Totality of Scientific Evidence in the Development of ABP 798, a Biosimilar to Rituximab

Background: ABP 798 is a proposed biosimilar to rituximab reference product (RP). Regulatory approval of biosimilars is based on a totality of evidence (TOE) approach which includes similarity in analytical characteristics (structural and functional), in mechanism of action (MoA) -clinical pharmacology, pharmacokinetics (PK), pharmacodynamics (PD), and in comparative clinical studies. Evidence from analytical, functional, clinical PK/PD, and efficacy and safety evaluations indicate that ABP 798 is similar to rituximab RP. Similarity of efficacy and safety in patients with non-Hodgkin lymphoma (NHL) was demonstrated in a randomized, double-blind, comparative clinical study in which the primary analysis was based on data from central, independent, and blinded assessments of disease and the sensitivity analysis was based on data using investigator assessments of disease. Here we present in vitro chemo synergy, preclinical in-vivo tumor efficacy data and show how the results of the sensitivity analysis using investigator assessments of the response rate from the NHL clinical study further add to the TOE supporting similarity of ABP 798 with the RP.

Methods:

In vitro synergy with chemotherapeutic agents was investigated by assessing the effect of docetaxel, gemcitabine, or doxorubicin with ABP 798 and rituximab RP in SU-DHL-6 cell lines and measured using the ATP-based cell viability endpoint (ATPLite). In vivo efficacy was assessed by evaluating the anti-tumor activity of ABP 798 and rituximab RP in NHL xenograft models in NOD/SCID mice using CD20-expressing RL and Raji cell lines. A sensitivity analysis was conducted to assess the robustness of the results of the primary efficacy endpoint of the proportion of patients with the best overall response of complete response, partial response, or unconfirmed complete response (ORR) in the NHL comparative clinical study using the investigator's assessment of disease on the full analysis set. Waterfall plots of the maximum percent reduction in the sum of the products of the greatest diameters (SPD) for index nodal lesions based on both the central, independent, blinded assessment of disease and the investigator's assessment of disease were used.

Results:

In vitro study results demonstrated synergistic activity of the combination of docetaxel and ABP 798 or the RP (synergy score = 6.35±0.69 [with ABP 798]; 6.79±0.12 [with RP]). In contrast, combinations with gemcitabine (synergy score = 2.68±0.20 [with ABP 798] and 2.54±0.29 [with RP]) or doxorubicin (synergy score = 1.95±0.56 [with ABP 79]) and 3.23±0.53 [with RP]) were additive. In RL cell lines, ABP 798 inhibited tumor growth similarly to rituximab RP, with no statistical difference between dose groups at 3 and 30 mg/kg (p = 0.951 and p = 0.996, respectively). In Raji cells, ABP 798 and rituximab RP significantly and similarly inhibited tumor growth when compared to an IgG1 isotype control (p < 0.001 for 3 and 30 µg/kg doses). ABP 798, at both dose levels, inhibited tumor growth similarly to rituximab RP by 27-41%, with no statistical difference. In the NHL comparative clinical study, risk difference (RD) of ORR with 1-sided 95% lower and upper confidence limits from the adjusted linear model was 7.7% (-3.2%, 18.6%) using central, blinded, independent assessments of the modified full analysis set based on randomized treatment assignment. The sensitivity analysis using investigator assessments of disease of the full analysis set of all randomized subjects based on the randomized treatment assignment received was RD = 0.5% (-7.8%, 8.8%). Waterfall plots of the maximum percent reduction in the SPD for index nodal lesions further demonstrate similarity of efficacy with central or local assessments of disease.

Conclusions: In vitro and in vivo preclinical studies demonstrated chemo synergy and in vivo tumor efficacy. The sensitivity analysis of ORR using local assessments in the NHL comparative clinical study was consistent with results from the primary efficacy analysis; the maximum percent reduction in index nodal lesions was similar for central and local assessments of disease. These results further support that ABP 798 is similar to the RP across a range of efficacy evaluations as well as primary and secondary mechanisms of action and add additional supporting data to the TOE.