An FDA Inspection
So you’ve been visited by the FDA and whether you’re a seasoned QA manager at a major pharmaceutical manufacturer or the pharmacist in charge of a small compounding pharmacy you’ll more than likely receive (or have received) a 483 form after the visit. It’s how you respond that decides if any further action by the FDA will take place. In this article I’ll talk about the basis for receiving a 483, the FDAs systematic way for conducting inspections including their expectations and finally I’ll present a case study of a pharmacy that received a 483 from the FDA and how they were able to close it out; with a very small intervention on my part.
FDA Form 483
A 483 is simply the agency’s way of documenting and notifying a firm’s management of observations made during an inspection. Typically, the observations are significant enough for an inspector to document when they have noticed something that would cause the firm’s products to be adulterated and/or bring harm to the public. However, the 483 is not meant to be all inclusive; meaning there may be other issues with the facility that the inspector hasn’t noted. The inspections are pre-planned and conducted with a particular focus in mind. If there’s been a specific complaint or publicly known issue then the inspection may be very targeted on certain aspects of the operation.
This leads to the two types of inspections that the FDA conducts: full and abbreviated. A new firm opening its doors for the first time will receive a full inspection as well as previously inspected manufacturers that have had significant deficiencies from abbreviated inspections. A full inspection is a top to bottom inspection of most of the major systems the FDA monitors (more on those systems in a bit). An abbreviated inspection is warranted only when a firm has previously demonstrated that they have an ongoing record of compliance and has implemented a risk management program in their manufacturing operation.
The Basis for 483 Observations
I’ll be touching on this below but a brief explanation for why your firm received a 483 is based on adherence to the FDAs Quality Systems approach to manufacturing. The takeaway here is that you have control over the six main systems the FDA observes: the Quality, Facilities and Equipment, Materials, Production, Packaging and Labeling and Laboratory Control systems. The key is in how you demonstrate and maintain control over each of these systems; more on those below.
Responding to a 483
One of the most important things about your response to the FDA is the timeliness. You MUST respond, in writing, within 15 days from the receipt of the form 483. In your response be sure to include:
- restatement of the issue
- activities being pursued to contain the issue
- root cause investigation (if necessary)
- actions taken to prevent recurrence
- documented evidence of corrective measures
- verification of effectiveness of corrective measures
More than likely, if your response is well written with a good action plan for remediation and a timeline for completion, your chances of receiving a warning letter will be greatly reduced. A best practice to illustrate your action plan on a timeline is to provide the agency with a Gantt chart with your response. This gives the agency a visual timeline for correcting all of the observations from the 483. If you’re going to include a nice, fancy Gantt chart with a timeline of corrections be sure to make the necessary changes in a timely fashion but be realistic. Whatever you do, DON’Toverpromise and under deliver. This is the quickest way to receive follow up actions (i.e. warning letter).
Here are two gantt chart templates; one from google docs and the other is for excel:
FDAs inspection process
If you’re unfamiliar with how the FDA conducts inspections and what they are observing, read on. If you’re already familiar with their quality systems approach to inspections then feel free to skip ahead. Let me first say that nothing the FDA says or does is covert; it is ALL publicly available. Their expectations, how they inspect, what they observe is published online; with the exception of the 483s for large manufacturers. That’s correct, there is a disparity between what is shown online for small compounders and large manufacturers. However, that’s NOT to say it isn’t publicly available; anyone can make a request to see specific 483s through a freedom of information act request. At the conclusion of the article I will include all of the FDA web addresses where important documents can be found including their inspection manuals and guidances pertaining to compounding and sterile drug manufacturing.
There are six quality systems the FDA looks at in total but typically in routine inspections they may only look at as few as two; but possibly more (the quality system is always mandatory for being inspected). The six systems are:
- Quality System
- Production System
- Facilities and Equipment System
- Materials System
- Packaging and Labeling System
- Laboratory Control System
Through this inspection process the Agency is able to quickly and efficiently ascertain whether a company has control over its systems to maintain a safe, quality product is being made. While compliance with cGMPs is the main focus of the inspection process; quality is at the center of the cGMPs. It must be said that while these systems seem to be split up and disjointed they are very much interrelated. A disruption in one system may very easily lead to faults in another.
