Immunotherapy with the PD1 inhibitor pembrolizumab acts as first-line therapy in patients with advanced NSCLC and PD-L1 expression ≥ 1% better than standard chemotherapy. Patients lived 4 to 8 months longer in the median and suffered fewer side effects. These results were presented by Gilberto Lopes, Sylvester Comprehensive Cancer Center at the University of Miami, Miami, at the plenary session of the 2018 ASCO Annual Meeting in Chicago on June 3, 2018.
"The treatment of advanced or metastatic non-small cell lung cancer (NSCLC) has changed very rapidly," Lopes explained, "Targeting substances can be used in patients with certain driver mutations and immunotherapeutics such as pembrolizumab are effective in a variety of clinical situations. Pembrolizumab can be used alone or in combination with chemotherapy. For example, pembrolizumab monotherapy in the KEYNOTE-024 study was 50 % superior to platinum-based chemotherapy in untreated patients with metastatic NSCLC and PD-L1 expression.
The KEYNOTE-042 study now investigated whether pembrolizumab is superior to 1 % platinum-based chemotherapy as first-line therapy in patients with PD-L1 expression. The Phase 3 study enrolled 1,274 untreated patients with NSCLC and PD-L1 expression ≥ 1% who did not have EGFR and ALK mutations. Randomized 637 patients were treated with pembrolizumab (200 mg every three weeks up to 35 cycles) and 637 patients with up to 6 cycles of carboplatin/paclitaxel or carboplatin/pemetrexed. Primary endpoints were overall survival (OS), secondary endpoints included progression-free survival (PFS), response rates (ORR) and tolerability.
Lopes presented the results of the second pre-planned interim analysis with a median follow-up time of 12.8 months. The average age of the patients was 63 years, about 70% were men. Nearly 30% of patients were recruited in East Asia. About 39% of patients suffered from squamous cell carcinoma, 78% had smoked. PD-L1 expression ≥ 50 % was found in 599 patients, ≥ 20 % in 818 patients.
In the group with PD-L1 expression ≥ 50 % the OS median was 12.2 months under chemotherapy and 20.0 months under pembrolizumab (hazard ratio 0.69, p = 0.0003). In a PD-L1 expression ≥ 20 % the median OS was 13.0 months under chemotherapy, 17.7 months under pembrolizumab (HR 0.77, p = 0.002) and in a PD-L1 expression ≥ 1 % it was 12.1 months with chemotherapy and 16.7 months with pembrolizumab (HR 0.81, p = 0.0018).
PFS was 50% with pembrolizumab in median 7.1 months in PD-L1 expression ≥, 6.4 months with chemotherapy (HR 0.81, p =0.017), 20% with pembrolizumab 6.2 months, 6.6 months with chemotherapy (HR 0.94) and 6.5 months with PD-L1 expression ≥ 1% with pembrolizumab 5.4 months, 6.5 months with chemotherapy.
Pembrolizumab was administered to 39.5% (PD-L1 expression ≥ 50%), 33.4% (PD-L1 expression ≥ 20%) and 27.3% (PD-L1 expression ≥ 1%) of patients, respectively, and chemotherapy to 32.0%, 28.9% and 26.5%. At 20.2 months, however, the response to pembrolizumab lasted much longer than with chemotherapy at 8.3 months.
Patients in the pembrolizumab group received 9 median doses, those in the chemotherapy group 6 doses. Adverse effects from grade ≥ 3 were observed in 17.8 % under pembrolizumab and in 41 % under chemotherapy. However, as expected, immune-mediated side effects were more frequent in the pembrolizumab group (27.8%) than in the chemotherapy group (7.2%).
These data show that pembrolizumab significantly improves OS in first-line treatment of patients with advanced or metastatic NSCLC with PD-L1 expression ≥ 1 % compared to platinum-based chemotherapy. The stronger the PD-L1 expression, the greater the benefit of pembrolizumab. Pembrolizumab did not have a significant effect on PFS, so the external data monitoring committee recommended that the study be continued.
"Keynote 042 is the first study with the primary endpoint OS showing the superiority of pembrolizumab over platinum-based chemotherapy in untreated patients with advanced or metastatic NSCLC with PD-L1 expression ≥ 1% but without EGFR mutations or ALK translocations," summarizes Lopes. He concluded: "These data confirm the role of pembrolizumab monotherapy as a standard in the first-line treatment of patients with PD-L1 expressing tumors.
Discutant Leena Gandhi, NYU Perlmutter Cancer Center, New York City, emphasized that PD-L1 inhibitors have "changed the landscape" but that it is not clear which patients should be treated with these substances. The distribution of PD-L1 expression in KEYNOTE 042 was as expected. However, the good effect of pembrolizumab on OS was mainly due to the group of patients with PD-L1 expression ≥ 50 %. Gandhi underwent a trial according to the ASCO guidelines for a significant outcome in clinical trials. According to ASCO's recommendations, an improvement in OS of 3.25 to 4 months and a hazard ratio of 0.76 to .08 should be achieved in non-plate-epithelial carcinoma; in KEYNOTE 042, HR was 0.86. In squamous cell carcinoma, an HR of 0.77 to 0.8 should be achieved, which was 0.75 in KEYNOTE 042. The costs are another factor, according to Gandhi in the USA for a 12-week treatment with pembrolizumab are a good 38,000 US dollars, for carboplatin/paclitaxel 350.00 US dollars.
Biomarkers play an important role in the choice of therapy. The determination of PD-L1 expression could select a population of patients who could benefit from first-line treatment with pembrolizumab, but not those who would not benefit from the treatment. There would also be differences between the different assays. The mutation load of the tumor (TMB) as a marker could usefully complement the marker PD-L1 expression. Together they would have an additive predictive significance. A further refinement of the biomarkers will allow better predictions in the future as to which patient would best benefit from which therapy.
Gandhi no longer sees chemotherapy alone as the standard in the first-line treatment of NSCLC, but pembrolizumab alone is not the standard for all patients. The higher the PD-L1 expression, the better the effect of pembrolizumab. PD-L1 is a suitable, though not a perfect biomarker and should be used. She summarized her statements as follows: "Lung cancer is no longer one size fits all: Uses biomarkers to select the best individual option for each individual patient".
Lopes G et al. Pembrolizumab versus platinum-based chemotherapy as first-line therapy for advanced/metastatic NSCLC with a PD-L1 tumor proportion score (TPS) ≥ 1%: Open-label, phase 3 KEYNOTE-042 study. 2018 ASCO Annual Meeting, Chicago, June 1-5, 2018, Abstract LBA4. https://meetinglibrary.asco.org/record/165950/abstract