Amid rapid-fire developments in non–small cell lung cancer (NSCLC) research, the role of checkpoint immunotherapy continues to mature.
Suresh S. Ramalingam, MD
Amid rapid-fire developments in non—small cell lung cancer (NSCLC) research, the role of checkpoint immunotherapy continues to mature. Experts from the United States, France, and Italy discussed the evolution of therapy for patients with advanced nonsquamous disease during a recent OncLive® Peer Exchange® panel. Moderator Suresh S.
Ramalingam, MD, opened the discussion by noting that there has been a “sea change in the way advanced lung cancer is treated.” Along with the many molecularly targeted therapies now available for patients with mutations, immunotherapy agents have been approved in several countries. In all, the FDA has approved has approved 3 checkpoint blockade immunotherapies for patients with NSCLC: pembrolizumab (Keytruda) and nivolumab (Opdivo), both monoclonal antibodies that inhibit PD-1, and atezolizumab (Tecentriq), which inhibits PD-L1. Another PD-L1 inhibitor, durvalumab (Imfinzi), is under review for patients with stage III unresectable NSCLC.
The approvals issued thus far differ in disease settings, patient populations, and use of PD-L1 expression levels to select candidates for treatment.Pembrolizumab was originally approved by the FDA for the second-line treatment of patients with metastatic NSCLC whose tumors express PD-L1 at any level, regardless of histology. Indications for pembrolizumab monotherapy were subsequently expanded to include treatment-naïve patients whose tumors express PD-L1 with a tumor proportion score (TPS) ≥50%.1 This was based on the results of the randomized phase III KEYNOTE-024 trial in which pembrolizumab was compared with investigator’s choice of chemotherapy for advanced NSCLC tumors with PD-L1 expression with TPS ≥50%.2 Patients in the chemotherapy arm were permitted to cross over to pembrolizumab after disease progression. Approximately 30% of the more than 1600 patients screened had the enrollment biomarker and 305 were enrolled.2 The primary endpoint was median progression-free survival (PFS), which reached 10.3 months in the pembrolizumab arm versus 6 months in the chemotherapy arm (HR for disease progression or death, 0.50; 95% CI, 0.37-0.68; P <.001).2 Investigators presented updated findings at the 2017 American Society of Clinical Oncology Annual Meeting, which showed significantly better median PFS in the pembrolizumab arm than in the chemotherapy arm (HR, 0.50; P <.0001) and significantly better overall survival (OS; HR, 0.60; P = .005).3
“The primary endpoint was, at least for me, shocking because there was a significant clinical improvement, not only significant improvement in PFS,” said Giorgio V. Scagliotti, MD, PhD. “More importantly...there was also an improvement in the overall survival despite the study allowing the crossover.”
The OS benefit was more pronounced in updated results presented during the 18th World Conference on Lung Cancer (WCLC) in October, which was after the Peer Exchange® program took place. In the WCLC findings, the median OS for patients treated with pembrolizumab was 30.2 months versus 14.2 months with chemotherapy, representing a 37% reduction in the risk of death (HR, 0.63; 95% CI, 0.47-0.86; P = .002).4
In May 2017, the FDA granted accelerated approval to pembrolizumab in combination with the chemotherapy agents pemetrexed (Alimta) and carboplatin for the first-line treatment of advanced nonsquamous NSCLC.5 Approval was based on results from the KEYNOTE-021 trial, which did not stratify patients according to PD-L1 expression.6
Patients were chemotherapy-naïve and had no targetable EGFR or ALK genomic aberrations. Participants were randomly assigned to pembrolizumab plus chemotherapy with carboplatin and pemetrexed for 24 months or to 3 to 4 cycles of chemotherapy only.6 After completing initial treatment assignments, all patients in the study were switched to pembrolizumab maintenance therapy for an indefinite amount of time. Interim results showed a significantly higher objective response rate (ORR) in the combination arm than the chemotherapy-only arm, which was the primary objective of KEYNOTE-021 (55% vs 29%, respectively; P = .0016).6
Borghaei and colleagues presented updated data at the European Society for Medical Oncology (ESMO) 2017 Congress.7 The ORR was 57% in the pembrolizumab arm versus 32% in the chemotherapy-only arm (P = .0029), and the median PFS was significantly longer in the pembrolizumab arm than in the chemotherapy arm (19 vs 8.9 months, respectively; P = .0067).7
OS was not reached in either arm, but the authors said the HR for OS increasingly favored the pembrolizumab arm (HR, 0.90-0.69 to 0.59, respectively) over time.
When considering the findings from KEYNOTE-021, Scagliotti said it is important to remember that it is a phase II study that enrolled only 123 patients. “We are still not able to say that chemotherapy plus immunotherapy is definitely better in terms of OS, but we are walking on the boundary of it being statistically significant and probably need to wait a little longer,” Scagliotti said. He said data are needed from a study comparing chemotherapy plus immunotherapy versus immunotherapy alone.
Benjamin Besse, MD, PhD, said, “First-line pembrolizumab in high PD-L1 NSCLC patients is probably the standard of care because the study with pembrolizumab is really convincing—it’s a phase III study.” He praised the PFS findings of the phase II study but said it was too small and the patient population too highly selected to be practice changing. “I was quite surprised that the FDA could approve the drug based on such limited data,” Basse said. Summarizing the panel’s recommendations, Ramalingam said that for a patient with squamous cell lung cancer, “It’s clear now that the first-line paradigm is, if the tumor is overexpressed in PD-L1 at 50% or greater, you give monotherapy with pembrolizumab. If it’s not, then you give platinum-based chemotherapy.
