Emil W. Ciurczak, DoraMaxx Consulting09.06.17
Current Good Manufacturing Practices (cGMPs) are, in short, at the heart of every move we make in pharmaceuticals. Quite literally, every step from simple notebook record keeping through final release of the product is overseen by GMPs. Since the majority of people in the industry do not remember a time without them, much as millennials can’t grasp a world without smart phones or cable TV, we tend to assume they’ve always been in force. In fact, GMPs are relatively new.
So, from my perspective, the updated, comprehensive edition of 1978, cGMPs only became an issue eight years after I became a full-time worker in the industry. As I am wont to point out that, by the time we were told by the U.S. FDA how to properly conduct research, we had already walked on the moon, had color TVs, were routinely using polio vaccines, penicillin, streptomycin, had performed organ transplants, and were using computers. As you might imagine, a number of older scientists considered them intrusive.
But are they all that bad? Well, much as a razor blade: are you getting a clean shave or a nasty nick on the chin. cGMPs are based on the initial contract of the FDA with industry: make the product safe and effective. Sounds straightforward, no? However, consider another set of “simple” rules: The Ten Commandments. You would think that “don’t steal” or “don’t murder” would be self-explanatory, wouldn’t you? Nonetheless, we have tens of thousands of priests, rabbis, mullahs, and ministers making a living, telling us what these “simple” words mean.
Likewise, the simple command, “Make the product safe and effective,” has spawned tens of thousands of SOPS, guidances, guidelines, consent decrees, observations, and such. Never mind that we have a massive bureaucracy centered around Washington, DC to “help” us understand what “good” means. Please, do not misunderstand. Had our “distinguished” industry not killed so many people, there wouldn’t have been a need for the FDA. (Fun fact: the initial Pure Food & Drug Act only specified that commercial pharmaceuticals could not hurt people. Only a later edition of the Act suggested that the product should also work as advertised.)
So many of cGMPs are logical, posting them makes as much sense as telling food handlers to wash their hands after using the restroom: 1. Permanent notebooks must be kept in pen, with an index; cross-outs must be a single line with an explanation of why an entry was changed; all pages should be dated and witnessed; no pages may be left blank, etc. All things we learned in freshman science class; 2. No cross-contamination; 3. Proper venting must be utilized; and 4. Labels must contain proper (pre-approved) information.
However, there are useful rules that are not always taught in beginning classes. These are important, but restrictive, so companies do them grudgingly. For example:
From the point of view of rapidly gaining approval from the Agency, a set of templates—in any given company, most dosage forms have similar DNA for formulation purposes—allows for rapid development of a new product: merely insert a NCE (new chemical entity) into one of the proven tablet, caplet, or capsule templates and crank out the product. We can change the color, coat or not coat, emboss or imprint, but whatever we generate, it will not be too different from previous products. The physical parameters of the new dosage form can almost be written before actual production +/- a slight variation.
With all that being true, when the FDA—fueled by Ajaz Hussain—held a massive workshop/information gathering event in early 2002 to encourage modernization and better control of the finished product through control of the process, a number of “traditionalists” balked at the radical changes. Perhaps the one phrase that chaffed the Quality Assurance people was “using their best judgement” in lieu of “written in stone” SOPs. The draft Guidance took much longer than “typical” Guidances to be approved. The document drew up, until that time, a record number of comments from industry.
Keep in mind that two “products” emerge from production for every lot produced: the tablets/capsules that are sent to the hospital/pharmacy for distribution to patients; and the MMF (master manufacturing formula) with all the weights and signatures for the batch.
Now, the former could have an absolutely wonderful distribution of API, excellent dissolution, color, weights, etc., but, if there is a tiny error—missed weight, signature, set of initials—on the MMF, and the batch “fails.” Conversely, should the analytical results show a terrible lot, needing to be destroyed, but every line on the MMF is filled in, QA considers the batch a success.
In the face of this mindset, we are encouraging manufacturers to implement PAT, with monitors/controllers at each step. As the PAT Guidance states, “each batch may be considered a validation batch.” Not three lots and chisel the batch instructions in stone, but perform each step until the pre-determined quality is achieved. E.g., we are to mix until blended, not for a set time.
How, you might ask, can we justify PAT and QbD under the law—cGMPs are the law. Well, to quote/paraphrase the GMPs, there are two requirements that are begging for more measurements and control:
Emil W. Ciurczak
DoraMaxx Consulting
Emil W. Ciurczak has worked in the pharmaceutical industry since 1970 for companies that include Ciba-Geigy, Sandoz, Berlex, Merck, and Purdue Pharma, where he specialized in performing method development on most types of analytical equipment. In 1983, he introduced NIR spectroscopy to pharmaceutical applications, and is generally credited as one of the first to use process analytical technologies (PAT) in drug manufacturing and development.
So, from my perspective, the updated, comprehensive edition of 1978, cGMPs only became an issue eight years after I became a full-time worker in the industry. As I am wont to point out that, by the time we were told by the U.S. FDA how to properly conduct research, we had already walked on the moon, had color TVs, were routinely using polio vaccines, penicillin, streptomycin, had performed organ transplants, and were using computers. As you might imagine, a number of older scientists considered them intrusive.
