Mutations in circulating tumor DNA (ctDNA) may indicate which patients are likely to respond to checkpoint inhibition, according to research published in Clinical Cancer Research.1

While immunotherapy with PD-1, PD-L1, and CTLA-4 inhibition is improving outcomes among patients with cancer, responses do not occur in the majority of patients. And while tissue biopsies may help predict which patients are likely to benefit from checkpoint inhibition, these procedures can be invasive and expensive.

For this study ( Identifier: NCT02478931), researchers collected ctDNA samples from patients with varying cancer types to determine whether liquid biopsies may help to predict response to immunotherapy.

Continue Reading

Next-generation sequencing (NGS) was performed on liquid biopsy samples from 69 patients. The median patient age was 56 years, 79.7% of patients received PD-L1 or PD-1 monotherapy, and the most common malignancies in the patient sample were melanoma, lung cancer, and head and neck cancer.

Related Articles

Over 90% of patients had one or more alteration detected in the ctDNA sample. A mean of 3 variants of unknown significance (VUSs) were detected in all patients; 29% of patients had more than 3 VUSs and 71% had 3 or fewer.

Patients with more than 3 VUSs tended to have better outcomes, with a stable disease/partial response/complete response rate of 45% vs 15% for patients with 3 or fewer. Progression-free survival (PFS) was longer among patients with more than 3 VUSs (3.84 months) than among those with fewer than 3 (2.07 months; P = .019).

Median overall survival (OS) was not reached among patients with more than 3 VUSs vs 10.72 months among those with 3 or fewer (P = .042).

The authors concluded that “ctDNA-determined hypermutated states predict improved response, PFS, and OS after checkpoint inhibitor therapy across histologies.”


  1. Khagi Y, Goodman AM, Daniels GA, et al. Hypermutated circulating tumor DNA: correlation with response to checkpoint inhibitor–based immunotherapy. Clin Cancer Res. 2017 Oct 2. doi: 10.1158/1078-0432.CCR-17-1439 [Epub ahead of print]