This issue’s cover story asks a simple question: has drug manufacturing become more of a science and less of an art over the past 10 years? The consensus of a number of experts who work in and follow pharma closely wasn’t all that positive.
Yes, they acknowledge, there have been some areas of improvement, and some processes are now up and running at six sigma levels. Tools such as statistical process control and process capability analysis are becoming more common in pharma than they were 10 years ago, while FDA’s updated Process Validation Guidance promises to answer critical tech transfer and design questions, and keep the spirit of process analytical technology (PAT) alive.
By and large, though, manufacturing is still well behind potato chip and soap flake making, as former FDA Commissioner McClellan noted almost a decade ago, in its use of modern methods and technology.
Operational Excellence programs have become “flavor of the month” exercises at some pharma companies, says a consultant, where some Lean projects gradually morph into Six Sigma and everyone loses interest. In addition, there is little understanding of the cost of poor product quality.
IT investments are focusing on the transactional level, not on areas which would aid quality and continuous improvement, while regulatory uncertainty still prevents many manufacturers from using the best technical solutions available, day to day.
But there are even more fundamental problems than these, the biggest of which is variability in practice. There is no set of standards for pharmaceutical equipment design or operation, or even what constitutes “best practices” for unit operations such as blending or drying. Each company, or even facility, develops its own.
Without standard practices it becomes difficult to think systematically, and to improve. Then, when dealing with processes that are used to make less profitable products such as over-the-counter medications, things can (and do) fall apart.
Considering the fact that the consumer is depending much more on generics and OTC drugs, shouldn’t robust standards and specifications be made available, globally and industrywide?
If cars were made the way drugs are today, one expert notes, the auto industry would still be at that pre-model T stage where every automobile was a unique creation. If petroleum were processed this way, filling those cars would become a validation exercise each and every time, as each gas station struggled to work out the issues presented by different blends using differently processed crude oils from different parts of the world.
Given the potential risks to the public, should this random approach still be accepted in the 21st century?
ASTM’s E55 committee is addressing the need for standards, but some feel that more basic efforts are needed before going to ASTM.
Prabir Basu, head of NIPTE, believes that the key is cooperation and technology sharing. Pharma, he says, should establish a consortium analogous to Sematech, which was set up by U.S. semiconductor manufacturers over 20 years ago, in the wake of serious threats to competitiveness from Asia.
Another potential role model, he says, could be the American Petroleum Institute, whose members share knowledge and best practices proactively, or the Fraunhofer Institute in Germany, where industry and academia work on fundamental research.
As it looks to improve its manufacturing, pharma has no shortage of inspiration, or role models from other industries, as McClellan noted in 2003. However, where it was acceptable a decade ago to discuss the need for improvement, it has since become extremely unpopular.
Is the predominating voice to become that of Dr. Pangloss, the character in Voltaire’s Candide, who, despite catastrophe after catastrophe, insisted that “all was for the best in the best of all possible worlds?” Or will the critical voice of reason and continuous improvement prevail? Ajaz Hussain, former PAT team leader at FDA, now with Philip Morris International, notes the need for rigorous criteria, as established by the philosopher Karl Popper, that would bring science to manufacturing. There must be an agreement of peers, based on evidence, and the difference between practice and theory aggressively analyzed and reduced.
It can’t be easy to work with the level of variability and risk, and basic questions of industry and job stability that many in the industry face today. But here’s hoping that your efforts bring more modern methods to day-to-day pharmaceutical manufacturing operations.