The regulation of similar biological medicinal products ("biosimilars") is governed by article 10(4) of Directive 2001/83 as amended, supplemented by the Annex to the Directive. The Annex in turn refers to a guideline published by the EMA that governs how biosimilars are to be assessed. There are three guidelines that apply to all biosimilars: an "overarching" guideline which was adopted in September 2005; and two further guidelines which were both adopted by the EMA on 22 February 2006, one relating to non-clinical and clinical issues and one to quality issues. These guidelines are supplemented by a number of product or class specific guidelines, also issued by the EMA.

Since 2006 the EMA has gained much experience in biosimilars; it has authorized 14 biosimilars1 and issued scientific advice relating to many more. In 2011 the EMA therefore started a review of the overarching guideline, the non-clinical and clinical issues and quality issues guidelines so that they could be revised to reflect the way that applications for marketing authorisations for biosimilars are currently being considered. The EMA has now adopted draft versions of the revised guidelines and released them for consultation.

The issues addressed by these guidelines are overlapping but distinct. It is interesting at this stage to consider what the EMA's thinking is in relation to the big issues that were raised for each of the guidelines.

Overarching guideline

The concept paper for the overarching guideline identified certain areas that required clarification. In particular, the principles on which biosimilars should be assessed should be identified more clearly.

One issue was whether a biosimilar, with adequate additional data, could be of a different pharmaceutical form, strength or route of administration from the reference biological product. The draft guideline clarifies that a biosimilar should have the same posology and route of administration but that changes in the formulation or excipients included are possible but must be justified or further studies conducted.

Often, biological reference products might be sold globally and a company may wish also to market a biosimilar globally. The draft guideline gives the EMA the discretion to allow comparative studies to be conducted with a reference product that is not authorised in the EU but in a country with similar regulatory standards as the EMA. Such a possibility is designed to avoid unnecessary repetition of clinical trials but must be scientifically justified.

Non-clinical and clinical issues guideline

The issues raised in the non-clinical and clinical issues concept paper were: the relevance of animal models for biological products; the clinical trial design for comparability studies; and whether or not extrapolation from one indication to another was possible.

In vitro studies, which in many cases will be more sensitive in detecting any differences between the biosimilar and the reference medicinal product are paramount for the non-clinical comparability exercise. The draft guideline recognises that there is not always an appropriate animal model in which non-clinical in vivo studies can be conducted. The recommendation is first to carry out studies in vitro and then to consider whether or not (on a risk-based approach) it is necessary to move into animal models. Factors that are identified as relevant to this decision include: the presence of relevant quality attributes in the biosimilar, such as glycosylation patterns, that are different from the reference product; and any relevant differences in formulation. The guideline makes clear that animal studies will not always be considered necessary.

The guidance on clinical studies is split into sections relating to pharmacokinetic (PK) and pharmacodynamic (PD) studies and clinical efficacy. PK and PD studies should be carried out first in the most sensitive patient population, which is most likely to show any differences between the biosimilar and the reference product. In some rare circumstances, where there is a clear dose response and there is a PD marker that is accepted to be related to patient outcome and so can be used as a surrogate endpoint, it may be unnecessary to carry out clinical efficacy studies.

In most cases, however, clinical efficacy studies will be required. Such studies are not intended to demonstrate efficacy per se but rather to identify any clinically meaningful differences between the biosimilar and the reference product. The studies will typically be equivalence studies and will assess whether the biosimilar is neither better nor worse than the reference product. In situations where increased efficacy can be excluded because of the mechanism of action, it may be possible to design non-inferiority studies which would require fewer patients to be enrolled in the study but this should be agreed with the EMA in advance.

In some cases, biologic products are indicated for a number of different indications. Carrying out trials into the clinical efficacy of each and every indication for a biosimilar would be time-consuming and expensive. One area of debate has been whether or not it is possible to carry out clinical efficacy studies in relation to one indication and then to extrapolate the results to the other indications. The guideline states that the inclusion of each indication must be demonstrated (by studies) or justified. Extrapolation to a different indication is therefore possible but must be justified. The justification will be different for each biologic and must take into account clinical experience, literature and mechanisms of action. In particular, any justification for extrapolation of indication must consider different effects of the drug in different tissues and patient factors such as co-medication and co-morbidities.

Quality issues guideline

During the life cycle of a biologic product, the manufacturing process is likely to change and this may lead to changes in the product. The concept paper for quality issues therefore set out to clarify the expectations in relation to the structure, expression system and formulation of a biosimilar for the comparability exercise and throughout its lifecycle.

The manufacturing process may introduce different molecular variants and isoforms and any potential changes, as compared to the reference product, that might be introduced by the manufacturing process should be considered carefully. The use of novel expression systems is highlighted as something in particular to consider as this could lead to different glycosylation patterns and impurity profiles as compared to the reference product.

The formulation of the biosimilar does not have to be identical to that of the reference medicinal product but any new formulation and excipient must be justified.


Overall, the guidelines provide some indication as to the likely approach of the EMA but the details of the necessary studies will be different for individual cases. The recommendation in the guidelines is to discuss the issues with the EMA at an early stage in the development of the biosimilar. There is the opportunity to provide comments on these draft guidelines before 31 October 2013 for the overarching guidelines and before 30 November 2013 for the non-clinical and clinical issues and quality issues guidelines.