As the debate rumbles on about biosimilars, DrugBaron cannot help wondering if they are not in any case the wrong solution to a very real problem. After all, when it comes to biologic drugs is biosimilar similar enough?
The principle of generic drugs mostly works very well, ensuring that at the end of the state-sponsored monopoly period, which allows super-normal profits to compensate and incentivize innovators, normal market economics applies to drug pricing.
It works because almost everyone (medical professionals and patients) believe the generic to be functionally identical to the proprietary drug. Not just similar, but identical. The initial high price was pay-back for discovering and developing the drug in the first place, not a premium for quality.
Biosimilars attempt to extend the same principle to biologic drugs. But the core tenet of the generics system (functional identity of the generic) is demonstrably absent. Healthcare professionals and patients will not, and should not, accept that similar is good enough.
Yet no-one can doubt the need to find a solution to the problem of indefinite premium pricing for biologics. As their ranks swell on the medicine cabinet shelf, it’s a pressing problem for the world’s healthcare providers.
Maybe, then, its time to look for a different solution? There is an alternative to biosimilars. Enshrine the principle of patent law directly into the regulatory framework for biologics: in return for approving the drug, ten years after approval the registrant must license others on pre-agreed terms to make the identical product. Such a system, creating true ‘biogenerics’, would allow market economics to determine the price of biologic drugs while providing the patients with the real product, backed by extensive hard end-point clinical trial, not some cheap imitator.
The name captures the nub of the issue – bioSIMILAR. Unfortunately, when it comes to biologic drugs that similarity is only sequence-deep. Or, in the case of antibody drugs, not even that similar.
Our understanding of post-translational modifications of proteins barely scratches the surface of a breath-takingly diverse field. Relatively well understood modifications, such as glycosylation, are dwarfed at least in number if not frequency by a growing array of amino acid modifications that are found almost everywhere they have been looked for. Make a protein in a different cell line, and it is quite literally a different protein.
DrugBaron’s proposal for true ‘biogenerics’ protects patients and lowers healthcare costs, while still granting the industry the innovation premium that is essential for continued investment
Proponents of biosimilars argue that such differences amount to little more than the different excipients employed in generic small molecule drugs compared to their proprietary ancestors. But that dramatically underestimates the functional consequences of many post-translational modifications. Humans have an amazing diversity of natural antibodies against carbohydrates, which vary from one individual to the next, and these antibodies can change the functional properties of the protein in ways that are impossible to predict and much more significant than might be imagined for a minor change in glycosylation pattern itself.
And if post-translational modifications can have a major effect on properties, that is nothing compared to the difference between two antibodies against the same target protein.
Some differences are obvious: some monoclonal antibodies against a signaling receptor, for example, have agonist properties, while others are antagonists preventing ligand binding; and yet others bind without appreciable agonist or antagonist properties – but perhaps have biological effects through immune cell engagement.
The biosimilar fans argue that such gross differences in functionality can be readily tested, and excluded in simple preclinical tests. For the most part that is true, but it misses the most salient point: for antibody drugs, as most small molecules, we simply don’t have certainty about the molecular mechanism of action. The relative contribution, for example, of direct antagonism versus induction of receptor internalization (or perhaps intracellular phosphorylation to switch off activity) versus killing the target cell, is usually completely unknown. How then do you know which biological activity assays have to show similar behavior in order to conclude the biosimilar drug will behave identically in humans?
The interests of patients and of the industry – in restricting biosimilar development – are at least partially aligned
The subtlety of these differences is already amply illustrated by the four biological drugs targeting TNF-alpha that are marketed today (none as biosimilars – all four as independent proprietary products that happen to share the same target). Different kinetics and affinity for the ligand give each of the drugs differing clinical profiles: for example Remicade™ infliximab (JnJ) is effective in Crohn’s Disease, but Enbrel™ etanercept (a receptor fusion protein from Amgen) is worse than placebo – even though both neutralize TNF-alpha effectively. The benefit of infliximab is now attributed to induced cell death of TNF-alpha expressing T lymphocytes, an activity that etanercept lacks.
Could that have been predicted preclinically? Presumably not, since no-one did make the prediction prior to running large scale trials of etanercept in Crohn’s Disease.
Had Enbrel™ been approved simply on the basis of bioequivalence to infliximab in a TNF-alpha neutralization assay, the prize of reduced costs to the healthcare system would have come at the cost of treating patients with a medicine that was worse than ineffective.
The differences between antibodies targeting the same receptor are not just linked to efficacy – side-effects can differ too. Erbitux™ cetuximab (BMS/Lilly) shows a much higher incidence of severe injection reactions (including anaphylaxis) than Vectibix™ panitumumab (Amgen), another anti-EGFR antibody. It also shows more efficacy, the reasons for which are entirely unclear.
