When Does Immunotherapy Work Best in Lung Cancer?
Chicago, IL—Response to immunotherapy in non–small-cell lung cancer (NSCLC) differs depending on the driver mutation involved and on the amount of PD-L1 expression within the tumor. Immune checkpoint inhibitors in NSCLC—PD-1 and PD-L1 inhibitors—have similar efficacy and adverse event profiles. These were among the topics discussed at the 2017 ASCO annual meeting in a session on immunotherapy and lung cancer.
Who Should Receive Front-Line Immunotherapy?
Based on outcome studies exploring the efficacy of immunotherapy compared with chemotherapy by PD-L1 status, “patients with staining for PD-L1 in at least half of tumor cells using the [PD-L1 IHC] 22C3 assay should receive front-line pembrolizumab [Keytruda],” said Edward B. Garon, MD, MS, Director of Thoracic Oncology, David Geffen School of Medicine at the University of California, Los Angeles. “Patients who do not have this level of staining should receive frontline chemotherapy,” he added.
Lung Cancer Mutations and Immunotherapy
Tumor mutation burden is predictive of benefit to immuno-oncologic agents. Tumor mutation burden correlates with the degree of DNA damage or germline mismatch repair deficiency. In the CheckMate-026 clinical trial, nivolumab (Opdivo) improved progression-free survival (PFS) compared with platinum-based doublet chemotherapy in the first-line setting in patients with NSCLC with a high tumor mutation burden but not in patients with low or medium mutation burden, said Benjamin Creelan, MD, MS, a thoracic oncologist at the H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL.
However, tumor mutation burden is simply a surrogate for neoantigen expression, “and not all neoantigens are made the same,” said Dr Creelan. Clonal mutations or neoantigens may permit sustained responses to immuno-oncotherapy, but patients with lung tumors with a high fraction of subclonal neoantigens do not have durable benefit from immune checkpoint blockade. Subclonal neoantigens often overtake the original antigen and create clones that escape T-cell recognition, explained Dr Creelan.
When outcomes with immune checkpoint inhibitors are stratified by the presence of an EGFR driver mutation, patients with EGFR mutation–positive tumors have significantly worse PFS and no improvement in overall survival (OS) after receiving PD-1 or PD-L1 inhibitors compared with docetaxel; the inverse is true in patients with EGFR wild-type tumors.
However, in patients with NSCLC and high PD-L1 expression, EGFR-positive cancers respond to PD-L1 monotherapy to those with wild-type tumors. “The only problem is that PD-L1 is rarely expressed in EGFR-mutant tumors,” said Dr Creelan.
Therefore, an EGFR tyrosine kinase inhibitor (TKI) or chemotherapy is generally preferred over PD-L1 monotherapy for EGFR mutation–positive NSCLC, but PD-L1 monotherapy is a reasonable option for EGFR mutation–positive NSCLC if it is strongly PD-L1–positive, he said.
Combinations of immune checkpoint inhibitors and an EGFR TKI are being studied in patients with locally advanced or metastatic NSCLC with or without EGFR mutations, but early data suggest that these combination therapies are associated with high toxicity rates.
Data also suggest that STK11 mutation–positive lung adenocarcinoma may not be appropriate for immuno-oncotherapy, because these tumors contain fewer CD8-positive tumor infiltrating lymphocytes and less immune gene activation versus patients with other driver mutations. STK11 is frequently comutated with KRAS, noted Dr Creelan. “Multiple retrospective studies indicate worse outcome with PD-L1 monotherapy for STK11-mutant NSCLC,” he said.
Conversely, patients with MET exon14 short variants, which represent 2% to 4% of NSCLC, may be good candidates for immuno-oncotherapy. Patients with MET exon14 mutations tend to have more T-cell infiltration, antigen-presenting cells, gene expression, and PD-L1 expression than patients with other driver mutations. Retrospective data have shown favorable results with PD-L1 monotherapy in patients with MET exon14 short variants.
Are Immunotherapy Drugs Interchangeable?
Different PD-1 inhibitors bind in different ways to PD-1. This variability underscores that “these drugs are not carbon copies of one another; rather, there are structural differences between them,” said Melissa L. Johnson, MD, Associate Director, Lung Cancer Research, Sarah Cannon Research Institute, Nashville, TN.
Yet adverse event rates are largely similar between the PD-1 and PD-L1 inhibitors, ranging from 61% to 76%. In a meta-analysis involving patients with NSCLC, the rate of immune-related adverse events was slightly higher with PD-1 inhibitors than with PD-L1 inhibitors (P = .04), although there have been no direct comparisons between any of the agents.
The overall response rates and OS data from key phase 3, randomized controlled clinical trials of anti–PD-1 or anti–PD-L1 therapy in patients with NSCLC in the second-line setting are also comparable, with improvements in the median OS of 3 to 4 months compared with docetaxel, said Dr Johnson.
Outcomes between nivolumab and pembrolizumab are “quite similar between a whole host of different histologies between the 2 drugs, including unselected patients with previously treated NSCLC,” said Dr Johnson. “It is hard to imagine why the drugs would be very different in terms of efficacy, specifically in frontline NSCLC,” she added. In the absence of differences in efficacy or adverse event profiles, “it comes down to cost and convenience,” said Dr Johnson.
Pembrolizumab and atezolizumab (Tecentriq) are given every 3 weeks, whereas nivolumab is administered every 2 weeks. Clinical trials investigating every-4-week dosing and every-6-week dosing of some checkpoint inhibitors are ongoing. “If those trials are positive, then I would say then that would become a competitive differentiator,” she added.
The cost of each checkpoint inhibitor is approximately $21,000 for 6 weeks of therapy, but over 6 months, atezolizumab offers approximately 5% cost-savings over the other 2 approved immune checkpoint inhibitors in the second-line setting for NSCLC.