Quality by Design – the new manufacturing paradigm

pharmafile | May 31, 2011 | Feature | Manufacturing and Production |  Quality by Design, pharma manufacturing 

Pharmaceutical manufacturing currently operates using processes that are widely seen as inefficient, inflexible and outdated.

In the current system of ‘batch processing’ product quality is tested at numerous staging posts in the manufacturing process.

Firstly by raw material testing, and then by in-process material testing during manufacturing. Finally, end product testing is conducted against a number of FDA approved parameters. If a batch does not meet the required specification then it is discarded as OoS (Out of Specification), resulting in a waste of resources and money.

Furthermore, any small changes to the manufacturing process require FDA or EMA permission, and endless small process revisions are not helpful to the quest for better quality. In essence, both quality and performance for pharmaceutical manufacturing are currently achieved predominantly by restricting flexibility in the manufacturing process, and by end product testing.

The current regulatory review system places little scope on how the design of the manufacturing process can ensure good and reliable product quality. Because of this, the inherent difficulties of process scale-up – particularly for complex dosage forms – are often not fully recognised.

Product specifications are often derived using test data from one or more batches, but rarely at production scale, and sadly, full mechanistic understanding does not play a large part.

Quality by Design

In contrast to this approach is the Quality by Design (QbD) approach. The thinking which underpins this is that there should be more emphasis on learning and continuous improvement to drive process improvements, and hence – quality.

QbD is part of a broad push to bring about a complete transformation in how pharma manufacturing is conducted.

In August 2002, the FDA launched a new initiative called ‘Pharmaceutical CGMPs for the 21st Century: A Risk-Based Approach’. A major component of this new approach is Process Analytical Technology (PAT) – a very similar but complementary concept to QbD.

PAT is defined as a system for designing, analysing, and controlling manufacturing through timely measurements (i.e. during processing) of critical quality and performance attributes of raw and in-process materials and processes, with the goal of ensuring final product quality.

Then in November 2003, the ICH agreed to work on a harmonised plan to develop a pharmaceutical quality system based on ‘‘an integrated approach to risk management and science”.

ICH set up two Expert Working Groups (EWGs) on pharmaceutical development.

ICH Q8 EWG seeks to incorporate elements of risk and quality by design throughout the life cycle of the product. 

The ICH Q8 EWG defined the ‘desired state’ for pharmaceutical manufacturing in the 21st century as: ‘‘Product quality and performance achieved and assured by design of effective and efficient manufacturing processes.”

The second working group (ICH Q9 EWG) is trying to better define the principles by which risk management will be integrated into decisions by regulators and industry, regarding quality and including CGMP compliance. The outcome should be a risk management framework intended to lead to more consistent science-based decision-making across the lifecycle of a product. These two expert working groups work in parallel and exchange information on a regular basis.

The principles of Quality by Design have been adopted by a number of industries with great success since being developed by management consultant Joseph Duran in the 1970s. A notable example is the success of the Japanese car industry and how the top three spots in car reliability surveys are consistently held by Japanese companies. Juran believed that quality could be planned into a process, and that most quality crises and problems directly related to the way in which the quality of the final product was planned in the first place.

Regulators to collaborate

In February 2011 the FDA and the EMA announced a new three-year pilot programme for the joint review of the QbD component of new drug marketing applications.

The two regulators say the pilot programme has been created because of concern that some ICH guidelines were being interpreted differently in Europe and the US.

Commenting on the launch of the joint initiative, Janet Woodcock, director of the FDA’s Center for Drug Evaluation and Research, said: “As the number of applications that follow the QbD approach steadily increases, collaborative assessments will enhance understanding of QbD concepts. She added: ‘‘The tools used by FDA and EU reviewers will increase information sharing and reduce redundancy. To fully implement QbD, we need to further harmonise the implementation of the guidelines, work collaboratively, and provide scientific, risk-based regulatory decisions in a timely manner.”

The pilot programme applies to NDAs and MAAs, some supplements, and CMC meeting requests that include QbD elements submitted to both agencies at about the same time. Review of QbD applications does not change statutory deadlines. The pilot will end on 31 March, 2014.

The pilot is only being used in chemical drugs and not biologics, but nevertheless represents a major shift for pharmaceutical manufacturing. Company participation will be fully voluntary and will involve companies prepared to submit a filing simultaneously in the US and Europe.

The new ICH Q8-10/QbD paradigm is now being developed and will provide guidance to companies in how to adopt QbD principles to achieve the best quality standards. For the early adopters, this will mean recognising the concept first mentioned in ICH Q8 guidance – ‘quality cannot be tested into products, it should be built in by design’.