These examples already demonstrate beyond doubt the impossibility of predicting whether a supposed biosimilar is similar enough. Nothing short of full development can protect the needs of the patient population, and any kind of abbreviated development pathway for biosimilars is lowering the bar under pressure from payors, and to the detriment of patients.
It is unfortunate that this argument, which is based on sound biology and impeccable clinical evidence of the NON-similarity of ostensibly similar biologicals, has been rendered political. Big pharmaceutical companies pointing to this data have been accused of self-interest, protecting their profits on existing biological franchises rather than looking out for the patients.
In the pervasive, and mostly unfair, Bad Pharma culture of today, there seems a very real danger that the public may join the argument on the side of the biosimilar supporters – against their own interests – simply through lack of trust of large corporations.
Governments should listen to these well-made arguments no less well because they are supported by those who profit from them
When the unstoppable force meets the immovable mountain, its time to look for another solution. Innovators cannot expect to earn super-normal profits on biological drugs indefinitely. Quite apart from the cost to the taxpayer, this state of affairs introduces an unintended bias into drug discovery and development: there is an economic incentive to discover biological drugs rather than conventional small molecules simply because of the extra barriers to competition that extend indefinitely into the future.
If society needs to break the monopoly on a particular biological drug, and biosimilars are not the answer, what other options are there?
For small molecule drugs (which typically have less ‘know-how’ protecting their synthesis) the monopoly that yields super-normal profits to reward innovation has to be artificially enforced: through patents. Patents are an artificial construct based on a deal between the inventors and the state. In return for providing all the necessary details that allow others to practice the invention in perpetuity, the state grants the inventor a period of exclusivity. In effect, society gains the benefit of an invention that might otherwise be kept secret, and the inventor gains the certainty that no-one can reverse-engineer their invention and break the monopoly for that period.
The only reason the same contract does not work for biologic drugs is that the patent contract calls for full disclosure of information – and not materials. According to patent law, the information provided by the inventor must be enabling. In other words it must be sufficient to allow a sufficiently experienced practitioner to use the invention described.
But with biologics, the principle protection against copycats is control of the producing cell lines. In this respect, the inventor is better off not disclosing all the details to earn a patent, since control of the producing cell line itself is sufficient to guarantee an indefinite monopoly.
Changing patent laws is not the answer – for a start, such a herculean task would be challenging to pull off in any one jurisdiction, and getting anything approaching consensus across major markets would likely prove impossible. But more importantly, inventors of biologic drugs don’t need to engage with patent offices – protected by control of their cell lines, they use patents as adjunct rather than primary protection for their monopolies. Changing patent laws to require material (as well as informational) enablement would only result in the inventors of biologics relying even less on patents than they do today.
But they do absolutely rely on regulators for marketing approvals.
The solution that DrugBaron favours, then, is to incorporate the core principle of patent system into the regulatory process
In order to win marketing approval a company must provide not only the information (in the form of the NDA) but also the materials required to allow others to produce their medicinal product. This information is kept confidential (and the materials held in legal escrow) for a defined period, ensuring an effective monopoly as is the case today. As for patented small molecules, the inventor enjoys a period of super-normal profits to reward their innovation.
The key difference, however, is that at the end of the agreed period, the information and materials are made publically available. True generic versions of the biologic drug (very different from biosimilar versions) will then be available, with the same impact on pricing economics for biological drugs as we already enjoy for small molecules.
There is no reason why the exclusivity period granted by regulators under such a scheme has to be the same period as for patents. Governments must make a judgment as to the length of exclusivity that best balances the need to provide an incentive for innovation with the need to control healthcare costs. A ten-year exclusivity period after registration looks like a good starting point for the discussion.
When it comes to biosimilars the similarity is only sequence-deep
Such a system may be granted with howls of protest from the industry, who stand to gain nothing from such an arrangement. Indefinite periods of super-normal profits on biologic drugs would be curtailed without any obvious gain (for them). But at least it would be clear that those complaints were driven entirely by self-interest. With the present debate, the interests of patients and of the industry are at least partially aligned – there is some truth in the argument that easing the regulatory path for biosimilars will damage patients as well as profits.
DrugBaron’s proposal for true ‘biogenerics’, by contrast, protects patients and lowers healthcare costs, while still granting the industry the innovation premium that’s essential for continued investment. It is (relatively) simple to implement, robust and easily policed. Given the turmoil around the biosimilar debate, its surely time to look at imaginative ways round the problem rather than driving a juggernaut through the mountain – and through the cherished principle of looking after the interests of the patient.
Total Scientific Ltd is a preclinical CRO based near Cambridge, UK. We specialise in developing and characterising bespoke in vitro assays for discovery and development, including enzyme assays, binding assays and immunoassays together with biomolecule interaction services (Biacore) Total Scientific is a niche contract research organisation that offers a range of in vitro laboratory-based …