A wide variety of reagents and materials, many of which are unique or complex, are required for the manufacture of cell, gene, and tissue-engineered products. These materials include plasma- or serum-derived products, biological extracts, antibiotics, cytokines, culture media, antibodies, polymeric matrices, separation devices, density gradient media, toxins, conditioned media supplied by “feeder cell layers,” fine chemicals, enzymes, and processing buffers. Many of these items are used to ensure the survival and promote the growth of certain cell populations, although their mechanism of action may not be entirely understood. Examples include fetal bovine serum (FBS) and various media supplements. Other items, such as highly purified cholera toxin, are introduced into the processing stream during manufacturing to exert a specific biochemical effect and are immediately washed out in subsequent processing steps to avoid unwanted toxicity at a later point. The finished biological products produced in such processes are often complex mixtures that, in some cases, cannot be completely characterized. Careful scrutiny of the materials used in manufacturing is necessary to prevent the introduction of adventitious agents or toxic impurities, as well as to ensure the ultimate safety, effectiveness, and consistency of the final product.

In cell, gene, and tissue-engineered product manufacturing, these reagents and materials are collectively called ancillary materials (AMs). AMs have also been referred to as ancillary products, ancillary reagents, processing aids, and process reagents. AMs were first discussed under the synonym ancillary products in the U.S. Food and Drug Administration Notice, “Application of Current Statutory Authorities to Human Somatic Cell Therapy Products and Gene Therapy Products” (Federal Register 58(197), October 14, 1993, pp. 53248–53251). This document established the FDA's authority to regulate human somatic cell therapy products and gene therapy products. AMs are also synonymous with “processing materials” that were defined in 21 CFR Part 1271, “Current Good Tissue Practice for Manufacturers of Human Cellular and Tissue-Based Products; Inspection and Enforcement; Proposed Rule” (Federal Register 66(5), January 8, 2001, pp. 1508–1559). AMs can be analogous to “components,” and in some cases, “containers” as described in the current good manufacturing practice (cGMP) regulations for finished pharmaceuticals as outlined in s21 CFR 211.80 through 211.94 and 211.101(b) and (c).

The defining property of AMs is that they are not intended to be present in the final product. They are materials used as processing and purification aids or agents that exert their effect on the therapeutic substance. Materials or components that are intended to be in the final product dosage form (e.g., genetic materials, biopolymeric supports, physiological buffers) are not AMs. Cell banks and virus banks are also not considered AMs; there are a number of guidances that describe requirements for their certification. However “helper” viruses and “helper” plasmids may be considered AMs when they are not intended to be part of the final product.

The quality of an AM can affect the stability, safety, potency, and purity of a cell, gene, or tissue-engineered product. For example, the mechanism by which an AM exerts its effect may not be known, and the impact of normal variation of the AM on the quality and safety of the therapeutic product may not be understood. Alternatively, AMs of human or animal origin may present an infectious disease transmission risk. Other AMs, if administered to humans, may cause an immune reaction. Finally, an AM with toxic properties that is introduced into a manufacturing process and is not adequately removed in subsequent processing steps will expose the patient to a toxic substance and may impair the effectiveness of the therapeutic entity. These risks to the quality and safety of the therapeutic product are often heightened with cell, gene, and tissue-engineered products, due to the limited ability to conduct extensive in-process and release tests. For example, lack of in-process holding steps or limited shelf life may create the need to administer the cell, gene, or tissue-engineered products before in-process or final-release testing results are available. In other cases, the scarcity of suitable donor tissue or the complex logistics in the transport of biological materials may limit the amount of material available for testing. To minimize these risks, whenever possible, it is necessary to implement rigorous material qualification and prudent application of manufacturing process controls.

