Emil W. Ciurczak, DoraMaxx Consulting05.05.16
Why, you might ask, do I always include fiscal thoughts and rules when I am chatting about science? Well, back when I was rebuilding the QC lab at Sandoz in 1983, I was told by a vice president that I wrote excellent capital requests. Why? The other scientists would go on and on about the technical merits of a device; I would explain that the newest sample injection system held 96 instead of 48 samples, doubling the number of samples run overnight. That’s what management wants to hear: what’s in it for the bottom line?
So, economics rule every business; yes, the company may make useful products or give useful service, but they must also make a profit. As a young, idealistic scientist, I originally considered making a profit as possibly something evil, eschewed instrument sales reps, and didn’t read application notes thinking them sales propaganda.
However, as I engaged more with instrument vendors and even was part of two start-up businesses, I came to understand that “profits,” far from being evil are the only way a company can stay in existence. Not the grossly exaggerated profits of a particular Pharma company in the news recently, just good, old-fashioned ROI.
Obviously, I am referring to pharmaceutical manufacturing companies as well as analytical instrument companies. In the case of CMOs, there has been a paradigm of them simply doing a manufacturing process the way it’s always been done. In other words, the initiator usually farms out the process soon after obtaining approval for a NDA or ANDA (new drug application or amended NDA) or, in the case of a proprietary product, immediately before or soon after its patent expires. To save time and expense, the CMO production line closely mirrors the original set-up, in hopes of quickly producing a salable product with little fear of making changes that require validation.
Unfortunately, validating a new site, new equipment, and new operators, even if it mirrors the original, could take every bit as long as moving to different presses, mixers, or even a continuous manufacturing system. Pfizer has begun to construct CM units in shipping containers at their home base, placing the units on flat-bed trucks and delivering them to outsourcing locations. This may work in some cases, but takes planning, infrastructure, and capital, not always available to smaller companies. And, by analogy, transferring and validating an analytical method might take as long as simply moving to a new technology as a replacement. What? Try something not from the 1980s? That makes us sound like NASA, where only computer technology with 10 years’ experience is used in rockets. For example, the space shuttles were equipped with Tandy Shack or Commodore 64 era computers. In other words, your current smart-phones are more powerful than the brain of the multi-million dollar shuttles. So much for the tried and true approach. Keep in mind that liquid chromatography was developed in 1900 by Tswett1, and not “discovered” by Pharma in 1990. Even more to the point, in 1970, the FDA was still not allowing that new-fangled technology of HPLC to be included in NDAs.
What are the top arguments being used for a CMO not developing their own new analytical methods for existing products? I will address each point in parentheses.
The existing method has already been approved by the FDA so why change?
Yes, the existing method has been verified/validated, but, if the CMO is located more than a short distance from the client, then it may likely have a host of different suppliers for chemicals, LC units, columns, solvents. These subtle and sometimes not so subtle differences will make method transfer just hard enough that the CMO may take just as long to adapt a new method as verify the transfer, when all new substances including API reference material are being used.
In reality, by the time all the subtle and not-so-subtle changes/tweaks are made to the submitted method, it is, for all intents and purposes, a new method. The validation SOP/procedure/documentation will be no less stringent for an entirely new method then for the heavily modified original method.
Also, keep in mind that every analytical method had to be introduced to the FDA for the first time. For example, HPLC was not accepted as a primary method in 1970.
We don’t have the hardware/software for any new technology
As your business expands—more clients, more batches, more buildings—more equipment will be required, in any case. So, why not make both small and major changes up-front? The Netherlands built Schiphol Airport twice as large as was needed at the time, assuming future inflation; when the space was needed, the cost savings of having done construction a decade earlier were significant. A single UPLC (ultra-high pressure/performance) unit can easily perform the same number of assays as several HPLC units, often in less time. The savings in solvents, space, columns, and time more than makes up for doing the same-old HPLC.
We don’t have the experience/scientists to perform the new technology
If a CRO/CMO/generic company is serious about continuing to flourish, it will have to grow. Growing, by definition, means adding headcount. If a company is hiring, why not hire scientists familiar with both classic method development and are conversant with newer technologies? As more and newer products come off patent, they will begin to appear with newer, more modern release technologies. It will become more and more necessary to be conversant with more than GC and LC, or you will begin to be passed over as sites for production.
