Eleni Dokou

	The International Consortium on Innovation and Quality in Pharmaceutical Development (IQ) formed in 2010, and is an association of more than 25 pharmaceutical and biotechnology companies with a mission to advance science-based and scientifically driven standards and regulations for medicinal products worldwide. In the June 2012 issue of P

, a paper written by the IQ Consortium's GMPs in Early Development Working Group described the desire and rationale for more clear and consolidated recommendations for GMPs in early development (Phase 1 through Phase 2a) (1). A consequence of the absence of clarity surrounding early phase GMP guidances has been varied interpretation and application of existing GMP guidances within different companies according to its own culture and risk tolerance. Internal debates often result in conservative "one-size-fits-all" interpretations that rely on International Conference on Harmonization (ICH) guidelines that are mostly relevant to commercial product development and do not distinguish differences in requirements between early development and later stage development (Phase 2b and beyond). A key driver of this working group (WG), therefore, has been to collectively define the minimum acceptable practices within the industry regarding GMP expectations in early development that allow for added flexibility, are consistent with existing guidance and statutes, and which assure product quality and patient safety (2–4).

	The second paper in this series addressed recommendations for analytical method validation of both drug substances (DS) and drug products (DP) (5). In this third article of the series, the authors address the application of GMPs in early development as they pertain to drug-product manufacturing.

	As noted previously in the introduction to the paper series, due to high attrition in early development, most companies try to minimize the resource required to advance compounds into human clinical trials by employing simple formulations and manufacturing processes (1). For example, drug substance filled in bottle, drug in capsule, or simple powder blends filled in capsule are frequently used. Even for more challenging molecules, for example, those with low aqueous solubility, prototype formulations designed to investigate bioavailability enhancement (e.g., amorphous solid dispersions, lipid-based drug delivery systems, nanosuspensions) are evaluated in as simple a manner as possible. For example, on-site preparation at the clinic site of an amorphous solid dispersion by weighing into a capsule, or by suspension in a suitable vehicle may be utilized instead of a finished tablet or capsule dosage form. Whatever approach to formulation/process is taken, there is a need for flexibility in manufacturing due to limited product and process understanding in early development.

Based on the collective industry experience of the members of this working group, the authors believe that pharmaceutical companies could be making better use of available guidances as they pertain to drug-product manufacturing in early development. This belief was reflected by the mixed responses to a survey of IQ member companies conducted in late 2011 (see 

). Accordingly, in this article, the authors provide pragmatic recommendations related to GMP drug-product manufacturing focusing on those areas where it is believed there is opportunity for added clarity and flexibility, without added risk to product quality and patient safety.

	Following a brief explanation of the role and importance of quality systems, the areas highlighted in this paper include facilities, equipment, materials (receipt and approval for use), and batch documentation. Not covered is cleaning verification/validation, as this is a topic that is addressed in detail in the industry (6). The scope of this position paper has purposely been limited to traditional small molecules that are formulated into solid oral dosage forms intended for US regulatory filings with the desire to build consistency across all worldwide regulatory regions. However, it is believed that the concepts presented can be easily adapted to other dosage forms and routes of administration. Although designed as an industry position, it is recognized that each company needs to evaluate these recommendations for drug-product manufacturing practices based on individual business needs and risk culture.

In early development, relationships between material attributes, process parameters and product quality attributes are typically not well understood. It should be anticipated that even during GMP manufacture, there will likely be a need to deviate from the process conditions specified in a batch record. Quality systems should have the flexibility to allow these changes to be documented in batch records without formal prior approval from the Quality department. However, these changes should be reviewed by Quality after production, to assess the potential impact to product quality and patient safety. It is crucial to summarize and track process changes because much of this information will be used to develop process understanding and may be included in development history reports. In a similar manner, unplanned deviations from the written manufacturing procedures should be documented and justified, but may not need to be addressed in a formal corrective and preventive action (CAPA) system, because the process is changing continuously as part of the development process. Unplanned deviations that are likely to impact product quality and patient safety, such as cleaning failures, contaminations, and certain equipment failures must be investigated and corrective actions put in place to prevent recurrence.

