Prior anticancer treatment influences the genomic landscapes of patients with advanced cancer, as indicated by an analysis of a large pan-cancer, whole-genome, transcriptome dataset, according to a study published in Nature Cancer.
“Our data offer a rich resource for investigation of advanced cancers and interpretation of whole-genome and transcriptome sequencing in the context of a cancer clinic,” the authors wrote.
In the clinical setting, most sequencing approaches use panels. In this study, a whole-genome, transcriptome approach was used to analyze advanced and metastatic cancers to evaluate the interaction between prior drug treatment and the genomic landscape of different cancer types.
The study included samples from 570 patients with advanced and metastatic cancer representing 25 different histologies who were treated at a tertiary cancer center. Most tumor samples were taken from metastatic sites (77%), with the remaining from local recurrences or refractory disease. Some rare cancers were included, such as eccrine porocarcinoma, carcinoma ex pleomorphic adenoma, and ghost cell odontogenic carcinoma.
Sequencing was performed as part of the Personalized OncoGenomics Program at BC Cancer, and this population was termed the POG570 cohort.
Systemic therapy had been previously given to 82% of the cohort, with 72% of patients receiving more than 1 drug. There were a total of 110 different drugs used, and the duration of drug treatment was 4 days to over 4 years.
There were 7,441,311 somatic substitution and 701,166 small insertions or deletions identified. Viral or microbial sequences were detected in 4% of cases. There were 58,638 structural variants comprised of deletions (28%), inversions (24%), duplications (22%), and interchromosomal events (26%).
TP53, NF1, RB1, KRAS, CDKN2A/B, and MYC were the most commonly altered oncogenes and tumor suppressor genes, which was consistent with findings from The Cancer Genome Atlas. Among specific cancer types, the most commonly altered genes were KEAP1 in lung; SF3B1, GATA3, and SOX10 in breast; and MAPK8/JNK1 and NCOA4 across cancer types.
There were 357 patients with a somatic mutation in DNA-repair and damage-response pathways, with the most common including TP53 (36%), ATM (2.6%), and BRCA2 (2.3%).
Small coding mutations and copy-number variants were also detected and were more frequent among patients who were previously treated compared with those who had not received prior anticancer treatment. There were 13 clustered or truncating mutations of the 35 important small coding mutations, which were associated with drug-mutation associations for ESR1, EGFR, SMAD4, TP53, ARID1A, OR5H2, ANKRD12, and TCF7L2. Moreover, 3 of these associations were known to confer resistance to aromatase inhibitors in breast cancer, EGFR inhibitors in lung cancer, and EGFR-sensitizing mutations.
Among patients who previously received treatment with an immune checkpoint inhibitor, longer treatment duration was associated with high exonic tumor mutational burden and high T-cell signatures, and T-cell receptor diversity, regardless of tumor type.
“In addition to contributing fundamental research insights,” the authors concluded, “the availability of our rich dataset serves as a foundation for understanding the genomic landscape and treatment impacts in advanced tumors and brings whole-genome and transcriptome sequencing closer to the cancer clinic.”
Pleasance E, Titmuss E, Williamson L, et al. Pan-cancer analysis of advanced patient tumors reveals interactions between therapy and genomic landscapes. Nat Cancer. 2020;1:452-468.
This article originally appeared on Cancer Therapy Advisor