PF-06439535 (a Bevacizumab Biosimilar) Compared with Reference Bevacizumab (Avastin®), Both Plus Paclitaxel and Carboplatin, as First-Line Treatment for Advanced Non-Squamous Non-Small-Cell Lung Cancer: A Randomized, Double-Blind Study

Abstract

Background: PF-06439535 is a bevacizumab biosimilar. We aimed to compare the efficacy and safety of PF-06439535 with that of reference bevacizumab (Avastin^®) sourced from the EU (bevacizumab-EU), each with paclitaxel and carboplatin, in the first-line treatment of advanced non-squamous non-small-cell lung cancer (NSCLC).

Methods: In this double-blind, parallel-group study, we recruited patients from 159 centers in 27 countries. Participants were randomized 1:1 to receive PF-06439535 plus paclitaxel and carboplatin or bevacizumab-EU plus paclitaxel and carboplatin on day 1 of each 21-day cycle for 4-6 cycles, followed by blinded monotherapy with PF-06439535 or bevacizumab-EU until disease progression, unacceptable toxicity, withdrawal of consent, or the end of the study. Randomization was stratified by region, sex, and smoking history. The primary endpoint was objective response rate (ORR) in accordance with RECIST 1.1, based on responses achieved by week 19 and confirmed by week 25.

Results: Between 21 May 2015 and 14 November 2016, 719 patients were randomized to the PF-06439535 group (n = 358) or the bevacizumab-EU group (n = 361). As of data cutoff for analysis of the primary endpoint (8 May 2017), 45.3% (95% confidence interval [CI] 40.01-50.57) of patients in the PF-06439535 group and 44.6% (95% CI 39.40-49.89) of patients in the bevacizumab-EU group achieved an objective response by week 19 that was confirmed by week 25. The unstratified ORR risk ratio was 1.015 (95% CI 0.863-1.193; 90% CI 0.886-1.163), and the unstratified ORR risk difference was 0.653% (95% CI - 6.608 to 7.908); all three CIs fell within pre-specified equivalence margins. Using final data after study completion (22 December 2017), no notable differences in progression-free survival or overall survival were observed between the groups. The most frequently reported grade 3 or higher treatment-emergent adverse events were hypertension, neutropenia, and anemia. There were no clinically meaningful differences in safety, pharmacokinetics, or immunogenicity across treatment groups.

Conclusion: Among patients with advanced non-squamous NSCLC, PF-06439535 demonstrated similarity to bevacizumab-EU in terms of efficacy. Safety profiles for the two treatments were comparable.

Trial registration: ClinicalTrials.gov, NCT02364999.

Funding: Pfizer.

Fig. 1

Participant flow diagram. ^aOne patient received paclitaxel and carboplatin but withdrew before receiving PF-06439535. ^bPatient was indicated as “study terminated by sponsor” by the investigator; however, this patient was considered to have met the definition of study completion (i.e. patient was alive and had completed survival follow-up as defined by the protocol). ^cThe most frequently reported reason for patients not being included in the PP population was due to being randomized but never dosed with PF-06439535 or bevacizumab-EU (3 [0.8%] patients in the PF-06439535 group and 3 [0.8%] patients in the bevacizumab-EU group). Bevacizumab-EU reference bevacizumab sourced from the European Union, COPD chronic obstructive pulmonary disease, ITT intent-to-treat, ORR objective response rate, PK pharmacokinetics, PP per-protocol

Fig. 2

Treatment comparison (PF-06439535 group vs bevacizumab-EU group) for ORR, based on responses achieved by week 19 and confirmed by week 25 in the intent-to-treat population. Panel a depicts the unstratified ORR risk ratio with 90% and 95% CIs, and panel b depicts the unstratified ORR risk difference with 95% CI. Dashed lines indicate equivalence margins agreed with regulatory authorities in the EU, Japan, or the US. Data cutoff date 8 May 2017. Analyses based on 2-sided Miettinen and Nurminen method without stratification variables. Bevacizumab-EU reference bevacizumab sourced from the European Union, CI confidence interval, ORR objective response rate

Fig. 3

Kaplan–Meier plots of a progression-free survival and b overall survival in the intent-to-treat population. Final data after study completion on 22 December 2017. ^aCI based on product-limit method. ^bCI based on Brookmeyer and Crowley method. ^cHazard ratio based on Cox proportional hazards model stratified by region, sex, and smoking history. ^d2-sided p-value based on log-rank test stratified by region, sex, and smoking history. Bevacizumab-EU reference bevacizumab sourced from the European Union, CI confidence interval, NE not estimable

Fig. 4

Mean serum concentrations of PF-06439535 and bevacizumab-EU in the PK population. Final data after study completion on 22 December 2017. Triangle or square and bar represent the mean with standard deviation. Concentrations are pre-dose (0-h time point) unless otherwise noted. Summary statistics calculated by setting concentration values below the LLOQ (< 250 ng/mL) to 0. Unplanned readings excluded. Samples with a time deviation of > 20% or any positive time deviation from the 0-h planned time point excluded. End of treatment and early withdrawal samples excluded. Standard deviation not shown for the 0-h time point at cycle 1, day 1 because the bars would be shorter than the height of the triangle and square symbols. There were 20 patients in the PF-06439535 group and 17 patients in the bevacizumab-EU group with measurable pre-dose concentrations above the LLOQ on cycle 1, day 1. Patients with measurable pre-dose concentrations > 5% of apparent C_max (serum concentration at 2.5-h time point) on cycle 1, day 1 were excluded from the summary analysis. After this pre-specified exclusion, six patients in the PF-06439535 group and seven patients in the bevacizumab-EU group with measurable pre-dose concentrations on cycle 1, day 1 were included in the summary. Bevacizumab-EU reference bevacizumab sourced from the European Union, C cycle, C_max maximum concentration, h hour(s), LLOQ lower limit of quantification, PK pharmacokinetics