The week that I started working at the FDA, I was asked to review and comment on the draft guidance on stratified sampling, which was still being drafted but not yet made available to public for review. I was very critical of the draft: it was the true embodiment of the old and arcane thinking; would not make Deming proud; was not workable; and worst of all seemed to have been written by a committee more intent on giving validity to old practices that had failed industry. My handwritten comments in the margins bore testimony to my displeasure.
A gentleman with a smiling, welcoming face entered my office. He introduced himself, offered his hand of friendship, and welcomed me to the FDA and to my empty office. On noticing the bloodied document on my desk—the only piece of paper in my office—he asked my views. My vitriolic assessment was offered wholeheartedly and with conviction. As the man left my office, I looked down on the offending document and, to my horror, realized that I had been talking to the man responsible for the draft guidance.
A decade later he is one of my best friends, and in agreement as to the uselessness and non-scientific nature of the draft guidance. So, you may ask, why has the guidance not been withdrawn?
The process for withdrawing the draft guidance is an ongoing effort. And since this is still a draft guidance, the process is not afforded great urgency.
The draft lacks good scientific rigor and is the fruit of the “input and efforts of a committee from industry.” What was the objective and the mission? To challenge Judge Daniel Wolin and the Barr ruling, or to create a shelter for the old guard against the winds of change which have been blowing in the form of the “drug product initiative for the 21st century”?
The scope of the draft guidance promises the following: “Traditional powder blend sampling and testing, in conjunction with testing for uniformity of content in the finished product, can be used to comply with current good manufacturing practice requirements (CGMPs).” Traditional powder sampling in the pharmaceuticals industry is borrowed practice from the agricultural industry. It relies on a sample thief, which, by its nature, introduces variability and produces non-representative samples. The draft guidance promises that combining a bad method for sampling granules when combined with testing a number of samples offers compliance with the GMPs.
At the heart of the GMP’s, the most appropriate requirement for any process is stipulated in 21 CFR 211.110(a). This regulation demands that control procedures “be established to monitor the output and to validate the performance of those manufacturing processes that may be responsible for causing variability in the characteristics of in-process material and the drug product.” A not-too-detailed but careful study of the mechanism and dynamics of withdrawing samples, using a thief, would clearly show that the old practice does not have a place within the drug product quality paradigm of the 21st century.
Sampling of the finished dosages using the USP methods—as stipulated by the USP, never intended as a means of batch release—devoid of any statistical or scientific rigor cannot satisfy the requirement to “monitor the output and to validate the performance of the process.”
Yet the draft guidance is adhered to and followed religiously by many firms as they develop blending and blend control strategies. Tacit reviews have also established the practices contained therein as common practice; an industry bent on copying examples rather than learning from them has established these practices as “current best practice.”
So why are we so adamantly wedded to these archaic thoughts, practices and methodologies? Why are we so comfortable following draft guidance—FDA’s unfinished and incomplete (hence draft) current thinking?
Are the firms wittingly using the practices recommended, and counting on the reviewers unwitting approval?
Clearly following the easiest aspects of the draft guidance has provided industry the means to manufacture and market products. Is it the best science? Are we actually serving consumers by using the worst possible practices? Are we, the consumers, happy with the status quo?
The FDA’s 2011 guidance on Process Validation will eventually address this situation if the FDA will live according to the text of the guidance. The CFR clearly demands understanding and managing the process.
So what are the options? One can wait for the 483 observation, which will eventually come. As the Agency begins to align its practices more and more with science and statistics, then citations will be inevitable.
Another option: Two standards were developed by ASTM specifically to address this issue. ASTM E2709 and E2810. Following either one will address the sample size issue, as well as provide the confidence in the process to produce consistent quality blends and doses. In the future I will write regarding these standards.
In the meantime my plea to the FDA is this: Please withdraw the guidance as a matter of urgency. It is causing conflict between the Field and Review in public; it is misleading industry by offering incorrect advice. It is costing industry, the Agency and consumers.
