Kantar Health Blog

Seeing the Forest from the Trees: Atezolizumab in the Rapidly Shifting NSCLC Landscape

by Stephanie Hawthorne | Sep 28, 2015
Stephanie Hawthorne
Co-authored by Len Kusdra, Ph.D. 

Without a doubt, PD-1 and PD-L1 inhibitors have taken the oncology field by storm. With its approval in Japan in July 2014 for unresectable melanoma, Opdivo® (nivolumab, Bristol-Myers Squibb/Ono Pharmaceuticals) was the first PD-1 inhibitor to be approved worldwide and represented the culmination of a longstanding quest to harness the body’s inherent immune system to combat cancer. With the proverbial “cat out of the bag,” the approval also sparked a furious race between Opdivo and its closest competitor, Keytruda® (pembrolizumab, Merck), with the U.S. Food and Drug Administration (FDA) granting Keytruda accelerated approval in relapsed unresectable/advanced melanoma (September 2014) and subsequently approving Opdivo in the same setting (December 2014). All eyes turned to non-small cell lung cancer (NSCLC) in March 2015, when Opdivo became first-in-class in this indication with its approval in second-line squamous histology, and Merck quickly followed suit filing for Keytruda in NSCLC (PDUFA date: October 2, 2015). It seemed initially that these two agents would dominate the field of immuno-oncology; however, agents such as atezolizumab (Roche/Genentech), avelumab (Merck Serono/Pfizer), and durvalumab (MEDI4736, MedImmune/AstraZeneca) have entered the fray, and their entry into late-stage development, particularly in NSCLC, has intensified competition with no signs of abating in the near future. A question (among many) now becomes how will all these agents equilibrate in the market and what factors will guide physicians in their treatment paradigms for NSCLC?

Atezolizumab is Roche/Genentech’s gateway into this hot area. While a slight latecomer, atezolizumab is one of the most advanced immuno-oncology agents, representing a complementary mechanism by being an inhibitor of the PD-L1 ligand. The companies have initiated a broad NSCLC development program for atezolizumab, with three pivotal Phase II trials (FIR, NCT01846416; POPLAR, NCT01903993; and BIRCH, NCT02031458) and one Phase III trial (OAK, NCT02008227) for atezolizumab in second-line NSCLC in an effort to catch up with Opdivo and Keytruda; recently five Phase III trials also were initiated in the first-line setting as well as another Phase III trial in the adjuvant setting. FIR (n=138) and BIRCH (n=667) are single-arm trials of atezolizumab in PD-L1-selected patients in metastatic NSCLC. In both trials, atezolizumab was given at 1200 mg/kg IV every three weeks, and the primary endpoint was objective response rate (ORR). POPLAR (n=287) was a randomized Phase II trial that enrolled all-comers with relapsed metastatic NSCLC, with an accompanied stratification by PD-L1 expression; in this trial, atezolizumab (1200 mg/kg IV every three weeks) was compared with standard-of-care chemotherapy docetaxel (75 mg/m2 IV every three weeks). Roche has consistently guided that these three Phase II trials will form the basis for regulatory applications in the U.S. and Europe.

Interim results for POPLAR and FIR were first reported at the American Society of Clinical Oncology (ASCO) 2015 annual meeting.1,2 Patients in POPLAR who had higher levels of PD-L1 expression in either the tumor (TC1/2/3) or immune cells (IC1/2/3) showed increased overall survival with atezolizumab compared with docetaxel (median not reached versus 9.1 months (HR 0.63, p=0.024)) and the magnitude of overall survival (OS) benefit was greater for patients who more strongly expressed PD-L1 (immunohistochemistry score (IHC) 2/3, HR 0.56; IHC3, HR 0.46); this survival benefit did not significantly differ among patients with TC or IC IHC0. The FIR trial showed that patients expressing a high level of PD-L1 (TC or IC IHC2/3) exhibited an ORR of 29% for first-line patients, 17% for second-line or later patients without brain metastases, and 23% for second-line or later patients with brain metastases. After POPLAR and FIR reported data at ASCO 2015, the last of the Phase II trials, BIRCH, was highly anticipated. In August 2015, Roche announced that BIRCH had met its primary endpoint and induced tumor shrinkage in patients who expressed PD-L1, and that PD-L1 expression correlated with response; the results were presented Sunday at the 2015 European Cancer Congress in Vienna.3

