A bevacizumab biosimilar developed by Boehringer Ingelheim has comparable safety, efficacy, immunogenicity, and pharmacokinetics to the reference product in patients with recurrent or metastatic non-small cell lung cancer (NSCLC), according to a new study.

In a phase 3 trial of BI 695502, the primary end point of best overall response rate (ORR) until 18 weeks was 54% in the group receiving the biosimilar and 63.1% in the bevacizumab RP (Avastin) group. The study demonstrated a 90% confidence interval (CI) for the primary efficacy end point falling within the prespecified range, the authors say.

Similar efficacy and toxicity have been observed in clinical studies of other bevacizumab biosimilars compared with bevacizumab RP. Two biosimilars, Pfizer’s Zirabev and Amgen’s Mvasi, have been approved for indications including unresectable, locally advanced NSCLC and recurrent or metastatic nonsquamous NSCLC. BI 695502 is no longer under development by Boehringer Ingelheim.

“Studies like the one reported here are…integral to the increased use of biosimilars across a range of indications,” the authors write. “These results enrich the evidence that bevacizumab biosimilars are a safe and efficacious alternative to bevacizumab RP.”

The trial screened 1030 patients at 190 locations, and randomized 671 patients from 161 centers in 28 countries between July 2015 and November 2018. They received up to 18 weeks of induction treatment with BI 695502 or bevacizumab RP 15 mg/kg plus paclitaxel and carboplatin. Slightly more patients in the BI 695502 group had ECOG PS of 1 (63.0% versus 60.4%), and slightly more patients had brain lesions (6.6% versus 3.7%) than in the bevacizumab RP group.

After a manufacturing issue with one batch of the biosimilar in December 2017, investigators switched the 13% of patients who had not yet completed maintenance to bevacizumab RP. Comparisons of the 2 products were based primarily on preswitch data.

The 90% CI for the between-group ratio of best ORR (0.770 to 0.951) was within the prespecified range for equivalence (0.736–1.359). The 95% CI for the ratio of best ORR (0.754–0.970) also fell within the required range (0.727-1.376) to satisfy the Japanese and European regulatory criteria for equivalence.

Median progression-free survival was 8.3 months in patients receiving BI 695502 and 9.0 months in those treated with bevacizumab RP (HR = 1.22, 95% CI: 1.02–1.45). Median overall survival was 15.6 months in the biosimilar treatment group and 19.5 months in the RP group (HR = 1.23, 95% CI: 1.00–1.51).

Plasma concentrations were slightly higher in the bevacizumab RP group, but the differences were not considered clinically relevant. Among patients with negative or unknown anti-drug antibody (ADA) status at baseline, 4 patients (1.2%) in the BI 695502 group and 8 patients (2.4%) in the bevacizumab RP group had at least 1 positive ADA result post-baseline.

The proportions of patients with at least 1 adverse event (AE) and with drug-related AEs were similar in the 2 groups during the preswitch period. The most common AEs were similar and occurred at slightly higher frequency in the biosimilar group. All-cause and drug-related grade 3/4 AEs occurred in a higher proportion of patients in the BI 695502 group compared with the bevacizumab RP group.

Among the AEs leading to death, they were drug-related in 4 patients in the BI 695502 group and 2 patients in the bevacizumab RP group. Signs and symptoms of immunogenicity were rare.

AEs of special interest (hepatic injury, anaphylactic reactions, GI perforations, pulmonary hemorrhage) were reported in a higher percentage of patients in the BI 695502 group (3.6%) compared with the bevacizumab RP group (0.9%). The 95% CI of the risk ratio (BI 695502/bevacizumab RP; the calculated value of 1.16) was 0.99 to 1.37, indicating the treatments were comparable.

Reference

Kim ES, Balser S, Rohr KB, Lohmann R, Liedert B, Schliephake D. Phase 3 trial of BI 695502 plus chemotherapy versus bevacizumab reference product plus chemotherapy in patients with advanced nonsquamous NSCLC. JTO Clin Res Rep. 2021;3(1):100248. doi:10.1016/j.jtocrr.2021.100248