Fortunately, many larger Pharma companies are reducing their staffs. It would be smart to look into acquiring these analysts for your facility. Careful recruiting will enable you to find trained analysts, knowledgeable in the technology you are seeking to add to your labs. Discuss potential analysts with instrument manufacturers; they might be able to give some idea of who they feel is adept at using their instruments, helping you interviewing process. If you do not already have the skills, it means no one in your organization can assess a candidate’s skill set.
It will take too long to adapt to the new technology
Considering that adding scientist/analysts will imply training and testing with either existing or new instruments. If you hire experienced scientists/technicians in NIR, Raman, etc., you will be halfway there. When you are comfortable with your headcount, the experienced person can have a dual role: develop methods for that particular instrumentation, running the analyses, as well; train other analysts in the running of these technologies.
NIR, for instance, is notably simple to operate, while development of methods is where the expertise is required. As such, a single, well-versed analyst can carry the development/validation load, while second, third, etc. analyst(s) become familiar enough with the software to run routine analyses. All GMP/21part11 compliant software contains signature safeguards, allowing a less experienced operator to be blocked from changing methods or data.
We would have to get approval from the governing agency(s)
Considering the number of permutations in the original method, mandated by input variables (see point 1), the approval time for getting approval from any agency would scarcely be larger than for a heavily modified older method. If, for example, the client/originator can analyze several lots, then send samples quickly to the remote site for analysis by the new method. A paired t-test would indicate the new method was not different—most statistics are not used to prove positive correlation—and lead to replacement with the new technology/method.
If Raman or NIR is used on intact tablets/capsules, they could first be run by the new method and the exact samples re-run via the usually destructive reference method, achieving an even better correlation.
In addition, since the first PAT Guidance, the USFDA has been encouraging all Pharma companies to move to newer technologies for monitoring the process, not just analyzing the end product.
We might lose customers if we don’t do things exactly as the client does
The U.S. FDA’s recent communications are encouraging not only new products to be prepared by continuous manufacturing (CM), but suggest that existing products be converted from step-by-step batch processing to a continuous mode. In light of that attitude, we don’t believe there will be any problems adapting technologies used in CM. In light of the industry trend to CM, it would be fatal to not update the available monitoring equipment.
In the very near future, I would not be surprised to find that client companies will begin shopping around for CMOs that are using cutting-edge for their newer products and converting older, existing not-yet-off-patent products to CM methodology. Yes, you might be passed over by older, smaller, traditional clients, but you will be in a more competitive position to attract the more progressive clients who are moving to PAT, QbD, and CM for new and existing products. In fact, they would be encouraged to do so, if only they knew that such products could successfully be transferred to a CMO, at some point.
References
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, economics rule every business; yes, the company may make useful products or give useful service, but they must also make a profit. As a young, idealistic scientist, I originally considered making a profit as possibly something evil, eschewed instrument sales reps, and didn’t read application notes thinking them sales propaganda.
However, as I engaged more with instrument vendors and even was part of two start-up businesses, I came to understand that “profits,” far from being evil are the only way a company can stay in existence. Not the grossly exaggerated profits of a particular Pharma company in the news recently, just good, old-fashioned ROI.
Obviously, I am referring to pharmaceutical manufacturing companies as well as analytical instrument companies. In the case of CMOs, there has been a paradigm of them simply doing a manufacturing process the way it’s always been done. In other words, the initiator usually farms out the process soon after obtaining approval for a NDA or ANDA (new drug application or amended NDA) or, in the case of a proprietary product, immediately before or soon after its patent expires. To save time and expense, the CMO production line closely mirrors the original set-up, in hopes of quickly producing a salable product with little fear of making changes that require validation.
Unfortunately, validating a new site, new equipment, and new operators, even if it mirrors the original, could take every bit as long as moving to different presses, mixers, or even a continuous manufacturing system. Pfizer has begun to construct CM units in shipping containers at their home base, placing the units on flat-bed trucks and delivering them to outsourcing locations. This may work in some cases, but takes planning, infrastructure, and capital, not always available to smaller companies. And, by analogy, transferring and validating an analytical method might take as long as simply moving to a new technology as a replacement. What? Try something not from the 1980s? That makes us sound like NASA, where only computer technology with 10 years’ experience is used in rockets. For example, the space shuttles were equipped with Tandy Shack or Commodore 64 era computers. In other words, your current smart-phones are more powerful than the brain of the multi-million dollar shuttles. So much for the tried and true approach. Keep in mind that liquid chromatography was developed in 1900 by Tswett1, and not “discovered” by Pharma in 1990. Even more to the point, in 1970, the FDA was still not allowing that new-fangled technology of HPLC to be included in NDAs.
