Comments to FDA on 503B Quality Standards Draft Guidance
On September 2, Pew submitted comments to the U.S. Food and Drug Administration in response to draft guidance on the implementation of Title I of the Drug Quality and Security Act.
The comments highlight the need for more robust safeguards when compounding occurs on a larger scale and reference the Pew commissioned report “Ensuring the Safety of Compounded Drugs: Study highlights key quality standards.”
September 2, 2014
Division of Dockets Management (HFA-305)
Food and Drug Administration
5630 Fishers Lane, Rm. 1061
Rockville, MD 20852
Re: FDA Draft Guidance: Current Good Manufacturing Practice — Interim Guidance for Human Drug Compounding Outsourcing Facilities under Section 503B of the FD&C Act.
Dear Sir or Madam,
The Pew Charitable Trusts is an independent, nonpartisan research and policy organization dedicated to serving the public. Pew has a longstanding focus drug quality, including pharmaceutical compounding. We supported federal legislation to improve compounding oversight in 2013, and have conducted research and convened stakeholder summits.
In 2014 we commissioned a study that compares Current Good Manufacturing Practices (CGMPs), the standard applied to pharmaceutical companies, with United States Pharmacopeia Chapter 797, the standard used by compounding pharmacies. The study, Quality Standards for Large-Scale Sterile Compounding Facilities, authored by Clinical IQ LLC, highlights the more robust safeguards needed when compounding occurs on a larger scale.
We appreciate the opportunity to comment on FDA draft guidance on CGMPs for outsourcing facilities, which will be permitted to compound not pursuant to a patient prescription. Such facilities will register and submit to FDA oversight, maintain strict quality standards, and adhere to certain other requirements. This draft guidance provides a roadmap that will help outsourcing facilities understand how the agency intends for them to comply with CGMPs, and what standards are of particular importance to sterile compounding activities. This clarity is important for facilities that elect to join the new 503B regulatory category, which is voluntary, and will help support successful implementation.
In this guidance the FDA has recognized the need to make certain accommodations for valid compounding practices that differ from drug manufacturing. FDA should continue to take this approach, and should work closely with 503B facilities to refine the guidance where necessary to ensure they provide meaningful protections but also make reasonable accommodations and set appropriate requirements for activities of higher and lower risk. Currently the draft guidance does not differentiate in most cases between higher and lower risk compounding, but such differentiation may be warranted in areas such as release testing.
In our comments below, we also suggest that the FDA add greater specificity in some areas to give outsourcing facilities more clarity on how the FDA would like them to comply with relevant cGMP provisions. CGMP assumes a company has already developed detailed specifications, production systems, and testing regimens for their products, and that further detail and tailoring will be agreed upon between the manufacturer and the agency through the application process. But this mechanism is not available to FDA in regulating 503B compounders. Greater specificity from the FDA will help ensure all outsourcing facilities have the ability to meet the Agency’s expectations.
We support FDA’s emphasis on proper facility layout and air flow to prevent contamination. Controlling air flow is very important to maintaining a controlled environment for sterile production, and should include ongoing air flow studies under dynamic conditions.
Control Systems and Procedures for Maintaining Suitable Facilities
We support FDA’s emphasis of the need for detailed cleaning and disinfecting procedures, and for use of sterile disinfectants. We also support the recommended use of sporicidal disinfectants. Humans are the greatest source of contamination, and processes should minimize this risk, such as by minimizing additional human manipulations.
Environmental and Personnel Monitoring
We support FDA’s emphasis of proper environmental monitoring, including monitoring during actual operations. Regular surface, air, and personnel monitoring are essential to control of the production environment.
Equipment, Containers, and Closures
We support the expectation to qualify equipment, as well as processes to ensure the quality and sterility of containers and closures prior to use. Particulate matter from glass vials and other packaging have caused recent drug recalls.
Recommendation: Guidance should require periodic assessment of supplier reliability. Guidance should state that in order to rely on a supplier’s certification of sterility and forego separate sterility testing of each lot of containers and closures, outsourcing facilities should assess the reliability of the supplier’s analysis at appropriate intervals, as the guidance stipulates for components (every two years).
The quality of starting drug components has serious implications for the quality assurance of a finished product intended to be sterile, including the need for additional end-stage testing or other measures to ensure sterility. We do not object to foregoing component testing if the starting component is an FDA-approved drug. For components that are not approved drugs, we support the reliance on a supplier’s certificate of analysis if a separate identity test is conducted, and the reliability of the supplier’s analysis is assessed at appropriate intervals.
Recommendation: Clarify in guidance that endotoxin testing is needed for all incoming non-sterile components used in compounded sterile products. While line 353 of the draft guidance indicates this is required, line 336 seems to suggest this might not be required depending on intended use. Guidance should clarify that for any product that must be sterile, its intended use would necessitate such testing. The bioburden of starting ingredients has an impact on the effectiveness of sterilization methods. Endotoxin testing to assess bioburden is particularly important for incoming nonsterile active ingredients when a facility is not conducting overkill terminal sterilization but rather is using sterile filtration, which is a more fallible process.
