Draft WHO Guidance Provides Two Pathways For Biosimilars

Clinical trials would be required for most biosimilars under two approval scenarios that are being developed by the World Health Organization.

Guidelines for the two biosimilar pathways - a biosimilars approach and a clinical comparability approach - have been drafted by a working group and will be reviewed by the WHO Expert Committee on Biological Standardization during a meeting in Geneva, Oct. 13-17.

The working group met in May 2008 and the draft guidelines have been fluid as work has continued on the document.

Under the first option, similar efficacy of the biosimilar and a reference product "will usually have to be demonstrated in adequately powered, randomized, parallel group clinical trial(s)," notes a recent version of the draft guidelines.

These studies preferably should be double-blind or, at a minimum, observer-blind, the draft says.

In the clinical comparability approach, head-to-head clinical studies with the reference product must be conducted to demonstrate comparable safety and efficacy, the draft says. The studies could have either a non-inferiority or equivalence design.

"The ultimate demonstration that the product is clinically comparable and non-inferior to the comparator product will be obtained from the required head-to-head comparative clinical studies."

The importance of clinical trials was stressed by the American Society of Clinical Oncology in its comments to the House Energy and Commerce Committee on a potential biosimilars pathway in the U.S. (¹ (Also see "Most Biosimilars Would Need Clinical Trials, ASCO Tells House Panel" - Pink Sheet, 9 Jun, 2008.), p. 27).

Domestic Or Foreign Reference Products

Both pathways developed by the working group for approving a follow-on biologic would require "demonstration of similarity" to a reference product.

But acknowledging that some countries may not have licensed biologics, the draft suggests the use of a reference product from another country. This "will increase opportunities for better access to alternatives to innovator biologics and/or entry of biosimilar products to some markets."

In such cases, clinical trials will provide the national regulatory authority with important data for assessing a biosimilar under the clinical comparability approach, the draft points out. NRAs should establish additional criteria for using foreign reference products, the draft says, but provides no guidance in this regard.

The draft guidance focuses on scientific aspects of the program. It does not discuss patents or data exclusivity and concludes that "decisions on interchangeability [substitution of an equivalent medication] should be the responsibility of a national regulatory authority and based on appropriate scientific and clinical data."

While International Nonproprietary Names should be the same for biologics and their biosimilars, the draft says regulatory agencies should ensure that a product related to an adverse event is readily identifiable. This implies that information on an adverse event should include indicators such as a proprietary name, manufacturer name, lot number and country of origin, the draft says.

In principle, an abbreviated pathway to approval should apply to all biologics, but the working group notes that "in practice, it will mostly be applicable to well-established and well-characterized recombinant DNA-derived therapeutic proteins."

More Similarity Enables Extrapolation

With the biosimilar approach, a similarity of all characteristics with the reference product must be demonstrated in terms of quality, safety and efficacy, the draft says. Studies must be comparative in nature and analyses must be sensitive enough to detect potential differences.

The advantage to this approach, the draft points out, is that "it opens the possibility that the product may be approved for use in more clinical indications than were directly tested in clinical trials with the biosimilar product."

A quality comparison between the biosimilar and innovator product "is essential to allow extrapolation of clinical safety and efficacy data of the reference medicinal product to the biosimilar product," the draft guidance stresses, noting that "differences in critical product quality attributes may prevent" using the biosimilars approach.

To extrapolate, a sufficiently sensitive test model must be used, the mechanism of action and/or involved receptor(s) of the reference product and biosimilar must be the same, and safety and immunogenicity must be sufficiently characterized.

Because non-clinical studies are part of the overall comparability exercise, these should be conducted to detect differences in the response of the biosimilar and reference product. This also is true of pharmacokinetic studies, the draft adds.

As part of the efficacy trials, the "frequency and type of antibodies induced, as well as possible chemical consequences of the immune response, should be compared" for the two products.

With the clinical comparability approach, "It is not necessary to show that all molecular features of the product are nearly identical to the reference product," only that it has the same types and degrees of biological activity as the reference product.

This could be shown through in vitro biological characterization and a demonstration of comparable efficacy and safety during the confirmatory clinical trials, the draft points out. Considerations for immunogenicity testing are the same as those described in the biosimilar approach.

The clinical comparability approach should be taken only when there is extensive knowledge about the pharmacological properties, safety and efficacy of one or more reference products and the substance class's biological activities are well-known, the draft advises.

Clinical trials would be required for each indication. Extrapolation to other indications of the reference product generally is not possible because of "incomplete assurance" that no clinically relevant differences would occur in other indications or at other doses, the draft notes.

The amount of safety and efficacy data required at the non-clinical level will be dependent on clinical experience with the product class. It usually should include the spectrum of studies needed for the biosimilar approach, although they need not be comparative, the draft says.

- Cathy Dombrowski ([email protected])