Frequently, these novel therapeutic products are created using complicated biological processes. The AMs employed in these procedures may be selected primarily for their unique functional contributions or biological effects. Whenever possible, it is preferable to source AMs that are approved or licensed therapeutic products because they are well characterized, have an established toxicological profile, and are manufactured according to controlled and documented procedures. Conversely, the AM may be intended “for research use” and may, therefore, lack the level of qualification necessary for use in the production of a therapeutic product. In either case, the manufacturer of the cell, gene, or tissue-engineered product should develop comprehensive and scientifically sound qualification plans to ensure the traceability, consistency, suitability, purity, and safety of the AM. In cases where AMs are products approved for use for therapeutic purposes, the level of qualification will probably be less extensive than that for a material intended for research purposes. However, their suitability in the manufacturing process will still need to be established when the AM is being used beyond the scope of its intended use or labeling. The purpose of this chapter is to provide guidance in developing appropriate qualification programs for AMs employed in cell, gene, and tissue-engineered product manufacturing.

Qualification is the process of acquiring and evaluating data to establish the source, identity, purity, biological safety, and overall suitability of a specific AM. The responsibility for AM qualification resides with the developer or manufacturer of the cell, gene, or tissue-engineered product. This section outlines the basis by which a manufacturer can establish rational and scientifically sound programs for qualifying AMs, although the broad nature of the cell, gene, and tissue-engineered products and of the AM used to produce these products make it difficult to recommend specific tests or protocols for a qualification program. Thorough documentation is the cornerstone of any qualification program.

A well-designed qualification program becomes more comprehensive as product development progresses. In the early stages of product development, safety is the primary focus. In the later stages, AM production and qualification activities should be comprehensively developed to support eventual licensure of the cell, gene, and tissue-engineered product. On some occasions, complex or unique substances that have been shown to be essential for process control or production may not be available from suppliers that produce them in compliance with cGMP. In these situations, the manufacturer will have to develop a scientifically sound strategy for qualification. A qualification program for AMs used in cell, gene, and tissue-engineered product manufacturing should address each of the following areas: (1) identification, (2) selection and suitability for use in manufacturing, (3) characterization, (4) vendor qualification, and (5) quality assurance and control.

Some AMs that are biological in nature may be difficult to fully characterize. Because these materials exert their effects through complex biological activities, and biochemical testing may not be predictive of the AM's process performance, functional or performance testing may be needed. Performance variability of such materials may have a detrimental impact on the potency and consistency of the final therapeutic product. Examples of complex functionality testing for AMs include growth promotion testing of individual lots of FBS on the cell line used in manufacturing, performance testing of digestive enzyme preparations, and in vitro tissue culture cytotoxicity assays. (see aspects of Performance Testing).

A scientifically sound and rational qualification program should be designed for each AM and should take into account the source and processes employed in its manufacture. Whenever available, AMs that are approved or licensed therapeutic products are preferable because they are well-characterized with an established toxicological profile and are manufactured according to controlled and documented procedures. Licensed biologics, approved drugs, and approved or cleared medical devices or implantable materials that have been incorporated into cell, gene, or tissue-engineered product manufacturing processes present a known or more favorable safety profile for the patient than nonapproved or nonlicensed versions. Qualification programs for these AMs should reflect the extensive scrutiny that these items were subjected to in their development and manufacture. Consequently, greater emphasis should be placed on the investigation of the impact of inherent variability of these AMs on final product function. For instance, a manufacturer may utilize human serum albumin, intended for human administration, as a supplement to a cell cultivation medium for a cell-based product. Because the cell-based product is marketed as a licensed biological, one need not repeat all the testing already performed by the supplier as part of material qualification. In contrast, the impact of lot-to-lot variability on cell growth rate or maintenance of an important differentiated cellular property may be a prudent area of investigation. Alternatively, the stability of this material at the concentration employed in processing or its potential for interaction with other processing components may also be areas worthy of investigation. Such approaches to AM qualification therefore focus on the AM as a potential source of variability that may influence final product potency and safety. Qualification programs for these AMs should be comprehensive to minimize consumer risk and ensure that unacceptable lots or adulteration will be detected.