 To assist in the application of risk-based decision making in the development and manufacturing of drug product, the authors recommend that companies apply a scientific and risk-based approach, similar in principle to that of the ICH Q9 

guideline (7). Annex I of the guideline describes various methods and tools that can be useful to determine the relative risks related to system risks, such as facility and people; organizations (including quality systems); process risks and product risks (safety and efficacy). There are many ways of identifying, qualifying, and mitigating operational and quality risks. Regardless of the methodology used, a documented strategy and good records of risk-based decisions are important in ensuring that the appropriate factors are considered for the protection of patients and product quality.

	Regardless of the scale of manufacturing, the facility used for manufacturing clinical trial supplies must meet the basic GMP requirements as described in the regulations and guidance documents. Below are three scenarios for early development and the advantages of each as pertaining to early development. The first involves a pilot plant facility designed and equipped for routine GMP operations. The second scenario aims to establish a GMP area within a laboratory environment. The third example focuses on conducting GMP manufacturing or leveraging the practice of pharmacy in close proximity to the clinical site.

GMP facility for drug-product manufacture. 

The traditional approach in GMP drug-product manufacture is to use a dedicated facility (often called a pilot plant) for early phase clinical trials. Advantages of this approach include that the quality systems for the facility (i.e., maintenance, calibration, cleaning, change management, CAPA, and documentation) are well defined, and that training and other activities required for maintaining GMP compliance are centralized. Other drivers to use a pilot plant in early development may be the need for specialized equipment, or larger batch sizes in special situations.

In some cases, it may be advantageous to establish a GMP area within a "laboratory setting" (i.e., a drug-development facility not dedicated to the production of clinical supplies) for the manufacture of drug product in early development. The rationale for this approach might be to avoid the significant investment in setting up a dedicated facility and to create simpler, more flexible systems that meet GMP requirements but are tailored for the specific activity envisioned. Examples where this approach might be considered include the need for special containment not available in the pilot-plant; the need to work with radioactive or hazardous materials, use of controlled substances and the production of "one-off manufactured" product used for proof of concept. The business rationale should be documented and approved by the manufacturing and Quality groups. As long as the appropriate GMP controls are maintained, especially as related to operator safety, cleaning, and prevention of cross-contamination, there is no compliance barrier to using "lab-type" facilities for the manufacturing of early phase clinical batches. Before GMP manufacturing is initiated, however, a risk assessment should be conducted and documented. Inclusion of representatives from Quality, analytical, clinical manufacturing, product development, and environmental health and safety would be prudent. When selecting/designing an early development clinical manufacturing facility, consideration should be made for the receipt, storage, dispensing, and movement of materials. The manufacturing processes in the nondedicated area must protect the product, patient, and the manufacturing operators.

	Additionally, companies should consider what items are appropriate for the manufacture. For example, the use of a certified laminar flow hood may be a better choice for manufacturing than a fume hood, because the former is designed to prevent contamination of the product, protect the operator, and the laboratory environment. In addition, with the appropriate cleaning, a laminar flow hood can more easily be used for multiple products. Small scale/manual equipment or procedures may be the best approach because the space is likely to be limited. With a small batch size, the use of small scale or manual equipment/procedures will minimize yield loss. Additional measures to be assessed include appropriate gowning and operator personal protection devices, area and operator monitoring for potent or radiolabeled drug exposure, and so forth.

	Documentation of the facility preparation, product manufacture, and the return of the facility to the previous state, if needed, is recommended. This documentation should describe the rationale for the manufacture in the nondedicated area, risk assessment, preparation of the area, cleaning procedures, and list of responsible persons. This documentation can reference existing procedures or standard operating procedures (SOPs) along with documents associated with the meetings and preparation for the manufacture of the batch. Batch records and cleaning records should be part of the documentation and should follow the company's data-retention policy.

