Disclaimer

All opinions are my own.

Tuesday, November 28, 2017

Innovation In Pharmaceuticals: What Would It Take & Who is Responsible?

Humans from time immemorial through their imagination, creativity and different teachings have innovated in every aspect of life and things they have done. All of this should have happened for every aspect of pharmaceutical manufacturing also. However reading FDA’s PAT guidance (1) [excerpts shared below] one gets the regulatory perspective and their snap shot that suggests pharmaceutical manufacturing is trailing in innovation.

I have difficulties accepting their conjecture. We need to understand the basis and come up with a potential solution to remedy regulatory supposition. It will not be easy but we need to start a discussion. Sides will be taken and that would be healthy.

Unfortunately, the pharmaceutical industry generally has been hesitant to introduce innovative systems into the manufacturing sector for a number of reasons. One reason often cited is regulatory uncertainty, which may result from the perception that our existing regulatory system is rigid and unfavorable to the introduction of innovative systems. For example, many manufacturing procedures are treated as being frozen and many process changes are managed through regulatory submissions. In addition, other scientific and technical issues have been raised as possible reasons for this hesitancy. Nonetheless, industry's hesitancy to broadly embrace innovation in pharmaceutical manufacturing is undesirable from a public health perspective. Efficient pharmaceutical manufacturing is a critical part of an effective U.S. health care system. The health of our citizens (and animals in their care) depends on the availability of safe, effective, and affordable medicines.

…. pharmaceutical manufacturing will need to employ innovation, cutting edge scientific and engineering knowledge, along with the best principles of quality management ….

Nonetheless, industry's hesitancy to broadly embrace innovation in pharmaceutical manufacturing is undesirable from a public health perspective.......

Manufacturers are encouraged to use the latest scientific advances in pharmaceutical manufacturing and technology


1.     Multivariate tools for design, data acquisition and analysis
2.     Process analyzers
3.     Process control tools
4.     Continuous improvement and knowledge management tools

Like medical doctors who take a Hippocratic Oath, chemists and chemical engineers a take an “invisible” oath and apply fundamentals of chemistry, chemical engineering, their imagination and creativity to best of their ability to develop, design, justify and commercialize pharmaceutical manufacturing processes that are ecofriendly and produce quality products first time and all the time. They are graduates of the same schools that have sent man on the moon and back and sent Voyager and Cassini. Thus we should not doubt their capabilities. Human instinct of continuous improvement always sets in after the processes have been successfully commercialized.

There is a latent conjecture in regulatory guidance that the chemists and chemical engineers have not done the best to create the existing processes. I have difficulty accepting regulatory perspective that pharma needs to innovate. Something is a miss some place.

Any layperson reading guidance excerpts would think that the API (Active Pharmacceutical Ingredients) and their formulation processes were designed by seat of the pants rather than by incorporating fundamentals of chemistry and chemical engineering. I hope that was never the case and is not the case even today.

Could it be that the regulations are standing in the way of continued pharmaceutical manufacturing technology improvement and innovation? PAT (1) guidance has few self-confessions in this regard when it suggests, “our existing regulatory system is rigid and unfavorable to the introduction of innovative systems.” Second clue from the guidance is “many manufacturing procedures are treated as being frozen and many process changes are managed through regulatory submissions”

There are few other clues, at least to me. 


1.     Regulators don’t trust company’s continuous process improvements and innovation for the existing processes on the belief that every improvement would change the product performance and its efficacy. Thus, the process has to be re-examined for product quality and performance. This admission could be a challenge.

2.     Many times chemists and chemical engineers see potential improvements only after the process has been commercial. In-house time and effort needed to prove equivalence, even if not necessary, along with paperwork needed to assure same product performance is either too expensive, cumbersome and not worth the effort as it takes too long to get the necessary regulatory approval. 

Suggestion of use of certain tools and methods mentioned in guidance is bothersome because without the use of these the basic design of a manufacturing process cannot be started. Thus, I am not sure of the rationale of the suggestion in the guidance.

Chemists and chemical engineers have the aspiration and take pride in designing processes that produces quality product from the get go. They have done this for as far back as time can tell. Quality by repeated analysis (QbA) (aggravation) has disastrous financial impact on the whole business process. One bad move can have catastrophic domino effect. It is Economics 101. Since it is being done in pharmaceutical manufacturing, it suggests that there are extraneous constraints forcing this scenario. If the business is highly profitable all ignored.    

