Approximately 30% of patients with non-small cell lung cancer (NSCLC) have inadequate tumour tissue for molecular testing at diagnosis, therefore, blood-based tumour mutational burden (TMB) is a biomarker that is being explored to assess its predictive value for patients being treated with checkpoint inhibitors.
PD-L1 unselected patients with high levels of blood TMB demonstrated higher response rates and improved survival following first-line atezolizumab monotherapy forl ocally advanced or metastatic NSCLC, according to findings presented at the ESMO 2018 Congress in Munich, Germany.
Previously, authors have reported on the retrospective analyses from the randomised, phase III OAK and phase II POPLAR studies wherein the TMB detected in blood and tissue showed promise in predicting benefit from atezolizumab administered as second-line therapy in patients with NSCLC.
At the ESMO 2018 Congress, Professor Kim of the Department of Solid Tumour Oncology and Investigational Therapeutics, Levine Cancer Institute, Atrium Health, Carolinas HealthCare System in Charlotte, United States of America, presented primary efficacy results from the prospective phase II B-F1RST (NCT02848651) trial evaluating the utility of a novel blood-based TMB assay as a predictive biomarker for atezolizumab administered as first-line therapy to PD-L1–unselected patients with locally advanced or metastatic NSCLC.
In B-F1RST, 152 patients comprised the intent-to-treat (ITT) population and received intravenous atezolizumab at 1200 mg every 3 weeks until disease progression or loss of clinical benefit. Of this group, 119 patients were included in the biomarker evaluable population (BEP). Atezolizumab clinical efficacy was assessed according to a prespecified blood TMB cut-off set at ≥ 16, depicting high TMB or < 16, which was considered low TMB. The co-primary endpoints were objective response rate (ORR) and progression-free survival (PFS).
In the ITT population, 119 patients had adequate circulating tumour (ct)DNA, at a maximum somatic allele frequency (MSAF) ≥ 1% and were included in the BEP. Twenty-nine patients had inadequate ctDNA tumour shedding into the blood (MSAF < 1%) and comprised the non-BEP population.
Results
At a minimum follow-up of 6 months, the ORR in the overall ITT population was 14.5% and in the BEP was 10.1%. In the non-BEP, which had favourable prognostic characteristics, the ORR was 34.5%.
At the prespecified blood TMB cut-off ≥16, assessment of the response to atezolizumab in patients with high (≥16) versus low (<16) blood TMB revealed an ORR of 28.6% compared to 4.4%. For PFS, TMB low patients demonstrated median PFS of 3.7 months compared to 4.6 months in patients with high blood TMB, HR 0.66 (90% CI 0.42 – 1.02).
Median overall survival (OS) was not estimable (NE) in patients with blood TMB high compared to 13.1 months in blood TMB low patients, HR 0.77; 90% CI, 0.41 – 1.43 (p = 0.48).
In the ITT population, treatment-related serious adverse events (AEs) occurred in 13% of patients and 20% of patients had treatment-related grade 3/4 AEs. Adverse events led to atezolizumab treatment discontinuation in 15% of patients.
Analysis of data by prespecified TMB cut-offs further defined atezolizumab response
At increasing blood TMB cut-offs ranging from 10 to 20, the hazard ratio for PFS improved from HR 1.09 (blood TMB ≥10) to 0.48 (blood TMB ≥20), emphasizing the proportional relationship between increasing blood TMB score and improved clinical outcomes.
Discussant points
Prof. Benjamin Besse, Head of the Cancer Medecine Department, Gustave Roussy and Head of the EORTC Lung Cancer Group who discussed the study findings said that a role of liquid biopsy might be a key. B-F1RST has major findings but blood TMB may not capture the more sensitive tumours.
Conclusions
The B-F1RST primary analysis represents the first prospective dataset evaluating the clinical utility of blood TMB as a predictive biomarker for first-line atezolizumab monotherapy in patients with advanced lung cancer. The authors concluded that at the prespecified blood TMB ≥16 cut-off, patients showed numerical benefit for PFS, ORR and OS, which was consistent with interim data. Study follow-up will continue until the final analysis.
They reported that follow-up will continue for ≥18 months, per protocol.
Disclosure
This trial was sponsored by F. Hoffmann-La Roche AG.
Reference
LBA55 – Kim ES, Velcheti V, Mekhail T, et al. Primary efficacy results from B-F1RST, a prospective Phase II trial evaluating blood-based tumour mutational burden (bTMB) as a predictive biomarker for atezolizumab (atezo) in 1L non-small cell lung cancer (NSCLC).