Experience With Manufacturing Changes Underscores Importance of Stepwise Approach to Biosimilars

When biologic sponsors propose to make changes to manufacturing, they must perform comparability studies to confirm that there are no significant differences in structural or functional attributes of the drug that could cause clinically meaningful changes in safety or efficacy. If a change is complex, or if clinically relevant differences are possible, bridging studies are often required to confirm overall comparability. Such studies, say authors of a recent paper published in Journal of Pharmaceutical Sciences, have implications for demonstrations of biosimilarity.

In the case study, the authors report findings from a comparability assessment of epoetin alfa before and after a proposed manufacturing process change for which, they say, nonclinical assessment of structure-function relationships were not sensitive enough to identify clinically relevant differences resulting from differences in the drug’s glycosylation profile.

Epoetin alfa is manufactured using mammalian cell culture technology, and changes to manufacturing conditions can affect quality attributes. To assess possible effects, structural, functional, and nonclinical pharmacology studies were conducted to evaluate the drug prechange and postchange, as were clinical bridging studies: one was a single-dose crossover study in patients with chronic kidney disease (CKD), in which the predetermined equivalence criterion was that the geometric ratio of area under time concentration-time curve (AUC)_-t for the postchange versus the prechange product should have a 90% confidence interval (CI) between 80% and 125%. The other was a phase 3, multicenter, randomized, double-blind hemoglobin and dose-equivalence trial that enrolled adult patients with CKD who were receiving hemodialysis.

The analytical comparability studies determined that the postchange product was comparable with the prechange material with the exception of minor quantitative differences in certain glycosylation structures that were not suspected to have an effect on biological function or safety of the drug. In vitro and in vivo functional assays showed the postchange product to be equivalent to the prechange product in terms of potency and pharmacodynamics.

In the single-dose crossover study, 30 patients initiated and completed the study, and the geometric ratio of AUC the from time zero to the last time sampled for the prechange and postchange product met predefined criteria for demonstrating bioequivalence (90% CI, 94.8%-104.3%).

However, in the phase 3 study in 462 patients, the log dose-ratio difference was −0.15 (90% CI, −0.25 to −0.06), demonstrating the postchange drug was more potent than the prechange drug, with the 90% CI values falling outside the prespecified equivalence margin. The researchers ultimately determined that the difference in potency was attributable to an increase in lactosamine content in the postchange epoetin alfa.

“These results illustrate that initial assumptions about the clinical relevance of certain structural attributes...may be incorrect,” write the authors. “We found in this case study that the common approach to assess the relevance of these observed structural differences using common in vitro functional assays may result in misleading conclusions and contribute to confirmation bias.”

This case study underscores the importance of the stepwise development exercise needed for biosimilar approval; the scope and burden of proof for each step in the exercise, say the authors, is informed by the residual uncertainty in the previous stages, as well as by limitations and sensitivities of any part of the clinical program to detect relevant differences.

Reference

Grampp G, McElroy PL, Camblin G, Pollock A. Structure-function relationships for recombinant erythropoietins: a case study from a proposed manufacturing change with implications for erythropoietin biosimilar study designs [Published online February 4, 2018]. J Pharm Sci. doi: 10.1016/j.xphs.2018.01.018.