On-site preparation of formulations, sometimes referred to as "extemporaneous preparation" or "compounding," is not considered manufacturing, but is an effective method to prepare early clinical supplies using the local laws governing the practice of pharmacy. The processes used for on-site preparation can be as simple as preparing and diluting solutions or filling capsules with API to more complicated processes such as blending and compression of tablets. An IQ working group addressing current practices in extemporaneous preparation has conducted a survey to be communicated at a later date, which confirms that only a few companies have adopted practices to take advantage of more advanced formulation approaches at the clinic site (8). This is a missed opportunity, because on-site preparation of formulations has the potential to dramatically speed industry's ability to answer critical questions related to pharmacokinetic parameters, absolute or relative bioavailability, feasibility of extended release and bioavailability enhancement approaches for difficult to deliver molecules. Currently, each company must develop its own best practices to assure product quality and patient safety; often, these practices are based on compendial or professional association publications (9–11). The authors believe this topic warrants further discussion between the industry and regulatory authorities to determine when dose preparation at the clinical site is appropriate and the level of quality controls required. The results of such a discussion could help to increase the utilization of on-site preparation of clinical trial materials to facilitate quick answers to critical questions in early drug development and ultimately bring the best possible products to patients in the shortest time.

Most equipment used to manufacture early GMP drug product is be managed under a qualification, preventive maintenance, and calibration program for the GMP facility. However, in early development, there may occasionally be a need to use equipment that is not part of such a program. Rather than performing a comprehensive qualification for a piece of equipment not expected to be frequently used, an organization may choose to qualify it for a single step or campaign. Documentation from an installation qualification/operational qualification (IQ/OQ) and or performance verification at the proposed operating condition is sufficient. For example, if solution preparation needs a mixer with a rotation speed of 75 rpm, then documentation in the batch record using a calibrated tachometer to verify that the mixer was operating at 75 rpm will suffice.

	The use of dedicated or disposable equipment or product contact parts may be preferable to following standard cleaning procedures to ensure equipment is clean and acceptable for use. However, not all equipment or equipment parts are disposable or may have a substantial cost that makes disposal prohibitive. In that case, the product contact parts could be dedicated to a specific drug substance for use in drug product manufacture. Dedicating product contact parts to a compound may be costly and may be avoided in some cases by carefully considering product changeover and effective cleaning methods when purchasing equipment.

	Another item to consider with respect to equipment, is that the more complicated the equipment is to run or maintain, the less desirable it might be for early GMP batches. In most cases, simple equipment is adequate and will uses less material and consume less total time for preparation, operation, and cleaning activities.

Part 211.42 state that buildings or areas used in the receiving, storage, and handling of raw materials should be of suitable size, construction and location to allow for the proper cleaning, maintenance, and operation (7). The common theme for this section of 

 Parts 210 and 211 is the prevention of errors and contamination. In principle, the requirements for buildings and facilities used in early phase manufacturing are not significantly different than those for later phases or even commercial production. However, there are some areas that are unique to early clinical trial manufacturing.

 regulations under Part 211.80 provide good direction with respect to lot identification, inventory, receipt, storage, and destruction of materials (7). The clear intent is to ensure patient safety by establishing controls that prevent errors or cross-contamination and ensure traceability of components from receipt through clinical use. In general, the requirements for the control of materials are identical across all phases of development, so it is important to consider these requirements when designing a GMP facility within a laboratory setting.

