A high tumour mutational burden in the plasma (bTMB) is a clinically actionable biomarker for atezolizumab (Tencentriq®, Roche) therapy in patients who failed 1st-line platinum-based treatment for advanced non-small cell lung cancer (NSCLC). Prospective validation of the bTMB assay in the 1st-line NSCLC is currently ongoing in the Phase III BFAST trial.

Several immune-checkpoint inhibitors targeting the programmed death 1 (PD-1)–axis have demonstrated significant overall survival (OS) benefit in the 2nd-line treatment for advanced NSCLC. Tumour- and immune-cell PD-1 ligand (PD-L1) expression has shown to partially predict the clinical benefit in patients treated with PD-1 or PD-L1 inhibitors; however, the expression alone does not adequately explain benefit in all patients.

Recently, multiple studies have linked a high-TMB to improved outcomes with immune-checkpoint inhibitors. A less invasive diagnostic method, such as the bTMB assay, may help to select patients who are more likely to benefit from anti-PD-1 or anti-PD-L1 monotherapy in advanced NSCLC.

Up to 30% of patients with NSCLC do not have adequate tissue available at diagnosis for standard biomarker testing.

The researchers, led by Drs David Gandara, Tony Mok, and David Shames, reported on a novel, technically robust, blood-based assay to measure plasma tumour mutational burden (bTMB) in over 1,000 prospectively obtained samples of patients who received anti-PD-L1 treatment with atezolizumab in the Phase II POPLAR and Phase III OAK studies.

In the POPLAR study, 211 of  273 samples obtained at different time-points in treatment were available and sufficient for the bTMB assay. Using the POPLAR samples, the researchers tested three bTMB cut-points (≥10, ≥16, and ≥20) and correlated these with clinically meaningful outcomes. The results from the POPLAR study related the TMB ≥16 cut-point to stronger progression-free survival (PFS) treatment effects. The TMB ≥16 cut-point was subsequently validated using 583 of 797 selected OAK samples which excluded plasma of patients with EGFR or ALK alterations.

“The ability to analyse tumour genomes through a simple blood draw has distinct advantages compared to tissue biopsy collection: the blood offers a ready, contemporaneous source of diagnostic material and such testing may be less susceptible to potential sampling biases that are associated with single-site tissue biopsies.”

Results

A TMB ≥16 was found in 30% and 27% of POPLAR and OAK NSCLC patients, respectively. The bTMB ≥16 cut-point correlated to improved PFS (HR=0.57; 95% CI, 0.33–0.99) for atezolizumab vs docetaxel in the POPLAR trial. Similarly, patients in the OAK trial with bTMB ≥ 16 had a significantly improved PFS for atezolizumab vs docetaxel (HR=0.65; 95% CI, 0.47–0.92); p=0.013).

Important limitations of the blood-based assays remain the requirement for a minimum amount of circulating tumour DNA (ctDNA), which is partially dependent on overall tumour burden. Moreover, the current bTMB assay calculation only accounts for single nucleotide variants (SNVs) and sensitivity for tumours with a relatively large number of indels and few SNVs may be low.

References

Gandara DR, et al. Nature Med. 2018; doi: 10.1038/s41591-018-0134-3.