Julie Warner, VP of Regulatory Affairs at Boyds, has co-edited the latest issue of TOPRA‘s Regulatory Rapporteur. In this issue, there is an article co-authored by Julie and Kathryn Parsley of Gyroscope Therapeutics, entitled ‘The critical role of a CMO in advanced therapy clinical and commercial product manufacture’ which considers some of the complexities of ATMP manufacture and recommends aspects to consider when selecting a CMO.
The following article was published in the Regulatory Rapporteur, Volume 17, Number 2, February 2020.
Recent years have seen a significant increase in the number of advanced therapy medicinal products (ATMPs) both in development and reaching the market (see Table 11 and 2). This increase has resulted in a distinct shortage of contract manufacturing organisations (CMOs) specialising in ATMP manufacture, despite the fact that a number of companies have recently either expanded existing, or built new, facilities. This, coupled with the manufacturing challenges associated with these types of products at all stages, can result in several difficulties for companies. In addition, as technology has advanced and the number of these types of products has increased, the existing regulations have also evolved (resulting in the Guidelines on Good Manufacturing Practice [GMP] specific to ATMPs)2 and this combination introduces a number of unique concepts that present fresh challenges for traditional (new chemical entity [NCE]) manufacturers, who cannot meet the requirements of GMP for ATMPs. It is clear that many CMOs have had to significantly invest in facilities, technical teams and expertise, and company infrastructure and processes to ensure they remain competitive and achieve growth targets. In some cases, this may mean that effectively competing in this fast-moving arena may require the forging of collaborations and partnerships and the acquisition of other companies that can fill gaps in expertise compared to key competitors.
Table 1. Increase in number of clinical trials per product type (2015–2018).1
|Year||In vivo gene therapy||Ex vivo gene-modified cell therapy||Cell therapy||Tissue-engineered||Total|
The large volume of ATMPs reaching late-phase clinical development are able to benefit from the many incentives introduced by regulators in the EU to expedite the development of medicinal products that have the potential to address significant unmet needs. These are not all specific to ATMPs but the specificity of these products (a targeted mechanism of action and often a small, well-defined niche indication) means that they are generally well suited to such procedures (as indicated in Table 2).
Table 2: Summary of EU-approved ATMPs and incentives
|Tradename||Product type||ATMP classification||Orphan drug designation||PRIME scheme access?||Type of MA granted|
CTMP = cell therapy medicinal product; GTMP = gene therapy medicinal product; MA = marketing authorisation; NA = Not available; TEP =- tissue engineered product; ? = possible
Biopharmaceutical manufacturers are focused on rapid deployment of products into the marketplace. Their efforts include developing facilities and supply chains to handle increasing patient demands worldwide, reducing manufacturing costs, increasing the efficacy and potency of these novel modalities, and increasing safety for patients by reducing potential side-effects of gene therapy products. Major components that address these challenges involve the implementation of novel, efficient, and robust manufacturing and analytical technologies for the production of gene therapy products and developing a well-defined global regulatory framework for process validation and batch release.
The manufacture of ATMPs is a specialist area. ATMPs require often complex, critical starting materials compared to NCEs and standard biologics. These are unique to the end product, can be specific to the patient and treatment indication. In addition, they are likely to incorporate additional factors that are built in to optimise the manufacturing process and/or quality, safety and overall efficacy of the end product such as plasmids incorporating promotors/enhancer sequences as well as engineered cell lines and human derived autologous or allogenic cells that introduce significant complexities compared with the manufacture of NCEs. In the creation of viral vectors, for instance, more than one plasmid is often required to introduce these elements predominantly for safety purposes. Further, different viral elements confer different properties and this also needs to be accounted for. Multiple cell lines, each with their own characteristics (and therefore strengths and weaknesses), may also be used for production. Finally, some analytical methods are similar across product types, but there are often key differences and complexities that impact the final decision on which to use (and their appropriateness at early and late stages of development).
Significant consolidation is starting to occur in the ATMP CMO space. Recent months have seen the acquisition of Cobra Biologics by Cognate Biosciences, and of Paragon Bioservices by Catalent, to name but a few. This obviously represents a huge opportunity for the sector to “get it right”, with investment in highly skilled staff (who are appropriately experienced and trained), and high specification technological facilities (up to date manufacturing equipment, provision for adaptation to accommodate future technological developments, and in-built operational efficiency and quality).
