EMA Offers Edits on FDA Draft Guidance on Biosimilars
Posted 29 August 2019 | By
Although regulators rarely offer public comments on another regulator’s guidance, the European Medicines Agency (EMA) late last month offered line-by-line comments and edits on a recently released draft guidance from the US Food and Drug Administration (FDA) on comparative analytical assessments for biosimilars.
While noting the close collaboration between EMA and FDA on the statistical aspects of comparative analytical assessments, the EMA comments feature questions for clarification, recommendations to omit and areas of commonality. EMA pledges its support for the sections of the guidance related to the preparation of a “(prospective) comparative analytical assessment plan” and “accounting for lots/batches to be used for comparison.”
But EMA’s Biostatistics, Biosimilar and Biologics Working Parties also assert areas where FDA may differ from EMA.
“When guiding upon comparative analytical assessment, FDA keeps the focus on biosimilar development issues, whereas EMA aims at keeping a broader scope in the revision of their reflection paper, also e.g. including aspects concerning pre-post change comparisons. This difference in scope is of relevance for further guidance development, in particular in relation to differences in complexity and differences in required rigor to demonstrate similarity in quality attributes (QAs) in different regulatory settings,” EMA said.
The 28-page draft guidance revises a final guidance from 2015 on quality considerations
for demonstrating biosimilarity and serves as a replacement for a 2017 draft guidance on statistical approaches to evaluating biosimilarity that was withdrawn
last year after industry questions
Other commenters on the draft included Novartis, Pfizer, Roche’s Genentech and Celltrion.
Novartis’ Florian Bieber and Jessica Rizzo took issue with the draft’s requirement that companies cannot pool data across US and non-US-licensed products.
“The draft guidance states that as a scientific matter, it is not acceptable to pool data obtained with US and non-US-licensed products when determining the acceptance criteria or when performing the comparative analytical assessment. We propose the Agency to strike the phrase ‘as a scientific matter’ since we believe that this requirement represents regulatory policy and is not a scientific matter. In cases where US and non-US-licensed product are made by the same process and employ the same control strategy, they provide a single data population with respect to quality attributes. Therefore, we believe that pooling of data should be allowed in cases where it is possible to provide evidence that the manufacturer of the reference product is using the same manufacturing process for the US reference product and the non-US comparator product,” they wrote.
Pfizer’s Laura McKinley similarly criticized FDA’s stance and said such pooling of data “should be acceptable if bridging has been established.”
Novartis’ Bieber and Rizzo also took issue with the distribution of quality attributes, noting that one line in the draft runs “a very real risk that manufacture of biosimilars will be held to stricter requirements than those that are applied to their respective reference products. Depending on the degree by which this potential requirement is applied by the Agency, this may, in the worst case, lead to an unintentional bias in market competition between companies that may ultimately limit patient access to biologic medicines.”
Genentech/Roche also criticized FDA’s use of the quality ranges (QR) method for data analysis. “Since the QR method encourages smaller sample sizes and does not exemplify scientifically rigorous regulatory policy,
Genentech/Roche recommends that FDA withdraw its endorsement of the QR method,” the company said.
Celltrion, meanwhile, noted that under section 351(k)(2)(A)(i)(IV) of the Public Health Service Act
, an applicant “must demonstrate that the ‘strength’ of the proposed biosimilar product or proposed interchangeable product is the same as that of the reference product.” But the new draft, Celltrion notes, “is silent on ‘strength’ and on the label claim, and [Celltrion] recommends that FDA clarify their position, or otherwise include reference to” other applicable guidances.