Plasma-Based TMB May Predict Pembrolizumab Response in NSCLC

Charu Aggarwal, MD, MPH

Plasma-based tumor mutational burden (TMB) has potential as a predictive biomarker for responses to immunotherapy or immunotherapy combinations in patients with non—small cell lung cancer (NSCLC), explained Charu Aggarwal, MD, MPH.

“[TMB] may be a novel, relatively less-invasive biomarker that we can incorporate into our clinical practice in the future to guide treatment decisions,” said Aggarwal.

To assess the role of plasma TMB in NSCLC, patients with newly diagnosed NSCLC were tested for TMB using a 500-gene next-generation sequencing panel as they were starting first-line treatment with pembrolizumab (Keytruda) either alone or in combination with chemotherapy.

The study evaluated 66 patients, of whom 52 were evaluable for plasma TMB analysis, showing a median plasma TMB of 16.8 mutations per megabase (mut/Mb). Moreover, the median plasma TMB was significantly higher for patients achieving durable clinical benefit (DCB) compared with no durable benefit at 21.3 mut/Mb versus 12.4 mut/Mb, respectively (P = .003). Patients who had a plasma TMB ≥16 mut/Mb had a median progression-free survival (PFS) of 14.1 months compared with 4.7 months for patients with plasma TMB <16 mut/Mb (HR, 0.30; 95% CI, 0.16-0.60; P <.001). The median overall survival (OS) for patients with plasma TMB ≥16 was not reached versus 8.8 months for patients with plasma TMB <16 mut/Mb (HR, 0.48; 95% CI, 0.22-1.03; P = .061).

Further, mutations in ERBB2 exon 20, STK11, KEAP1, or PTEN were more common in patients with no DCB. A combination of plasma TMB ≥16 and the absence of a negative predictive biomarker was associated with PFS (HR, 0.24; 95% CI, 0.11-0.49; P <.001) and OS (HR, 0.31; 95% CI, 0.13-0.74; P = .009).

The next steps of the trial will include exploring the predictive utility of this biomarker in NSCLC in larger, prospective studies.

In an interview with OncLive, Aggarwal, the Leslye Heisler Assistant Professor of Medicine in the Hematology-Oncology Division at the University of Pennsylvania’s Perelman School of Medicine, discussed the design and findings of this analysis examining the role of TMB as a predictive biomarker for pembrolizumab-based therapy in NSCLC.

OncLive: Could you give an overview of the trial design?

Aggarwal: We know a lot about TMB and its effects in tissue. We have seen those both in retrospective and prospective data. Tissue TMB can predict response to immunotherapy and combination immunotherapy. However, the value of plasma TMB as a biomarker in patients with lung cancer receiving pembrolizumab-based therapy has not been clearly assessed or defined. We sought out to examine the correlation between plasma TMB and outcomes in patients who receive pembrolizumab-based monotherapy or in combination with chemotherapy in patients who receive therapy for first-line metastatic NSCLC.

What were the results of the study?

This was a prospectively conducted clinical trial where samples were collected from patients with metastatic NSCLC who were treatment-naïve and starting first-line pembrolizumab-based therapy. They could not have any of the activating mutations, such as EGFR, ALK, BRAF, and ROS1. All these patients underwent selection of 2 tubes of blood that were collected prior to therapy, and then they were followed for response and survival. The TMB analysis was conducted using the Guardant Health 500-gene panel that has 2.1 Mb of coverage. We included 66 patients, 52 of which were evaluated for plasma TMB, and our median TMB in the whole group of our patients was 16.8 mut/Mb. Patient who had plasma TMB ≥16 mut/Mb have significantly improved progression-free survival and also had a tendency to have better durability of response, meaning that they are more likely to have either a complete response, partial response, or stable disease at 6 months.

There was also a trend noted toward improved survival for the patients who had high plasma TMB. This is true, even for patients who received immunotherapy and chemotherapy, which is an important finding because that's the first time we have any studies that have shown [those results].

Next, we went in to look at some of the other mutations and how they play a role in interaction with TMB. We evaluated the role of ERBB2 exon 20, STK11, KEAP1, and PTEN mutations, which were all more common in patients who had no durable benefit from immunotherapy. This leads us to believe that we can come up with a more comprehensive combined score with these negative predictors, which could lead to a significant improvement in survival. This is a very small exploratory analysis with small numbers, but these are interesting findings. Plasma TMB is a much less invasive way to assess another biomarker that should be validated in larger clinical studies.

How do you anticipate that these results will impact clinical practice for NSCLC? What are the next steps of the trial?

These kinds of tests and biomarkers should be incorporated into further clinical trials as we design new clinical trials for patients with metastatic lung cancer.

At this point in time, this is not ready for primetime. We need validation of these results and a larger prospective study before we can start using [TMB as a biomarker]. These were thought-provoking data that give us another biomarker to evaluate. We are thinking about prospectively evaluating the role of TMB in patients with NSCLC.

Aggarwal C, Thompson JC, Chien AL, et al. Baseline plasma tumor mutation burden predicts response to pembrolizumab-based therapy in patients with metastatic non-small cell lung cancer [published online ahead of print February 26, 2020]. Clin Cancer Res. doi: 10.1158/1078-0432.CCR-19-3663