Bevacizumab Biosimilar Demonstrates Similar Response Rate, Survival as European Originator

Mark A. Socinski, MD

The bevacizumab (Avastin) biosimilar PF-06439535 demonstrated similarity to the European Union (EU) bevacizumab reference product for overall response rate (ORR), with similar pharmacokinetic and immunogenicity profiles for patients with advanced non-squamous non-small cell lung cancer (NSCLC), according to results presented at the 2018 ASCO Annual Meeting.

Data from an ongoing double-blind, randomized, global clinical trial (NCT02364999) showed ORR rates of 45.3% (95% CI, 40.01-50.57) for firstline PF-06439535 combination with paclitaxel and carboplatin compared with 44.6% (95% CI, 39.40-49.89) for bevacizumab-EU at 19 weeks (confirmed by week 25) in the intent-to-treat population, the primary endpoint of the trial.

Lead author Mark A. Socinski, MD, executive medical director, Florida Hospital Cancer Institute, Orlando, added that the confidence intervals that fell within the prespecified therapeutic equivalence margin. The risk ratio of ORR, the ratio of ORR between the two arms, was 1.015, with a 90% CI of 0.886 to 1.163, which was contained within the prespecified therapeutic equivalence of 0.73 to 1.37. The risk difference, difference of ORR between the arms, was 0.653, with a 95% CI of -6.608 to 7.908, which was also within the prespecified equivalence margin of -13% to 13%.

In addition, 1-year rates of progression-free survival (PFS) and overall survival (OS) were not different between the treatment groups. One-year PFS was 30.8% and 29.3% with the biosimilar and source product, respectively, and 1-year OS rates were 66.4% and 68.8%, respectively.

“These results confirm similarity demonstrated in earlier analytical, nonclinical, and clinical studies of PF-06439535 and bevacizumab-EU,” said Socinski.

Current treatment guidelines recommend the use of bevacizumab plus chemotherapy as a first-line option for advanced nonsquamous NSCLCC in patients with a performance status of 0 or 1 without EGFR/ALK mutations, he noted.

Analytical studies have shown similar structure and function of PF-06439535 to bevacizumab-US and bevacizumab-EU. In nonclinical studies, the biosimilar had similar toxicity, toxicokinetics, and immunogenicity to bevacizumab-EU.

“In comparative clinical pharmacology studies, it had similar pharmacokinetics [PK], safety, and immunogenicity to both bevacizumab-EU and bevacizumab-US in healthy male volunteers,” Socinski said. “In a comparative clinical study, the question is does it have similar efficacy, safety, PK, and immunogenicity to bevacizumab-EU in patients with advanced nonsquamous NSCLC.”

A total of 719 patients with newly diagnosed histologically or cytologically confirmed stage IIIB or IV NSCLC or recurrent NSCLC were randomized in a double-blind fashion 1:1 to PF-06439535 (n = 358) or bevacizumab-EU (n = 361) plus paclitaxel/carboplatin on day 1 of an every 3-week cycle. That was followed by PF-06439535 or bevacizumab-EU blinded monotherapy as maintenance until disease progression or unacceptable toxicity.

Patients could not have known central nervous system metastases nor known sensitizing EGFR mutations or EML4-ALK translocations. The study was conducted in 159 centers in 27 countries. It was ongoing at the time of this analysis, with a data cut-off of May 8, 2017.

Sixty-five percent of the study population was male, 89% were white, and median age was 61 years. Approximately 30% were never smokers. Histologic classification was adenocarcinoma in 97% and 76% had stage IV NSCLC. About 30% of patients had prior treatment for NSCLC, primarily surgery (18%).

The median PFS was 9.0 months in the PF-06439535 arm compared with 7.7 months in the patients assigned to originator bevacizumab-EU (HR, 0.974; 95% CI, 0.798-1.190, P = .814). The median OS was 18.4 months and 17.8 months (HR 1.001, 95% CI 0.786-1.304, P = .991), respectively.

Treatment exposure was not different between the two arms, with a mean of 10.7 cycles of PF-06439535 and bevacizumab-EU administered. Bevacizumab treatment was discontinued in 73% in each arm, and the major reason for discontinuation was objective progression or relapse in each arm (43.0%, PF-06439535; 47.5%, bevacizumab-EU).

The rate of adverse events was 14.6% in the PF-06439535 arm and 10.9% in the bevacizumab-EU arm, and the rate of death was 4.8% and 5.3%, respectively.

“No clinically meaningful differences in safety profile was shown in this trial, and similar PK and immunogenicity results were seen across treatment groups,” said Socinski.

In the safety population, grade ≥3 rates of hypertension, cardiac disorders, bleeding/hemorrhage, venous thromboembolism, and other adverse events (AEs) associated with bevacizumab were similar between the two arms. Grade ≥3 serious AEs in the safety population occurred in about 19% in each arm and were not statistically different between the two arms. The proportion of patients with grade-5 (fatal) serious adverse events was 5.3% with PF-06439535 and 5.9% with bevacizumab-EU.

Peak and trough serum drug concentrations overlapped between the two arms in the PK population.

The overall rate of post-treatment anti-drug antibodies and neutralizing antibodies was low and comparable between groups.

Socinski MA, Von Pawel J, Kasahara K, et al. A comparative clinical study of PF-06439535, a candidate bevacizumab biosimilar, and reference bevacizumab, in patients with advanced non-squamous non-small cell lung cancer. J Clin Oncol. 36, 2018 (suppl; abstr 109)

<<< 2018 ASCO Annual Meeting