A Tumor-Agnostic Prognostic Tool?
ctDNA tumor fraction acts as a prognostic biomarker across four advanced cancer types
Georgia Hulme | | News
Liquid biopsies are increasingly revolutionizing the way people with cancer are diagnosed, monitored, and treated. Clinical adoption of blood-based circulating tumor DNA (ctDNA) to track cancer progression is now widespread and many healthcare systems use ctDNA levels to help guide individualized therapy. In a new study, researchers investigated the prognostic value of elevated circulating tumor fraction (TF) calculated using ctDNA analysis.
The study used a multi-tumor cohort of patients from a nationwide clinicogenomic database. Each participant had one of four advanced cancer types: metastatic breast cancer, metastatic colorectal cancer, castration-resistant prostate cancer, or advanced non-small cell lung cancer. A total of 1,725 liquid biopsies were collected and TF levels were analyzed based on single-nucleotide polymorphism aneuploidy across the genome.
TF levels of at least 10 percent were closely associated with lower overall survival in univariate analyses across all cancer types; the biomarker’s prognostic impact is independent of most clinical features on multivariate analysis. However, certain clinical features formed an exception to this rule; the authors found that brain metastases, for instance, increased patient morbidity, but did not increase TF levels because these tumors may not shed ctDNA. . Exploratory analysis further found that TF is a successful prognostic biomarker across a range of different cutpoints, although more research on relevant cutpoints within each cancer type is still needed.
The researchers concluded that ctDNA TF can act as a biomarker for prognosis across multiple common cancer types in a real-world dataset. In the future, they hope that liquid biopsy for ctDNA TF testing will be incorporated into point-of-care settings to provide cancer therapy based on patient-level tumor biology.
- ZR Reichert et al., Ann Oncol, 34, 111 (2023). PMID: 36208697.
Associate Editor for the Pathologist