The Quality System
The quality system is the system that sort of oversees all of the others; in their (FDA) words:
“The system includes the quality control unit and all of its review and approval duties (e.g., change control, reprocessing, batch release, annual record review, validation protocols, and reports, etc.). It includes all product defect evaluations and evaluation of returned and salvaged drug products.”
Change control is worth further explanation. If you’ve made a change to a process or procedure, have you kept track of the change, made everyone aware and documented as such? You need to have a system that documents the change and that everyone is aware of and following. Something else to mention though is what was the purpose of the change? Did this come about as a result of your CAPA (corrective action/preventive action) program? What factors led to this change and was an evaluation of risk performed? After the change was made was there a determination made that the overall system improved as a result?
The point that really needs to be driven home here is that with ALL of these systems, they require inputs. A system is only as good as its inputs. If you’re scratching your head, you’re not alone. By inputs I mean data points with which you monitor your systems performance. In the case of your quality system this may mean taking a look at your records for recalls. How many recalls did the firm have over the past “x” number of months? What was the reason(s) for the recall? Is there a trend and conclusion that can be made to fix some kind of issue (corrective action) or maybe a policy change to prevent the problem going forward (preventive action)?
The Facilities and Equipment System
The facilities and equipment system is fairly self explanatory, in the FDA’s words:
“This system includes the measures and activities which provide an appropriate physical environment and resources used in the production of the drugs or drug products.”
An example of this may include a remark on a compounding pharmacy’s HVAC system and the lack of a reliable way of detecting and warning of a loss of pressure during production. Obviously this is just one example and there are many systems that may come under scrutiny including any piece of equipment used in the direct production of a compound or indirectly. One of the key factors here is that if there’s a piece of equipment you’re using anywhere in your facility its use, calibration and maintenance should be documented.
The Production System
The production system deals with all activities related directly or even indirectly to production;
“This system includes measures and activities to control the manufacture of drugs and drug products including batch compounding, dosage form production, in-process sampling and testing, and process validation.”
I think some of the more serious observations can be made of the production system. To put this in context though, let’s look at some examples of some higher risk compounds purported to be sterile and each compound’s route of administration:
- Phenylephrine: typically given intravenous or intrathecal
- Succinylcholine: intravenous infusion
- Adenosine: intravenous
- Hydromorphone (labeled as intrathecal)
- Fentanyl (labeled as intrathecal)
If performing a quality risk assessment of your processes and products you’d find that products injected into the spine in particular have the highest risk involved due to the potential outcome if something was to go wrong during production. The risk of contamination in a product that’s injected into the spine can honestly be deadly.
This system is fairly self explanatory in that it deals with the monitoring of the various materials that come in contact with the final product at some point in the process. Materials could be ingredients, containers, water systems, closures or process gases. Let’s take an example from a recent 483:
The firm does not have procedures or processes that provide sufficient control and storage over non-sterile glass vials prior to their use in compounding. The firm purchases non-sterile glass vials. Throughout the inspection I observed these glass vials being stored uncovered in unclassified areas with the opening up. The firm does not document the reported sterilization of these containers by (15) (4) j. The firm does not depyrogenate these containers before use.
The outsourcing facility cleans and sterilizes non-sterile vials for use in production. Attempting to clean, sterilize, and depyrogenate vials is a massive undertaking that requires quite a bit of work to validate the process. More to the point, it’s necessary to have the right equipment and documentation of the process to prove it is being done correctly. Not impossible, but this is such a critical process for obvious reasons that smaller operations should consider just outsourcing this unless you can justify the costs of doing it right. To just cut to the chase, ISO 13408-1 Annex A has an example flow chart of the various materials used in aseptic processing that would be considered to have an effect on the Critical Quality Attributes (CQA) of your compound. CQA is defined as:
A physical, chemical, biological, or microbiological property or characteristic that should be within an appropriate limit, range, or distribution to ensure the desired product quality (ICH Q8)
The CQA in the above example would be the depyrogenation of the final container and its effect on the endotoxin levels of the product. Remember the 4 things the FDA is trying to ensure in all products: identity, strength, purity, quality (quality meaning without microbial contamination).
While the cost of ready-to-fill (cleaned, depyrogenated and sterilized) vials may be more expensive than buying non-sterile and attempting to process yourself; it’s worth the money to have that assurance level and not having to do a recall of those compounds (or all).