Innovation in pharmaceutical manufacturing can only happen when the technocrats are given the freedom to innovate. However a qualification has to be attached to process improvements and that is the product quality and performance will not deviate from the approved product. With this freedom there should be string attached and that is if the performance is different from the approved product, company has to abandon the process change and if they don’t, the production has to be shut down. Only a financial constraint will be a deterrent to unscrupulous process innovation adventures. 

Reality is that the systems in place today are not conducive to pharmaceutical innovation once the processes have been commercialized. To change a process step, change has to be submitted for every product and can face the approval ordeal. Due to limited patent life brand products may not have the patent life for the change to come alive. Generics just do the best they can but then shy away for additional innovations for approval expense and delays.

If we want to have continuous innovation in pharmaceutical manufacturing then we need to change the current system. To change the current system regulatory bodies and the industry have to tango together. There has to be mutual trust from each side. Pharmaceutical manufacturing and regulatory product quality landscape will have to be re-sculptured.

Since better than 90% of the pharmaceuticals are produced by batch process, industry has become dependent on taking a sample almost after every step and check it for quality. Quality by analysis (aggravation) has become the norm even if the sample taken meets the specs. This has essentially established a culture and a business model that is very different from other manufacturing enterprises. Unlike other industries in pharma we see low inventory turns of raw materials, finished goods and intermediates. They require storage space and lead to poor use of assets. Whole business process is inefficient and cumbersome. Brigades vs. battalions are needed. Processes that are designed to produce quality product from the get go become victim of quality by aggravation process. Current business model has basically destroyed planned simplicity that works for every efficient manufacturing operation.  


I would ask another question to all associated with pharmaceuticals and that is “have we progressed with respect to innovations or are we progressing”. Most likely the answer is “no”. Regulators are telling the industry of what and how to innovate and the type of manufacturing process to use e.g. continuous processes (2). These are distractions. Regulatory abstinence of making suggestions on methods and types of processes that companies should use will have a very positive impact on pharmaceutical innovation.

I am not sure of the basis why the regulators are making suggestions of types of processes companies should use. Do they hands on experience in process development, design, scale-up, commercialization, justification and management experience in the manufacturing processes for products that are marketed? It takes significant rigor to make the suggestions.

Same wonderment also applies for many trade journal authors who postulate certain types of manufacturing processes for pharma. A process on paper is a speculation and is different from a lab process and miles apart from an actual commercial process. When rubber meets the road reality sinks in. My intent here is not to knit pick but face reality.

Industry knows and plays on its landscape. They know and need to justify every investment of new process technologies and every improvement. As suggested earlier, regulatory bodies need to facilitate these innovations and investments by reducing the approval time. As suggested earlier manufacturing process suggestions made by the regulators are a distraction. Most of the process methods and technologies have been existed and are used in industries other than pharma. My conjecture is lack of financial justification in the current business model along with regulations has prevented use of such processes and innovation.

If regulators want industry to innovation then they have to have a defined road map (3) that companies can follow for process and quality approval. Regulators need not know the process design and operating parameters. However, companies have to document every change. There has to be a time period in which companies can expect approval of their submission. Regulators have laid out the cGMP practices that companies have to follow. In a previous blog I suggested three months for generic ANDAs (4). New brand drug manufacturing approval process of 18 months or less could be a target. Such expectations will present challenges for the regulators. There has to be a trust established and if the trust is violated shutting down of the operation has to be the only recourse. 
Regulatory bodies should layout the expectations for every new product specifications in a certain specified time and let the innovators innovate. This will help and lower drug commercialization time.

Lack of trust and re-checking the checker seems to be the problem also. Such situations develop only when the deliverables change from the defined specifications because someone dropped the ball. Trust has to be earned and cannot be taken for granted. Any deviation from expected specifications and processes as suggested earlier should have strict penalty like closure of the facility with no “ifs and buts”.

To recap it is in the best interest of the industry to innovate and it should be allowed to continuously improve their existing processes even after the regulators have approved the produced products and as long as the expected deliverable quality is not compromised. Review of improvements made after initial approval should be on faith, trust and desire. If regulators find excursions outside the strictly defined boundary conditions, such operations should be shut down. Only strict financial constraint will keep everyone on their tows.  


Process of continuous improvement/innovation (5) will make drugs affordable and improve pharma revenue and profits. I just wonder why pharmaceutical industry shies away from win-win opportunities. Inaction suggests that the current business model needs change.

