Authorized manufacturing changes for therapeutic monoclonal antibodies (mAbs) in European Public Assessment Report (EPAR) documents

Abstract

Background The quality of biologicals, including biosimilars, is subject to change as a result of manufacturing process modifications following initial authorization. It is important that such product changes have no adverse impact on product efficacy or safety, including immunogenicity.

Objectives The aim of this study was to investigate the number and types of manufacturing changes for originator mAbs (the reference for the comparability exercise to confirm biosimilarity) according to European Public Assessment Report (EPAR) documentation and to ascertain the level of risk these changes might impart. The extensive body of evidence contained in the EPAR documents can help support the EMA during the EC marketing authorization approval process for biosimilars, since it provides a broad base of scientific experience.

Research designs and methods For EPAR-listed mAbs, details of all changes listed chronologically in the EPAR were evaluated and described. Based on these descriptions the manufacturing changes can be categorized by risk status (low, moderate or high).

Results Entries for 29 mAbs with publicly available EPAR reports were reviewed. These contained details of 404 manufacturing changes authorized by the European Medicines Agency (EMA): 22 were categorized as high risk, 286 as moderate risk and 96 as low risk manufacturing changes. A limitation of this analysis is that it only summarizes publicly available data from EPAR documents.

Conclusions Manufacturing change data indicate that the EMA has significant experience of process changes for originator mAbs, and the impact they may have on the efficacy and safety of biologicals. This experience will be useful in biosimilar product development to ensure adherence to sound scientific principles. Compared with the established manufacturing process for a reference product, the production of biosimilars will usually be different. Consequently, in addition to a comprehensive comparative functional and physicochemical characterization analysis, clinical data is required to confirm mAb biosimilarity.

Keywords:

• Biologicals
• Biosimilarity
• Biosimilars
• Comparability
• European Public Assessment Report
• Manufacturing
• Monoclonal antibodies

Introduction

It has been estimated that more than 250 biologicals derived from recombinant DNA technology have been approved for medicinal use in humans¹. The high cost of these agents has created a barrier to their routine use. The introduction of biosimilar medicinal products (biosimilars), non-identical but highly similar copies of previously approved biological therapeutics (referred to as the reference products), provides a method of improving access to these drugs in a more cost-effective manner². This might facilitate wider usage in a greater number of individuals who could benefit from such treatments. However, biological agents are generally large complex proteins that are difficult to characterize and difficult to copy. Differences in structure between the biosimilar and reference product, small or large, may impact the properties of the biosimilar³. The European Medicines Agency (EMA) has developed a regulatory framework for the development of biosimilars to ensure that they are highly similar to the reference product, in terms of structure and function, pharmacodynamics and mechanism of action, pharmacokinetic properties, and clinical efficacy and safety (including immunogenicity)⁴. A landmark in biopharmaceutical research was achieved in September 2013 when, based on the scientific recommendation of the EMA, the EC granted marketing authorizations for two biosimilar infliximab monoclonal antibodies (mAbs) for use in the same indications as the originator reference infliximab product^5,6. The Food and Drug Administration (FDA) in the United States has issued draft guidelines for the development of biosimilars⁷, and in March 2015 Zarxio*(filgrastim-sndz) was the first biosimilar product to be approved in the United States⁸.

Approval of the first biosimilar agents in Europe and the United States has brought the issue of variability of biological medicines to the attention of practicing clinicians. The quality attributes of all biotechnology products are subject to change as a result of manufacturing process modifications following the initial authorization. This applies to reference products as well as to biosimilars³. According to the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use guideline (ICH Q5E), a comparability exercise should provide analytical evidence that, based on appropriate comparison of relevant quality attributes, pre- and post-change products are highly similar and considered comparable. Such product changes should have no adverse impact on product efficacy or safety, including immunogenicity⁹. The comparability exercise has become the scientific norm for biosimilar development, and is based on a combination of chemical and analytical testing, biological assays, and, in some cases, additional non-clinical and clinical (safety and/or efficacy) data³. In a review of biosimilars development Schneider presented details of the number of manufacturing changes after authorization for 11 mAbs and fusion proteins (e.g. etanercept) approved for use in rheumatology indications. The number of changes ranged from 0 (rilonacept) to 37 (infliximab) with a median of four. Three products had in excess of 15 changes in manufacturing (infliximab, etanercept and rituximab)³. The documented changes could be as simple as a change of address of the authorization holder to more complex changes in the manufacturing process.