What are the top arguments being used for a CMO not developing their own new analytical methods for existing products? I will address each point in parentheses.
The existing method has already been approved by the FDA so why change?
Yes, the existing method has been verified/validated, but, if the CMO is located more than a short distance from the client, then it may likely have a host of different suppliers for chemicals, LC units, columns, solvents. These subtle and sometimes not so subtle differences will make method transfer just hard enough that the CMO may take just as long to adapt a new method as verify the transfer, when all new substances including API reference material are being used.
In reality, by the time all the subtle and not-so-subtle changes/tweaks are made to the submitted method, it is, for all intents and purposes, a new method. The validation SOP/procedure/documentation will be no less stringent for an entirely new method then for the heavily modified original method.
Also, keep in mind that every analytical method had to be introduced to the FDA for the first time. For example, HPLC was not accepted as a primary method in 1970.
We don’t have the hardware/software for any new technology
As your business expands—more clients, more batches, more buildings—more equipment will be required, in any case. So, why not make both small and major changes up-front? The Netherlands built Schiphol Airport twice as large as was needed at the time, assuming future inflation; when the space was needed, the cost savings of having done construction a decade earlier were significant. A single UPLC (ultra-high pressure/performance) unit can easily perform the same number of assays as several HPLC units, often in less time. The savings in solvents, space, columns, and time more than makes up for doing the same-old HPLC.
We don’t have the experience/scientists to perform the new technology
If a CRO/CMO/generic company is serious about continuing to flourish, it will have to grow. Growing, by definition, means adding headcount. If a company is hiring, why not hire scientists familiar with both classic method development and are conversant with newer technologies? As more and newer products come off patent, they will begin to appear with newer, more modern release technologies. It will become more and more necessary to be conversant with more than GC and LC, or you will begin to be passed over as sites for production.
Fortunately, many larger Pharma companies are reducing their staffs. It would be smart to look into acquiring these analysts for your facility. Careful recruiting will enable you to find trained analysts, knowledgeable in the technology you are seeking to add to your labs. Discuss potential analysts with instrument manufacturers; they might be able to give some idea of who they feel is adept at using their instruments, helping you interviewing process. If you do not already have the skills, it means no one in your organization can assess a candidate’s skill set.
It will take too long to adapt to the new technology
Considering that adding scientist/analysts will imply training and testing with either existing or new instruments. If you hire experienced scientists/technicians in NIR, Raman, etc., you will be halfway there. When you are comfortable with your headcount, the experienced person can have a dual role: develop methods for that particular instrumentation, running the analyses, as well; train other analysts in the running of these technologies.
NIR, for instance, is notably simple to operate, while development of methods is where the expertise is required. As such, a single, well-versed analyst can carry the development/validation load, while second, third, etc. analyst(s) become familiar enough with the software to run routine analyses. All GMP/21part11 compliant software contains signature safeguards, allowing a less experienced operator to be blocked from changing methods or data.
We would have to get approval from the governing agency(s)
Considering the number of permutations in the original method, mandated by input variables (see point 1), the approval time for getting approval from any agency would scarcely be larger than for a heavily modified older method. If, for example, the client/originator can analyze several lots, then send samples quickly to the remote site for analysis by the new method. A paired t-test would indicate the new method was not different—most statistics are not used to prove positive correlation—and lead to replacement with the new technology/method.
If Raman or NIR is used on intact tablets/capsules, they could first be run by the new method and the exact samples re-run via the usually destructive reference method, achieving an even better correlation.
In addition, since the first PAT Guidance, the USFDA has been encouraging all Pharma companies to move to newer technologies for monitoring the process, not just analyzing the end product.
We might lose customers if we don’t do things exactly as the client does
The U.S. FDA’s recent communications are encouraging not only new products to be prepared by continuous manufacturing (CM), but suggest that existing products be converted from step-by-step batch processing to a continuous mode. In light of that attitude, we don’t believe there will be any problems adapting technologies used in CM. In light of the industry trend to CM, it would be fatal to not update the available monitoring equipment.
In the very near future, I would not be surprised to find that client companies will begin shopping around for CMOs that are using cutting-edge for their newer products and converting older, existing not-yet-off-patent products to CM methodology. Yes, you might be passed over by older, smaller, traditional clients, but you will be in a more competitive position to attract the more progressive clients who are moving to PAT, QbD, and CM for new and existing products. In fact, they would be encouraged to do so, if only they knew that such products could successfully be transferred to a CMO, at some point.
References
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.