Recommendation: Guidance should include a direct reference to USP Chapter 85 on endotoxin testing, which describes detailed best practices. Omitting such a reference could create confusion about what methods are acceptable. If USP 85 is not directly referenced, FDA should make clear what specific endotoxin testing methods should be used.
Recommendation: FDA should pursue the proposal to allow for discretion regarding periodic testing if the ingredient supplier submits a Drug Master File (DMF) to FDA that describes testing protocols. Suppliers of ingredients to outsourcing facilities should be conducting their own robust testing, and we support a system where suppliers describe these methods to the FDA in a DMF. However, the FDA must provide sufficient oversight of these suppliers, such as through physical inspections and records review, to ensure meaningful confidence in component quality.
Production and Process Controls
Strong SOPs are important to ensure process uniformity and consistency. We support FDA’s call for written procedures for production and process controls.
We support the specific points made in the draft guidance about the need for sterile filtration if the drug product intended to be sterile is not terminally sterilized, as well as the emphasis placed on limited hold times and assessing the effects of hold times on bioburden and endotoxin levels.
We support the emphasis on sterile garb, including sterile and non-particle shedding gowning.
We support the call for introductory and ongoing personnel training by conducting media fills, and for the media fills to be performed in the same area where production occurs, closely simulating aseptic manufacturing operations and, as appropriate, worst-case activities and conditions.
Recommendation: FDA should include a frequency for ongoing training in guidance that, at minimum, matches USP requirements. Personnel engaged in low and medium-risk activities should train by conducting media fills at least once a year, and personnel engaged in high-risk activities (e.g. compounding a sterile drug from a nonsterile starting ingredient) should do media fills at least twice a year. FDA should also consider providing greater specification about the length of each media fill training, such as the minimum number of media fills per day over a minimum number of consecutive days that constitutes an adequate training.
Although the best assurance of quality, including sterility, is robust, validated sterile production methods, a release testing program still provides an important check, particularly for high-risk compounding activities.
The testing accommodations in the draft guidance are reasonable, including foregoing a sterility test if the drugs have been terminally sterilized using a validated method, and allowing release of product before a sterility test is complete if appropriate notification mechanisms are in place.
Recommendation: Batches of fewer than 10 dosage units could reasonably be held to the same testing standards whether or not they are associated with an individual patient prescription. If forgoing sterility testing is permitted for these batches as long as certain beyond use dates are used, this standard should not change due to the presence of a prescription, which has no impact on the quality or risk of the product or process.
Recommendation: Sterility testing expectations should be different for high risk and lower risk activities. As currently written, the guidance treats low- and high-risk activities the same with regard to sterility testing. The FDA should consider whether stricter standards should apply to sterile drug compounding that begins with nonsterile ingredients, or whether further accommodations should be granted when the starting material is sterile FDA-approved products.
Recommendation: Aside from differences stipulated in the guidance, the FDA should set the expectation that USP Chapter 71 be followed for sterility testing methods. As written, the guidance references USP 71 but does not suggest it must be followed. FDA should make clear that in the absence of other independently validated sterility testing methodologies, USP 71, which contains best practices for sterility testing, should be applied. If USP 71 is not set as the expectation, FDA should make clear what specific sterility testing methods should be used.
We support FDA’s focus on laboratory controls, particularly in light of recent challenges identified at external contract labs.
Recommendation: FDA should pursue the proposal to reduce the need for in-house final release testing if the outsourcing facility uses an external laboratory that submits a DMF to FDA describing testing protocols. This proposal could help ensure external labs are using robust testing methodologies. However, the FDA must provide sufficient oversight of these external labs, such as through physical inspections and records review, to ensure meaningful confidence in testing reliability.
Stability Program and Expiration Dating
Expiration dates for a drug should ideally be established through objective data from a robust stability testing program. We support the expectation that outsourcing facilities conduct stability studies that includes evaluation of at least three batches of product in order to set extended expiration dates.
When an expiration date is not established through stability studies, the guidance appears to set appropriate maximum beyond use dates for different scenarios. We support that beyond use dates are the most restrictive when an aseptically produced product is not terminally sterilized and is released before the sterility test is completed – mimicking the USP 797 storage periods indicated for drugs produced under high-risk conditions in the absence of separate testing results.
Packaging and Labeling
We support the expectation that packaging must appropriately maintain sterility and integrity, and that labeling operations must include controls to prevent mix-ups.
Quality Assurance Activities/Complaint Handling
We support the expectations set by this section. The independence of quality units is important to ensure their work is not affected by production or financial considerations.