The aim of the current analysis was to update the findings of Schneider in relation to mAbs, and to collect and thoroughly investigate the number and types of manufacturing changes for originator mAbs according to their European Public Assessment Report (EPAR) documentation¹⁰. The EMA’s EPAR documents provide a broad level of scientific experience which can be used in the future to evaluate process changes for biological drugs (including biosimilars) so that the necessary comparability tests can be applied to help obtain EC marketing authorization approval.

Methods

In October 2014 publicly available EPARs for all mAbs authorized by the EMA between 1998 and October 2014 were analyzed¹⁰. Details of all changes listed chronologically in the ‘Procedural steps taken and scientific information after the authorisation’ section were evaluated and categorized by risk status (low, moderate or high), as defined by Lee et al. (Figure 1)¹.

Figure 1. Manufacturing changes for biotechnology products were categorized as low/moderate/high risk¹. Adapted from Lee et al. Comparability and biosimilarity: considerations for the healthcare provider. CMRO 2012;28:1053-8.

Below we include examples of statements from EPARs registered with the EMA. The changes were assessed by regulatory experts and, depending on the significance of the change, they were rated as low, moderate or high risk. The same change (e.g. change in manufacturer) can be rated at different risk levels depending on the significance of the change. The following published (EPAR) examples are categorized as high-risk status^1,10:

• Change in the purification of the active substance.

• Change in synthesis or recovery of a non-pharmacopoeial or novel excipient.

• Change in the manufacturer of AS (active substance) or of a starting material/reagent/intermediate for AS – the proposed manufacturer uses a substantially different route of synthesis or manufacturing conditions.

• Change in batch size (including batch size ranges) of AS or intermediate – the change requires assessment of the comparability of a biological/immunological AS.

• Change to in-process tests or limits applied during the manufacture of the AS – widening of the approved in-process test limits, which may have a significant effect on the overall quality of the AS.

We categorized from the EPAR documents the following examples as moderate-risk manufacturing changes^1,10:

• Change(s) to the manufacturing process for the active substance.

• Changes in the manufacturing process of the AS – the change refers to a [-] substance in the manufacture of a biological/immunological medicinal product and is not related to a protocol.

• Change in the manufacturer of AS or of a starting material/reagent/intermediate for AS.

• Change to in-process tests or limits applied during the manufacture of the AS.

• Change in the manufacturer of AS or of a starting material/reagent/intermediate for AS – the change relates to a biological AS or a starting material [-] used in the manufacture of a biological/immunological product.

Low-risk manufacturing changes are based on the following EPAR examples^1,10:

• Change in the manufacturing process of the finished product (FP).

• Replacement or addition of a manufacturing site for the FP – site where any manufacturing operation(s) take place, except batch release, batch control, and secondary packaging, for biological/immunological medicinal products.

• Change to in-process tests or limits applied during the manufacture of the finished product.

• Change in immediate packaging of the finished product – qualitative and quantitative composition.

• Changes to the manufacturing facility of the finished product.

Results

A total of 36 entries for mAbs were found published on the EMA website and 29 of these had publicly available EPAR reports (Table 1). As of October 2014, these 29 EPAR reports contained details of 404 manufacturing changes authorized by the EMA. Of these, 22 were categorized as high-risk, 286 as moderate-risk and 96 as low-risk manufacturing changes (Figure 2). The number of changes ranged from 0 (pertuzumab) to 50 (infliximab) with a median of 11 per product, and five products had in excess of 25 changes in manufacturing (infliximab, basiliximab, trastuzumab, adalimumab and eculizumab) (Figure 2). Since the elapsed time after authorization differs significantly (1998–2013) the annual number of manufacturing changes was introduced for comparison. The annual average number was 1.8 (range 0–3.71) (Figure 3).

