Atezolizumab First-Choice Immunotherapy in Second-Line NSCLC?

Many experts have long considered the various immune checkpoint inhibitors targeting PD-1 or PD-L1 as essentially clinically interchangeable options that have not readily distinguished themselves. Perhaps because the PD-1 inhibitors nivolumab and pembrolizumab were previously approved in 2015 as second-line therapies for advanced non-small cell lung cancer (NSCLC)—nivolumab regardless of PD-L1 expression, pembrolizumab only in patients with a threshold level of tumor PD-L1 expression—the clinical world wasn't abuzz with excitement following a recent press release announcing that the PD-L1 inhibitor atezolizumab had also been superior to docetaxel in the randomized phase 3 OAK trial of previously treated patients.^[1] Notably, we had already seen the same outcome in earlier trials with nivolumab^[2,3] and pembrolizumab,^[4] as well as the smaller phase 2 POPLAR trial of atezolizumab vs docetaxel as second-line therapy for patients with advanced lung cancer.^[5]

Without an acute need for yet another remarkably similar agent with a very similar indication, how might atezolizumab distinguish itself as a compelling treatment option with an incremental benefit over the agents that preceded it in this space?

The OAK Trial

Let's review the results from the OAK trial, presented by Fabrice Barlesi, MD, PhD, head of Multidisciplinary Oncology and Therapeutic Innovations at APHM (Assistance Publique - Hopitaux de Marseille), at the 2016 European Society for Medical Oncology (ESMO) meeting,^[6] to see how atezolizumab has emerged as what I would now narrowly favor as my preferred choice among immune checkpoint inhibitors for second-line therapy in previously treated patients with advanced NSCLC.

The OAK trial enrolled 1225 patients with locally advanced or metastatic NSCLC and a performance status of 0-1 who had received one to two prior lines of chemotherapy, including a platinum-based therapy, without any required level of PD-L1 expression. It randomly assigned them to either atezolizumab at a fixed dose of 1200 mg IV every 3 weeks or docetaxel at the standard dose of 75 mg/m² IV every 3 weeks. Treatment was continued until progressive disease by standard RECIST criteria or, for atezolizumab, loss of clinical benefit as defined by the treating investigator. The two co-primary endpoints were overall survival (OS) in the intent-to-treat (ITT) population, as well as OS in patients with > 1% PD-L1 expression on tumor cells (TC) or host immune cells (IC) using the SP142 immunohistochemistry assay as the relevant biomarker; secondary endpoints included objective response rate (ORR), progression-free survival (PFS), duration of response (DoR), and safety. This presentation was based on a prespecified analysis on the first 850 patients enrolled.

Atezolizumab was associated with a significantly superior overall survival, with a hazard ratio (HR) of 0.73 favoring the atezolizumab arm (P = .0003) and a longer median OS by more than 4 months (13.8 vs 9.6 months). Landmark analyses showed far more patients alive on the atezolizumab arm at landmark time points of 12 months (55% vs 41%) and 18 months (40% vs 27%).

These OS results were also broken down by level of PD-L1 expression. Among patients with PD-L1 expression of > 1% on TC or IC, which accounted for 55% of enrolled patients, atezolizumab was associated with an OS benefit comparable to that of the entire trial population, with an HR of 0.74 (P = .0102) and a median OS of 15.7 vs 10.3 months for recipients of atezolizumab or docetaxel, respectively.

Results were slightly less favorable in both arms among the 45% of patients with < 1% PD-L1 expression, though atezolizumab was still associated with an HR of 0.75 for OS (HR, 0.0205) and a median OS of 12.6 vs 8.9 months. The OS difference was especially marked in the 16% of patients with the highest level of TC or IC staining for PD-L1 (designated as TC3 or IC3), in whom the HR for OS was 0.41 (P < .0001) and median OS was 20.5 vs 8.9 months.

The atezolizumab OS benefit was identical across NSCLC histologies, with an HR for OS of 0.73 for both squamous and non-squamous NSCLC. Subgroup analyses revealed that all clinical variables showed similar benefit in favor of atezolizumab, except for the approximately 10% of patients with an activating EGFR mutation, in whom the HR for OS was 1.24, a nonsignificant trend that favored docetaxel.

There was no cross-over design in the trial, and only 17% of patients on the docetaxel arm received subsequent immunotherapy upon progression.

Interestingly, there was no significant difference in PFS in the broad trial population, with an HR of 0.95, but there were clear differences based on PD-L1 expression; specifically, the HR for PFS was 0.63 in the patients with TC3 or IC3, 0.76 in patients with TC2/3 or IC2/3, 0.91 for patients with TC1/2/3 or IC 1/2/3, and 1.00 for patients with TC0 and IC0.

The comparison of ORR was similar, with no difference by overall ITT (14% vs 13% for atezolizumab and docetaxel, respectively) but ORR of 31% vs 11% favoring atezolizumab for the TC3 threshold or IC3 population, 22% vs 13% for TC2/3 or IC2/3 patients, 18% vs 16% for TC1/2/3 or IC1/2/3, and 8% vs 11% for patients with no PD-L1 staining by TC or IC measurement. Duration of response was far longer with atezolizumab, at 16.3 months, compared with 6.2 months for docetaxel in the ITT population.

Safety also favored atezolizumab by a wide range of measures. Among the most relevant, 43% vs 15% of patients demonstrated treatment-related grade 3-4 adverse events (AEs) and 19% vs 8% experienced AEs leading to treatment withdrawal on docetaxel and atezolizumab, respectively. Nearly all specific AEs were more common with docetaxel, and immune-mediated AEs such as pneumonitis, hepatitis, and colitis were 1% or less in both arms, with no significant differences between them.

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.