Manufacture is usually the rate-limiting factor in the development of ATMPs. These products are often on an accelerated or condensed clinical development programme as they are generally for rare diseases with a high unmet medical need. To keep pace with the accelerated clinical development, manufacture needs to take place in around 30–50% of the time of that of a “standard” (ie, NCE or standard biologic) development programme. Incentives are available to help expedite development; however, these do not, on the whole, help to reduce manufacturing time and the only option that could (ATMP quality certification from the EMA) has not been widely taken up. Further, ATMP quality certification is not a guarantee, and it is only open to small and medium-sized enterprises (SMEs), who unfortunately often do not have the resources to pursue this with so many other competing priorities. In addition, investment in an SME by a non-SME (in the case of a licensing deal or development partnership) often results in loss of SME status, and thus access to ATMP certification as a tool, for the developer.
It is apparent that “CMC preparedness” is rapidly becoming a hot topic for ATMPs, as evidenced by the significant number of publications, communications and meetings in this regard by organisations such as the Alliance for Regenerative Medicine3 and the ATMP Manufacturing Community.4 This article presents some of the critical items that a regulatory affairs professional can identify and raise during selection of a CMO for, and subsequent manufacture of, an ATMP, largely from the perspective of the in-licensing of an existing product from an academic environment. However, many of the principles also apply to the general ATMP development setting.
It is well recognised that there is currently a lack of regulatory CMC professionals with extensive experience of ATMPs at this time. This has presented several challenges for industry and, as a result, schemes now exist to encourage and build regulatory experience in the manufacture and control of ATMPs; the past one to two years have seen the emergence of, for example, apprenticeships, training courses, and dedicated on-the-job training as the importance of this role is increasingly recognised. It stands to reason that it cannot just be industry facing these challenges, as the pool of potential employees is also the same pool available to regulators. This situation will, of course, improve over time as the number of ATMPs increases and experience is gained, but this will take a few years.
One of the biggest challenges facing ATMP developers looking to in-license products is that many of those projects originate from academia or hospital research and the method of manufacture used is largely unsuitable for, or unable to support, later-stage development. Although it is well recognised that academics have the scientific expertise (having often designed the construct and performed at least proof of concept pharmacology studies, and possibly even toxicology studies), their background is not traditionally strong in the technical aspects of development and manufacture. As such, the technical expertise, staff and facilities are not often appropriate for development beyond the initial stages owing to a lack of industry experience; in addition, their end goal often differs from that of industry (being focused on publishing papers rather than compliance with GMP and the long-term big picture) and groups frequently deal with limited financial support, which is often only in the form of research grants.
Based on experience, the academic origins and lack of GMP compliance often make it difficult to obtain information on key aspects of the ATMP’s development. For instance, when development work has been conducted in the academic setting, the nature of the material studied may not be clear (eg, the vector may not have been fully sequenced in compliance with the quality standards required for investigational medicinal product [IMP] development). Critically, any licensee should seek to determine who owns the intellectual property. Plasmids are often provided as “gifts” from one academic group to another, without any licence or rights to use them, which can result in challenges later on in development when information is required to support genetically modified organism (GMO) notifications for clinical trials, and marketing authorisation applications (MAAs). This information is very difficult to source years down the line if not documented, or is insufficiently recorded in research laboratory notebooks. The answer is often just to chemically re-synthesise the plasmid(s) to provide a definitive starting point.
However, this introduces another key topic: that of comparability. With nonclinical toxicology programmes often extremely abbreviated for ATMPs, consisting of just one or two studies in a single relevant species, completion of this work in the academic setting prior to in-licensing may not always be beneficial. Such studies are often not conducted to good laboratory practice (GLP) and may be done with research-grade material. As such, unless it can be justified, those studies may be considered insufficient by regulators to support clinical trials or MAAs, and comparability certainly needs to be discussed with regulators (via scientific advice) if the manufacturing process, site or starting materials change over the course of development, as this may mean that such work would need to be repeated. This, of course, requires close and detailed collaborations between the academic group and their regulatory counterparts within the licensee organisation to obtain the relevant information to support the project, and then between regulatory and product development functions to put in place a project strategy from drug discovery through to licensing.
