Comparability Considerations for Cellular & Gene Therapy Products

C&GT products—also known as advanced therapy medicinal products (ATMPs)1—can present developers with novel circumstances that create technical barriers or otherwise impact their approach to assessing comparability. These products fall under the regulatory framework of biologicals and include a wide array of medicinal products such as gene therapies (both in vivo and ex vivo gene therapies, gene editing technologies, etc.), somatic cell therapies, and tissue-based products. The scope of this article encompasses all C&GT modalities at a high level.

Assessing Comparability

Because C&GT products encompass a broad range of modalities with widely different properties, there is no single broadly applicable approach to assessing their comparability; instead, more tailored fit-for-purpose approaches are needed. For example, many C&GT products are made in limited quantities (by necessity) and there may not be sufficient drug product to evaluate in the usual manner.2^, 3

Comparability assessments are crucial for life cycle management of all biological products, including C&GT, and are used to ensure that manufacturing changes will not have an adverse effect on product quality, safety, or efficacy. The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Q5E guideline4 provides sound principles for assessing comparability and has been implemented for many years for biotechnological and biological products (e.g., monoclonal antibody products).

The same principles should be leveraged for C&GT products using a risk-based approach, with the appropriate flexibility to account for the extenuating circumstances often posed by these innovative therapeutics. Flexibility is needed to maintain the high standards of C&GT quality and, in some situations, the usual data packages and/or practices for demonstrating comparability of pre- and post-change product may not be suitable.

Manufacturing changes are inevitable throughout the life cycle of a medicinal product and are necessary to ensure continuity of supply and enable best practices for biopharmaceuticals (such as dual sourcing of raw materials). It is generally necessary to scale up or scale out the manufacturing process or introduce new manufacturing facilities to produce enough C&GT product to treat all patients.

Manufacturing processes for C&GT are often complex, but improvements and innovation should be encouraged. In addition, C&GT production can involve several biologically active input materials and, because of their intrinsic variability (from different vendors or batch to batch), focusing a comparability exercise on a particular stage of manufacturing or incoming material can be appropriate. When manufacturing changes are made, the risks associated with the changes should always be assessed and their potential impact on subsequent process steps should be evaluated.

Comparability assessments are needed throughout the life cycle of a medicinal product, from preclinical through commercialization to postapproval.2^, 3^, 4^, 5^, 6^,7 During early development, comparability exercises generally focus on safety, and in late development focus more on efficacy. However, clinical development of C&GT products is often compressed and may lack the usual distinctions between early and late stages of development.

For C&GT products, the understanding of their mechanisms of action, manufacturing processes, and product quality attributes is evolving.

For C&GT products, the understanding of their mechanisms of action, manufacturing processes, and product quality attributes is evolving. Techniques that enable detection and measurement of product quality attributes may include methods that are more commonly used in research settings and thus need to be adapted to the development environment, and to quality control settings for release assays. Given the current level of understanding of many C&GT products and the inherent complexity of the products themselves, evaluating the impact of manufacturing changes is often a complicated endeavor and may involve multidisciplinary studies, such as in vivo assessments in nonclinical and/or clinical studies 2^, 3^, 8 more often than for conventional biologics.

Overall, it’s important to remember that a comparability exercise is based on scientific principles and does not simply follow a checklist. All available knowledge about the manufacturing process and the medicinal product should be leveraged appropriately. The potential impact and risk of any manufacturing process change should be thoughtfully considered. There is no “one size fits all” approach to assessing comparability of all C&GT given the wide diversity of these products. Regulatory requirements are also evolving, and manufacturing changes may be needed to keep pace with these evolving expectations.

General Principles and Challenges for C&GT

The guideline ICH Q5E “Comparability of Biotechnological/Biological Products Subject to Changes in Their Manufacturing Process” 4 sets the regulatory expectations. Developers should evaluate the relevant product quality attributes to show that any changes that would adversely impact the safety and efficacy of the drug product did not occur. The evaluation for conventional biologicals is typically done using a stepwise approach that starts with physicochemical and biological properties of the product to indicate whether nonclinical and/or clinical studies would be appropriate.

The developer’s comparability study plans should include lists of predefined manufacturing process changes, analytical methods to be performed, and product quality attributes to be monitored (with well-defined target ranges). The testing plan should include release tests, in-process controls (IPC), stability data, and extended characterization studies. To conduct a meaningful comparability assessment, well-controlled, sensitive, and quantitative assays are needed. Acceptance criteria should be derived by statistical analysis of historical data when it will be meaningful. For C&GT products, tailored approaches are needed.

As described in ICH Q5E, the developer should consider the risk posed by each manufacturing change, including the extent of the change, the particular step in the manufacturing process, and the potential impact to downstream manufacturing steps. When multiple manufacturing changes are to be implemented, a plan should be developed to evaluate the changes stepwise (if appropriate) and/or end to end. The developer must also assess the ability to detect changes in product quality given the status of their analytical methods.

