New England Journal of Medicine November 10, 2016

Pembrolizumab versus Chemotherapy for PD-L1–Positive Non–Small-Cell Lung Cancer

Martin Reck, Delvys Rodríguez-Abreu, Andrew G. Robinson, et al.

Bottom Line

In previously untreated patients with advanced non-small-cell lung cancer and PD-L1 expression on at least 50% of tumor cells, first-line pembrolizumab significantly prolonged progression-free and overall survival with fewer adverse events compared to platinum-based chemotherapy.

Key Findings

1. Median progression-free survival (PFS) was significantly longer with pembrolizumab compared to chemotherapy (10.3 vs. 6.0 months; HR 0.50, 95% CI 0.37-0.68; P<0.001) [1.2].
2. The estimated 6-month overall survival (OS) rate was higher in the pembrolizumab group (80.2%) than in the chemotherapy group (72.4%), representing a 40% reduction in the risk of death (HR 0.60, 95% CI 0.41-0.89; P=0.005).
3. The objective response rate (ORR) was 44.8% for pembrolizumab versus 27.8% for chemotherapy.
4. Grade 3, 4, or 5 treatment-related adverse events were substantially less frequent with pembrolizumab (26.6%) compared to chemotherapy (53.3%).

Study Design

Design
RCT
Open-Label
Sample
305
Patients
Duration
11.2 mo
Median
Setting
Multicenter, international
Population Patients with previously untreated advanced non-small-cell lung cancer (NSCLC) with a PD-L1 tumor proportion score (TPS) ≥50% and no sensitizing EGFR mutations or ALK translocations.
Intervention Pembrolizumab 200 mg administered intravenously every 3 weeks for up to 35 cycles.
Comparator Investigator's choice of platinum-based doublet chemotherapy for 4 to 6 cycles (with optional pemetrexed maintenance for non-squamous histology).
Outcome Progression-free survival (PFS) as assessed by a blinded, independent central radiologic review.

Study Limitations

The open-label design could potentially introduce bias, although independent, blinded central review was used to assess the primary radiologic endpoint [1.3].
The findings specifically apply to a biomarker-selected subset of patients (approximately 30% of advanced NSCLC cases with PD-L1 TPS ≥50%) and cannot be generalized to those with lower PD-L1 expression.
A high crossover rate (43.7% of the chemotherapy arm crossed over to pembrolizumab upon progression) could confound long-term overall survival analyses, though the initial interim OS benefit remained robust.

Clinical Significance

KEYNOTE-024 established pembrolizumab monotherapy as the new standard of care for the first-line treatment of advanced NSCLC in patients with a PD-L1 TPS ≥50% and no targetable EGFR or ALK mutations. This landmark finding catalyzed the integration of upfront PD-L1 biomarker testing into routine clinical practice algorithms.

Historical Context

Prior to KEYNOTE-024, platinum-based doublet chemotherapy had been the undisputed standard of care for the first-line treatment of advanced NSCLC without targetable oncogenic driver mutations. While immune checkpoint inhibitors had already demonstrated a survival benefit in the second-line setting (e.g., KEYNOTE-010, CheckMate 017/057), KEYNOTE-024 was the pivotal trial showing that an anti-PD-1 therapy could surpass standard chemotherapy in the first-line setting for a highly selected PD-L1-positive population.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How does pembrolizumab exert its anti-tumor effect in non-small cell lung cancer, and why is testing for PD-L1 expression using the Tumor Proportion Score (TPS) biologically relevant before starting therapy?

Key Response

Pembrolizumab is an IgG4 monoclonal antibody that binds to the PD-1 receptor on T-cells, blocking its interaction with PD-L1 on tumor cells. This prevents T-cell exhaustion, allowing the immune system to recognize and destroy cancer cells. Testing for TPS >= 50% ensures the tumor relies on this specific immune evasion pathway, predicting a higher likelihood of response.

Resident
Resident

A 65-year-old patient with newly diagnosed metastatic adenocarcinoma of the lung has a PD-L1 TPS of 60%. Molecular testing reveals an EGFR exon 19 deletion. Based on the inclusion criteria and findings of KEYNOTE-024, should pembrolizumab monotherapy be the initial treatment of choice?

Key Response

No. KEYNOTE-024 specifically excluded patients with sensitizing EGFR mutations or ALK translocations. These patients typically have poor responses to single-agent immunotherapy and should receive targeted tyrosine kinase inhibitors (TKIs) like osimertinib as first-line therapy, regardless of their high PD-L1 expression level.