One of the first critical decisions is to bring a CMO on board as soon as possible. Although this is a huge cost commitment at an early stage of development, it is an invaluable activity. However, it is vital that the right organisation is chosen; with comparability being a significant issue with the potential to result in repeat work, it is best to select a CMO that will be able to support all stages of development through to commercial supply to avoid the need, for example, for (transfer and) re-validation of analytical methods. When selecting a CMO, the company should be aware that a different approach to that taken for conventional medicinal products is required, as a result of manufacturing complexity. Therefore, it is best to be as transparent as possible to identify and avoid potential issues up front (for instance, the use of a particular antibiotic resistance gene such as ampicillin in a plasmid may result in a CMO refusing to handle the material in its facility). The CMO will also need to know the provenance of the starting materials; for example, in the case of immortalised cells, the CMO will expect to be told how these were immortalised and what the quality of that material was. Companies need to be prepared to accept that a CMO may not be willing to manufacture the product if this information cannot be provided.
This transparency should continue across the whole development programme. As mentioned before, it is advisable to avoid changing manufacturers for ATMPs to avoid the need to demonstrate comparability. Therefore, companies should clearly communicate the overall programme timings to the CMO to help it ascertain whether it will have enough capacity to support all manufacturing development activities through to commercialisation. This includes process validation on a certain number of batches – the plans for which, again, should always be discussed with regulators as early as possible. If information from regulators is available to inform the plans, this could also be shared (as it is recognised that the FDA has allowed companies to launch with clinical trial material; however, process validation is still required).
Several factors can contribute to the successful selection of the right CMO for an ATMP. Use of a CMO with a regulatory chemistry, manufacturing and control (CMC) function is of immense help to expedite access to documents and manufacturing information. In addition, flexibility within the CMO facility is a bonus; modular cleanrooms are increasingly being used, which allows for the moving of equipment from one location to another to support larger scale production runs or take advantage of manufacturing slots. This can be especially useful for ATMPs as the downtime between manufacturing runs maybe longer than for NCEs, usually owing to the unpredictability of their manufacture, unforeseen problems with batches, and cleaning down procedures.
There is also the potential for contamination of the facility, especially when working with patient cells, as guaranteeing their quality can be difficult (owing to the potential for presence of viruses). It is critical that ATMP CMOs have validated cleaning procedures in place for managing patient cells used as starting material, but also for managing different types of viral vector systems to mitigate the risks of cross-contamination between products. In the case of ex vivo gene and cell therapy products, patients are usually treated with the product “at risk” without full release testing being completed. This is due to the nature of the product and the need to administer the product as quickly as possible, avoiding increased morbidity or even mortality.
With regard to release testing, companies need to recognise that analytical assays can also be very variable and specialised. They may have been developed initially in academic research facilities and are unlikely to be qualified, requiring modification to meet GLP/GMP requirements. It is common practice to outsource such assays to specialised analytical laboratories, which are also in high demand at the current time. These vendors may need careful management and the tests maybe labour-intensive and complex, with turnaround times slower than conventional tests. Further, it is not uncommon for very specialised assays (eg, in vivo potency testing) to be performed by an academic group with the appropriate expertise and animal model but the test may not be done to GMP. This can be accepted if justified but it is recommended to agree this approach with regulators in advance of a clinical trial or MAA. The principles of GMP should always be followed where possible and, of course, the method should be qualified or validated as appropriate for the stage of development. It is also recognised that it may not be possible to perform very specialised release testing in the relevant jurisdiction; indeed, there is precedent for deviation from the EU requirement to test a commercial product on import into the EU for an approved ATMP (European Public Assessment Report for Luxturna®;5 voretigene neparvovec). This likely also recognises the fact that these assays are difficult or impossible to transfer successfully.