Comprehensive comparability assessments are expected in late-stage development and postapproval, and manufacturing changes should be avoided, when possible, during pivotal trials. In addition, the principles described in ICH Q129for established conditions and postapproval change management protocols are applicable for postapproval life cycle management of C&GT products.

Challenges arise for C&GT products when there is wide variability in an assay (e.g., infectivity assays for viral-vector-based gene therapies) or wide inherent variability in the final drug product (e.g., autologous cell-based gene therapies or tissue-based products). This can make it difficult to compare drug product batches that have been analyzed in separate analytical test sessions or to set acceptance criteria statistically. Side-by-side testing of pre- and post-change C&GT product in the same test session may mitigate assay variability, though the availability of the product may be limited and/or the shelf life of the product may be short (e.g., 72 hours for tissue-based or cellular products that cannot be cryopreserved).

The analytical techniques for release and stability testing and extended characterization of C&GT should be established as early in development as possible with a strong emphasis on meaningful potency assays. It is not unusual to use a matrix of potency assays to address various aspects of the C&GT mechanism of action. Developers need to carefully introduce new analytical methods in a well-controlled manner and conduct proper method bridging studies to ensure continuity with earlier results.

Comparability guidelines 2^, 3^, 4^, 5 also call for process improvements that are not expected to adversely affect product quality, so this leaves room for improved product quality (e.g., lower levels of product- or process-related impurities). When significant benefits, including potential safety benefits, would result from manufacturing changes, such changes should be properly enabled.

When conducting a comparability exercise, an adequate number of representative batches should be included. Although GMP batches are produced to supply clinical trials and the market, non-GMP batches (e.g., engineering runs) may be suitable if they are representative of the process being evaluated. Variability in manufacturing processes should be considered, and the more variability, then the more batches that are needed. It can be challenging to identify and manage sources of variability in C&GT production given the complexity of these medicinal products, their manufacturing processes and testing methods, and the incoming materials used in their production.

Points to Consider

Although there is a need for regulatory flexibility, this should not imply that quality standards can be lower for C&GT products, but rather that alternative approaches may be needed to ensure appropriate standards. Comparability assessments can be particularly important for C&GT products, given that many of these are one-time treatments and the opportunity to re-dose patients is currently limited. C&GT products pose many new challenges and uncertainties, but they also bring new concepts and opportunities. Considerations are herein provided for assessing the comparability of pre- and post-change C&GT products.

C&GT Product Characterization

C&GT products can be complex, and characterization at a molecular level may be achievable for some modalities (e.g., messenger RNA therapeutics) but may be impractical for others (e.g., tissue-based therapeutics). There has been considerable progress in the characterization of viral-vector-based gene therapies. Briefly reflecting on cellular products, these are “living drugs” that are dynamic; their therapeutic effect may be linked to numerous different structures and they may undergo additional changes, such as cell division or migration or engraftment, upon administration to the patient.

For genetically modified cells, extensive characterization is expected, including the off-target and the intended on-target gene editing events. For each genetic modification, analytical tools are needed to assess the expression level, the distribution of expression, and the function for each component. These should be considered when assessing comparability of cell-based gene therapies.

With the emergence of individualized cellular and gene therapies (i.e., products that are custom made for a specific patient where the manufacturing begins with the patient’s cells or tissue, like autologous CAR-T cell products and individualized neoantigen-specific immunotherapies), it is not possible to generate reference material of the same composition as the respective individualized product, but analytical standards can be established to ensure method performance.

It’s necessary to account for the intended variability of individualized products during comparability assessments. Each batch is highly influenced by patient material characteristics. The patient-specific product quality attributes vary with the corresponding patient and should not be the focus of a comparability assessment. Instead, the product-specific quality attributes should be comparable after manufacturing changes.

These are just a few examples of the complexity presented by these innovative therapeutics. Given their product complexity, C&GT products need to be defined early by the developer. The use of a draft quality target product profile (QTPP) by the developer is encouraged to establish and maintain boundaries for their product as they develop the manufacturing process along with the analytical methods for characterization, release, and stability testing.

Having such a QTPP document in place early in development (for example, a draft during phase I) will help raise awareness of the boundaries of the defined product (and when they may have been exceeded and the developer may possibly have a new product). Potential critical quality attributes (CQAs) should be flagged early, as they will be the focus of the comparability assessment. Because the ability to detect and quantify product quality at-tributes is often limited for C&GT, there can be limited understanding of the impact of manufacturing changes on product quality, safety, efficacy, and duration of response, and the changes may be challenging to justify in some cases.

