November 2015, Vol. 2, No. 6

← Back to Issue

Another Immunotherapy for NSCLC?

European Cancer Congress

Immunotherapy is poised to become a game changer for patients with non–small cell lung cancer (NSCLC). Nivolumab is now approved by the FDA for the treatment of NSCLC, and atezolizumab is the second checkpoint inhibitor to show promise as second- line or later therapy in NSCLC.

Results of the POPLAR and BIRCH trials, presented separately at the 2015 European Cancer Congress, were very encouraging, and in both trials programmed death-1 ligand 1 (PD-L1) was found to be a predictive biomarker for response.

POPLAR enrolled all comers, whereas BIRCH enrolled only patients who expressed PD-L1.


The randomized, phase 2 POPLAR trial was mounted to compare atezolizumab, a PD-L1 antibody, to standard therapy with docetaxel in 287 patients with advanced NSCLC that progressed on prior platinum therapy. Treatment with atezolizumab was continued until loss of clinical benefit and until disease progression in the docetaxel arm. The primary analysis was on efficacy, safety, and predictive biomarkers.

“POPLAR demonstrated significant improvements in overall survival for unselected patients treated with atezolizumab versus docetaxel, reflecting a 27% improvement. Higher PD-L1 expression was associated with improved overall survival,” said lead author Johan Vansteenkiste, MD, PhD, University Hospitals Leuven, Belgium.

PD-L1 expression was assessed by a proprietary assay called SP142 that works on both tumor cells (TCs) and immune cells (ICs). POPLAR study subjects were stratified according to PD-L1 expression: TC3 and/or IC3 = high expressors; TC2/3 and/or IC2/3 = intermediate to high expressors; TC1/2/3 and/or IC1/2/3 = “any expressors,” and TC0 and IC0 = nonexpressors.

At a minimum follow-up of 13 months, overall survival (OS) was significantly improved in patients treated with atezolizumab: median OS was 12.6 months with atezolizumab versus 9.7 months with docetaxel (P = .04).

Median OS was higher in patients categorized as TC3 or IC3 (high expressors). The magnitude of survival benefit correlated with level of PD-L1 expression, with a 41% improvement with atezolizumab in patients categorized as TC3 and IC3 and no OS benefit in nonexpressors.

Progression-free survival also correlated with the intensity of PD-L1 expression.

Patients taking atezolizumab exhibited fewer treatment-emergent grade 3/4 adverse events: 11% with atezolizumab versus 39% with docetaxel. The rate of grade 5 adverse events was 4% in both treatment arms. Withdrawals due to treatment-related adverse events were higher in the docetaxel arm: 22% versus 8% for atezolizumab.


The BIRCH trial was a single-arm study of first-line or subsequent therapy that enrolled only PD-L1–positive patients with advanced or metastatic NSCLC (brain metastasis was an exclusion criteria).

“The BIRCH trial found clinically meaningful results with atezolizumab monotherapy in PD-L1–selected patients with advanced NSCLC. The majority of responses are ongoing. Overall survival data are not mature. Six-month overall survival correlates with POPLAR results in second- and third-line therapy,” said lead author Benjamin Besse, MD, Institut Gustave Roussy, Villejuif, France.

Similar to POPLAR, higher levels of PD-L1 correlated with higher response rates in BIRCH.

The study enrolled 667 patients: cohort 1 was treated with first-line atezolizumab until disease progression (n = 142); cohort 2 was treated with second-line atezolizumab after 1 prior platinum-containing therapy (n = 271); and cohort 3 was treated with third-line and higher atezo­lizumab (n = 254).

Using the SP142 assay to assess PD-L1 expression, patients were classified as TC2/3 and/or IC2/3. Forty-six percent of the overall population was classified as high PD-L1 expressors (ie, TC3 or IC3).

The overall response rate with atezolizumab was 19% in cohort 1 and 17% in cohorts 2 and 3. Response rates were higher among high PD-L1 expressors: 27%, 24%, and 26%, in cohorts 1, 2, and 3, respectively. The majority of responses are ongoing.

Six-month OS was 82%, 76%, and 71% in cohorts 1, 2, and 3, respectively, having TC2/3 or IC2/3 expression levels and by 79%, 80%, and 75%, respectively, of patients in cohorts 1, 2, and 3 having TC3 or IC3 expression levels.

Grade 3/4 adverse events were reported in 11% of patients enrolled in the trial. The rate of discontinuations due to treatment-related events was 5%. Common adverse events include fatigue, diarrhea, nausea, pruritus, pyrexia, and decreased appetite.

Adverse events of special interest occurred in 26% of patients; most were grade 1/2. The most frequently occurring grade 3/4 adverse events of special interest were pneumonitis (1.5%), increased aspartate aminotransferase (0.8%), colitis (0.5%), and hypothyroidism and rash (0.3% each). One grade 5 event occurred (pneumonitis).

More ongoing studies of atezolizumab are being conducted in first- and second-line settings, and results are eagerly awaited.

PD-L1 needs to be validated in a phase 3 trial that enrolls unselected patients, according to formal discussant Luis Paz-Ares, MD, PhD, Professor of Medicine at the Hopital Universitario 12 de Octubre, Madrid, Spain.

POPLAR and BIRCH are part of a large development program at Roche. Results from a phase 3 trial in unselected patients with NSCLC (OAK trial) are expected within 1 year.

16th International Lung Cancer Congress - November 9, 2015

Looking into the Future of Lung Cancer Research

Each year, the Bonnie J. Addario Lung Cancer Foundation (ALCF) recognizes luminaries in the quest to cure lung cancer with the Addario Lectureship Award. This year, Fred R. Hirsch, MD, PhD, Professor of Medicine and Pathology, University of Colorado School of Medicine, Denver, delivered his keynote lecture on the many [ Read More ]

16th International Lung Cancer Congress - November 9, 2015

The Next Generation of Agents for Lung Cancer

At the 16th Annual International Lung Cancer Congress, Tony Mok, MD, Professor of Clinical Oncology, The Chinese University of Hong Kong, provided an overview of the next generation of agents, taking stock of the recent past for insights into the future of drug discovery. Drug discovery is rarely, if ever, [ Read More ]