Girish Malhotra, PE
EPCOT International



1.     Guidance for Industry PAT—A Frame work for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance http://academy.gmp-compliance.org/guidemgr/files/PAT-FDA-6419FNL.PDF
2.     Malhotra, Girish: Batch, Continuous or "Fake/False" Continuous Processes in Pharmaceutical Manufacturing, Profitability through Simplicity, July 20, 2017
3.     Malhotra, Girish: ANDA (Abbreviated New Drug Application) / NDA (New Drug Applications) Filing Simplification: Road Maps are a Must, Profitability through Simplicity, May 11, 2017
4.     Malhotra, Girish: Can the Review and Approval Process for ANDA at USFDA be Reduced from Ten Months to Three Months? Profitability through Simplicity, March 25, 2017
5.   Malhotra, Girish: An Alternate Look at the Pharmaceutical World and drug affordability, Prospects, Analysis and Trends in Global Pharma, CPhI Annual Industry Report 2017, page 36-41, Accessed November 28, 2017

Thursday, July 20, 2017

Batch, Continuous or "Fake/False" Continuous Processes in Pharmaceutical Manufacturing


Use of “Fake and/or False” in our vocabulary has become very pervasive since 2016. I thought it would be useful and helpful if we re-visit definitions of Batch and Continuous production processes and try to see do either of them fit in the realm of False or Fake “B or C” process. My observations are based on my education, experiences in chemistry and chemical engineering and the prevailing practices established more than 100 years ago. They are also based on actual process design and development, scale up of developed processes and management of manufacturing processes and operations in fine/specialty chemicals, coatings, resins that were produced using organic chemicals. Many of the products were produced by reacting chemicals and others were formulated by blending different chemicals.

In recent months certain pharmaceutical processes have been labeled as continuous process but no process information has been shared. However, based on public information about the products their viability of being operated as a continuous process is extremely doubtful. Additional details are shared later.

Established batch and continuous process definitions and examples are reviewed after a brief overview of what chemical engineers are taught. I do not want to go in details but the following are part of the Ch.E. curriculum.  

Fundamentals taught in Chemical Engineering:

Chemical engineers are taught that every process should be safe, sustainable, and economic and should produce quality product first time and all the time. If a product has to be reworked or its stoichiometry has to be adjusted during the process to produce a quality product, its cost goes up and the profitability of the company is lowered.

It is well known and practiced in every manufacturing industry that the market size demand influences the type of process used. Irrespective of the process used product quality is a must for the market success of every product. These expectations are most stringent for the pharmaceuticals as they influence human life.

Chemical reactive processes are called Unit Processes (1) that produce a product that could be the final saleable product or could be an intermediate for another reactive or formulated product. Unit operations (2) involve a physical change or chemical transformation that are used in the chemical and allied industry to facilitate unit processes to produce the desired products. The combination of these makes a process and depending on demand it could be a batch or continuous process. 

McGraw-Hill has published series (3) of books that cover unit processes, unit operations and the economic considerations that allow chemical engineers and chemists to develop, design and commercialize processes that are economic and profitable to any individual or company. Besides these books there are many other books that have been written on the subject. Excellent books have been written about process control methods and strategies (4)

Most of these books have been written by what I would call “Hall of Famers” of Chemical Engineering and are considered Bibles of Chemical Engineering profession. None of these books are recommended or sponsored. There is no affiliation or financial relationship with the author.

Batch production:

Wikipedia’s (5) definition for batch production: Batch production is a technique used in manufacturing, in which the object in question is created stage by stage over a series of workstations (steps), and different batches of product/s are made. Batch production is most common in bakeries and in the manufacture of sports shoes, specialty/fine chemicals, resins, pharmaceutical ingredients (APIs), purifying water, inks, paints and adhesives. There are other formulated products and most prominent are finished pharmaceutical drug dosages.    
Batch production processes generally require much lower investment and have an advantage because several products can be produced in the same equipment. However, if the same equipment is to be used for different products, productivity will be significantly lower (6) compared to a process where the equipment is dedicated to produce a single product. If different products are to be produced in the same equipment, cleanliness becomes critical, especially for the APIs and their formulations. Idle (down) time can be high. Down time in the pharmaceutical industry is extremely high (7).  
For the pharmaceutical industry batch processes are the main stay in the manufacture of active ingredients (API) and their formulations. This is due to two inherent reasons. Micrograms to milligrams of API are needed in every dose. One kilogram of an active produces ONE million tablets of one milligram each. Thus, not a large quantity of the API is needed to satisfy the need. Table 1 is an illustration of the needed API and potential production methods for different dosages and number of patients. Additional analysis (8) is available elsewhere.