Continuing the theme of release, the company should identify whether the CMO has a qualified person (QP) on site, or whether the services of a remote QP are used. Regardless, it is imperative that this individual is experienced in ATMP manufacture and release, as they may need to be comfortable with releasing the product quickly, and without all release testing being completed, in order to meet a short treatment window (especially in the case of cell- and tissue-based products where some tests may take weeks to report out). Therefore, a good understanding of biological processes (in particular, their unpredictable nature) and a flexible and pragmatic approach are invaluable. In an ideal situation, the QP for an ATMP would be supported by a quality assurance (QA) team that has experience with both ATMPs and NCEs, to provide the best balance between scientific rigour and sufficient quality data for release. It can be particularly challenging to make decisions for ATMPs; for example, owing to the inherent variability in biological systems, experience is required to know when cells are ready for harvesting, and this requires a different skill- and mindset compared with not just NCEs but also standard biologics. Flexibility to take into account the variability seen within biological systems is critical; however, a balance between process variability and the established critical quality attributes needs to be defined to ensure product quality. This is where close links with the academic groups or original developer are essential, as those individuals/teams will often have the technical expertise needed to help troubleshoot.
Other practical considerations of ATMP manufacture are often overlooked when selecting a CMO. As the quality of the raw materials can have a significant impact on manufacture and the finished product, the company should ensure raw material suppliers can meet product demands and that raw materials are verified and highly controlled (meaning they are stored appropriately prior to use but also meet a defined specification for use in the manufacturing process). This concept continues through to retained samples, for which the company should identify storage facilities early on, taking into consideration storage conditions and stability (an appropriate facility, from both a cleanroom and support perspective). Finally, and relatively unique to ATMPs, is to ensure the CMO has sufficient fill and finish capabilities on site. This avoids the need to transport material that usually needs to be maintained at ultra-low temperatures (ie, on dry ice), and the need for freeze-thaw studies to support its formulation. A similar concept should be considered for cell-based products which realistically need to be manufactured in close geographical proximity to viral vector manufacture for both ease and consistent quality.
Many ATMPs are associated with devices for their administration (eg, a novel catheter) and, although outside the remit of CMO selection, companies need to plan for studies to assess compatibility and in-use stability with those devices if applicable. Available support for such studies can be discussed with the CMO at an early stage to assist in the identification of additional vendors, resource, or support for method transfer as needed to ensure a seamless and timely generation of data.
Finally, it is noted that the European Commission has issued guidance on GMP for ATMPs. This guidance has been informative and recognises the specific challenges associated with the manufacture of these complex products.
The choice of CMO for an ATMP is not just a decision for a CMC team; it can have far-reaching implications. From experience, the initial manufacture of these complex products can require a significant amount of troubleshooting if the manufacturing process needs to be transferred from an academic facility. It is, therefore, a multidisciplinary decision requiring functional input from CMC, regulatory, and business colleagues and should be informed by the long-term strategy for the development and registration of the product.
It stands to reason that the CMO should be chosen carefully. Ideally, the final choice will be a CMO with a modern facility, an experienced, well-trained team, good management infrastructure and an experienced, pragmatic QP. With this foundation, and with work to build a transparent working relationship, a company will have taken all available steps to mitigate issues during manufacture that would impact on the ability to provide GMP-compliant ATMPs for the conduct of clinical trials and for registration and commercialisation.
With that said, issues do arise, so regulatory professionals should make use of all available assistance from various sources – be it via scientific advice, ATMP classification and certification, conferences, newsletters etc – to stay up to date with current technical (eg, analytical method improvements) and other developments (such as market changes).
- Alliance for Regenerative Medicine. https://alliancerm.org/publications-presentations/
- European Commission. https://ec.europa.eu/health/sites/health/files/files/eudralex/vol-4/2017_11_22_guidelines_gmp_for_atmps.pdf
- 2019 ARM CMC Summit. http://alliancerm.org/wp-content/uploads/2019/11/2019-CMC-Summit-Draft-Agenda-v1-05Nov2019.pdf
- The Advanced therapy medicinal products Manufacturing Community (AMC). https://atmpmanufacture.org
- European Medicines Agency. https://www.ema.europa.eu/en/documents/assessment-report/luxturna-epar-public-assessment-report_en.pdf
To view the February issue of Regulatory Rapporteur, click here.