Identification of actionable targets using DEPArray-based sorting of pure carcinoma and stromal populations from formalin-fixed paraffin-embedded tissues followed by shallow whole-genome sequencing.
journal contribution
posted on 2025-12-01, 11:45 authored by Georgios Nteliopoulos, Evie Wren, Amelia Rushton, Marc K Wadsley, Daniel Fernandez-Garcia, Floriana Manodoro, Otis Agbaimoni, Ritika Chauhan, Zhao Cheng, Darren P Ennis, Karen PageKaren Page, Rebecca AllsoppRebecca Allsopp, Joel Bautista, Ignazio Puccio, Nik Matthews, Kelly Lt Gleason, Rehman Farah, Laura Kenny, Iain A McNeish, Jacqueline ShawJacqueline Shaw, R Charles CoombesFormalin-fixed paraffin-embedded (FFPE) tissue specimens represent precious resources for clinical genomic profiling studies, especially when coupled with comprehensive medical records. Even though next-generation sequencing (NGS) is an effective tool to detect somatic mutations and somatic copy number alterations (sCNA), the biggest challenges in unlocking clinically translatable genomic information from FFPE tissue are low DNA yields and degraded DNA, affected by variable formalin fixation. Another issue is that the proportion of carcinoma and other noncarcinoma cells is variable and can be confounded by intratumoral heterogeneity. To explore these challenges, we isolated pure carcinoma and stromal cells using the DEPArray™ NxT system, a microchip-based digital sorter that allows isolation of pure, homogeneous subpopulations of cells from FFPE samples. We isolated pure carcinoma and stromal cell populations from 12 FFPE tissues, including tissues from nine primary and metastatic breast cancer and three primary ovarian high-grade serous carcinomas. This was followed by downstream shallow whole-genome sequencing (WGS) for copy number landscape profiling (10 samples) and/or a targeted panel for somatic mutation and sCNA analysis (seven samples), subject to cell availability. Seven out of 10 samples (even some with low tumour content or of old age) produced good-quality genomic data, detecting sCNA in all carcinoma population samples but not in the stromal populations. Mutation analysis was performed successfully in 6/7 samples and somatic mutations were detected in all of them. Our workflow enabled the identification of clinically actionable targets, including PIK3CA, ERBB2, FGFR1/2, CDK6, CCNE1, KRAS amplifications and RB, BRCA1/2 losses in patients that would direct therapy. © 2025 The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.<p></p>
Funding
Cancer Research UK. Grant Number: A23464
History
Author affiliation
University of Leicester College of Life Sciences Medical SciencesVersion
- VoR (Version of Record)
Published in
The Journal of PathologyPublisher
Wileyissn
0022-3417eissn
1096-9896Copyright date
2025Available date
2025-12-01Publisher DOI
Spatial coverage
EnglandLanguage
engPublisher version
Deposited by
Dr Karen PageDeposit date
2025-11-21Usage metrics
Keywords
Licence
Exports
RefWorksRefWorks
BibTeXBibTeX
Ref. managerRef. manager
EndnoteEndnote
DataCiteDataCite
NLMNLM
DCDC