Fellow
Fellow

KEYNOTE-024 demonstrated a significant overall survival benefit for pembrolizumab despite a high crossover rate (over 50%) from the chemotherapy arm to pembrolizumab upon disease progression. How does this high crossover rate impact the interpretation of the OS data, and what does it suggest about the timing of immunotherapy?

Key Response

The fact that an OS benefit was still observed despite a >50% crossover rate heavily underscores the superiority of upfront immunotherapy. It suggests that delaying immunotherapy until after chemotherapy failure misses a critical window of immune competence or clonal evolution, emphasizing that sequence matters significantly in PD-L1 >= 50% NSCLC.

Attending
Attending

While KEYNOTE-024 established pembrolizumab monotherapy as a standard of care for TPS >= 50%, subsequent trials like KEYNOTE-189 showed benefits of chemo-immunotherapy regardless of PD-L1 status. How do you decide between pembrolizumab monotherapy versus chemo-immunotherapy in a highly symptomatic patient with a PD-L1 of 80%?

Key Response

In patients with very high PD-L1, monotherapy is often preferred to spare chemotherapy toxicity, as per KEYNOTE-024. However, in a highly symptomatic patient with high disease burden or impending organ compromise, adding chemotherapy (the KEYNOTE-189 approach) might be chosen for a faster, more reliable objective response and rapid cytoreduction, weighing the need for rapid symptom relief against added toxicity.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The trial utilized a stratified randomized design and an unstratified Kaplan-Meier analysis for survival curves, but employed a stratified Cox proportional-hazards model for hazard ratios. What are the statistical implications of using the stratification factors (e.g., ECOG performance status, histology) in the Cox model, and how might unmeasured confounding within the PD-L1 >= 50% threshold affect the true treatment effect?

Key Response

Stratification in the Cox model reduces variance and accounts for prognostic heterogeneity, yielding a more precise hazard ratio. However, because PD-L1 >= 50% represents a wide biological gradient, an uneven distribution of ultra-high expressers (e.g., 90-100%) between arms could confound results. A continuous assessment of PD-L1 as a covariate might reveal a nonlinear treatment effect.

Journal Editor
Journal Editor

The control arm in KEYNOTE-024 consisted of the investigator's choice of standard platinum-based chemotherapy, but maintenance chemotherapy was not mandated. How does this lack of standardized maintenance therapy in the control arm introduce a potential threat to the study's internal validity when assessing survival endpoints?

Key Response

If a significant proportion of the control group did not receive optimal maintenance therapy (such as pemetrexed for non-squamous histology), the control arm might have underperformed compared to optimal real-world standards. A strict reviewer would flag this as potentially exaggerating the comparative PFS and OS benefits of the experimental arm.

Guideline Committee
Guideline Committee

Based on the KEYNOTE-024 data regarding both survival benefit and adverse event profiles, how should guidelines (e.g., NCCN or ASCO) grade the recommendation for single-agent pembrolizumab in patients with advanced NSCLC, PD-L1 >= 50%, and no driver mutations, and what specific contraindications must be highlighted?

Key Response

Guidelines upgraded pembrolizumab to a Category 1, preferred recommendation for this specific subpopulation based on the robust OS benefit and lower toxicity profile compared to chemotherapy. The committee must explicitly caveat this recommendation by noting contraindications such as active autoimmune disease, underlying interstitial lung disease, or the presence of targetable driver mutations, where TKIs take precedence.

Clinical Landscape

Noteworthy Related Trials

2017

CheckMate 026

n = 541 · NEJM

Tested

Nivolumab monotherapy

Population

Stage IV or recurrent NSCLC with PD-L1 expression of 1 percent or greater

Comparator

Platinum-based chemotherapy

Endpoint

Progression-free survival

Key result: Nivolumab did not improve progression-free or overall survival compared to chemotherapy in previously untreated NSCLC.
2018

KEYNOTE-189

n = 616 · NEJM

Tested

Pembrolizumab plus platinum-based chemotherapy

Population

Metastatic nonsquamous NSCLC without EGFR or ALK mutations

Comparator

Placebo plus platinum-based chemotherapy

Endpoint

Overall survival and progression-free survival

Key result: The addition of pembrolizumab to standard chemotherapy significantly prolonged overall survival and progression-free survival.
2019

KEYNOTE-042

n = 1,274 · Lancet

Tested

Pembrolizumab monotherapy

Population

Locally advanced or metastatic NSCLC with PD-L1 TPS of 1 percent or greater

Comparator

Platinum-based chemotherapy

Endpoint

Overall survival

Key result: Pembrolizumab significantly improved overall survival compared with chemotherapy in patients with PD-L1 TPS of 1 percent or greater.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis