GMPs for Small-Molecule Drugs in Early Development: Workshop Summary-Part VI

Published on: 
, , , ,
Pharmaceutical Technology, Pharmaceutical Technology-11-02-2014, Volume 38, Issue 11

The authors explore and define common industry approaches and practices when applying GMPs in early development.

The International Consortium on Innovation and Quality in Pharmaceutical Development (IQ Consortium) is a technically focused organization of pharmaceutical and biotechnology companies with a mission of advancing science-based and scientifically driven standards and regulations for pharmaceutical and biotechnology products worldwide.  In previous issues of Pharmaceutical Technology, papers 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 I through Phase IIa) (1-5). In this paper, the IQ Consortium presents a summary of key analytical method validation, stability, and manufacturing discussions that were part of the IQ Consortium’s recent workshop, “Best Practices and Applications of GMPs for Small Molecule Drugs in Early Development,” which was based on these earlier papers. The workshop was held on Feb. 4–5, 2014 in Washington, D.C.  Attendees included more than 70 analytical; formulation development; quality assurance; and chemistry, manufacturing, and controls (CMC) regulatory scientists, representing more than 20 companies and FDA.

Workshop presentations consisted of industry representatives summarizing the previously published IQ Consortium papers (2–5) as well as FDA representatives who spoke on the same topics.  The breakout sessions were designed to stimulate deeper discussion on specific topics and sharing of best practices across the industry.  The presentation materials and key messages from general presentation sessions and the related breakout sessions are available on the IQ Consortium website (6).  Although there were no specific agreements reached, there were constructive discussions throughout the workshop.

A summary of the key discussions related to analytical method validation, stability, and manufacturing is outlined in the following sections.  A summary of the key discussions related to specifications in early development will be the subject of a future article.

Analytical method validation in early development workshop output

Workshop discussions regarding method validation in early development focused on validation parameters for compound-specific methods and general methods, as well as industry terminology.  There was general agreement with phase-appropriate method validations as outlined in the position paper (2) as long as sound scientific judgment is used. The group recognized that method validation is not a singular event performed to satisfy a regulatory requirement. Rather, validation should be thought of as an ongoing exercise that evolves every time the method is used. Although there was agreement that evaluating robustness and intermediate precision did not have to be part of the method validation process in early development, consensus on some other validation elements was not reached. Specifically, there was some discussion that, even though it is routinely evaluated by most companies, linearity may not be a critical parameter to evaluate in early development since the criteria are almost always easily met. In addition, there was discussion regarding whether the use of system suitability in lieu of a separate method-validation process could ensure that the method performance (e.g., linearity, precision, specificity, sensitivity, etc.) was suitable for the intended use.

There was considerable discussion during the presentations and breakout session regarding the terminology used to describe validation activities. Specifically, the apparent interchangeable use of “validation” and “qualification” was discussed.  The lack of consistency in the terminology has led to unnecessary debate and confusion across the industry and in communications with regulatory authorities. Harmonization of clarifying terminology such as Stage 1 validation, Stage 2 validation, or fit-for-purpose (FFP) validation might provide greater clarification on the stage of validation.

Approaches for validating general methods, such as gas chromatography (GC) residual solvent, heavy metal, and compendial methods varied within the industry. In most cases, compendial methods are evaluated to determine suitability of use, but are not validated.  However, some companies performed validation of FFP general methods (e.g., GC residual solvents) for each drug substance (DS) and/or drug product (DP) sample matrix as compound specific methods.  In contrast, other companies validated these methods for each individual analyte (e.g., solvent or metal), but did not perform additional validation for every DS and/or DP matrix.

Drug product manufacturing in early development workshop output

Results from the 2011 survey on early development manufacturing practices (3) highlight two areas that could benefit from open discussion: batch documentation and change-control systems.  There was consensus at the workshop that batch records and change-control systems need to be designed to accommodate manufacturing flexibility in early development. Feedback from participants indicated that many companies have implemented approaches that allow this flexibility.  The following are some examples of these approaches as well as discussion notes:

  • Only providing process parameter ranges in batch records when sufficient process knowledge has been accumulated. This approach reduces the number of process “deviations” that need to be addressed during or after a campaign.  While health authorities would like to see some targets/ranges specified, they are mainly interested in knowing that the manufacturer has adequate controls and knows what steps must be taken to assure high quality product for the clinic.  It is important, however, to provide enough instruction in the batch record such that actual set points or values observed during processing are recorded and deviations from expected performance are clear.

  • Several companies are moving away from having the quality unit preapprove batch records. In the 2011 survey, 100% of the 10 companies that responded indicated that their quality units preapprove batch records.  Since the survey, one company has removed the requirement for packaging operations.  Another company performs daily, concurrent batch record review with the quality unit during campaigns and is in the process of removing the preapproval requirement from their quality system.

  • Change control approaches vary between companies, but there was agreement that systems in early development must be able to quickly address deviations and changes. Several companies manufacture externally and rely heavily on the CMO’s quality systems for documenting deviations.  In cases where the CMO manufactures both early- and late-phase/commercial supplies, it is advantageous to track the deviations separately for corrective action and preventive action (CAPA) purposes. It was noted during discussion that direct and timely quality involvement in early development should be encouraged to ensure that potential manufacturing concerns are quickly captured and addressed or dismissed.

A second breakout session focused on the practices of extemporaneous preparations (EP). On-site formulation preparation is an effective means of preparing early clinical supplies.  In addition to shorter timelines, advantages of this approach include: lower DS demand, reduced analytical method and stability support, flexibility to quickly adjust dosing in response to clinical data, and less resource demand. According to an IQ working group survey, approximately 80% of companies already have the capability to do EP.  This approach, however, is only being used in approximately 20% of applicable applications. Full results from this IQ survey will be published separately.  Some obstacles companies have faced in implementing EP include: access to sites that can handle hazardous compounds or complex formulations, interpretation of regulatory requirements and regulatory authority expectations, and internal resistance toward “non-traditional” approaches.


The breakout session included three short presentations covering the history of EP, regulatory requirements/strategies and case studies of successful on-site preparation of immediate- and controlled-release dosage units.

While the preparation of EP material might not take place in traditional GMP manufacturing areas, it remains crucial that appropriate controls are in place to ensure subject safety and to safeguard product quality. Participants at the workshop agreed that safety is top-priority and that this responsibility ultimately resides with the study sponsor.  The sponsor quality unit should audit the compounding site. Industry participants discussed the importance of carrying out one or more practice preparations at the compounding site.  Industry participants discussed the practice of testing during and after these mock runs and using these data to validate preparation instructions, potentially eliminating the need to perform end-product testing on the actual preparations that will be given to study subjects. This approach allows for rapid dosing and de-risks changes that may occur in the product while waiting for analytical testing results.  If this approach is taken, it is still important to perform stability studies on the mock preparations to support a practical “use-period” for the product.  FDA representatives did not provide comment, but referred to the FDA guidance on testing expectations (7).

The EP approach is not limited to simple powder-in-bottle or solution/parenteral preparations for ADME and absolute bioavailability. Companies have successfully used EP to study the effects of drug release rate on pharmacokinetics using controlled-release preparations.  Case studies from on-site preparations of matrix tablets as well as osmotic capsules were presented.  Results of these types of EP studies can be used to quickly answer questions about food effect, colonic absorption, or potential for reducing adverse events-while using only a small amount of DS.

There was considerable discussion during the breakout session about regulatory expectations and filing strategy. EP practices rely on standards of quality in a clinical pharmacy, which are governed by local laws and good clinical practices.  Requirements for CMC submission vary country-to-country. For Singapore studies, no CMC information is typically provided to the Health Sciences Authority (HSA). Instead, someone from the HSA will visit the clinic or pharmacy to inspect preparation activities. For US studies, section P.3.3 of the investigational new drug (IND) application should include all of the key preparation steps and controls.  It was noted that in Europe, all compounding pharmacies are cGMP compliant, and EP are released by a qualified person (QP).  From a submission perspective, most companies treat EP clinical studies as “Phase I” even if the active ingredient itself is in later development or commercially available.  The studies are short in duration, are closely monitored, and often use low or sub-therapeutic doses. The EP studies are also often the first time that a formulation has been evaluated, and usually do not represent the intended Phase III or commercial product.

Stability in early development workshop output

Presentations and breakout sessions on stability in early development followed the elements and recommendations of the IQ paper on the subject (4). Some participants acknowledged that the industry may be currently performing a lot of non-value-added stability studies, and that most of the resistance to reducing stability testing comes from within the companies rather than from regulatory authorities. It was pointed out that companies should focus on accumulating product/process knowledge and use risk assessment to design only necessary stability studies, while being mindful of regulatory environments.

Strategies in leveraging solid stress testing for drug substance and simple drug products like PiB (Powder in Bottle) and PiC (Powder in Capsule) were discussed. This practice is typically applied to stable and moderately stable DS, which is usually stressed at 70°C/75%RH for up to three weeks and at the 1X International Conference on Harmonization (ICH) Q1B photostability testing condition. Open containers to simulate the bulk (DS) least-protective packaging configuration are typically used, although some companies also stress the DS in closed containers to simulate packaged PiB and PiC configurations. Usually an initial DS retest period and an initial DP shelf life can be extrapolated to 15 months at 25 °C, with a range of 1218 months depending on companies’ internal policies and risk tolerance. The assigned retest period and shelf life (called “use period” for brevity) is then verified with reduced ICH long-term and accelerated stability testing of a “representative batch.”  With this approach, one company has achieved seven successful clinical trial application (CTA)/IND submissions with an EU country and FDA.

Companies also shared their experiences in using the software ASAPprime, an Accelerated Stability Assessment Program (ASAP) modeling software, to enable rapid stability evaluations in early development. Employing several combinations of temperature and humidity conditions, ASAP is typically used to evaluate the chemical stability and thus to extrapolate a use period of DS and formulated DP. This model is also commonly used to assess the potential impact of packaging changes on material stability. Depending on the model data, it is possible to change to a less protective packaging without having to repeat stability testing. Physical changes are harder to predict with ASAP. Two companies have included ASAP stability data in regulatory filings with FDA and some EU countries. One company noted that some EU countries have expressed discomfort with ASAP-based extrapolation.

How stability data from representative batches of DS or DP are used differed among companies.  Most companies use a non-GMP representative batch of DS or DP to get a more rapid assessment of the material’s stability. How stability studies are performed on the first GMP batch, however, varies across the industry. In some cases, when a representative non-GMP batch is placed on stability, GMP batches may be placed into stability chambers without testing. This approach would make the GMP materials on stability available for testing when there is a need to respond to regulatory queries, or if further development information gathered suggests that the batch is, in fact, not similar to the representative batch. Other companies conduct ICH long-term and accelerated-stability studies on the first GMP batches to use those batches as the stability representative batches.

It was also noted that while changes to DS or DP processes may affect the quality of a new batch, they may not affect its stability. When changes to processes or materials are made, any potential changes to the stability-related quality attributes should be considered to determine if stability studies are warranted for the new material. This is commonly done using a stability risk assessment (RA). In general, companies use RAs to make stability-testing decisions for new batches of DS and DP after changes. The formality of the procedures used to follow and capture RAs varies across the industry. RAs tend to be less formal for early-stage development and more formal for late-stage development.


The authors thank Linda Ng, Stephen Miller, Mahesh Ramanadham, and Ramesh Sood from FDA for their participation and contributions to the workshop and this summary.


1. A. Eylath et al., Pharm. Technol. 36 (6) 54-58 (2012).
2. D. Chambers et al., Pharm. Technol. 36 (7) 76-84 (2012).
3. R. Creekmore et al., Pharm. Technol. 36 (8) 56-61 (2012).
4. B. Acken et al., Pharm. Technol. 36 (9) 64-70 (2012).
5. M. Coutant et al., Pharm. Technol. 36 (10) 86-94 (2012).
7. FDA, Guidance for Industry: cGMP for Phase 1 Investigational Drugs (CDER, July 2008).

About the Authors

Q. Chan Li, senior principal scientist, Boehringer Ingelheim
Jackson D. Pellett, scientist, Genentech
Michael Szulc, principal scientist, Biogen-Idec
Mark D. Trone, director, Alkermes
Kirby Wong-Moon, principal scientist, Amgen