“In your response, you state that you have informed your clients on the importance of validating the methods, but they have chosen not to validate the methods. In addition, you state that you will inform them again in writing. Your response, however, is inadequate because you do not provide your firm’s planned corrective actions for this CGMP violation. You are responsible for ensuring that the test methods used by your firm are validated.” (emphasis added)If there was any doubt about who is ultimately responsible for method validation, this letter seems to clarify it very succinctly. If you are running the method, you are responsible for assuring that it is validated. It doesn’t matter what your client wants. It doesn’t matter what they are willing to support. It doesn’t matter what they are willing to pay for, or how you complete the work. You must validate the methods that are being run by your laboratory.
The situation sounds simple enough, but it is far from easy to resolve. Method validation is time consuming, takes significant resources, is technically demanding, and receives a great deal of scrutiny from global regulators. The ICH Q2(R1) guideline gives guidance on how to accomplish the validation of analytical procedures in order to demonstrate suitability for their intended use. The sponsor of the application is ostensibly the owner of the control strategy, and determines what the intended use of the method will be. It is usually presumed that the client is the subject matter expert with regards to its product, and therefore it should have oversight of the ultimate disposition of the material. The contract laboratory is providing a defined service for the client, with specific, but limited, technical expertise. Ideally, this combined responsibility should result in an analytical control strategy that is well defined and well controlled, as run under CGMPs. The specifics of this shared effort should be spelled out in the Quality Agreement, and both parties should understand their responsibilities and accountability.
It is not always the case, however, that the laboratory and the client agree on the specifics. The client may have a very different expectation of the control of analytical methods than the testing lab. Should all raw material testing methods be validated, even for Phase I clinical production? What is appropriate at each phase of development, and how does the testing lab assure that its data is not being applied incorrectly after the fact? If the agency is holding the lab responsible, how do they assert that responsibility?
Biopharma Sponsor Perspective
I do not foresee a time when most innovative biopharma firms would willfully abdicate responsibility for their product control strategy. I do know that expectations are continuously evolving, and late-lifecycle products are often supported with aging analytical methods. Investment in re-validating, or otherwise modernizing, a product dossier is carefully measured, and often controlled to minimize cost. The situation with generic firms, and perhaps biosimilar firms of the future, may be slightly different. Costs become a primary concern, and method and process validation are very expensive and time-consuming. There is inevitable pressure to minimize the cost of validation. If the quality unit does not insist on a complete validation package, the business unit could very well assume some risk. In fact, based on the Warning Letter, they can now pass this risk directly to their contract lab, which then will be held responsible for the validation of the methods run in their laboratories. The client may now have little incentive to support a more thorough and in-depth method development and validation plan.
For the contract lab, the situation presents a real dilemma. If you insist that a validation package is insufficient and your client disagrees, your choices are significantly narrowed. You can refuse the project, leading to loss of current and potential revenue. You can go forward at risk, realizing now that your cGMP compliance will possibly be called into question. Or you can remediate the validation yourself, probably at your own expense, and with little support from your client.
Method validation is difficult enough, even with client support. The availability and suitability of reference standards, determining appropriate acceptance criteria, performing many replicates, and the review and approval of documentation takes a team of individuals, including those who have an in-depth knowledge of the product and methods. Further, this requires effort from both technical staff and within the quality unit. It is not feasible to believe that this expertise lies solely in the contract lab, unless there is significant experience already on hand. A poorly designed and executed validation will not be very likely to pass scrutiny from today’s inspectors, and that leaves the business even more vulnerable.
The Paradigm Shift
I think this Warning Letter should be a wake-up call to the quality unit in every contract lab. It is clear that the FDA expects you to retain accountability for all aspects of cGMP compliance. If there is a disagreement with a client, you must determine the acceptable practice within your organization, and hold the line. It is incumbent upon the service providers to be clear in setting expectations; clients that are unwilling or unable to support this level of commitment may not be the best clients to have. Anything less will be put the business at increased risk, and you will end up being well short of meeting the expectations as indicated in the Warning Letter.
Morrey Atkinson is the Chief Scientific Officer at Cook Pharmica, in Bloomington, IN, a full-service CDMO for biological drug substance and drug product development and manufacturing. Prior to this he spent nine years at Eli Lilly & Co., as both Head of Biotechnology Manufacturing Sciences and Technology in Kinsale, Ireland and Director of Bioprocess R&D in Indianapolis, IN. He can be reached at email@example.com.