Dose, mg
Patients    Millions
API Kilograms needed/ year @ one tablet per day per year
API Production
Preferred Process Type
Number of plants
1
500
182,500
Batch
One or More
200
0.1
7,300
Batch
One
10
100
365,000
Batch
One or More

100
50
1,825,000
Batch or Continuous
Could be a single continuous plant but most likely batch
500
20
3,650,000
Continuous

Could be a single continuous plant but generally batch due to multiple sites

Table 1: Typical API need, Number of Patients and API Production Process
Process economics, chemistry and execution method will determine type of process used. Typical available hours for production at any plant are about 8,400 (24x7x50) hours per year. How many of these available hours are used for the production of a single product define the production process. If less than 8,400 hours are used to produce a product such process according to the established norms (5) will be a batch process. This holds true for API production and their formulations.
Table 2 is an illustration of how many hours per year are needed to produce at @200,000 tablets per hour at different doses. Production equipment of higher tableting rates are commercial and available. It is most likely most of the products would be formulated using batch processes even when they could be operated continuously.     

Dose, mg
Patients Millions
Tablets Used per day
Tablets, Millions/ year
Run Time Needed, hours @200,000 tablets/hr.
Number of plants
operating 7,140 hours per year
Preferred Formulation Process
1
50
1
18,250
91,250
13
Continuous but will be multiple batch plants
10
20
2
14,600
73,000
10
Continuous but will be multiple batch plants
100
15
3
16,425
82,125
12
Continuous but will be multiple batch plants
500
10
4
14,600
73,000
10
Continuous but will be multiple batch plants
200
0.1
1
36.5
183
1 (Run time less than 200 hours
BATCH ONLY
Table 2: Tablets Needed, Tablet Run time, Number of Plants and Patients Served
Continuous production:

The definition of continuous (9) production has been long established. Continuous usually means operating 24 hours per day, seven days per week with infrequent maintenance shutdowns, such as semi-annual or annual. Time could be allocated for unexpected shutdowns. Any process that does not meet the defined and established definition and is operated fewer hours than the established definition would be difficult to be justified as a continuous process. Excellent examples of continuous process are Earth’s rotation and our heartbeat. Can we imagine “stop and go” movement of earth and human heart? 

Throughout the pharma landscape most likely there are ten or less APIs that are being produced by continuous processes. There are additional APIs that could be produced continuously (10) but effort is needed to develop and commercialize such processes. Different business model would be needed. Alternate business model and consolidation can convert batch processes to continuous processes. There is a downside of the continuous processes. It will result in consolidation of operations especially among the contract API producers and formulators.

McNeil, a Johnson and Johnson subsidiary, due to the formulation volume of Tylenol could have built a continuous process but they did not, a missed opportunity. It seems that chemical engineers and chemists at some companies forgotten FRUGAL SCIENCE and FRUGAL ENGINEERING.


Fake/False Continuous processes:

In the current pharmaceutical landscape, processes that are in reality batch processes are being called continuous by the chemical engineers working in public and private sectors. We all need to understand and recognize that if multiple products can be processed in the same calendar year in the same equipment by re-arranging the reactive chemistry or the formulations, their processes would not be called “continuous manufacturing”. Any process that has a wide spot in the manufacturing line [material is held for any time period], the process is a batch process. Processing step before such hold can be called by any name but if it is different from established definition, we have are making fun of science and engineering.

FDA (11) does not have an established definition and I am told one would be forthcoming. I wonder how different it would be from what is established and practiced methods.  Even the press (12) has chimed in. Numerous articles have been written that include “continuous manufacturing” but no one has put forth its definition or names of the product or their production rates. I am sure many, me included, want to see the proof. I am sure many of us remember movie Jerry Maguire (13).

API Example:

Analyzing Table 1, we have an example of a drug that has 100,000 patients. They have to take 200 mg dose every day of the year. Total API needed is about 7,300 Kg. per year. This product would and should be produced using a batch process in the available equipment. If each batch reaction produced about 250 kilos per batch and about 30 batches per year would be needed. If each batch took 30 hours per batch all of the API would be produced in about eight weeks. If they campaigned the whole production for first two months of every year, this would not make their process a continuous process. If this campaign run was called a “continuous process” then it would be an incorrect characterization and I will label this process “fake/false continuous process”.  

Formulation Example:

If the example of API above was to be formulated and tableted, the product can be produced at single plant in less than 200 hours at 200,000 tablets per hour. However, if the tableting were done at 10,000 tablets per hour, it would take them six months to produce the total demand. It still does not make the process continuous, as the equipment would be sitting idle for six months every year. It would be a wasted investment earning no return if every reader of this post had to invest his or her own money. Again if the process is called a continuous process, it again would be an incorrect characterization and I will label this process “fake/false continuous process”.