Analytical Methods

Analytical methods for characterization, release, and stability testing tend to evolve in parallel with manufacturing process development. With new methods/techniques, product quality attributes that can be detected and quantified often change over the course of development. The comparability assessment should be focused on the most relevant quality attributes of the product and not simply on which attributes can be measured.

Understanding of product quality attributes and the maturity of analytical methods should increase throughout the product development life cycle, so advance planning to reserve appropriate amounts of product for later evaluation is recommended, with the caveat that sample stability over time needs to be kept in mind. Analytical methods for C&GT products are often product-specific, non-compendial, and complex. Early implementation of reference materials and/or assay controls is recommended to enable bridging to new and improved analytical assays.

Potency assays are a pillar of comparability assessments because they measure the bioactivity of the product. Potency measurements are generally challenging for C&GT products because of their complex mechanisms of action, and multiple orthogonal methods are often needed to measure relevant aspects of the product’s biological activity. For cellular products, the cells may continue to divide while they are being prepared for and during potency analysis, so there can be inherent variability in these bioactivity measurements.

The C&GT field holds the potential of better correlations of chemistry, manufacturing, and controls (CMC) and clinical data if or when appropriate data and analytics ecosystems are established.

Product Amounts Available

It is common for C&GT products to be manufactured in limited amounts because of manufacturing constraints or limited amounts of cellular starting material. For example, there are limited amounts of patient cells available for the manufacture of autologous CAR-T cell products. Therefore, the analytical approach taken must accommodate the product and patient needs. There are often only small amounts of material available to develop analytical techniques and to conduct routine analyses and characterization studies, including comparison of pre-and post-change product. For certain compendial assays (e.g., microbiological), there may not be sufficient material available to both treat the patient and conduct a compendial assay, in which case non-compendial methods are required.

Given the small volumes of C&GT production, developers should carefully consider whether it’s possible to collect sufficient sample volumes for meaningful analysis and, if so, establish sample retention best practices from early on so that they have retain samples for comparability exercises later in development. It is often not possible to follow standard guidelines on the number of retain samples to keep, especially for individualized products. However, it may be possible to utilize otherwise unused clinical material that was not administered to patients (e.g., for autologous cell-based products).

Assay variability may be inherent in some cases but must be minimized. One approach to overcoming assay variability is to conduct side-by-side analysis with all samples tested in the same analytical run. However, side-by-side analyses may not always be feasible, so the use of established assays with understanding of intermediate precision may be used as a means of analytical comparability testing.

Manufacturing Technologies and Materials

Manufacturing technologies for the production of C&GT are often rather innovative or have transitioned from research settings to GMP manufacturing in recent years. The requirements for pharmaceutical production are much more rigorous than for research, and they require demonstration of process reproducibility, which is important for maintaining continuity throughout clinical development and commercial supply.

In addition, many C&GT manufacturing processes involve multiple biologically active and sourced materials. The purity of incoming materials (raw materials, starting materials, etc.) needs to be verified and documented, and their impact on manufacturing process performance and final product quality and safety need to be evaluated. The designation of incoming materials and the expectations for their quality are areas where additional regulatory guidance and harmonization is being recognized by regulators globally.10

Many of the incoming materials used for C&GT manufacturing are produced by vendors with limited experience in biopharmaceutical production. Because they may be unfamiliar with medicinal product manufacturing requirements—such as the need to minimize the use of animal-derived materials or the need to minimize the risk of infectious agents, including transmissible spongiform encephalopathy (TSE)—vendors may need guidance from the developer.

In general, the sourcing of raw materials should be done methodically, accounting for the potential impact of raw material quality or changes in their production processes on C&GT manufacturing and product quality. The developer should consider their ability to detect differences in the incoming materials themselves and they may need to conduct their own characterization studies on incoming materials. The ability or robustness of the C&GT manufacturing process to accommodate incoming material variability or changes should also be evaluated. It can be helpful to focus on the most relevant manufacturing steps where the change can be evaluated.

For autologous cell-based products, there is variability in the cellular starting material derived from patients that depends on the disease state of each patient, comorbidities, prior treatments, and other factors. Therefore, the use of surrogate material (i.e., healthy donor cells) should be considered for comparability studies. Nonetheless, material from healthy donors is also heterogeneous, and their representativeness of material from patients needs to be justified. The developer also needs to consider the ability of the C&GT manufacturing process to perform as it is designed to perform in the presence of the surrogate material. Overall, the use of suitable surrogate material allows for better assessment of process-related variability that can be controlled but doesn’t address product-related variability.

Split manufacturing

There are several circumstances in which split manufacturing can be an effective approach for assessing comparability of C&GT. For example, when introducing new vendors of raw materials or when conducting a manufacturing site transfer for an individualized C&GT product. The manufacturing stream is split at the point of the change (e.g., starting material) and run downstream. Then head-to-head comparisons can be conducted on, say, the resulting pairs of drug substance or drug product batches, and would generally involve the usual evaluations of release testing and extended characterization.