Packaging and Labeling System
There are a few aspects of packaging and labeling the FDA is looking for when looking at this system. First, they want to ensure the product won’t introduce risk to the product integrity (damage to container). Next, the container isn’t susceptible to damage during shipping or isn’t stored out of range of the temperature appropriate for the product. They do understand that some containers may be stored for a period of time unlabeled, however they want assurance that the system in place accounts for this and is able to prevent mix-ups. The safest thing to do here is to leave no ambiguity and nothing to chance because after all we’re human; this would be an easy to mix-up situation. Label each unit.
Laboratory Control System
This is a decent sized system to keep control over. Without listing every aspect that the FDA will look at inside this system it can be summarized by saying anything that a firm is utilizing to have assurance over the sterility of the product, materials, personnel or environment it’s best to have a fair amount of scientifically backed data to support that claim. This includes the analytical lab, environmental monitoring, biological indicators, disinfectants and growth media being used.
As I eluded to at the beginning of this article, how you respond to 483 will determine whether or not the matter is brought to a successful conclusion. If you respond in an argumentative disagreeing tone, this will not draw anything to a close. If you respond in a manner where you have a plan of action correcting each observation with a timeline for implementation and truly change your policies and procedures as a result, this will conclude the situation, at least perhaps until the next time the FDA visits your facility. However as I’ve said, the FDA certainly does favor the response to a 483 in a particular format. I gave a couple of examples above but recently I was directly involved, to a small degree, in responding to the FDA and it resulted in the cancellation of an in-person meeting and receiving an FDA FMD-145 (Field Management Directive – Procedure for Release of Establishment Inspection Report to the Inspected Establishment, which means the matter is considered closed).
In 2016, this pharmacy (a 503A) was visited by the FDA for a total of 5 days. The pharmacy had responded in a timely manner however the FDA wasn’t satisfied with their response for one observation in particular and had sent a letter requesting a meeting to discuss the pharmacy’s remediation plan. This is the point at which I was asked my opinion on the situation. I was sent all of the information that had been presented to the FDA thus far. The observation that was still under scrutiny by the FDA was (NOTE: (b)(4) denotes where the FDA has redacted a piece of information, I have also redacted information to protect the identity of the pharmacy):
There is a failure to thoroughly review any unexplained discrepancy and the failure of a batch or any of its components to meet any of its specifications whether or not the batch has been already distributed.
Specifically, Out-of-Specification results were obtained during environmental monitoring during sterile processing on (REDACTED DATE)
a. On (REDACTED DATE) a result of 1 CFU/m3 was recorded for the Viable particle air sampling within your Anteroom Clean Room (ISO 7 environment) and the fingertips of the left hand of the sterile processing pharmacist working within your ISO 5 LAFW exceeded the limit of(b)(4) < l CFU/m3 and CFU/plate, respectively, for fungus. The fungal contamination was identifled as Penicillium spp. The sterile processing pharmacist can move between LAFW Buffer Room (ISO 7), Anteroom (ISO 7) and Hazardous Materials Room (ISO 7). On this day various sterile products were produced including (REDACTED COMPOUND NAME AND LOT NUMBER).
b. On (REDACTED DATE) the limit of(b)(4)YCFU/plate for fungus was exceeded as 1 CFU/plate was observed on the surface of the ISO 5 LAFW where the pharmacist conducts sterile drug production/repackaging operations within the ISO 7 Buffer Room. The growth was identified as a non-sporulating hyaline fungus. Additionally, 1 CFU/plate was observed on the sterile processing pharmacists’ left hand that was identified as Penicillium spp. On this day various sterile products were produced including (REDACTED COMPOUND NAME AND LOT NUMBER), which was recalled on (REDACTED DATE).
C. Environmental Monitoring sampling on (REDACTED DATE), completed during room recertification by a third party, showed an OOS result of “UTQ” (Unable To Quantitate), which is equivalent to a result of Too Numerous To Count, for quantity inside the ISO 7 Hazardous materials Room, on the (b)(4). This (b)(4) is used to stage materials, container closures, etc. by the pharmacist that is working in the ISO 5 (b)(4) Identification showed that the bacteria was Bacillus with no speciation.