Figure 2. Number of manufacturing changes for monoclonal antibodies in their European Public Assessment Reports according to risk category (during the search period all non-proprietary names relate only to the trade named medicines listed in Table 1).

Figure 3. Average number of manufacturing changes per product year elapsed since registration (during the search period all non-proprietary names relate only to the trade named medicines listed in Table 1).

Table 1. Monoclonal antibody medicines and their EPAR* ‘Procedural steps…’ documents from EMA** webpage during the search period¹⁰.

	International non-proprietary name	Trade name***	First EU (US) approval year	Regulatory status in EU	EPAR documents in analysis^****	Last update of ‘Procedural steps taken and scientific information after the authorisation’ document was evaluated
1	Muromonab-CD3	Orthoclone Okt3	1986 (1986)	Voluntarily withdrawn	N.A.
2	Abciximab	Reopro	1995 (1994)	Voluntarily withdrawn	N.A.
3	Rituximab	MabThera, Rituxan	1998 (1997)	Approved	Included	09/07/2014
4	Basiliximab	Simulect	1998 (1998)	Approved	Included	02/10/2014
5	Daclizumab	Zenapax	1999 (1997)	Voluntarily withdrawn	Included	08/04/2008
6	Palivizumab	Synagis	1999 (1998)	Approved	Included	15/09/2014
7	Infliximab	Remicade	1999 (1998)	Approved	Included	25/09/2014
8	Trastuzumab	Herceptin	2000 (1998)	Approved	Included	01/09/2014
9	Gemtuzumab ozogamicin	Mylotarg	NA (2000)	Voluntarily withdrawn	N.A.
10	Alemtuzumab	MabCampath, Campath-1H	2001 (2001)	Not authorised	Included	15/08/2012
11	Adalimumab	Humira	2003 (2002)	Approved	Included	14/10/2014
12	Tositumomab-I131	Bexxar	NA (2003)	Not authorized	N.A.
13	Efalizumab	Raptiva	2004 (2003)	Voluntarily withdrawn	Included	04/08/2009
14	Cetuximab	Erbitux	2004 (2004)	Approved	Included	14/08/2014
15	Ibritumomab tiuxetan	Zevalin	2004 (2002)	Approved	Included	17/06/2014
16	Omalizumab	Xolair	2005 (2003)	Approved	Included	03/10/2014
17	Bevacizumab	Avastin	2005 (2004)	Approved	Included	15/08/2014
18	Natalizumab	Tysabri	2006 (2004)	Approved, indication extension withdrawn	Included	11/08/2014
19	Ranibizumab	Lucentis	2007 (2006)	Approved	Included	24/04/2014
20	Panitumumab	Vectibix	2007 (2006)	Approved	Included	02/07/2014
21	Eculizumab	Soliris	2007 (2007)	Approved	Included	11/04/2014
22	Certolizumab pegol	Cimzia	2009 (2008)	Approved	Included	28/07/2014
23	Golimumab	Simponi	2009 (2009)	Approved	Included	18/08/2014
24	Canakinumab	Ilaris	2009 (2009)	Approved	Included	01/10/2014
25	Catumaxomab	Removab	2009 (N.A.)	Approved	Included	23/04/2014
26	Ustekinumab	Stelara	2009 (2009)	Approved	Included	23/05/2014
27	Tocilizumab	RoActemra, Actemra	2009 (2010)	Approved	Included	29/09/2014
28	Ofatumumab	Arzerra	2010 (2009)	Approved	Included	05/08/2014
29	Denosumab	Prolia	2010 (2010)	Approved	Included	29/08/2014
30	Belimumab	Benlysta	2011 (2011)	Approved	Included	12/08/2014
31	Ipilimumab	Yervoy	2011 (2011)	Approved	Included	11/03/2014
32	Brentuximab vedotin	Adcetris	2012 (2011)	Approved	Included	29/08/2014
33	Pertuzumab	Perjeta	2013 (2012)	Approved	Included	30/07/2014
34	Raxibacumab	(Pending)	N.A. (2012)	Granted orphan design	N.A.
35	Trastuzumab emtansine	Kadcyla	2013 (2013)	Approved	N.A.
36	Vedolizumab	Entyvio	2014 (2014)	Approved	N.A.

