Background

Immunotherapy with or without chemo has improved survival vs chemo in 1L aNSCLC. NIVO + chemo showed encouraging activity in a phase 1 study in this setting. CheckMate 227 is a multi-part, randomized, open-label, phase 3 study evaluating NIVO-based regimens vs chemo. We present final results from Part 2, which evaluated NIVO + chemo vs chemo in 1L aNSCLC.

Methods

Pts (N = 755) with chemo-naive, stage IV or recurrent NSCLC, ECOG PS 0–1, and no sensitizing EGFR/ALK alterations were randomized 1:1 to receive every 3 weeks NIVO 360 mg + chemo or chemo. Pts were stratified by histology (squamous [SQ] vs nonsquamous [NSQ]), sex, and PD-L1 expression (< 1% vs ≥ 1%). Chemo was histology-based and continued for up to 4 cycles; pts with NSQ NSCLC could receive pemetrexed maintenance. Pts were treated until progression, unacceptable toxicity, or for 2 years for NIVO. The primary endpoint was overall survival (OS) with NIVO + chemo vs chemo in NSQ NSCLC. OS in all randomized pts (NSQ and SQ) was a secondary hierarchical endpoint.

Results

Baseline characteristics were generally balanced. Minimum follow-up was 19.5 mo. In pts with NSQ NSCLC, no statistically significant improvement in OS was seen with NIVO + chemo vs chemo (HR, 0.86 [95.62% CI, 0.69–1.08; P = 0.1859]); median OS was 18.8 mo vs 15.6 mo; 12-mo OS rates were 67.3% vs 59.2%. HR for OS was 0.81 (95% CI, 0.67–0.97) in all randomized pts; 0.69 (95% CI, 0.50–0.97) in pts with SQ NSCLC. Progression-free survival and objective response rates favored NIVO + chemo in NSQ, SQ, and all randomized pts (Table). Grade 3–4 tx-related adverse events occurred in 45% and 35% of all pts treated with NIVO + chemo and chemo, respectively.

[Note: Table emailed to Federica Frigerio]

Conclusion

CheckMate 227 Part 2 did not meet the primary endpoint of OS for NIVO + chemo vs chemo in NSQ NSCLC. Descriptive analyses showed longer OS with NIVO + chemo in all randomized pts and SQ NSCLC. No new safety signals were observed.

Clinical trial identification

NCT02477826; Release date: June 23, 2015

Editorial acknowledgement

Writing and editorial assistance was provided by Namiko Abe, PhD, of Caudex, funded by Bristol-Myers Squibb.

Background

The Phase III IMpower110 study (NCT02409342) is evaluating atezo (anti–PD-L1) monotherapy as 1L treatment (tx) in PD-L1–selected patients (pts) with NSCLC independent of tumour histology. IMpower110 met its primary endpoint with significant OS improvement in PD-L1–high (TC3 or IC3; ≥ 50% tumour cell [TC] or ≥ 10% tumour-infiltrating immune cell [IC]; VENTANA SP142 IHC assay) wild-type (WT; EGFR/ALK-negative) pts (HR, 0.59 [95% CI: 0.40, 0.89]; P = 0.0106). Efficacy analyses in prespecified biomarker subgroups by the SP263 and 22C3 PD-L1 IHC assays and bTMB are reported.

Methods

IMpower110 enrolled 572 chemo-naive pts with stage IV NSCLC, PD-L1 ≥ 1% TC or IC (TC1/2/3 or IC1/2/3; SP142), measurable disease by RECIST 1.1 and ECOG PS 0-1. Pts were randomised 1:1 to receive atezo 1200 mg IV q3w or platinum-based chemo (4 or 6 21-day cycles). OS (primary endpoint) was tested hierarchically in WT pts. Additional analyses included OS and PFS in the SP263 and 22C3 PD-L1 IHC and bTMB populations. PD-L1 cutoffs of ≥ 1% and ≥ 50% tumour proportion score (TPS) for 22C3 and ≥ 1% and ≥ 50% TC for SP263 were evaluated; bTMB cutoffs were ≥10, ≥ 16 and ≥ 20.

Results

Biomarker-evaluable populations (BEP) in the TC1/2/3 or IC1/2/3 WT population (SP142; 554) included 534 (22C3), 546 (SP263) and 389 (bTMB) pts. Baseline characteristics in the IHC and bTMB BEP subgroups were generally balanced. OS and PFS in the PD-L1–high (TC3 or IC3; ≥ 50% TPS; ≥ 50% TC) subgroups favoured atezo (OS data shown in table). Stepwise OS and PFS improvement, favouring atezo, was seen up to bTMB ≥ 16; no further benefit was seen at ≥ 20.

Conclusion

Pt subgroups defined as PD-L1–high by all 3 IHC assays (SP142, 22C3, SP263) had similar OS and PFS benefit with atezo, despite the different assay sensitivities and scoring algorithms. Enrichment in clinical benefit, favouring atezo, was also seen in bTMB positive subgroups. Atezo monotherapy is a potential new 1L tx option for pts with PD-L1–high NSCLC.

Clinical trial identification

NCT02409342

Editorial acknowledgement

Medical writing assistance for this abstract was provided by Kia C. E. Walcott, PhD, of Health Interactions, and
funded by F. Hoffmann-La Roche, Ltd

Background

Somatic KRAS mutations are detected in approximately 15%-30% of lung adenocarcinomas, with regional variation, and are associated with poor prognosis. We explored the prevalence of KRAS mutations and their association with efficacy in participants (pts) with nonsquamous NSCLC enrolled in the KEYNOTE-042 study of pembrolizumab monotherapy vs platinum-based chemotherapy as first-line therapy for advanced PD-L1-positive (TPS ≥1%) NSCLC (NCT02220894).

Methods

KRAS mutational status and tumor mutational burden (TMB) were assessed by whole-exome sequencing (WES) in pts with who had available tumor and matched-normal tissue. This exploratory analysis included descriptive analyses of the correlation between KRAS mutational status and shifts in distributions of TMB and PD-L1 expression and the association of KRAS and KRAS G12C status with efficacy.

Results

Of the 782 pts with nonsquamous histology, 301 (38%) were evaluable by WES and had matched tumor and normal DNA. KRAS mutations were identified in 69 (23%) pts, including 29 (10%) G12C carriers. Pts with vs without KRAS mutation tended to have higher PD-L1 TPS (median [IQR] 60% [10-95] vs 35% [10-80]) and TMB (median [IQR] 191 [129-288] vs 105 [56-226] mut/exome). Outcomes of pembrolizumab and of chemotherapy for pts with and without KRAS mutation and for KRAS G12C carriers are in the Table. Of note, CIs were wide given the modest frequency of KRAS mutation and low frequency of KRAS G12C.

Conclusion

Findings of this descriptive exploratory analysis suggest that pembrolizumab monotherapy should be considered as a standard first-line treatment option for PD-L1-positive advanced nonsquamous NSCLC regardless of KRAS mutational status. These findings also suggest that a pembrolizumab-containing regimen is a clinically relevant comparator for studies of KRAS -targeted therapy given as first-line treatment of NSCLC.

Clinical trial identification

KEYNOTE-042; NCT02220894

Editorial acknowledgement

Joanne Tomassini and Melanie Leiby, both of Merck & Co., Inc., Kenilworth, NJ, for writing support.