Split manufacturing requires that there is enough material to conduct two runs of the process in parallel. This may be possible with, for example, allogeneic cellular products, but may not be feasible with autologous cellular products because patient-derived material is limited in availability. In this case, alternative approaches may be preferable when a risk assessment is supportive, such as the transfer of a fully closed, automated manufacturing process to new sites of manufacture.

Manufacturing process comparisons

Given the inherent complexity and heterogeneity of cellular and tissue-based products, and the associated limitations in their characterization, a comparability assessment may need to focus on the manufacturing process, including IPC and the evaluation of process performance metrics. The manufacturing processes can involve multiple stages, such as expansion of cells, differentiation into a defined mature cell type, or enrichment steps to increase the population of the desirable cell type.

IPCs can provide valuable information about the robustness of the manufacturing process to accommodate modifications in manipulation of the cells. Analytical techniques to better define cellular phenotypes, subpopulations of cells, and cellular impurities are needed. Many of the assays show variability given the dynamic nature of the cells while they are being prepared for and undergoing analysis.

Considerations for Nonclinical Studies

When differences in product quality are detected during the analytical testing, nonclinical studies may be appropriate to assess the impact of manufacturing changes on product quality, safety, and efficacy. This will depend on the type of changes and extent of differences detected between the pre-and post-change product (e.g., product-related substances, process-related impurity profile). For example, new process-related impurities could warrant toxicological studies for qualification through additional nonclinical studies.

Nonclinical studies may provide supplemental information when the available analytical methods for assessing comparability are limited or the level of knowledge of the product is limited. For viral-vector-based gene therapies, this may include comparability of expression/functional assessments in animal models. However, for cellular and tissue-engineered products, there may be few or no meaningful animal models.

Considerations for Clinical Studies

Escalation of comparability assessments may call for clinical studies, though the strengths and limitations should be recognized. It is challenging to develop an understanding of the correlation and/or causation between product quality attributes and clinical outcomes for all medicinal products, and the C&GT field is rather early in developing this knowledge. When clinical bridging studies are performed to further assess comparability, they may provide information on safety and pharmacokinetics, but may not be able to address questions about the possible impact to clinical outcomes, including efficacy and duration of response.

Further, clinical comparability assessments may not be feasible or appropriate in all cases, such as for slowly progressing rare diseases where clinical effects could take years to be observed. When clinical bridging data are needed to evaluate comparability of pre-change and post-change product, there are longer timelines to the initiation of pivotal trials or approval of market applications.

Although C&GT products pose many challenges, they also pose opportunities. Because they tend to be manufactured in small batches for few patients (and even one batch for a specific patient for individualized products), clinical outcomes are more readily traceable on a per-batch basis than for conventional biological products (where one batch can be sufficient to treat thousands of patients). Thus, the C&GT field holds the potential of better correlations of chemistry, manufacturing, and controls (CMC) and clinical data if or when appropriate data and analytics ecosystems are established. It is therefore recommended to archive potentially relevant manufacturing and product quality data as well as clinical data in a searchable and retrievable manner.

Conclusion

Typically, comparability exercises are conducted to confirm the established safety and efficacy profile of a biologic product after incremental manufacturing changes have been made by the manufacturer, leveraging an extensive process and product history linked to clinical experience. With cellular and gene therapies, there is often limited process and product history and depth of understanding.

Although C&GT products have been studied for decades, relatively few products are commercially available, though the number is growing. Certain C&GT modalities are reaching a level of maturity where more detailed expectations for comparability assessments could be articulated in technical guidelines (e.g., viral vector-based gene therapies; genetically modified T cell products).

For C&GT products overall, the manufacturing technologies are often expensive, immature, and rapidly changing. Thus, once the initial feasibility of the medicinal product is demonstrated and supply increases become necessary, switching to state-of-the-art or scalable technologies is required to ensure the best product quality at an affordable cost.

The needed manufacturing changes can result in differences in product quality such that the post-change product may have an equivalent efficacy and comparable or improved safety but does not fulfill the usual expectations of an analytical comparability assessment as currently described in ICH Q5E. This suggests that ICH Q5E does not sufficiently account for the complexity of C&GT product and manufacturing knowledge. An addendum to ICH Q5E could address the novel circumstances faced when assessing the comparability of certain C&GT product modalities when changes are made in their manufacturing processes. In the meantime, developers should obtain feedback from regulators on their comparability study plans to ensure alignment on the approach.

Acknowledgments

We would like to acknowledge the contributions on this topic by the members of the Biotechnology Innovation Organization (BIO), a US trade association. We thank our colleagues and subject matter experts for their critical feedback and editorial input.