*European Public Assessment Report.

**European Medicines Agency.

***All medicine names are registered trademarks of their owner companies.

****Included or (N.A.) not available for analysis during the search period of 1 June 2014–15 October 2014.

Discussion

Biological products vary over time as a result of modifications to the manufacturing process. Consequently, manufacturers need to demonstrate the consistency and robustness of the manufacturing process by implementing good manufacturing practices, modern quality control and assurance procedures, in-process controls and process validation¹¹. This applies to both originator products and biosimilars, and the manufacturing process for a biosimilar should be optimized to minimize differences between it and its reference protein. As noted in this short review, the regulatory authorities have extensive experience of assessing manufacturing changes over time and their impact on the efficacy and safety of the final product, for a wide range of mAbs. These same principles, using state-of-the-art science and technology, are being applied to biosimilar manufacturing and quality control.

The scientific principles for assessing manufacturing process changes of mAbs are well established, and are based upon comprehensive and thorough physicochemical, analytical and functional comparative assessments to ensure comparability. These principles are encompassed in the ICH comparability guidance document ICH Q5E which has been used by the EMA and other regulatory bodies to develop their comparability guidelines⁹. This guideline defines the rationale and background for accepting changes in the production of biological products and acknowledges that… “the demonstration of comparability does not necessarily mean that the quality attributes of the pre-change and post-change product are identical, but that they are highly similar and that the existing knowledge is sufficiently predictive to ensure that any differences in quality attributes have no adverse impact upon safety or efficacy of the drug product.” This applies to both reference proteins and biosimilars. Interestingly, Schiestl and colleagues investigated the variation in quality of different batches of three marketed biological therapeutics (darbepoetin alfa, rituximab and etanercept) and found differences in the glycosylation profiles of all three products¹². The originators’ batch variants were predicted not to alter the clinical profiles of any drugs, and these were acceptable to the health authorities and all three products remained on the market.

Trust in the development of biosimilars is founded on the expertise of the regulatory authorities which has been gained from their experience with originator biologicals. Indeed, the entire biosimilar development process is built on a solid foundation of extensive analytical and physicochemical characterization which is robustly assessed. This is supported by the accumulated experience of the EMA over the last couple of decades from the scientific evaluation of manufacturing changes of originator biologicals and their impact on clinical outcomes.

The findings of this study are limited to publicly available data from the EU regulatory authority’s documents for mAbs for the time period 1998 to October 2014 (Table 1). Since the manufacturing processes of biologics are continuously changing during their life-time, monitoring and assessing the changes will provide examples of change consequences in the production process which can help guide future decisions. It should be noted that no information of process changes for biosimilar mAbs was available up to October 2014.

Conclusions

The manufacturing change data presented herein indicate that, prior to the authorization of the first biosimilar mAb, the EMA had extensively evaluated the manufacturing process changes of originator mAbs, and gained significant experience in the change process and its impact on the safety and efficacy of biologicals. These comparability exercises became the guiding principles of biosimilarity to confirm no meaningful differences in quality, safety and efficacy. These exercises have been employed in biosimilar product development to ensure that sound scientific principles are adhered to. Since the manufacturing process for biosimilars will likely be different from the reference product for proprietary reasons¹, physicians ought to be able to trust the expertise of regulatory authorities to confirm the similarity of previously approved originator products and their biosimilars akin to the pre- and post-manufacturing changes of original biologicals.

Transparency

Declaration of funding

This paper was funded by Egis Pharmaceuticals plc, Budapest, Hungary.

Declaration of financial/other relationships

B.V., Z.B., M.S., and Z.Z. have disclosed that, during the publication process, they were employees of Egis Pharmaceuticals plc.

CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Notice of correction

Please note that Figure 1, 2, 3 and Table 1 have been corrected since the article was first published online (25 February 2016).

Acknowledgments

The authors thank Dr Steve Clissold of Content Ed Net for editorial support.

Previous presentation: Parts of this paper were presented at 35th European Workshop for Rheumatology Research, 5–7 March 2015, Budapest, Hungary¹³.