Accounting for Lot-to-Lot Variability of Biologics in Biosimilar Development

In creating the 351(k) US biosimilar approval pathway, regulators had laid out a methodology in which the totality of evidence is weighted heavily by proof of structural and pharmacologic comparisons of the biosimilar to the originator product (unlike the most critical role of clinical trials in the conventional 351(a) regulatory pathway). Through the initial approvals by the Food and Drug Administration (FDA), the level of scrutiny given to these physical, pharmacodynamics, and pharmacokinetic evaluations has become clearer.

In its latest draft Lee 2guidance, the FDA has added some more direction, as well as emphasizing one of the key points of the biosimilar and biologic manufacturing process. They spotlight the level of variability of one biologic. Recognizing the potential for variation in one reference biologic, the 351(k) approval criteria include comparisons versus copies of the US-licensed (rather than EU-licensed) originator product. For some approvals, bridging studies, which also test the similarity of the EU- and US-licensed biologics, have been permitted. In the guidance, “Quality Considerations in Demonstrating Biosimilarity of a Therapeutic Protein Product to a Reference Product,” released earlier in September, the FDA further amplifies this requirement.

In the biosimilar development process, a prospective manufacturer must obtain samples of the biologic from the originator drug maker. However, as has been well documented, lot-to-lot differences in the biologic may well occur, though these are not expected to have implications for clinical safety or effectiveness. Proving structural similarity to this agent goes a long way to progressing down the path to approval. In its draft guidance, the FDA seems to expand its test for analytical similarity. The biosimilar manufacturer must obtain “a minimum of 10 reference product lots” and these lots “should represent the variability of the reference product,” against which the biosimilar drug is evaluated to allow for meaningful comparisons. In other words, the biosimilar manufacturer must consider the likelihood of variation in the local source of the biologic. The FDA is accepting public comments on its draft guidance for 60 days.

This has some basic implications for biosimilar manufacturers. To begin the process of engineering a biosimilar drug, they must obtain samples of the originator product from its manufacturer. This has not always been simple, as the drug maker defending its brand can delay the process or otherwise make it difficult to purchase. Some legislative proposals have been introduced to coerce the originator manufacturer to provide, in a timely manner, the samples required by the prospective biosimilar drug company (e.g. Fair Access for Safe and Timely Generics Act of 2017).

Taking the “Biobetter” or “Follow-on” Biologic Route

For conventional drug manufacturers, standard lifecycle management considerations include extending patent life through development of new dosage forms, formulations of extended- or sustained-release versions, or even the “next-generation” molecule (e.g., Prilosec® vs. Nexium®).

For manufacturers of biologics, the tools at their disposal for extending patent life are a bit more limited. A change in dosage forms is likely, as are extended-release versions. They can still attempt to develop next-generation versions that are more effective or safer than the parent molecule, which would offer at least some protection against biosimilar competition. These offspring are considered “biobetters.” Unlike the term, this concept is not really new. In the past, agents like colony-stimulating factors have been “pegylated” to produce enhanced effects.

Let’s consider one example. Roche’s (Genentech’s) product obinutuzumab (Gazya®) was approved by the FDA. This agent is a CD20-receptor antibody that is similar to the original CD20 product rituximab (Rituxan®). There are differences, however. The molecular structures differ slightly. Roche believes that Gazya has more antibody-dependent cell-mediated cytoxicity than Rituxan. This may translate into improved progression-free survival. Gazya has only one of Rituxan’s 7 indications (chronic lymphocytic leukemia), so it will not be a strict replacement for Rituxan. Rituxan may be facing biosimilar competition by 2018, and it will indeed be interesting to see how payers will utilize another version of rituximab. This may also be dependent on the extrapolation of indications for the original product.

Roche made the decision to seek approval for Gazya through the  supplemental biologics license application (BLA) pathway, and it will be able to take advantage of full exclusivity periods, and possibly premium pricing (at least until biosimilars arrive). That may well justify the cost of conducting the full clinical trial program required for BLA approval.

Others took the “follow-on biologic” route—in this case, the molecule is not considered substantially different. Teva filed for approval of Granix® (tbo-filgrastim) as a BLA, because at that time, the 351(k) pathway did not yet exist. As a result of its approval, Teva received the full market exclusivity benefits of an originator biologic even though it is truly a biosimilar version of filgrastim. Entering a competitive marketplace, Granix has eroded Neupogen®’s marketshare, but it is subject to biosimilar competition to Zarxio®, in addition to whatever future filgrastim agents are approved. That might make a follow-on biologic little more than a “me-too,” with limited lifecycle prospects.