Genzyme Warning Letter Update

I requested the Establishment Inspection Report (EIR) from the FDA, but I only received the Form FDA 483 with a letter that states,

“In order to help reduce processing time and costs, certain material has been deleted from the records furnished to you because preliminary review of the records indicated that the deleted information is not required to be publicly disclosed.”

FDA’s Field Management Directive 145 indicates EIRs are routinely released after an inspection is considered closed. Giving their response to my FOI request, I have to assume they don’t consider this inspection closed. This is further support by Genzyme’s 5/21/09 news release that states, “The FDA has begun the inspection process at the plant to determine if the deficiencies cited in its warning letter have been satisfactorily addressed. When the FDA issues its formal findings, Genzyme will provide an update.” (http://www.genzyme.com/corp/investors/GENZ%20PR-052109.asp)

The FDA 483 indicates that the inspection was lead by ORA and included CDER product specialist, which may account for the coordinated response (CR and Warning Letters delivered together).

My initial reaction to this coordinated response was surprise. I do not frequently see Warning Letters being issued as part of a Pre-Approval Inspection (PAI). In addition, the follow-up inspection to verify corrective actions prior to approval seems atypical.

It appears the FDA is stepping up their enforcement policies, which isn’t necessarily a bad thing. However, I am not sure about the severity of the action in this case, given there was no indication that adulterated product was found. It seems to me that the FDA should have completed an appropriate risk analysis before taking this position (see my blog posting, 5/20/09).

I still need to read the EIR before I can make any solid conclusions, but it looks like I will have to wait until the inspection is considered closed before I can get my hands on it.

In my previous GMP Notes, I neglected to include the link to Genzyme’s Feb 09 warning letter.

http://www.fda.gov/foi/warning_letters/s7123c.htm

Aseptic Processing: The Holy Grail?

The recent Genzyme warning letter depicts the focus and severity that regulatory bodies place on aseptic processing, but what are the real risks?

Is aseptic processing the greatest risk associate with sterile products, as is often implied by the governments?

The answer: it depends.

The risk associated with aseptic processing can only be properly assessed by looking at the big picture and should not be calculated strictly from a microbial perspective.

Although certain products are required to be sterile, we do not always manufacture them through sterile processing. Sometimes we use aseptic processing due to the sensitivity of the product and the near impossibility of keeping all microbes out of the processing stream. The definition for sterile is: “free from living germs or microorganisms”; while the definition for aseptic is: “free from the living germs of disease, fermentation, or putrefaction.” The subtle distinction between the two is: sterile means no bugs; aseptic means no potentially harmful bugs.

The definition of aseptic processing concedes that we can have bugs in our product. With this said, it is fair to conclude that everyday sterile units are leaving a manufacturing site somewhere in the world in a less than sterile condition. Oh my, isn’t this dangerous and aren’t we going to kill some? Again, it depends.

Even though certain body systems, areas, and/or organs are designed to be sterile (i.e., brain, blood, lungs), microbial infiltration into these areas does occur. As a defense to these intrusions, the body’s immune system leaps into action.

Consider the heroin junky in a back alley with a used syringe and tarnished spoon. He heats up the spoon to melt the heroin, uptakes the liquid heroin into the syringe, and promptly injects it into his veins. I think we can all agree that this process is not aseptic let alone sterile.

So, why aren’t all back-alley heroin addicts dying of septic shock? The answer is likely that the injection is small volume and their immune system is functioning as it was designed. It is reasonable to assume that junkies are routinely injecting themselves with contaminated product, but the injected microbes are not serious pathogens and/or their immune system is neutralizing the invaders before they become a serious issue. If these junkies were injecting large volumes into their blood stream, it is possible for their immune system to become overwhelmed and result in a hazardous outcome.

With this said, we can conclude that aseptic processing is not designed to completely eliminate risks associated with administering sterile products and some patients are receiving sterile products that are less than sterile.

So, what are the real risks? Is the real risk that we give someone an adulterated product (less than sterile) or that we give someone a product that overwhelms their natural defenses and causes them harm?

The real risks associated with aseptic processing are dependent on the product, route of administration, dose volume, dose frequency, patient’s immune system, and specific contaminating microorganism. Products that foster microbial growth and/or can not be fully subjected to a sterilization process carry greater risks (i.e., blood products); Intravenous injections carry greater risks than Intramuscular; Large volume parenterals carry greater risks than small volume parenterals; Multiple dose products carry greater risks than single dose products; Patients with compromised immune systems are at greater risks; and Product contamination with a pathogen carry greater risks than normal flora contaminations.

Obviously, our risks are lower when we have fewer microbes slipping through our processing cracks and therefore, putting less reliance on a body’s immune system. We should strive to control our aseptic processes as tightly as feasible, but we don’t need to over react when we find a crack in the process or a point-of-control deviation. When we encounter these situations, we need to do what 21st Century GMP and ICH guidelines tell us to do, complete a complete risk assessment by evaluating the type of contamination, extent of the contamination, and the product’s type, administration, volume, frequency, and target patient.

The principles of risk assessment require the specifics of a given situation to be assessed and the potential to cause harm to be assigned. When we treat aseptic processing as the “Holy Grail” of the medical manufacturing industry, we completely ignore these principles.

In conclusion, aseptic processing risks should be evaluated on a case-by-case basis and should encompass the complete picture and not just the microbiological perspective. We must recognize that microbial contamination are likely occurring everyday without detection and/or harm. Equally, we must acknowledge microbial contamination in sterile products have caused great harm and even killed people. We simply should not treat all situations the same by either over reacting or under reacting. With sterile products, the stakes can be high, but we shouldn’t automatically assume the worst-case outcome without completing a comprehensive investigation and an appropriate risk assessment.

Stop the Madness

Is your Quality System inundated and overwhelmed with investigations, documentation, and bureaucracy?

Does it seem that you spend your whole day jumping from one fire to another?

The best way to render a Quality System ineffective is to treat every issue equally.

When all things are treated equal, the Quality System can become ineffective at addressing critical issues because it is struggling to stay afloat in an ocean of documentation and investigations.

When everything is elevated to crisis level, people stop paying attention to anything. If the fire alarm in your office goes off everyday when there is no fire, then you start ignoring the alarm. The same applies to Quality Systems. It is very important for an effective Quality System to readily identify and separate critical issues from mere inconveniences. Inconveniences need to be addressed, but they don’t need to be addressed in the same manner and method as a potential product impacting issue.

Consider the triage process for medical units. They must quickly decide on which patients are in critical condition and which patients can wait for their care. When done correctly, all patients receive the appropriate attention with the appropriate urgency. Non-critical patients are not neglected nor are they treated before critical patients.

Quality Systems need to be monitored to evaluate their effectiveness at identifying and addressing critical issues promptly without neglecting inconveniences all together.

Of course, the best case scenario is to prevent non-conformance events from happening in the first place. We need to stop simply reacting to issues and inconveniences and start looking for better ways to prevent their future occurrence.

Over time, the Quality System and manufacturing process should be fine tuned to eliminate the background noise (inconveniences) so critical issues can be clearly identified and investigated in a timely manner.