Vertex (8) has claimed to have a continuous process for their cystic fibrosis (CF) drug. With less than 80,000 patients worldwide, they do not have large enough global patient (~80,000) base for all of their CF drugs and cannot operate their equipment for 8,400 hours per year per drug. Similar situation exists for Janssen Pharma drug Prezista.

To me, an incorrect characterization of established definitions seems to have become a new fashion especially in the pharma landscape. Is it because the US FDA is mentioning continuous manufacturing in their communications and the industry wants to look good by calling a naturally batch process as a “continuous process”? Thus, if we accept an incorrect definition then we should be also ready to accept 2+2=6, Sun can rise from the West and a mammal can be half pregnant (my apologies to all readers). If that were the case basic laws of science, math and anatomy would be defied.
Quality Assurance in Batch and Continuous Processes:
Meeting quality standards in batch as well as continuous processes have different rigors. Command of the processes is a must whether it is a batch or a continuous process. If command and understanding of the processing steps are lacking, invariably there could be a batch-to-batch and/or a lot-to-lot quality variation. Efforts to rework or bring the material to quality can and often results in waste i.e. higher product costs.
In batch processes due to their stop and go nature quality can be sometimes managed and the process adjusted to achieve the established quality benchmarks. Thus an absolute command of the batch process is less stressful but still is necessary. Batch processes are based on in-process quality checks and adjustments. This practice extends cycle time and adds to inventory [raw material, in-process, work in process and finished goods] challenges. If the batch process cannot be adjusted to correct process deviation, significant waste can result. In the simplest terms, batch processes have economic value for products that do not require continuous production but their quality testing can be an aggravation. Batch process = quality by aggravation (QbA) unless the process repeatability is strictly controlled.  

Compared to batch processes, continuous processes have much higher process control demand. Since the process is running with minimal/no stop time, it is extremely critical that process operating parameters do not deviate outside the established process operating control limits. If process deviates outside the established control limits, significant quantities of waste and financial loss can result. Continuous processes demand that the quality be established through robust process design when the process is developed, designed and commercialized. This would be a case of quality by desire (QbD).

I am sure the debate on what is a continuous or a fake/false continuous process will go on till the economic realities of investment that have been well established are understood by most. Chemical engineers and Chemists are taught everything, values and virtues, of batch and continuous processes there is to know about such processes but one thing is sure unless one has not justified, developed, designed or commercialized a process, it is difficult to discern value of the developed process. I know no investment would be made for namesake unless it can be justified and meets established norms of the science, economics and engineering.

As I said earlier a batch process cannot be continuous and vise versa. If we accept it otherwise then we have a case of “false/fake” science, engineering, economics and human intelligence.  


Girish Malhotra, PE
EPCOT International

  1. Shreve, R. Norris: Unit Process In Chemical Processing, Ind. Eng. Chem.195446 (4), pp. 672–672
  2. Unit Operation, https://en.wikipedia.org/wiki/Unit_operation, Accessed July 11, 2017
  3. McGraw Hill Chemical engineering series, https://www.librarything.com/tag/McGraw+Hill+Chemical+engineering+series, Accessed July 6, 2017
  4. Chemical process control, https://www.librarything.com/subject/Chemical+process+control, Accessed July 6, 2017
  5. Batch Production Wikipedia, https://en.wikipedia.org/wiki/Batch_production Accessed July 6, 2017
  6. Malhotra, Girish: Square Plug In A Round Hole: Does This Scenario Exist in Pharmaceuticals? Profitability through Simplicity, August 17, 2010 Accessed July 11, 2017
  7. Benchmarking Shows Need to Improve Uptime, Capacity Utilization, Pharmaceutical Manufacturing, Sep 20, 2007 Accessed July 7, 2017
  8. Malhotra, Girish: A Blueprint for Improved Pharma Competitiveness, Contract Pharma, September 2014, pp. 46-49
  9. Continuous Production, https://en.wikipedia.org/wiki/Continuous production, Accessed July 14, 2017
  10. Malhotra, Girish: Strategies for Improving Batch or Creating Continuous Active Pharmaceutical Ingredients Manufacturing Processes, Profitability through Simplicity, March 20, 2017, Accessed July 17, 2017
  11. Drug Making Email exchange with Dr. Janet Woodcock, FDA July 13, 2016
  12. Breaks Away From Its Old Ways, The Wall Street Journal, February 8, 2015, Accessed July 17, 2017
  13. Jerry Maguire, Accessed July 18, 2017