preheader BJMO

Banner website

Significant survival benefit with atezolizumab over docetaxel in patients with previously treated non-small cell lung cancer

Results of the phase III OAK study demonstrated a significant survival benefit of the PD-L1 inhibitor atezolizumab over docetaxel in previously treated patients with locally advanced or metastatic NSCLC. In this study, atezolizumab reduced the risk of death by 27% as compared to docetaxel, the current standard of care in this setting. The overall survival (OS) benefit with atezolizumab was seen regardless of the PD-L1 expression level (also in patients without PD-L1 expression), but was most pronounced in patients in the highest tertile of PD-L1 expression. In these patients, the median OS was increased from 8.9 months with docetaxel to 20.5 months with atezolizumab, corresponding with a 59% reduction in the risk of dying. Moreover, atezolizumab was well tolerated with a more favorable toxicity profile than what is seen with docetaxel.1

The OAK trial represents the first phase III trial with a PD-L1 directed antibody. Atezolizumab inhibits the binding of PD-L1 to its receptors PD1 and B7.1 and restores the tumor-specific T-cell immunity, while leaving the PD-L1/PD-L2 interaction intact. In the phase II POPLAR study, atezolizumab previously demonstrated a survival benefit over docetaxel in previously treated advanced NSCLC patients.2 In the phase III OAK trial, a total of 1,225 patients with locally advanced or metastatic NSCLC, who received one or two prior treatments, were treated with atezolizumab (1200mg IV q3w) or docetaxel (27mg/m2 q3w). Patients in the study were stratified by PD-L1 expression, histology and prior chemotherapy regimen. The primary endpoint of the trial was OS in the intent-to-treat (ITT) population, with OS in patients with PD-L1 expression (PD-L1 expression in 1% or more of the tumor [TC] or tumor infiltrating immune cells [IC]) as co-primary endpoint.

In the presented preliminary analysis, including data from 850 patients, atezolizumab was shown to be associated with a 27% reduction in the risk of dying as compared to docetaxel (median OS 13.8 versus 9.6 months; HR[95%CI]: 0.73[0.62-0.87]; p= 0.0003). In the subgroup of patients with PD-L1 expression on at least 1% of IC or TC cells (representing 55% of the ITT population), the median OS was 15.7 months with atezolizumab as compared to 10.3 months with docetaxel (HR[95%CI]: 0.74[0.58-0.93]; p=0.0102). The benefit of atezolizumab over docetaxel was seen regardless of the PD-L1 expression, but increased with an increasing PD-L1 expression. In fact, among patients in the highest tertile of PD-L1 expression, atezolizumab invoked a 59% reduction in the risk of death when compared to docetaxel (median OS: 20.5 versus 8.9 months; HR[95%CI]: 0.41[0.27-0.64]; p< 0.0001). In patients not expressing PD-L1, the median OS was 12.6 months with atezolizumab as compared to 8.9 months with docetaxel (HR[95%CI]: 0.75[0.59-0.96]; p= 0.0215). The observed OS benefit with atezolizumab was similar for patients with both squamous (HR[95%CI]: 0.73[0.54-0.98]; p= 0.0383) and non-squamous (HR[95%CI]: 0.73[0.60-0.89]; p= 0.0015) histology.

Looking at the secondary endpoints, no significant difference was seen in progression-free survival (2.8 months with atezolizumab versus 4.0 months with docetaxel), or objective response rate (13.6% with atezolizumab versus 13.4% with docetaxel) between both treatment arms. The duration of response was significantly longer with atezolizumab at 16.3 months as compared to 6.2 months with docetaxel (p< 0.001). The median duration of treatment was 3.4 months with atezolizumab and 2.1 months with docetaxel. Interestingly, one in five patients (20.5%) was on treatment with atezolizumab for more than 1 year.

The incidence of treatment-related adverse events was lower with atezolizumab (64%) than with docetaxel (86%). High-grade (grade 3 or more) adverse events were seen in 15% of patients in the atezolizumab arm compared to 43% in the docetaxel treated population. The incidence of adverse events leading to treatment withdrawal was 8% with atezolizumab and 19% with docetaxel. No new safety signals were observed for atezolizumab.

In summary, the results of this phase III trial confirm the positive results reported with atezolizumab in the phase II POPLAR study, with a significant OS benefit, regardless of the PD-L1 expression. Moreover, the toxicity profile of atezolizumab compared favorably to that of docetaxel. As such atezolizumab represents a new second-line therapeutic strategy for patients with advanced NSCLC.


  1. Barlesi F, Park K, Ciradiello F et al. Primary analysis of OAK, a randomized phase III study comparing atezolizumab with docetaxel in 2L/3L NSCLC. Presented at ESMO 2016; Abstract LBA44_PR.
  2. Fehrenbacher L, Spira A, Ballinger M et al. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. The Lancet 2016; 387(10030):1837-46.

Speaker Fabrice Barlesi


Fabrice Barlesi, MD, PhD,
Multidisciplinary Oncology and Therapeutic Innovations Department, Aix-Marseille University, Marseille, France


See: Keyslides

Back to Top