But are they all that bad? Well, much as a razor blade: are you getting a clean shave or a nasty nick on the chin. cGMPs are based on the initial contract of the FDA with industry: make the product safe and effective. Sounds straightforward, no? However, consider another set of “simple” rules: The Ten Commandments. You would think that “don’t steal” or “don’t murder” would be self-explanatory, wouldn’t you? Nonetheless, we have tens of thousands of priests, rabbis, mullahs, and ministers making a living, telling us what these “simple” words mean.
Likewise, the simple command, “Make the product safe and effective,” has spawned tens of thousands of SOPS, guidances, guidelines, consent decrees, observations, and such. Never mind that we have a massive bureaucracy centered around Washington, DC to “help” us understand what “good” means. Please, do not misunderstand. Had our “distinguished” industry not killed so many people, there wouldn’t have been a need for the FDA. (Fun fact: the initial Pure Food & Drug Act only specified that commercial pharmaceuticals could not hurt people. Only a later edition of the Act suggested that the product should also work as advertised.)
So many of cGMPs are logical, posting them makes as much sense as telling food handlers to wash their hands after using the restroom: 1. Permanent notebooks must be kept in pen, with an index; cross-outs must be a single line with an explanation of why an entry was changed; all pages should be dated and witnessed; no pages may be left blank, etc. All things we learned in freshman science class; 2. No cross-contamination; 3. Proper venting must be utilized; and 4. Labels must contain proper (pre-approved) information.
However, there are useful rules that are not always taught in beginning classes. These are important, but restrictive, so companies do them grudgingly. For example:
- Stability studies: a) downside: time-consuming and personnel-intensive; b) important: critical for many drugs, since some degradation products could be toxic; c) mixed review: some drugs can last longer than the sun and stability tests with outdating as a formality, costing money passed on to the patient.
- Activity vs. toxicity: a) necessary: the mode of action and dosage size is critical to know and any potential toxicity needs to be documented; b) overkill: every chemical has an LD50 (a “lethal dose” level, where 50% of test subjects die), even table salt and tap water. So, having to list any side-effect even if none have proven to be a direct result of the drug substance, invites both expensive lawsuits and generates fear of the drug, causing non-compliance on the part of a patient.
- Clinical studies: a) necessary: we need to know about activity (bioavailability), elimination, modes of action, and possible toxicity; b. negative aspects: most trials tend to use young, white males. Thus, an awful lot of new drugs are not tested on people with the actual illness or geriatric, female, or pediatric trial subjects. Also, long-term effects cannot be studies over the short-term of the study; the Agency depends on feedback from patients over time.
From the point of view of rapidly gaining approval from the Agency, a set of templates—in any given company, most dosage forms have similar DNA for formulation purposes—allows for rapid development of a new product: merely insert a NCE (new chemical entity) into one of the proven tablet, caplet, or capsule templates and crank out the product. We can change the color, coat or not coat, emboss or imprint, but whatever we generate, it will not be too different from previous products. The physical parameters of the new dosage form can almost be written before actual production +/- a slight variation.
With all that being true, when the FDA—fueled by Ajaz Hussain—held a massive workshop/information gathering event in early 2002 to encourage modernization and better control of the finished product through control of the process, a number of “traditionalists” balked at the radical changes. Perhaps the one phrase that chaffed the Quality Assurance people was “using their best judgement” in lieu of “written in stone” SOPs. The draft Guidance took much longer than “typical” Guidances to be approved. The document drew up, until that time, a record number of comments from industry.
Keep in mind that two “products” emerge from production for every lot produced: the tablets/capsules that are sent to the hospital/pharmacy for distribution to patients; and the MMF (master manufacturing formula) with all the weights and signatures for the batch.
Now, the former could have an absolutely wonderful distribution of API, excellent dissolution, color, weights, etc., but, if there is a tiny error—missed weight, signature, set of initials—on the MMF, and the batch “fails.” Conversely, should the analytical results show a terrible lot, needing to be destroyed, but every line on the MMF is filled in, QA considers the batch a success.
In the face of this mindset, we are encouraging manufacturers to implement PAT, with monitors/controllers at each step. As the PAT Guidance states, “each batch may be considered a validation batch.” Not three lots and chisel the batch instructions in stone, but perform each step until the pre-determined quality is achieved. E.g., we are to mix until blended, not for a set time.
How, you might ask, can we justify PAT and QbD under the law—cGMPs are the law. Well, to quote/paraphrase the GMPs, there are two requirements that are begging for more measurements and control:
- “Meaningful in-process tests shall be performed.” Well, in 1978, the best we had were friability, hardness, weight, disintegration, and appearance. All destructive and time consuming… and not actually controlling the process, since, by the time results are gleaned, many thousands of dosage units have been produced. Therefore, not really a control strategy.
- “A statistically significant number of units must be tested from every lot.” Now, I don’t know where you learned your statistics, but at my school, 20 tablets from a 1, 2, 5, or 10-million tablet lot is not to be considered significant, no matter how you trust your operators.
Emil W. Ciurczak
DoraMaxx Consulting
Emil W. Ciurczak has worked in the pharmaceutical industry since 1970 for companies that include Ciba-Geigy, Sandoz, Berlex, Merck, and Purdue Pharma, where he specialized in performing method development on most types of analytical equipment. In 1983, he introduced NIR spectroscopy to pharmaceutical applications, and is generally credited as one of the first to use process analytical technologies (PAT) in drug manufacturing and development.