Abstract
The feasibility of circulating tumor DNA (ctDNA) assays as a first-approach test in a pan-cancer setting is not well-established. Furthermore, low ctDNA levels limit assay sensitivity which challenges adaptation to clinical genomic profiling. A 33-gene next-generation sequencing (NGS) ctDNA panel was validated, and these issues were investigated using real-world clinical data. The clinical cohort included 123 patients with ctDNA testing performed as a first approach, and 48 patients for whom matched tissue was tested at the same time-point. The overall ctDNA testing failure rate is 0%. Insufficient tumor tissue was the main reason for liquid biopsy (69%). The most common cancer primary tested was lung (39.0%), followed by colon (13.8%), bile duct (8.9%), pancreas (8.1%), breast (4.1%) and prostate (4.1%). Using AMP/ASCO/CAP guidelines, Tier I variants were detected in 33.3% patients, and Tier I or II variants were detected in 65.0% patients (including 54.5% cholangiocarcinoma patients, for whom tissue biopsy is challenging due to anatomical location). Compared with matched tissue, ctDNA showed 76% sensitivity for Tier I variants. Concurrent testing increased the number of actionable variants by 14.3% versus tissue testing alone. ctDNA results preceded tissue results by an average of 21 days. In conclusion, high feasibility, actionability and sensitivity support ctDNA assays as a potential first-line genomic test, especially in specific tumor-types for advanced tumors when tissue is unavailable.

VIEW THE PUBLICATION