cfDNA Sequencing Shows Utility in Selecting KRAS G12C+ NSCLC for Sotorasib

Plasma-based cell-free DNA appears to be a viable strategy comparable with tissue-based testing for selecting patients with KRAS G12C–mutant non–small cell lung cancer for treatment with sotorasib.

Plasma-based cell-free DNA (cfDNA) appears to be a viable strategy comparable with tissue-based testing for selecting patients with KRAS G12C–mutant non–small cell lung cancer (NSCLC) for treatment with sotorasib (AMG 510), according to data from an analysis of the phase 2 CodeBreak 100 (NCT03600883) trial presented during the AACR Annual Meeting 2021.1

Investigators reported a high concordance of 81% (95% CI, 0.75-0.87) between the tissue- and plasma-based assays, as well as a negative percent agreement of 100% (95% CI, 0.95-1.00) and a positive percent agreement of 70% (95% CI, 0.61-0.78).

In the primary clinical efficacy analysis, wherein patients were selected for treatment by plasma, those who were evaluable for response (n = 77) demonstrated an objective response rate (ORR) of 36.4% (95% CI, 25.70%-48.12%) with sotorasib; this consisted entirely of partial responses (PRs; n = 28) and no complete responses (CRs). In the full analysis set, in which patients were selected by tissue, those who were evaluable for response (n = 124) experienced an ORR of 37.1% (95% CI, 28.60%-46.23%); this included 3 CRs (2.4%) and 43 PRs (34.7%).

Additionally, patients from the full analysis set who were selected based on tissue had a disease control rate (DCR) of 80.6% (95%, 72.58%-87.19%) vs 77.9% (95% CI, 67.02%-86.58%) among those who were selected based on plasma in the primary analyses.

The first-in-class small molecule sotorasib irreversibly inhibits the KRAS G12C protein, which is found in approximately 13% of all lung adenocarcinomas.2,3 In February 2021, the FDA granted sotorasib priority review for the treatment of patients with KRAS G12C–mutated locally advanced or metastatic NSCLC following at least 1 prior systemic therapy based on data from the phase 2 CodeBreak100 trial.4 Results showed that the agent yielded an ORR of 37.1%, a median duration of response of 10.0 months, a DCR of 80.6% (95% CI, 72.6%-87.2%), and a median progression-free survival of 6.8 months.5

Guardant360 CDx is a next-generation sequencing-based in vitro diagnostic assay that sequences cfDNA from plasma; in August 2020, the assay was FDA approved for the comprehensive genomic profiling in patients with any solid cancer. For the analysis presented during the meeting, investigators sought to clinically validate the assay in selecting patients with KRAS G12C–mutated NSCLC who may be candidates for sotorasib.

The full analysis set of the CodeBreak 100 trial population included those with NSCLC harboring a KRAS G12C mutation confirmed via tissue assay, while the primary clinical efficacy analysis set, which aimed to clinically validate the Guardant360 CDx, included a subset of patients from the trial who had their mutation confirmed through plasma-based testing.

Moreover, the assay agreement analysis set, which aimed to clinically validate the sensitivity of the plasma-based assay, included patients from CodeBreak 100 who were positive or negative for KRAS G12C; off-trial patients who had evaluable results from both tissue- and plasma-based assays to address the lack of tissue-negative patients within the on-trial patient population.

In total, 126 patients were enrolled to the trial, 112 of whom underwent testing with Guardant360 CDx. Of the patients who received plasma-based testing, 78 patients tested positive for the mutation, 31 tested negative, and 3 patients had failed testing with the assay. Among the additional 132 patients procured for the sensitivity analysis, 80 had valid plasma and tissue results and 52 were either not tested or had failed tissue or plasma testing. In total, 189 patients had available matched tissue and plasma results available for the sensitivity analysis.

The primary objective of the analysis was to demonstrate the comparability of the efficacy linked with sotorasib in those who tested positive for KRAS G12C via plasma-based cfDNA sequencing and in the overall trial population via tissue-based sequencing. The primary end point was ORR by RECIST 1.1 criteria, as assessed by independent central radiologic review.

Important secondary objectives included examining the concordance between the 2 assays in a sensitivity analysis, as well as evaluating the additional efficacy of sotorasib in those who tested positive for KRAS G12C via the plasma-based assay. Specific secondary end points included positive and negative percent agreements of plasma assay relative to tissue assay, overall concordance between the 2 tests, DCR, and assay agreement analysis set.

Among all 126 patients enrolled, the median age was 63.5 years (range, 37-80), with 53.2% of patients falling between the ages of 18 years and 64 years, 38.9% between 65 years and 74 years, and 7.9% between 75 years and 84 years. Fifty percent of patients were male and the other half were female. Moreover, 69.8% of patients had an ECOG performance status of 1, while 30.2% had a status of 0. Nearly all patients (96.0%) had stage IV disease while a minority of patients had stage III disease (4.0%). Additionally, 96.8% of patients had metastatic disease.

Just under 5% of patients were never smokers, 11.9% were current smokers, 81.0% were former smokers, and 2.4% had missing information. Moreover, 42.9% had 1 prior line of treatment, 34.9% has 2 prior lines, and 22.2% had 3 prior lines; no patients were previously untreated. The median number of prior lines of therapy received was 2.

Additional findings indicated that 42.5% of patients in the full analysis who were selected by tissue had stable disease, 16.1% had progressive disease, and 3.2% were not evaluable or were missing scans. By comparison, 41.6% of patients in the clinical efficacy analysis who were selected based on plasma achieved stable disease, 16.9% had progressive disease, and 5.2% were not evaluable or were missing scans.

Guardant360 CDx is concurrently under review as a potential companion diagnostic to sotorasib.


  1. Bauml JM, Li BT, Velcheti V, et al. Clinical validation of plasma cell-free DNA (cfDNA) sequencing in the phase 2 trial of sotorasib in patients (pts) with KRAS p.G12C mutated NSCLC. Presented at: AACR Annual Meeting 2021; April 10-15, 2021; Virtual. Poster CT181.
  2. Biernacka A, Tsongalis PD, Peterson JD, et al. The potential utility of re-mining results of somatic mutation testing: KRAS status in lung adenocarcinoma. Cancer Genet. 2016;209(5):195-198. doi:10.1016/j.cancergen.2016.03.001
  3. Hong DS, Fakih MG, Strickler JH, et al. KRASG12C inhibition with sotorasib in advanced solid tumors. N Engl J Med. 2020;383(13):1207-1217. doi:10.1056/NEJMoa1917239
  4. FDA grants sotorasib priority review designation for the treatment of patients with KRAS G12C-mutated locally advanced or metastatic non-small cell lung cancer. News release. Amgen. February 16, 2021. Accessed April 13, 2021.
  5. Li BT, Skoulidis F, Falchook G, et al. CodeBreaK 100: registrational phase 2 trial of sotorasib in KRAS p.G12C mutated non-small cell lung cancer. Presented at: International Association for the Study of Lung Cancer 2020 World Conference on Lung Cancer; January 28-31, 2021; Virtual. Abstract PS01.07.
  6. Guardant Health Guardant360 CDx first FDA-approved liquid biopsy for comprehensive tumor mutation profiling across all solid cancers. News release. Guardant Health, Inc. August 7, 2020. Accessed April 13, 2020.