the detection of variants with low allele fraction

Saumya D. Sisoudiya, Hanna Tukachinsky, Rachel B. Keller-Evans, Alexa B. Schrock, Richard S.P. Huang, Ole Gjoerup, Michael J. Pishvaian, Rachna Shroff, Ethan S. Sokol, Lucas Dennis, Garrett M. Frampton, Smruthy Sivakumar

Research output: Contribution to journalArticlepeer-review

Abstract

Tumor tissues obtained in the clinical setting typically have low purity and display treatment-associated genetic heterogeneity, contributing to variants at low variant allele fractions (VAF). We present a pan-cancer landscape and the therapeutic impact of somatic variants detected at low VAF (≤10%) from tumor tissues in 331,503 patients, across 78 tumor types, that received an FDA-approved comprehensive genomic profiling (CGP) test targeting ~324 genes during routine clinical care. 29% of patients had at least one variant detected at VAF ≤10% and 16% at VAF ≤5%. Among the frequently diagnosed tumors, several cases showed low VAF variants: pancreatic (37%), non-small cell lung cancer (35%), colorectal (29%) and prostate (24%). Treatment resistance-associated alterations had lower median VAF than driver alterations, although variants with VAF ≤5% comprised both driver and resistance alterations. This highlights the importance of CGP in detecting low VAF variants, to better inform clinical actionability and guide personalized treatment for patients with cancer.

Original languageEnglish (US)
Article number190
Journalnpj Precision Oncology
Volume9
Issue number1
DOIs
StatePublished - Dec 2025

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Fingerprint

Dive into the research topics of 'Tissue-based genomic profiling of 300,000 tumors highlights the detection of variants with low allele fraction'. Together they form a unique fingerprint.

Cite this