Abstract 4021
Background
Whole-exome sequencing of plasma cell-free DNA (WES-cfDNA) is an emerging liquid biopsy approach for identification of somatic mutations and yields high-concordance when compared to WES of the tumours of patients with advanced disease. Here we show that sequential WES-cfDNA provides insight regarding the evolution of metastatic breast cancer and that global mutation profiles relate to clinical response and progression to different lines of treatment.
Methods
A patient with inherited BRCA2 breast cancer was recruited at disease relapse and blood samples were then taken over a 2 year period on treatment until death. Whole-exome sequencing was performed on 12 DNA samples: primary tumor, 2 biopsies at relapse and 8 sequential cfDNA samples. Results were validated by targeted amplicon sequencing and ddPCR. Somatic Single Nucleotide Variants (SNVs) were called with Mutect2. Mutational signatures (MS) and their contributions to each sample were detected using the R package deconstructSigs.
Results
WES profiling showed that MS-3 (associated with germline BRCA2 status) was the predominant signature in the primary tumour, whereas the signature of the two relapse biopsies were MS-3, MS-4 and MS-5, and MS-3, MS-4 and MS-6. The first blood sample taken at relapse with bone disease also had a mixed MS profile. The MS-3 signature was detected in cfDNA with progression to lung and initially resolved in response to carboplatin treatment. MS-3 also predominated in cfDNA on subsequent disease progression and was the dominant MS at the time of death. SNV analysis identified a Tier-1 PIK3CA mutation in the primary and relapse tissue, whereas three other Tier-2 mutations in DNM2, DCC and ATP1A1 were specific to relapse, indicative of cancer evolution. All four variants were detected at high levels by WES-cfDNA. Lastly, an ESR1 emergent mutation was identified by targeted sequencing in cfDNA and predicted failure of letrazole therapy.
Conclusions
Our study demonstrates the capacity of sequential WES-cfDNA to capture the genomic profiles of advanced cancer. The MS in cfDNA can be used track tumour evolution, and give insight into the characteristic of the dominant and lethal clone.
Clinical trial identification
Legal entity responsible for the study
University of Leicester.
Funding
Hope Against Cancer, Cancer Research UK, Leicester ECMC.
Editorial Acknowledgement
Disclosure
A. Thomas: Honoraria: Amgen, Servier, BMS. S. Ahmed: Consulting: Novartis, Eisai, AstraZeneca, MSD, BMS. J.A. Shaw: Scientific advisory board: Qiagen; Board member: International Journal of Experimental Pathology. All other authors have declared no conflicts of interest.