	For example, all materials must be assigned a unique lot number and have proper labeling. An inventory system must provide for tracking each lot of each component with a record for each use. Upon receipt, each lot should be visually examined for appropriate labeling and for evidence of tampering or contamination. Materials should be placed into quarantine or in the approved area or reject area with proper labeling to identify the material and prevent mix-ups with other materials in the storage area. Provision should be made for materials with special storage requirements (e.g., refrigeration, high security). The storage labeling should match the actual conditions that the material is being stored and should include expiry/retest dates for approved materials. Although such labeling is inconvenient for new materials where the expiration or retest date may change as more information is known, this enables personnel to be able to determine quickly whether a particular lot of a material is nearing or exceeding the expiration or retest date. General expiry/retest dates for common materials should be based on manufacturer's recommendation or the literature.

	Finally, there are clear regulatory and environmental requirements for the destruction of expired or rejected materials. It is important to observe regional and international requirements regarding the use of animal sourced materials (12). It is recommended to use materials that are not animal sourced and that there be available certification by the raw material manufacturers that they contain no animal sourced materials. If animal sourced raw materials must be used, then certifications by the raw material manufacturers that they either originate from certified and approved (by regulatory bodies) sources for use in human pharmaceuticals, or that the material has been tested to the level required for acceptance by regulatory agencies (following US, EU, or Japanese guidelines, as applicable) is required.

It is a GMP requirement that all raw materials for the manufacture of drug product have appropriate specifications to ensure quality. The compendial requirements should be used for setting specifications provided the material is listed in at least one pharmaceutical compendium (e.g., 

). It is important that the use of materials meeting the requirements of a single compendium is acceptable for use in early phase clinical studies conducted in the US, Europe, and Japan. For example, a material that meets 

 criteria and is used in the manufacture of a drug product should be acceptable for use in early clinical studies in the European Union. In the absence of a pharmaceutical compendium monograph, the vendor specification and/or alternative compendial specifications such as USP's Food Chemical Codex should guide specification setting. In any case, the sponsor is responsible for the establishment of appropriate specifications. Therefore, it is the authors' position that good practice is to have at least a basic understanding of the manufacture, chemistry, and toxicology of the materials to guide appropriate specification setting.

The minimum testing required for incoming materials is visual inspection and identification. However, as mentioned above, the appropriate tests should be determined for the material based on the knowledge of the manufacture, chemistry, and toxicology. If the vendor is qualified, then the certificate of analysis may be acceptable in conjunction with the visual inspection and identification testing (see "Vendor Qualification" section below).

Ideally, manufacture of a bulk drug product should begin with approved material specifications and with materials that are fully tested and released. However, there are circumstances where it may not be feasible to start manufacture with approved specifications and fully tested and released materials, including API. Manufacturing prior to final release (sometimes called manufacturing "at risk") may be acceptable, however, because the quality system ensures that all specifications are approved, test results are within specifications, and all relevant documents are in place before the product is released for administration to humans. The "risk" must lie fully with the manufacturer and not with the patient.

Vendors supplying excipients, raw materials, or API must be qualified by the sponsor. Appropriate qualification should depend on the stage of development and an internal risk assessment. For, example if a vendor has a history of supplying the pharmaceutical industry and the material is to be used in early development, a paper assessment (e.g., a questionnaire) should be sufficient. If a supplier does not have a history of supplying the pharmaceutical industry, a risk assessment should be performed and depending on the outcome a site audit may be required prior to accepting material for use.

	Ideally, vendors should be qualified prior to using raw materials for manufacture. However, it is acceptable for qualification to proceed in parallel as long as documentation/risk assessments are available prior to product release and as in the previous section all risk lies with the manufacturer and not the patient.

Batch record documentation preparation. 

Manufacturing documentation is a basic requirement for all phases of clinical development. 21 

 Parts 211.186 and 211.188 describe master production and batch production records, respectively (7). The stated purpose of the master production record is to "assure uniformity from batch to batch." Although the record assurance is important for a commercial validated manufacturing process, it does not necessarily apply to clinical-development batches. Material properties, manufacturing scale, and quality target product profile frequently change from batch to batch. Therefore, batch production records are the appropriate documentation for clinical trial supplies. Batch production records for Phase 1 materials should minimally include:

		Name, strength, and description of the dosage form

		A complete list of active and inactive ingredients, including weight or measure per dosage unit and total weight or measure per unit

		Theoretical batch size (number of units)

		Manufacturing and control instructions.

