Tumor mutation burden is emerging as a biomarker for selecting non–small cell lung cancer (NSCLC) patients for immunotherapy, according to Naiyer Rizvi, MD, the Price Family Chair of Clinical Translational Medicine, Professor of Medicine, and Director of Thoracic Oncology at Columbia University, New York. At the European Society for Medical Oncology (ESMO) 2017 Congress, Dr. Rizvi described how tumor mutation burden is already identifying patients who are likely to benefit from programmed cell death protein 1/programmed cell death ligand 1 (PD-1/PD-L1) blockade. 

Tumor Mutation Burden as a Biomarker

While PD-L1 expression is a useful biomarker, PD-L1–negative patients can also respond to checkpoint inhibitors, so other biomarkers would be valuable. Evidence is emerging that tumor mutation burden is correlated with response to immunotherapy as well. The finding that nonsynonymous mutation burden is associated with the efficacy of anti–PD-1 agents is consistent with the hypothesis that recognition of neoantigens, formed as a consequence of somatic mutations, is important for the activity of these agents, he said.

Dr. Rizvi described one of his own patients with zero expression of PD-L1 though high tumor mutation burden. The patient responded well to first-line immunotherapy, suggesting that “clearly, PD-L1 is not the whole story,” he said. “Tumor mutation burden can be used as well to decide on first-line immunotherapy approaches.”

Clearly, PD-L1 is not the whole story. Tumor mutation burden can be used as well to decide on first-line immunotherapy approaches.

— Naiyer Rizvi, MD

Tweet this quote

The potential value of tumor mutation burden was shown in an exploratory analysis of CheckMate 026. Using PD-L1 ≥ 5% as a selection strategy, the study was negative for treatment with first-line nivolumab (Opdivo). However, matched tumor and germline exome sequencing (to identify somatic missense mutations) revealed strongly positive results in patients with high tumor mutation burden.¹ In the nivolumab arm, median progression-free survival was 4.2 months when tumor mutation burden was low, and 9.7 months when tumor mutation burden was high. Patients with low tumor mutation burden actually fared better with chemotherapy (6.9 months) than with nivolumab.

The use of tumor mutation burden together with PD-L1 identified an encouraging subset of patients. At 18 months, almost 75% of nivolumab recipients with both high tumor mutation burden and high PD-L1 expression were progression-free, compared with 25% of those with high tumor mutation burden but lower PD-L1 expression. Intriguingly, even if patients had PD-L1 ≥ 50%, in the setting of low/medium tumor mutation burden, fewer than 10% were progression-free when treated with nivolumab. PD-L1 expression on its own may not be informative enough for the selection of patients, Dr. Rizvi concluded.

Beyond tumor mutation burden, there may be specific mutations that confer advantages or disadvantages, and there may be other scenarios that signal a sort of hypermutation status, such as occurs with mismatch-repair deficiency. (Patients with mismatch repair deficiency tend to respond well to immunotherapy.) The wide range of mutation burden among both smokers and never-smokers in NSCLC implicates additional pathways contributing to the accumulation of somatic mutations, he said.

Some of these mutations are beginning to be identified. Two that are emerging as predictive of response are BRCA2 and POLE. Breakdowns in BRCA2 and POLE can compromise DNA-repair mechanisms, similar to what occurs with mismatch repair deficiency. 

Mutations were also identified in a number of other genes that are important in DNA repair and replication, in whole-exome sequencing on 34 advanced NSCLC patients by Dr. Rizvi and his team. Profiles differed between responders and nonresponders to pembrolizumab (Keytruda).² “For example, in three responders with the highest mutation burden, we identified deleterious mutations in POLD1, POLE, and MSH2. Of particular interest, a POLD1 E374K mutation was identified in a never-smoker with durable clinical benefit whose tumor harbored the greatest nonsynonymous mutation burden (n = 507) of all never-smokers in our series,” Dr. Rizvi said. 

It has also been observed that NSCLC tumors with mutations in both KRAS and LKB1 tend to respond poorly (< 10%) to anti–PD-1 therapy, compared to those with KRAS mutations alone (25%).³ 

“My point is that beyond whole-exome data and tumor mutation burden, there may be specific mutations that portend particular sensitivity to a drug or can be surrogates for hypermutation and thus inform treatment,” he said. “We may need to think with a genetic lens in terms of how we treat these patients.”

Treatment Algorithm Considerations

Ultimately, Dr. Rizvi said, the treatment algorithm for first-line NSCLC will take tumor mutation burden into account. He predicted guidelines could include the following -recommendations: 

• High tumor mutation burden and PD-L1 ≥ 50%: PD-1/PD-L1 blockade alone
• High tumor mutation burden and PD-L1 expression of 1%–49%: combination treatment, such as PD-1/PD-L1 blockade plus ipilimumab (Yervoy), an indoleamine 2,3-dioxygenase inhibitor, or other novel agent
• Low tumor mutation burden and any PD-L1 expression: PD-1/PD-L1 blockade plus novel therapy to enhance immunogenicity of the microenvironment, such as T-cell–redirecting antibodies, vaccines, or chemotherapy.

“The take-home message is that we are -doing a good job finding biomarkers for response to immunotherapy,” Dr. Rizvi concluded. “Mutation burden will be important, and what we are seeing today is just scratching the surface for first-line immunotherapy.” ■

DISCLOSURE: Dr. Rizvi is a consultant for Bristol-Myers Squib, cofounder of Gritstone Oncology, and is the co-inventor of a pending patent application related to tumor mutation burden and response to cancer immunotherapy.

REFERENCES

1. Peters S, Creelan B, Hellmann MD, et al: Impact of tumor mutation burden on the efficacy of first-line nivolumab in stage IV or recurrent non-small cell lung cancer: An exploratory analysis of CheckMate 026. 2017 AACR Annual Meeting. Abstract CT082. Presented April 3, 2017.

2. Rizvi NA, Hellmann MD, Snyder A, et al: Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science 348:124-128, 2015.

3. Skoulidis F, Hellmann MD, Awad MM, et al: STK11/LKB1 co-mutations to predict for de novo resistance to PD-1/PD-L1 axis blockade in KRAS-mutant lung adenocarcinoma. 2017 ASCO Annual Meeting. Abstract 9016. Presented June 3, 2017.