Detection and functional assessment of extrachromosomal DNA amplifications in FFPE lung tumor specimens using Hi-C sequencing
Reference:
Sikkink, et al.
AACR 2026 - San Diego, CA
Abstract:
INTRO:
Accurate detection of extrachromosomal DNA (ecDNA) has become increasingly important in oncology diagnostics, as ecDNA-mediated oncogene amplification drives aggressive tumor behavior, therapeutic resistance, and is associated with poor clinical outcomes. As the significance of ecDNA in tumor biology becomes clearer, sensitive methods for identifying and characterizing these structures are essential. Hi-C sequencing has emerged as a powerful approach for ecDNA detection, with our group first demonstrating accurate ecDNA identification from FFPE tumor specimens while others have developed Hi-C based tools for ecDNA detection, sequence reconstruction, and regulatory landscape analysis. Here, we leverage Hi-C to characterize the ecDNA landscape in advanced non-small cell lung cancer (NSCLC) tumors lacking other identifiable drivers by prior CGP analysis.
METHODS:
FFPE samples from NSCLC tumors (n=97) were retrospectively selected from patients with Stage III/IV disease whose prior CGP showed no EGFR/RAS mutations or gene fusions (“driver-negative”). Hi-C sequencing was performed by Arima Genomics, and rearrangements, gene fusions, and CNVs (including ecDNAs and other amplifications) were identified with Arima-SV pipeline. Functional characterization of oncogenes on ecDNAs was assessed by IHC. FISH studies are ongoing to corroborate Hi-C based predictions of ecDNAs and other chromosomal amplifications (e.g. homogeneously staining regions (HSRs)).
RESULTS:
95/97 (98%) Hi-C libraries passed QC and were deeply sequenced and analyzed. 33/95 (35%) tumors had amplifications containing at least one oncogene. 13/95 (14%) tumors carried oncogene amplifications predicted to be extra-chromosomal, including those involving MYCL (1), NFIB (1), PDGFD (1), CCND1 (1), CCND3(1), CCNE1 (1), EGFR (2), ERBB2 (1), KRAS (1), FGFR1 (1), and MYC (2). The remaining 20/95 (21%) were predicted to be chromosomally integrated amplifications including CCND1 (1), FGFR1 (12), KIT (2), MDM2 (1), MET (1), MYC (1), NTRK1 (1), and NRG1 (1). Twenty-six tumors had sufficient tissue remaining for functional characterization and a corresponding commercially available IHC test for protein expression. Of these, 7/7 (100%) amplifications predicted to be extra-chromosomal were expressed at the protein level, whereas only 8/19 (42%) not predicted to be extra-chromosomal were expressed at the protein level.
CONCLUSIONS:
These data demonstrate that Hi-C can detect ecDNA amplifications from routine FFPE lung tumor specimens and distinguish them from their chromosomally integrated amplification counterparts. This distinction is important because it correlates with amplified oncogene expression patterns and is necessary to inform clinical and translational research programs requiring precise understanding of underlying amplification mechanisms.