	These minimum requirements are consistent with the FDA 

Guidance for Industry: cGMP for Early Phase Investigational Drugs, 

which requires a record of manufacturing that details the materials, equipment, procedures used and any problems encountered during manufacturing (2). The records should allow for the replication of the process. On this basis, there is flexibility in the manner for which documentation of batch activities can occur, provided that the documentation allows for the post execution review by the quality unit and for the retention of these records.

Review and approval of executed batch records by the Quality unit is required per 21 

 Part 211.192 (7). This review and approval is required for all stages of clinical manufacturing. Pre-approvals of batch records should be governed by internal procedures as there is no requirement in 

 21 that the Quality unit pre-approves the batch record (though this is highly recommended in order to minimize the chance of errors). Indeed, 

shows that pre-approval of batch records by the Quality Unit is practiced by all 10 companies that participated in the IQ Consortium's drug-product manufacturing survey related to early development. Batch records must be retained for at least 1 year after the expiration of the batch according to CFR Part 211.180, but many companies keep their GMP records archived for longer terms.

 Part 211.130 requires inspection of packaging and labeling facilities immediately before use to ensure that all drug products from previous operations have been removed. This inspection should be documented and can be performed by any qualified individual.

	Although line clearance for bulk manufacture is not specifically mentioned in the 

, it is expected that a room clearance be performed. At a minimum, this clearance should be performed prior to the initiation of a new batch (i.e., prior to batch materials entering a processing room).

During the early stages of development, final dosage form release testing confirms product quality and support establishment of hold times later in the clinical development. There is no requirement to establish hold times for work in process in early development. Specific formulation and stability experience, which is usually limited at this stage of development, should be leveraged to assess any substantial variations from expected batch processing times. The data gathered from these batches and subsequent development can be used to help establish hold times for future batches. (Exceptions to this approach may include solution or suspension preparations used in solid dosage form manufacturing, where procedures typically govern allowable hold times to ensure the absence of microbial contamination in the final product.)

Changes to raw materials, processes, and products during early development are inevitable. It is not required that these changes be controlled by a central system but rather may be appropriately documented in technical reports and manufacturing batch records. Any changes in manufacturing process from a previous batch should be captured as part of the batch record documentation and communicated to affected areas. The rationale for these changes should also be documented as this serves as a source for development history reports and for updating regulatory filings. The authors recommend that those changes that could affect a regulatory filing be captured in a formal system.

Process parameters should be recorded but do not need to be predetermined because processes may not be fixed or established in early development. Given the limited API availability in early development, a clinical batch is often the first time a product is manufactured at a particular scale or using a particular process train. Therefore, process changes should be expected. Process trains and operating parameters must be documented in the batch record but changes should not trigger an exception report or CAPA. Changes should be documented as an operational note or modification to the batch record in real time. Such changes driven by technical observations should not require prior approval by the Quality unit, but should have the appropriate scientific justification (

 formulator/scientist) or the appropriate flexibility built into the batch record to allow for the changes. This documentation should be available for Quality review prior to product disposition.

Actual yields should be calculated for major processing steps to further process understanding and enable optimization of processes. Expected yield tolerances are not always applicable to early development manufacture. At this stage of early development, when formulation and process knowledge is extremely limited, there may be no technical basis for setting yield tolerances and, therefore, this yield may not be an indicator of the quality of the final product.

In-process tests and controls should follow basic requirements of GMPS to document consistency of the batch. For capsule products, these requirements may include capsule weights and physical inspection. For tablet products, compression force or tablet hardness and weights should be monitored together with appearance. R&D sampling, defined as samples taken for purposes of furthering process understanding but not utilized for batch disposition decisions, is a normal part of all phases of clinical manufacturing. In early development manufacturing, a sampling plan is required for in-process control tests, but not for R&D samples. However, for the purpose of material accountability, R&D sampling should be documented as part of batch execution. For these samples, testing results may be managed separately, and are not required to be included in regulatory documentation.