***END OF OBSERVATION FROM FDA FORM 483***
The pharmacy had responded within the timeframe they were supposed to but how they responded was in disagreement with the FDAs observation and without a timeframe for a plan of action or follow up with the FDA. This resulted in the FDA requesting an in person meeting at the pharmacy to “discuss the compliance status” of the pharmacy and “come prepared with your firm’s investigation into this contamination and corrective actions which have been completed.”
The pharmacy had contacted me and said that they were unsure of what more to do; all of the information had been presented to the FDA including a response from a microbiologist with a plan for remediation and ultimately concluded in the pharmacy ceasing sterile operations. I asked the pharmacy to send me everything that had been sent to the FDA thus far and after reviewing all of the documents I made one recommendation to them. For the brevity of this writing I won’t be sharing every piece of information that was shared between the pharmacy and the FDA but note that in the response that we crafted absolutely no new information was presented to the FDA. Rather the information was just reformatted with a systematic way of presenting, analyzing and correcting the issue. I certainly cannot take credit for the format but encourage you to use this as a means of identifying, analyzing and improving processes and procedures.
Six Sigma: DMAIC
If you’re unfamiliar with Six Sigma, it is a set of tools for process improvement. Six Sigma itself refers to a process that is well controlled with little variation and comes from statistics meaning six standard deviations from the mean and the specification limits. In plain English: a process that has less than 3.4 defects per million opportunities (very little error). Anyone and everyone involved in compounding should make themselves familiar with the improvement concepts and thought processes that Six Sigma has to offer as it will lead to higher productivity with less chance for error and variation; which in turn will increase efficiency and margins.
The DMAIC approach to process improvement is the framework that I shared with the pharmacy for presenting their plan of action to the FDA. DMAIC stands for Define, Measure, Analyze, Improve and Control. These are the steps for improving a process when a problem or variation has been identified. In the case of this pharmacy they had a contamination issue on the surface of their ISO 5 environment in which they compound. Any contamination that has been isolated from an ISO 5 area should be identified and remediated based on their own SOPs and standards within USP chapter <797> as this is the highest risk area for actually contaminating a preparation. Below is the actual DMAIC response that was sent to the FDA:
Summary of Investigation and Corrective Actions for Environmental Excursions Occurring on September 20, 2016.
DEFINE: USP <797> lists all molds and yeasts as “highly pathogenic mircroorganisms” and requires investigation and corrective action regardless of colony forming unit counts. Environmental excursions (fungi) occurred on (DATE REMOVED), on the surface of the ISO 5 laminar flow hood and on the left fingertip of the compounding pharmacist during routine environmental monitoring (EM). The following goals were identified in this investigation and improvement process:
– The environment where compounded sterile products (CSPs) are produced and personnel should remain in a state of control.
– The CSPs must be safe.
– The process for evaluating a state of control should be designed to protect patients receiving CSPs.
– The EM program should be designed to properly evaluate environmental control.
– Personnel and processes should be designed to maintain environmental control. This includes gowning, cleaning, aseptic technique, and EM technique.
MEASURE: Defect – The cleanroom environment is not in a constant state of control. Opportunity – Each time the cleanroom is used to make CSPs. Unit of Measure – EM results.
ANALYZE: The following areas were evaluated in the investigation and improvement process:
– Storage and staging of products into the cleanroom
o High traffic area.
o Plastic bags used to store and stage materials into cleanroom are stored in corrugated cardboard.
o Bioburden of process not evaluated.
– Gowning process
o Donning of sterile gown is performed after handwashing and prior to donning sterile gloves; bioburden may be transferred to gowning via clean bare hands.
– Environmental monitoring process
o Monthly cleaning occurs after 3rd party validation of cleanroom; it was discovered that routine validation was cancelled and rescheduled on (DATE REMOVED) after a significant amount of testing equipment had been staged into the cleanroom and monthly cleaning did not take place.
o Previous growth had been recovered the week before, on (DATE REMOVED). It is possible that bioburden was on specific testing plates (SabDex plates) that were only used on a weekly basis (DATES REMOVED) and were generally stored in the ante room.
o Monitoring of staged bags does not occur.
o Surface samples in ISO 7 areas on not performed on a weekly basis.
– Cleaning process
o Documentation of weekly and monthly cleaning is not well documented on logs.
o Sporicidal agents are rotated on a monthly basis.