For nonsquamous, again, the panel seems to agree that for a high expressor of PD-L1, you give pembrolizumab monotherapy. For patients who don’t have high expression, in the United States you have the option of chemotherapy plus pembrolizumab. In other parts of the world, chemotherapy continues to be the mainstay of treatment.”The field of available agents widens when it comes to patients who have progressed on prior therapy. The FDA has approved 3 checkpoint inhibitors in this space: pembrolizumab, nivolumab, and atezolizumab. According to Besse, studies comparing the checkpoint inhibitors with chemotherapy in patients previously treated with chemotherapy have established that immunotherapy is superior to chemotherapy in the second line and beyond. However, researchers are still trying to determine the appropriate duration for immunotherapy.
Data from a randomized trial that compared 1 year of nivolumab therapy versus continuous nivolumab (until disease progression or intolerability) in patients with advanced NSCLC who had already received 1 year of nivolumab salvage therapy were presented at the ESMO 2017 Congress.8 Besse said that the PFS curves showed continuous nivolumab was superior to 1 year of additional therapy and that the trend in OS curves favored continuous nivolumab. “For me, that really means that we should not stop immunotherapy after 1 year, even in the patient with special response. It will definitely change my practice,” he said. Scagliotti disagreed that the study was practice changing because of how few participants received additional nivolumab (218/1375 [15.9%]).8
He agreed with Besse that administering nivolumab for 2 years in the salvage setting was “probably the right way to move forward” but noted some oncologists are reluctant to do so. “I truly believe that if we are aiming to make lung cancer a chronic disease, we need to accept the idea of continuing treatment for a long period of time,” he said, comparing the approach to treating diabetes.
“I have some patients still in the CheckMate-017 and CheckMate-057 studies who are still on drugs after about 4 years or so,” said Marina Garassino, MD. “Sometimes you have the feeling that maybe you can stop, but we need further data for this.” Besse said that because nivolumab, pembrolizumab, and atezolizumab have a half-life exceeding 25 days, he tells patients, “We can skip 1 or 2 injections from time to time if they want to go on vacation or take 1 month off.”The use of PD-L1 expression levels as a biomarker for administering checkpoint im mu not herapy continues to be an unresolved question. Ramalingam pointed out that since the panelists agree patients with nonsquamous NSCLC and high PD-L1 expression should receive pembrolizumab monotherapy, tumors must be tested.
Garassino said that although the KEYNOTE-024 trial results show that it is “fundamental to test all patients for the presence of PD-L1,” efforts to do so are complicated by the fact that each of the checkpoint blockade agents has its own companion diagnostic test. She said there are 4 assays, which use different antibodies, available to test for PD-L1. “There are several papers suggesting that, starting with the Blueprint trial, at least 3 of the commercially available assays are more or less the same, but 1 is quite different, which is the VENTANA SP142 assay,” she said. The Blueprint PD-L1 Immunohistochemistry Assay Comparison Project found that the 22C3, 28-8, and SP263 assays were highly concordant in percentage of PD-L1 tumor cells stained, whereas the SP142 assay consistently stained a lower percentage of tumor cells than the other 3 tests.9
The 4 tests also use different cut-off points for PD-L1 positivity. Thus, each test must be interpreted according to its unique scale and the results cannot be used interchangeably. Garassino said efforts are underway to harmonize the tests, but that for now, “It’s important that, when you do PD-L1 [testing], to use at least 3 antibodies commercially available.”
Garassino said conducting all the necessary immunohistochemistry tests requires a pathology specimen with adequate tissue. Besse said when there is insufficient tissue to perform PD-L1 testing, smoking status is a good predictor of efficacy for second-line immunotherapy. “In a never-smoker, we should probably avoid giving second-line immunotherapy,” he said. Besse said evidence suggests a small percentage of patients with NSCLC are at risk of hyperprogressive disease when treated with PD1 or PD-L1 inhibitors but data are lacking to determine why.10
Garassino said mutation burden may be emerging as another predictive biomarker for immunotherapy efficacy.11 “Results from CheckMate-026 suggest that when you have a high mutational burden, nivolumab performs better than chemotherapy; but when the mutational burden is low, chemotherapy is much better than nivolumab,” she said.
Garassino said combining mutational burden with PD-L1 expression levels may turn out to be an even better tool for selecting patients for immunotherapy, although more data are needed.Although this Peer Exchange® discussionfocused on advanced and metastatic disease, the panel members noted the prospects for durvalumab in patients with stage III nonresectable NSCLC. This patient population is typically treated with concurrent chemotherapy and radiation and has a 5-year survival rate of about 20%, said Ramalingam.
The phase III PACIFIC trial compared durvalumab as consolidation therapy with placebo among patients whose disease had not progressed after 2 or more cycles of platinum-based chemoradiotherapy. Participants who received durvalumab (n = 473) demonstrated a median PFS of 16.8 versus 5.6 months for those who received placebo (n = 236; HR, 0.52; 95% CI, 0.42-0.65; P <.0001). The response rate also was higher with durvalumab than with placebo (28.4% vs 16.0%; P <.001), as was the median duration of response (72.8% vs 46.8% with an ongoing response at 18 months).12
Besse said the results are “quite impressive” and that durvalumab can be considered the new standard of care for this population. Scagliotti agreed. “The PACIFIC study is filling the gap in a segment of the treatment for non—small cell lung cancer that was totally empty for 20 years,” he said. Additionally, the addition of durvalumab for this population would be a form of maintenance therapy, Scagliotti observed. “It’s still a dream but [it’s helping] step by step to make lung cancer a type of chronic disease.”