OR WAIT null SECS
David E. Szymkowski is Senior Director of Biotherapeutics at Xencor.
Why the primary defence against biosimilars is to develop better molecules that can be launched quickly.
When faced with patent expiries and the risk of competition from biosimilars, innovator companies generally prefer to look forward by developing new and better drugs rather than looking back at biogeneric competition. That at least is the idealised business strategy. In reality, these companies take a more pragmatic approach to risk; this is where a biosuperior strategy can be advantageous.
The innovators' primary defence against a true biosimilar is to develop a better molecule that can be launched in the time it takes for the true biosimilars to reach the market (which can be nearly as long as the development of a novel biologic). In other words, innovators (by definition) seek to develop drugs to new targets with new mechanisms. However, it is not always possible to identify and exploit such new targets — that's where the "biobetter" or "biosuperior" strategy arises.
Pharma companies, though also innovators, are generally more conservative than smaller biopharmas and biotechs. In my opinion, pharmas are more willing to look sideways at the competition to identify their next blockbuster (the classic "me-too" strategy), and this holds true for biologics as well as small molecules. For examples, simply look at the large number of anti-CD20 antibodies in development following the success of rituximab and the five anti-TNF biologics on the market, with still more in development.
Also, keep in mind that the "threat" of competition is currently just that: an unquantifiable (possibly modest) threat. The general consensus is that the consequences of imminent patent cliffs are not expected to be nearly as dire as for small molecules, where market share can be devastated very quickly. In contrast, in a widely quoted analysis by the US Federal Trade Commission, it was estimated that innovator biologics would maintain the majority of market share (~70–90%) even well after patent expiry.1 There are many reasons for this, including the current lack of an abbreviated regulatory pathway (precluding significant cost or time savings for biosimilars), the potential new safety risks (e.g., due to differences in immunogenicity) and the unique supply chain for biologics (which are frequently handled by speciality and hospital pharmacies).
In summary, from my observation of innovators' response to the perceived threat, they have largely dismissed biogenerics as a viable strategy for themselves, while seeing them as an unavoidable yet modest external threat. Innovators are concerned about erosion of their market share by biogenerics, but are confident that this erosion will be much less severe and more delayed compared with small molecules. On the other hand, innovators are heavily investing in biosuperiors. Even the innovators are looking hard at biobetter strategies to fill their pipelines and provide a lifecycle management strategy for their blockbusters.
The differences between biogenerics, biosuperiors and next-generation innovative drugs are not clearly defined. I prefer "biosimilars" and "biosuperiors" as the most accurate descriptors. For example, is a humanised version of a chimeric antibody a biosuperior or next-generation molecule? It is clearly not a biogeneric, but on the other hand it is not really an innovative new drug, either. If a biologic has a superior formulation that allows subcutaneous administration of a formerly in vitro infusion, is this still a biogeneric? If a biologic is made using the original 1990s technology, can the biosimilar be manufactured in a state-of-the-art cell expression system?
What a biosimilar is not defines it more than what it is. There is a consensus that biosimilars (also known as followon biologics) must replicate as closely as possible the innovator biologic, as with small molecule generics. The critical point is that all characteristics of the innovator drug must be replicated, including mechanism of action, potency, immunogenicity (or lack thereof), identical pharmacokinetics, etc. By definition, a biogeneric cannot be better in any aspect than the innovator drug. This raises some interesting possibilities. For example, discovering that your biosimilar has a superior halflife in humans would be a damaging result. Ironically, although improved halflife would be an advantage, the result could kill your drug because it would demonstrate that it was not a true biogeneric. As another example, if the purity of the innovator biologic is less than ideal, should the biogeneric intentionally be manufactured to the lower specifications of the innovator drug? Since changes to biologics may result in unpredictable consequences, by definition, the biogeneric must be as good (or as bad) as the innovator drug in all respects. These types of issues are currently under debate and help illustrate why the biogenerics pathway to market is so much more complicated than that of small-molecule generics.
These biosimilar issues are much more transparent when the intent is to create a biobetter (also known as a biosuperior). In marked contrast with biosimilar development, developers are free, and in fact compelled, to improve on the original to generate clinical and market advantages over the innovator. Once the decision is made to improve rather than make a "me-too", there are many routes to generate the biosuperior. However, there is still a line that must not be crossed. For example, a biobetter cannot improve too much upon the original biologic, or it becomes a nextgeneration innovative drug. This certainly has greater rewards but also carries greater risks.
The divisions among these three classes (in my opinion) hinge on the nature of the improvements made over a true biogeneric. If changes result in an improvement in drug function, as opposed to drug properties, it is an innovative drug. As one example, the functional improvement could be in antibody tumour-killing potency through modification of effector functions (currently an area of intense commercial interest). Such an enhanced antibody drug would have the same target, and even the same epitope, but would be more efficacious than the original. On the other hand, if the change results in improved drug properties with no change in function, this would be a biosuperior drug. An obvious example is halflife improvement of biologics, perhaps by pegylation, or by antibody Fc engineering, or by many other strategies. Such drugs would be identical in terms of mechanism of action, but the longer halflife would lead to improved patient convenience and cost of goods. Finally, nextgeneration innovative drugs may incorporate some of the features of the biosuperiors, but can remain innovative in that they hit new targets, possibly with new mechanisms, leading to novel therapeutic activities.
