A multicenter, randomized, double-blind, phase III trial comparing denosumab biosimilar QL1206 and denosumab in patients with bone metastases from solid tumors.

Abstract

2526 Background: This phase III study compared clinical efficacy, safety of the first biosimilar QL1206 with denosumab in solid tumor patients with bone metastases. Methods: Patients aged 18-80 years, with bone metastatic solid tumors and ECOG performance status of 0-2 were randomly assigned 1:1 to receive subcutaneous QL1206 or denosumab (120 mg Q4W, both) based on stratification factors (tumor types, previous skeletal-related event [SRE], and current systemic anti-tumor therapy). The primary efficacy endpoint was the percentage changes from baseline to week 13 in urinary N-telopeptide/creatinine ratio (uNTX/uCr); Clinical equivalence would be confirmed if the two-sided 90% confidence intervals (CI) of the least-squares mean (LSM) difference of natural log-transformed ratio of week 13 uNTX/uCr to baseline calculated by analysis of covariance were within the margins of ±0.135. The secondary endpoints included the percentage changes from baseline to week 25 and 53 in uNTX/uCr, the percentage changes from baseline to week 13, 25 and 53 in serum bone-specific alkaline phosphatase (s-BALP) and the time to on-study SRE. Adverse events (AE), immunogenicity (IG) and pharmacokinetics (PK) were also assessed. Population PK was analyzed using nonlinear mixed effects model. Results: 717 patients were randomized to the QL1206 (n = 357) and denosumab (n = 360) groups, respectively. The median percentage changes at week 13 in uNTX/uCr were -75.2% for QL1206 and -75.8% for denosumab. LSM of natural log-transformed ratio of week 13 uNTX/uCr to baseline were -1.429 (standard error 0.130) and -1.441 (0.130) in the two groups. The LSM difference between the two groups was 0.012 (90% CI -0.078 to 0.103; P = 0.8208), within the equivalence margins. Analyses for subgroups of sex, age, and randomization strata showed almost no significant differences. The secondary endpoints were also similar between the two groups (median percentage changes in s-BALP from baseline to week 13: -38.0% vs -37.1%; hazard ratio of time to SRE: 1.264 [95% CI 0.670 to 2.383]; both P > 0.05). In the safety set, treatment-emergent AE occurred in 332 of 356 patients (93.3%) in the QL1206 group and 348 of 361 (96.4%) in the denosumab group. The proportions of positive anti-drug antibody (14/345 [4.1%] vs 16/352 [4.5%]), neutralizing antibody (three [0.9%], each), and PK characteristics were similar between the two groups. Conclusions: The results suggest that QL1206, the first biosimilar denosumab, is similar to denosumab with respect to clinical efficacy, acceptable safety, IG, and PK characteristics. The totality of evidence supports clinical equivalence of QL1206 and denosumab. Clinical trial information: NCT04550949.