The 'biogeneric' analysis problem

With the biogenerics market tipped by consulting firm Frost & Sullivan to reach $16bn (€11.9bn) by 2011 the need for advanced analytical techniques that can fully characterise biological drugs to a level that proves equivalency is growing.

With the continuing advances in analytical technology, some industry experts believe that some 'biogenerics' can already be characterised to a level that proves equivalency.

Although widely called biogenerics by the media, this is an imperfect term as the complexity of many biological drugs means that copycat versions are not necessarily substitutable in the same way as generic small molecule drugs.

Recognising this, the regulatory authorities in the EU have coined the phrase 'biosimilar' for these products, while the US favours 'follow-on biologics'.

The complexity of the drugs often leads to difficulties in fully characterising them and because of this, the approval procedures for biosimilars in Europe, while tuncated compared to a typical biologics license application still involves expensive clinical trials to prove equivalency, decreasing the potential cost savings that the drugs could deliver.

The US is still waiting for an approval pathway for 'follow-on biologicals' with the need becoming more pressing as patents for the first approved biopharmaceutical drugs are starting to expire.

If regulators can be convinced that biological drugs can be characterised as thoroughly as simpler small molecule drugs it could open the door for approval pathways similar to those used for small molecule generics where the active ingredients only need to be proved to be analytically equivalent.

Speaking at the 19 th Annual Drug Information Association meeting, held in Vienna earlier this year, Dr Richard Taylor, director of regulatory affairs at ERA consulting group said that current paradigm is that "the process is the product ".

He went on to explain that this is because European regulators currently believe that a biopharmaceutical product cannot be fully-defined by analytical characterisation alone, but is inexorably linked to the specific manufacturing process.

According to Kurt Leidner, head of communications at Sandoz - the generics arm of Novartis, in the past a static process for manufacturing biological drugs had to be fixed and validated within very tight limits often with a limited knowledge of how those process variables were interconnected.

"With the advances of analytical technology that have occurred in the past few years for the characterisation of biosimilar product development, it has became possible to limit the amount of pre clinical and clinical studies that need to be performed," he said " However, the biosimilar concept uses the data from pre clinical and clinical studies to verify and finally confirm the efficacy, and most importantly, the safety of a biosimilar drug product."

One of the reasons cited for many of the large pharmaceutical companies starting to produce biological drugs is the extra protection that the current regulations provide biological drugs even after there patents have expired.

Sandoz brought the first biosimilar, Omnitrope (somatropin [rDNA origin]), to market in Europe, which is a generic version of Pfizer's Genotropin.

Even after this landmark achievement, Sandoz teamed up with Momenta Pharmaceuticals to make use of their expertise in analysing complex biologics.

Dr Ganesh Venkataraman, co-founder and senior vice president for research at Momenta Pharmaceuticals, recently presented his views on the current level of analytical characterisation possible for biological drugs to a US House of Representatives Committee of Oversight and Government Reform hearing on "Safe and affordable biotech drugs - the need for a generic pathway".

In an interview, Dr Venkataraman said that the current techniques aimed at understanding protein structure and protein conformation could prove the equivalency of biologics to the same levels as required for small molecule generics, at least in some cases.

He continued by explaining that you have to frame your question around understanding the chemical components within a drug and to do that you have a long list of standard analytical techniques such as MS (mass spectrometry), NMR (nuclear magnetic resonance), HPLC (high performance liquid chromatography) etc and all the enzymes used to digest proteins.

"These analytical techniques are an integral part of the entire drug approval application, not just to be used for analysing the end product in terms of demonstrating equivalence but also making sure that you are able to develop an appropriate process," said Dr Venkataraman.

He said that there are some simpler proteins where a judicious collection of these techniques may allow the complete characterisation of a biological drug.

"But there are more complex proteins where one might argue that even with all the advances in today's techniques you are still looking for future innovations in analytical techniques to close the gaps between what information current techniques provide," he said.

Venkataraman believes there are techniques that are amenable to studying the more complex proteins, but highlighted the problems involved with fully characterising post translational modifications (PTMs) such as glycosylation and phosphorylation.

"As scientists we can hypothesise whether PTMs matter or not, but as the FDA using a risk management approach you need to be very careful that you are not making any assumptions."

He continued by saying that Momenta's core analytical technology is focussed at accurately and thoroughly characterising glycol (sugar) groups attached to proteins.

"Our approach is something of a novel approach from an analytics view-point because we don't pin our characterisation framework on one or two techniques, but rather we use a collection of tools - where each of these has some strengths and some blind spots - and we integrate the results to achieve thorough characterization," he said.

For example, he highlighted that while mass spectrometry techniques give certain types of information it is not the right tool to address questions about chirality or linkage information.

To answer these questions, he said that NMR, for example, may be more appropriate as it looks at the subtle differences between the electronic environments of atoms within a molecule.

"Momenta's entire platform is focused around leveraging the strength of these techniques and augmenting their blind spots by looking at the molecules through different technical perspectives."

This is exemplified that Momenta have dozens of techniques in their toolkit, such as ESR (electron spin resonance), CV (cyclic voltammetry), DSC (differential scanning calorimetry) and SPR (surface plasmon resonance).

"In small molecule drugs for example, there is usually just one active ingredient," said Dr Venkataraman.

"However, for a glycoprotein there are often very different glycoforms and one would argue that each of these glycoforms represents a component of the active ingredient."

This means that role of all these species needs to be understood and that the biogeneric product needs to have all these species present, with any differences needing to be substantially explained in any filing for approval.

"I would argue that if you're able to get to exactly the same combination of active ingredients then you've reduced the problem down to the same kind of framework as small molecule generics - although in terms of analytical detail it's many orders of magnitude more complex!"

he said.

He highlighted recent advances in MS, NMR, LC and sample handling to be particularly exciting, with the advances in MS over the past five years taking the information generated by the technique to a whole new realm.

"It's the same for NMR, with the cryoprobes and instruments getting close to 1GHz really increasing the resolution the technique can generate," he said.

"You can also almost do 2D LC and run separated fractions straight into MS or NMR instruments and I'm hoping these advances continue."

But important questions still remain.

For instance, even if the characterisation of biogenerics can be achieved to the levels that regulators would require to prove equivalency with the innovator products, will the cost of optimising the manufacturing process to such a level negate any potential cost savings?