	Organizations involved in manufacture of Phase 1 and 2a clinical supplies should recognize that process understanding is very limited in early development. Quality systems must be robust enough to ensure patient safety, but should also be flexible enough to accommodate accelerated timelines and process changes in real time. The special needs of small scale GMP manufacturing should also be considered when designing facilities, purchasing equipment, and selecting the type of dosage forms to use in early clinical studies. Companies are encouraged to apply quality risk management principles to support these risk-based decisions.

	The authors also believe that underutilized approaches exist (e.g., on-site preparation of more complex dosage forms) for quickly and efficiently answering formulation related questions about bioavailability, pharmacokinetics, and target release rates for controlled-release formulations. The potential benefits and risks of these approaches warrant further discussion. Finally, documentation of manufacturing operations should be risk-based. Manufacturing instructions in early development should not be overly prescriptive as to restrict process changes or discourage sampling for further process understanding. Changes should be expected, and able to be quickly reviewed and approved by the Quality department, or a qualified delegate.

	To further stimulate discussions on these approaches within the industry and with worldwide health authorities, the IQ Consortium's GMPs in Early Development Working Group is planning on conducting a workshop in the near future to promote robust debate and discussion on these recommended approaches to GMPs in early development. The group strongly believes that such dialogue will improve alignment between development, quality assurance, and chemistry/manufacturing/controls (CMC) regulatory groups within the pharmaceutical industry. In addition, agreement between the industry and drug regulatory authorities regarding acceptable approaches to applying GMPs in early phases of drug development would allow for a more nimble and flexible approach in early development, while still providing appropriate controls to ensure patient safety.

 is in pharmaceutical development at AstraZeneca (Wilmington, DE); 

is in pharmaceutical development at Vertex Pharmaceuticals (Cambridge, MA); 

 is in quality at ARIAD Pharmaceuticals (Cambridge, MA); 

is in pharmaceutical sciences and clinical supply at Merck (Summit, NJ); 

 is in pharmaceutical sciences at Pfizer (Groton, CT); 

is in small molecule pharmaceutical sciences at Genentech (South San Francisco, CA); and 

 are in drug product development at Abbott Laboratories (Abbott Park, IL).

	*To whom all correspondence should be addressed.

Guidance for Industry: CGMP for Phase 1 Investigational Drugs 

(Rockville, MD, July 2008).
	3. FDA, "Current Good Manufacturing Practice and Investigational New Drugs Intended for Use in Clinical Trials," 

Guidelines to Good Manufacturing Practice: Medicinal Products for Human and Veterinary Use,

 Annex 13: Investigational Medicinal Products (Feb. 2010).
	5. D. Chambers, 

 (7) 76–84 (2012).
	6. D.A. LeBlanc, in "Validated Cleaning Technologies for Pharmaceutical Manufacturing," (Interpharm/CRC, Washington, DC, 2000).
	7. 

, Title 21, Food and Drugs (Government Printing Office, Washington, DC).
	8. Personal communication with Mike Kress, Leader, IQ Consortium Drug Product Working Group on Extemporaneous Formulations (May 2012).
	9. 

 30 General Chapter <795> Pharmaceutical Compounding-Nonsterile Preparation, 344–350.
	10. 

 30 General Chapter <797> Pharmaceutical Compounding-Sterile Preparations, 350–387.
	11. L.V. Allen, The Art, Science and Technology of Pharmaceutical Compounding (American Pharmacists Association, 2008).
	12. See, for example: the USP Excipient Supplier Qualification Program, Section 10 (V. 1.1, Sept. 2008), and references therein, for specific guidance.

Articles

IQ Consortium representatives explore industry approaches and practices for applying GMPs in early development.