– Primary and Secondary controls
o Validated every 6 months, no current issues.
– Aseptic/compounding process
o Review of media fills and smoke visualization studies performed every 6 months did not reveal deficiencies.
– Patient safety
o Use of USP 797 dating and storage conditions reduces the chance of microbial growth if contamination of a CSP occurred.
o Variances in environmental monitoring procedures do not define the circumstances that would require a product recall.
IMPROVE: The following improvements were made:
– Storage and staging of products into the cleanroom
o Staging area was repositioned to reduce traffic.
o Plastic bags used to store and stage materials into cleanroom were changed to Ziplock bags that had not been exposed to corrugated cardboard.
o Procedure for EM was updated to include a weekly surface sample of a staged bag.
– Gowning process
o Procedure for gowning was updated for operator to don a pair of sterile gloves after handwashing and prior to donning sterile gown (followed by an additional pair of sterile gloves after completion of gowning).
– Environmental monitoring process
o Monthly and weekly cleaning logs were improved to properly show documentation of monthly and weekly cleaning.
o Use of SabDex plates was discontinued, plates will no longer be stored in the cleanroom for long periods of time.
o Monitoring of staged bags shall occur on a weekly basis.
o Surface samples in ISO 7 areas shall be performed on a weekly basis.
– Cleaning process
o Retraining and updating of logs for the documentation of weekly and monthly cleaning occurred, with increased frequency of reviewing the documentation of this cleaning.
o Procedures for cleaning were updated to require that sporicidal agents are rotated on a weekly basis.
– Primary and Secondary controls
o Validated every 6 months, no current issues. Ensure monthly cleaning occurs after validation process, including if a cancellation part way into the process occurs, in order to prevent increased bioburden.
– Aseptic/compounding process
o Continue current policies and procedures.
– Patient safety
o Policy for environmental excursions was updated to require a recall of products compounded on a date when an environmental excursion occurred.
CONTROL PHASE: To be determined if and when sterile compounding resumes.
There are many points to discuss further and I encourage the reader to educate themselves on using these process improvement tools. The main point that I would like to leave you with is that in order to make improvements in any process you must first be collecting data. In the case of preventing and controlling contamination of your clean room that data is in the form of environmental monitoring. However, data collection isn’t enough; using this data to improve using DMAIC is a systematic way for getting your processes under a state of control. Many pharmacies may already be collecting environmental information on their cleanroom but simply see this as a checkbox. There is some real actionable intelligence that can be gleaned from this information and if you’re unsure how to interpret I highly encourage you to reach out to a competent microbiologist who specializes in controlled environments. The certification company you use to classify your rooms may be your first lead for finding a microbiologist.
Generally speaking, beyond environmental monitoring, every critical process most likely has various data points that can be measured to assess the quality of your process. Fortunately with sterile compounding we should be doing testing on our final preparation in the form of sterility, endotoxin and potency. This may be a good starting point for measuring quality within our processes. The key is to break a process down as much as possible to find all of the points that can be measured. Below is a flow chart of an aseptic process that can be found in the appendix of ISO 13408-2008. As an exercise try to find all of the parts of that process that could be measured. As an example, at any point where “sterilization” is found, this can be tested for. The parts of the process where filtration is being performed an integrity test on the filter could be done. During filling operations particle counts and viable data can be collected. When performing visual inspection of the final preparation the number of vials that may have particulate can be accounted for. All of this information can be utilized as an opportunity to improve a process.
If you’re a pharmacist performing high risk compounding it is only a matter of time before the FDA will be at your doorstep; and it doesn’t need to be a frightening situation. While the FDA may seem like they’re in an adversarial posture with the compounding industry at the end of the day their main goal is to protect the public. The FDA has a systematic way for how they operate and we can learn a lot from each other starting with their thought process in how they inspect and correct sterile compounding operations. The Six Sigma tool presented isn’t intended to be used as a way to just get the FDA out of your hair so to speak but rather has great value in truly improving your compounding processes. Rather than looking at the FDA as a pest that is nosing its way into our businesses let’s use this opportunity to learn and truly improve our process for the better. These improvements can not only have lasting effects on the quality of our compounds but there is proof that it will increase your bottom line as well. Where’s the proof? Google: Toyota.