A novel high-intensity sequencing approach enabled broad detection of genomic variants in plasma with high rates of concordance with corresponding tumor tissue in patients with metastatic breast, lung, and prostate cancer, according to the results of a study (abstract LBA11516) presented at the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting, held June 2–6.
In 89% of patients in the study, at least one genetic change detected in the tumor was also detected in plasma. Almost three-quarters of genetic changes found in tumor samples were also found in blood samples using this approach called ultra-deep sequencing of plasma cell-free DNA (cfDNA).
According to Pedram Razavi, MD, PhD, a medical oncologist and instructor in medicine at Memorial Sloan Kettering Cancer Center in New York, the cfDNA approach scans a very broad area of the genome (508 genes and more than 2 million base pairs) with high accuracy (each region of the genome is sequenced 60,000 times).
“This combined approach resulted in approximately 100 times more data compared with commonly used methods,” Razavi said.
The researchers collected plasma and tissue samples from 161 patients with metastatic breast, non–small-cell lung cancer, or castration-resistant prostate cancer. Of the 161 eligible patients, 124 were evaluable, and genetic changes in the tumors were compared to those in circulating tumor DNA from plasma samples.
Tissue was sequenced using the MSK-IMPACT assay blinded to plasma. The cfDNA extracted from the plasma and, separately, the genome of white blood cells were then sequenced using the high-intensity, 508-gene sequencing assay.
At least one genetic change was detected in 97% of breast cancer samples, 85% of lung cancer samples, and 84% of prostate cancer samples. Overall, including all genomic variations present in most if not all tumor cells (clonal) as well as those present only in subsets of the cancer cells (subclonal) from tumor tissue, 864 variants were detected across the three cancer types and 73% of those were also detected in plasma without any prior knowledge of the tumor profile.
Razavi noted that without any prior knowledge of the analysis of tumor tissue, 76% of actionable mutations detected in the tissue were also detected in the blood.
Commenting on the results of the study, ASCO Expert John Heymach, MD, PhD, called the results of the study a clear advance for the field.