The implication of this classification is that biogenerics must avoid introducing superior commercial or clinical properties into the original drug, while biosuperiors by definition must do something better (e.g., convenient dosing, superior efficacy or safety, an improved formulation).
Finally, this leads to an intriguing opposing risk profile for biogenerics versus biosuperiors. Biogenerics are the low risk and low reward strategy, while biosuperiors are higher risk and higher reward. Biogenerics have minimal (though not absent) clinical risk relative to biosuperior approaches because the goal is to replicate exactly the clinical performance of the innovator. On the other hand, biosuperiors introduce more clinical risk because new features have been introduced into the molecule. Depending on the nature of the change, this risk can be quite modest (e.g., an improvement in half-life), or more substantial (e.g., a manipulation of immune effector functions to create a more cytotoxic antibody).
As mentioned above, perhaps the lowest risk and also most conservative biosuperior strategy is halflife extension. This can be achieved via pegylation, by antibody Fc engineering, or through the use of a stabilising fusion protein (such as albumin). The drug has the same target and same function, but with improved pharmacokinetics properties. Halflife extension introduces a low development risk but benefits from an easily quantified commercial differentiation. It is evident that less frequent dosing of a longer halflife biologic would benefit patients, while manufacturing groups should also be pleased that production requirements are reduced.
The immune effector strategy mentioned above introduces a higher level of both risk and reward. It is now very well accepted that therapeutic antibodies can be engineered to more efficiently engage the immune system to enhance the destruction of tumour cells and several companies are exploiting this approach to improve the cytotoxicity of marketed antibodies. This therapeutic concept pushes into nextgeneration territory, but in many ways these molecules will be developed and assessed like biosuperiors.
Another, more prosaic, example is the development of humanised or fully human versions of marketed chimeric antibodies. This is done typically to minimise immunogenicity and sometimes incorporates improved binding affinity as well. As examples, the fully human antiTNF antibody golimumab (developed by Centocor), is in some respects a biosuperior version of Centocor's own chimeric antiTNF, infliximab (though they are unrelated molecules).
In the US, at the moment at least, approval for all biologics (whether biosimilars, biosuperiors, or innovative) must be supported by clinical trials. These trials cannot be simple comparability studies (e.g., to assess pharmacokinetics), but must include a full safety and efficacy assessment. This situation contrasts markedly with the small-molecule generic regulatory pathway established under the Hatch-Waxman Act of 1984. It is likely that such a pathway will be developed for biologics, but in my opinion it is also likely that the approval process for biogenerics will be closer to innovator biologics than to small-molecule generics, which means that much of the time and cost savings that drive the development of generic small-molecules will not be achieved with biogenerics. Speed to approval is as critical for biogenerics as for generics; however, it is safe to assume that some clinical development will be required for all biologics, in contrast to smallmolecules. This alone is a significant barrier to entry for many companies and it will be very interesting to see how market forces stimulate (or perhaps suppress) entry of competitive biogenerics, even if a regulatory pathway is eventually defined.
Manufacturers of innovator biologics have said for many years that "the process is the product" compared with small molecule generics where "the product is the product". Given the great advances in biologics manufacturing and likely continued future improvements, the key questions is whether this statement remains true. If not, a "Hatch-Waxman-like" regulatory path may become feasible and stimulate development of biosimilar versions of many blockbuster biologics. However, I don't see this simple outcome developing for many years — if at all.
Regardless of the current or future regulatory environment for biosimilars, I do believe that the development of biobetters will have an impact on the future of biosimilars. If biosimilars and biobetters are channelled into the same regulatory path, then it is logical that a true superior molecule would be more attractive. The modestly higher risk of making a biosuperior can lead to a much higher reward; consider that the target risk remains low and the immunogenicity risk should be manageable versus the innovator drug. The only market advantage of the biogeneric would be in cost, but this advantage comes with a risk that your biogeneric is not comparable to the original drug. So if the development time and cost advantages are lost compared with a biobetter, then the wise choice would be to advance the biobetter, which can take market share based on true clinical superiority, rather than cost alone.
For example, consider immunogenicity. This is a significant risk for all biologics, with implications for both clinical safety and efficacy. Yet, immunogenicity cannot be predicted preclinically and even different batches of the same drug (for example, Eprex erythropoietin) can possess dramatically different immunogenicity risks in humans. This issue will also probably arise with biosimilars and without good predictors, the unknown risk of immunogenicity may negate any modest cost advantage.
In summary, in my opinion, true biosimilars do not offer a compelling business case. The risks of immunogenicity, of different pharmacokinetics, potency, stability, etc., more than negate the single advantage (cost) provided by such products. Thus, without an existing defined path forward, and with the likely future path probably demanding clinical trials, there is little incentive to create a biosimilar. Biosuperiors, on the other hand, have much to offer compared with biosimilars and, in many ways, are even more attractive than next-generation innovative biologics. Biosuperiors have many of the benefits of biosimilars yet have a dramatically lower development risk compared with truly novel biologics. Since examples of all three classes of biologics are moving through clinical development now, these questions should all be answered with realworld data over the next few years.
David E. Szymkowski is Senior Director of Biotherapeutics at Xencor.
1. Emerging Healthcare Issues: Follow-on Biologic Drug Competition (Federal Trade Commission, US, June 2009).