The qualification program must also take into account the quantity of the AM employed in manufacturing as well as its point of introduction in the manufacturing process. A relevant example is the use of FBS as a supplement to a tissue culture medium used to expand a stem cell population from a specific tissue for eventual administration to a patient (see Manufacturing Overview under Cell and Gene Therapy Products 1046). A qualification program for such an AM would include (a) assurance that the serum was sourced from a country or region known to be free of bovine spongiform encephalopathy (BSE); (b) assurance that the source herds are monitored and test negative for specific diseases relevant in agricultural settings (e.g., tuberculosis, brucellosis, foot and mouth disease); (c) testing of the serum for sterility, mycoplasma, endotoxin content, and adventitious bovine viruses known to be associated with the material;1 (d) the review and archiving of the supplier's certificate of analysis; (e) lot-to-lot assessment of the ability of the serum to consistently expand a representative cell population using a standardized cell culture quality control assay; and (f) on-site audit of the supplier to ensure that the material is sourced and processed in a manner deemed acceptable by a responsible QA unit.

To aid manufacturers and developers in the design of their qualification programs for a variety of AMs, tiers of sample risk categories are presented in Tables 1–4 and are provided as a guide. Risk is also dependent on the amount and the stage at which the AM is used in the manufacturing process. Tables 1–4 do not address the impact of quantity or stage of use.

In cases where AMs are chosen for their ability to provide a particular biological function in producing the therapeutic product, performance testing becomes an essential component of their overall qualification. This is especially true when the AM plays a critical role in modulating a complex biochemical effect and has a large impact on product manufacturing yield, purity, or final product potency. These AMs tend to be complex substances or mixtures, are frequently biologically sourced, and can exhibit significant lot-to-lot variability. As a result, these AMs usually have no simple identity test, nor can they be easily characterized by physical or chemical tests. The development of well-defined performance assays for complex AMs will not only ensure process reproducibility and final product quality, but in many cases will satisfy the identity testing criteria in accordance with 21 CFR 211.84(d).

In some cases, the initial qualification of an AM for use in manufacturing should be the investigation of the effect of the amount of the AM on the desired response (increased yield, purity, or potency of the therapeutic product). The amount of the AM used in manufacturing should be chosen to consistently yield the desired effect while minimizing issues by removing the AM in subsequent processing steps. Such testing frequently assesses the important functional attribute expected of the AM in a scaled-down or simulated manufacturing process. Some examples follow:

The actual assay used may well evolve as the manufacturing process is developed further and the critical relationships of the AM and the final product are better understood.

Because most performance testing yields relative results, it is often helpful to assay a new lot of AMs side by side with an approved lot of AMs or an official reference standard, if available. This simultaneous comparison helps to reduce the variability due to different lots of cells or vectors and will help discern variability associated with the different lots of AMs. If performance testing involves assays to demonstrate that the new lot of AMs does not affect the impurity profile of the final therapeutic product, either by generating new impurities or by increasing the level of existing impurities, it is helpful to assay both for the total level of impurities, as well as look for the presence of new impurities. An immunologically-based binding assay can typically assess only the total level of impurities. For example, a Western blot of the gene therapy product that is probed both with antibodies to the product and antibodies to host cell proteins is useful for detecting new protein species and significant increases in the levels of host cell impurities. This initial qualification is enhanced by a performance assay that has a quantitative readout with a clear change in the signal when a significant change in the amount of AMs is introduced into the assay (e.g., dose response). A threshold-type response (i.e., there are two levels of response to the AM and neither large changes in an AM below a certain dose nor above a certain dose change the response) can make it more difficult to select a concentration of AM that consistently results in the desired effect and minimizes the residual levels of the AM in the final therapeutic product.

AMs are not intended to be present in the final dosage form in cell, gene, and tissue-engineered products. Their presence in the final product could lead to undesired effects in the recipient or have a detrimental effect on product potency. Undesired effects in humans include direct toxicity of the AM or an unwanted immunogenic response. Some examples include the following:

These risks can be mitigated through the design of processes to include steps to adequately remove the AM through dilution, separation, or inactivation, as well as the development of analytical detection assays to assess the AM levels during processing and in the final therapeutic product. Assessment and removal strategies for residual AMs should be considered in the early phases of process development. There are two different approaches for assessing residual AM levels in the final therapeutic product: (1) Validation studies can demonstrate that the process is capable of removing more of the AM than would be present in a worst-case scenario. (2) The residual levels of an AM can be measured for each lot at an appropriate step in the manufacturing process.

