pTMB is a feasible predictive biomarker for chemoimmunotherapy

A biomarker which importance is clearly increasing, is the tumour mutational burden (TMB; number of somatic mutations per megabase [mut/Mb]). Furthermore, there has been some research into the role of plasma-based TMB as a biomarker, although this has been done only in anti-PD-L1 therapy studies or in combination therapy with PD-L1/cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) blockade. Aggarwal et al. investigated whether pTMB can be used as a predictive biomarker in chemoimmunotherapy.

The researchers used a cohort which consisted of 66 treatment-naïve patients with stage IV non-squamous NSCLC without EGFR/ALK/ROS1/BRAF-mutations. Patients were randomised to pembrolizumab monotherapy (n=31) and pembrolizumab + chemotherapy (n=36). At 9 weeks, pTMB was compared to Response Evaluation Criteria in Solid Tumors (RECIST) response and 6 months durable clinical benefit. High and low pTMB were defined at a cut-off of 16 mut/Mb [1].

The results showed that median progression-free survival in patients with high pTMB was 13.8 vs 4.7 months in patients who had low pTMB (HR 0.27, 95% CI 0.13-0.55, P0.001). In patients with high pTMB, median overall survival was not achieved compared to 8.4 months in patients harbouring low pTMB (HR 0.47, 95% CI 0.20-1.10, P=0.083).

It was concluded that it is feasible to assess pTMB in patients who receive first-line treatment (which included chemoimmunotherapy) for advanced NSCLC. Mutations in STK11/KEAP1/PTEN as well as ERBB2 seem to be able to predict a lack of clinical benefit of immunotherapy. However, they agreed that the role of pTMB needs to be validated in larger, prospective studies.

1. Aggarwal C, et al. MA25.04. WCLC 2019.

Posted on 8 November 20195 March 2021