The BIRCH trial evaluated atezolizumab in patients with locally advanced or metastatic NSCLC who expressed high levels of PD-L1 on either TC or IC, as determined by the VENTANA SP142 IHC assay. TC2/3 or IC2/3 was equal to or greater than 5% of PD-L1-positive TCs or ICs, while TC3 or IC3 was equal to or greater than 50% TCs or 10% ICs expressing PD-L1. Patients were stratified into three cohorts: Cohort 1 had received no prior chemotherapy (n=142), Cohort 2 had received one prior platinum chemotherapy (n=271), and Cohort 3 had received at least two prior chemotherapies including one platinum (n=254). Patients were treated until progressive disease or until loss of clinical benefit. The ORR in all three patient cohorts correlated with level of PD-L1 expression: ORR for IHC2/3 versus IHC3 was 19% versus 26% in Cohort 1; 17% versus 24% in Cohort 2; and 17% versus 27% in Cohort 3. Median PFS was 5.5 months, 2.8 months and 2.8 in Cohorts 1, 2, and 3 in patients with TC-IC2/3, while a similar trend was seen in patients with TC-IC3 (Cohort 1 5.5 months, Cohort 2 4.1 months, Cohort 3 4.2 months). OS data is not yet mature; however, six-month OS rates in Cohorts 1, 2, and 3 were 82%, 76% and 71%, respectively, in IHC2/3 patients and 79%, 80%, and 75% in Cohorts 1-3 in the IHC3 population. Adverse events seen with atezolizumab were similar to those seen in other studies, with the most common Grade 3/4 adverse events being rash, pneumonitis, elevated liver transaminases and colitis, all which had a frequency less than 2% each.

Roche has guided that it plans to file for regulatory approval in the U.S. and EU in 2016; in light of the Breakthrough Status granted by the FDA, combined results from FIR, POPLAR, and BIRCH should support an accelerated approval of atezolizumab in patients with PD-L1-positive, relapsed metastatic NSCLC. A question that stands out is how Roche will differentiate atezolizumab from its competitors and how will physicians adopt atezolizumab in their clinical practice. In terms of entry into market, Opdivo is ahead of atezolizumab, allowing Opdivo to establish itself in clinical practice. While cross-trial comparisons should always be made with caution, the survival curves in Opdivo’s CheckMate-017 and CheckMate-057 trials in second-line NSCLC and atezolizumab’s POPLAR trial in second-line NSCLC indicate very similar levels of efficacy between Opdivo and atezolizumab, so similar that the BIRCH trial discussant, Dr. Luiz Paz-Ares, stated that the trials “confirmed each other.” Given the similar level of benefit, it is unlikely atezolizumab will compete with Opdivo based on efficacy alone. Additionally, Keytruda is expected to receive FDA approval in relapsed NSCLC very shortly, thus putting atezolizumab third to market in the U.S.

One point of differentiation is in patient selection. The data showing increased response in patients with higher PD-L1 supports a regulatory filing in PD-L1-positive patients for atezolizumab, which is also supported by its FDA Breakthrough Therapy designation in this patient type; Opdivo is unlikely to have such a biomarker-restricted label, and whether or not Keytruda’s label will include a biomarker restriction is an area of hot debate. If atezolizumab becomes the first of these drugs to be approved with a PD-L1-positive biomarker limitation in its label, this could prove advantageous. Physicians may shift their preference toward atezolizumab if the efficacy is more pronounced in this patient population or if payers strictly enforce PD-L1 biomarker testing and subsequently choice of PD-1 or PD-L1 inhibitor. The loss of a fraction of patients who are PD-L1-negative if a restricted label is granted may be counterbalanced by the gain of greater market share among patients who are PD-L1-positive than might have been obtained by competing against two other drugs in an all-comers population.

Many questions remain (largely around implementation of the PD-L1 diagnostic), and physicians will soon find themselves with many arguably similar therapeutic choices for their relapsed NSCLC patients. Until more definitive data becomes available to better guide immuno-oncology treatment decisions, physicians may be left analyzing individual “trees” before the forest that is the clinical paradigm for immuno-oncology products can be fully understood.


  1. Spira AI, Park K, Mazieres J, et al. Efficacy, safety and predictive biomarker results from a randomized phase II study comparing MPDL3280A vs docetaxel in 2L/3L NSCLC (POPLAR). J Clin Oncol. 2015;33(suppl; abstr 8010).
  2. Spigel DR, Chaft JE, Gettinger SN, et al. Clinical activity and safety from a phase II study (FIR) of MPDL3280A (anti-PDL1) in PD-L1–selected patients with non-small cell lung cancer (NSCLC). J Clin Oncol. 2015;33(suppl; abstr 8028).
  3. Besse B, et al. Phase II, single-arm trial (BIRCH) of atezolizumab as first-line or subsequent therapy for locally advanced or metastatic PD-L1-selected non-small cell lung cancer (NSCLC). Vienna, Austria: European Cancer Congress; 2015. Abstract 16LBA.

Leave a comment