Validation of an AM removal is often best performed by spiking the impure product with “worst case” or higher levels of the AM and showing the purification process is capable of removing the AM to “undetectable levels.” Clearance factors can then be generated for each purification step in a manner analogous to that done in viral clearance studies. When designing the validation studies, the following three considerations should be included: (1) The assay should be able to accurately quantitate the AM in each sample matrix. (2) If the validation is conducted at a scale smaller than that used for routine lot production, the comparability of this smaller scale process to the full scale process needs to be demonstrated. This usually means that the smaller scale process is operated using the same critical parameters as the full scale process with the product generated at each step having a similar purity and yield. (3) As with any spiking study, one has to demonstrate that the additional, higher level of AM has not affected the purification process. If the second approach of measuring residual levels of the AM in each lot is used, the specification for the maximum amount of AM in the final therapeutic product is based on the amount of the AM in the lots used in toxicological or clinical studies or known toxicological data.

The development of sensitive and reproducible analytical assays for AMs is another important component of a risk reduction approach. Two types of assays are useful in assessing the levels of residual AM impurity: a limit test and a quantitative test. Either test should be accurate, precise, robust, and have a low limit of detection. Assays for residual AMs may be performed on the product before it is formulated (e.g., on the drug substance) to avoid any interference of the components used in the formulation with the assay for residual AMs or in the final drug product. Spike-recovery controls are often included in such assays to demonstrate that the sample matrix does not inhibit the detection of the AM. Preferably, assays should be designed to detect all forms of AMs including aggregates, fragments, or conjugates. Aggregated protein has been shown to be particularly immunogenic.

Immunoassays such as ELISA are most commonly used to assess residual levels of AMs. An ELISA for bovine serum albumin (BSA) has been used to assess residual levels of FBS. Polymerase chain reaction (PCR) technology has been employed to assess residual levels of host cell DNA. Labeling cells with 3H thymidine or performing PCR for a feeder cell-specific gene sequence are two ways to assess for residual levels of feeder cells. If “wash out” of the AM is achieved by exhaustive dilution associated with further processing activities, it may be useful to calculate the dilution factor for the AM during this processing. In some cases, this is sufficient to ensure that the AM has been reduced to safe levels for early clinical development. Data should be obtained later in clinical development to confirm the wash out of the AM at the expected step(s). This approach is particularly useful when there is pre-existing knowledge of the therapeutic levels and toxicity of the AM. In other cases, information regarding the safety and tolerability of the AM should be collected (in preclinical toxicology studies or later with human clinical studies) in order to determine the safe or nontoxic levels that must be achieved. These data may be needed even for an AM that is approved for use for therapeutic purposes if it is being used in a manner inconsistent with its intended use or labeling or if the route of administration or dosage level of the AM may present risks not previously encountered or considered.

While many types of AMs are used during the manufacture of cell, gene, and tissue-engineered products, they have received less emphasis than the final products. However, the importance of AM quality to the quality of the final product cannot be overstated. Good quality AMs should perform as intended in a consistent manner, batch-to-batch, if they are carefully selected and appropriately used. AMs of insufficient quality will affect the quality and the effectiveness of the final product and endanger the health of patients. Thus, implementing an AM qualification program that addresses the risks associated with the AM, the stage of manufacture at which it is used, and the amount of the AM used during manufacture will ensure the safety and effectiveness of the final product.

AMs used in cell, gene, and tissue-engineered products will be regulated in the context of the manufacturing process of the cell, gene, and tissue-engineered products. Certain AMs may already be approved for uses other than for cell, gene, and tissue-engineered product manufacture. It is preferable to source AMs that are approved therapeutic products when they are available because they are well-characterized with an established toxicological profile and are manufactured according to controlled and documented procedures. The following list of documents should provide relevant regulatory guidance and a description of best practices in product and process development, manufacturing, quality control, and quality assurance: