Considerations for Approving Generics Versus Biosimilars

Cornelius F. Waller, MD: Generic drugs are usually small molecules that are produced chemically and can easily be copied or reproduced. Therefore, if you have an active ingredient that you can chemically synthesize, there is no need for major clinical trials. However, for biosimilars, which are complex molecules of a much larger size, it is necessary not only to have the physical chemical analysis and preclinical data but also to show that due to the effect of these large molecules, which can have a certain amount of immunogenicity—meaning there are antibodies against the drug that is being developed—it is necessary to do clinical trials, such as a phase 1 trial, where you can show pharmacokinetic and, if possible, pharmacodynamic issues and attributes of the compound. And then, depending on the results of the phase 1 trial, you can do a phase 3 trial in a sensitive disease model, where you can show the similarity of the biosimilar to the reference product based on the so-called totality of evidence. Regulatory agencies such as the FDA and the [European Medicines Agency, EMA] decide whether this leads to the approval of a biosimilar.

In comparison with generic drugs, where it is possible to just chemically reproduce the originator product, the biosimilar production is different. We have large, complex molecules that you can’t copy in the sense of what you would usually want. There are certain differences between the reference and the embellishing of the product. To a certain degree, this is allowed by the regulatory agencies. Based on the data and from preclinical and clinical investigations, it has to be highly similar without quality differences to the reference product in order to be approved as a biosimilar.

As you can imagine, because of the fact that biosimilars are being produced by living cells, there is no complete equality to the reference product. Even the reference product, if you have several batches, might contain minor differences as long as the protein structure and certain parts of the manufacturing quality that is being asked for by the regulatory agencies is long. Differences can be considered for biosimilars. For example, posttranslational changes might lead to differences in efficacy of the drug later on or to increased immunogenicity. With these quality assurances, as a manufacturer, you ensure that this is not going to happen.

Hope S. Rugo, MD: One of the things that happens when these agents go for regulatory approval is that the regulatory agencies go to where they’re produced. They want the production companies to fit into very strict guidelines and want the reproducibility to show where the agents are made. This requires an enormous number of criteria. The regulatory agencies may go back—they have been known to go back and to say “You need to fix this process or that process.” They can go back in and reevaluate over time. In addition to that, they’ve put in this pharmacovigilance where the companies are obliged to continue to track any differences, toxicities, or unusual events over time. The information can be reported for long-term follow-up of these agents once they’re approved.

In terms of the clinical approval, they have to have no clinically meaningful differences in terms of short-term clinically sensitive endpoints, no differences in safety that are clinically meaningful, and no differences in immunogenicity. You have to be able to look at antidrug antibodies, as well as the common safety—both short and long term—and the short-term efficacy end point, as well.

Regulatory agencies don’t require that you look at longer-term efficacy end points. For example, if you do a neoadjuvant trial, you don’t have to show that survival is similar in the long term, because you have to follow those patients for a very long time. But they do like to have enough exposure to feel comfortable about safety.

In terms of the lot release, there may be variations in glycosylation and sometimes in activity in vitro between different lots that are released. And so, for every few lots, they have to end testing sporadically throughout a lot. You have to show that you’ve maintained your similarity throughout and that any changes that occur—any shift over time—is not clinically meaningful.