Despite regulators only piloting QbD programmes, experts in the field say there are compelling reasons for being in the vanguard of its adoption.

What can QbD achieve?

QbD ensures predictable and predefined product quality. The adoption of QbD will include defining a target product quality profile; design of the manufacturing process from basic principles with a very good understanding of the mechanism involved (good Design of Experiment); the identification of critical quality areas, process parameters and potential sources of variability; and finally good control of the manufacturing process to achieve the most consistent quality.

One of the achievable goals of QbD is to transform the Chemistry, Manufacturing, and Controls (CMC) review of an Abbreviated New Drug Application into a very precise and formal science-based quality assessment.

Alpaslan Yaman, a principal consultant at the quality auditing company Biotech, Pharma & Device Consulting, said: “The Quality by Design approach mandates that a company have a sound understanding of their product. The typical approach to product development relies heavily on using previous dosage form experience, even when the chemical entity may not be physico-chemically equivalent.”

This means specifications for a product may not be fully established or qualified with sound experimental design.

This sometimes becomes clear when technical issues arise once a product is launched. In fixing these problems, key principles that govern the product and process that are missing are then realised.

In the QbD approach, the product and process are understood prior to commercialisation so that when adjustments or issues occur, they are quickly dealt with without requiring further investigation.

Companies are enthusiastic about the new approach to quality because of the advantages for post approval changes to the process. Dr Alpasan says: “The QbD approach, if properly followed and the product development approved by the FDA, puts the product on a lower risk level with regards to the understood quality and consistent control of the process and thus the product.

“The difficulty is the lack of people in the industry with a good conception of the current approach to product development and what QbD actually involves.”

Is pharma ready to adopt QbD?

Most of the major pharma companies are now engaging with QbD, with numerous projects now underway. Despite this, Alpasan says many in the industry fail to understand just how big a change in culture and practices it represents.

“Companies are coming to terms with what this entails. It may be difficult or painful at the onset for the initial development projects, but ultimately the pay-off will be great, not just from a regulatory perspective but also in regard to long-term control during commercial operations for the market.”

He says an existing product approaching the end of its development, and on the verge of filling cannot be fitted into a QbD filing document for approval. Rather, it needs to be part of the plan for the product from the very beginning of its development. There is need to invest in new processes, but these costs can be recouped when QbD becomes established.

Alpasan does not believe that the QbD approach will necessarily increase the cost of manufacture unless the site has yet to implement PAT technologies. He thinks that PAT and QbD complement one another since a better understanding of the product and the process leads to a better real time control of the process. The QbD approach may increase the upfront cost of development, but there will be return on investment with better throughput for their processes and efficiency with tighter controls. Pharma should be able to avert issues with unplanned outcomes, because if they understand the product and are in control of the process and promote active trending, then uncontrolled outcomes, line shutdowns and recalls should in principle not occur.

Of course there may still be occurrences in production, but the time needed to understand and overcome problems should be less if the product and process are understood, at least with regards to the governing principles.

Adopting QbD will increase costs and development but this would be offset by more successful launches, less loss in production, fewer deviations, and fewer recalls – so there should be an overall net gain. Alpasan adds: “This expands the concept of ‘right first time’ from manufacturing and applies it to development. As with all quality initiatives, there are two approaches. One is to install culture like a piece of equipment and the other to grow it slowly from inside. I suspect the cost of QbD will be largely determined by which approach is taken.”

He says QbD should simply be seen as part of good business practice. ‘‘Companies need to realise they should be doing this because it is good business, not because the auditors are telling them to. If implemented correctly then this will decrease variation, increase quality, decrease waste, decrease costs, decrease Risk to the Patient and everybody wins.”

QbD promises to ultimately contribute to improving the safety of drugs compared to existing practices. It is widely acknowledged that the QC testing for final product release is not a statistically significant sampling, so relying on this for product quality is problematic. That is why validation of a process (with justified specifications and defined critical to quality attributes) and how that process is controlled, is key to product success.

But the implemented QbD-developed product promotes the continuous monitoring of each of the critical unit operation that is predefined, based on scientific rationale. This monitoring ensures the process is in control and also that the required product critical attributes are achieved. Thus in the long run, this approach should eliminate the need for final QC release testing and save money, eliminating QC testing and the constant investigations for OoS events that occur from lab or sampling errors.

Continuous processing

One new methodology currently being introduced is continuous processing, and Alpasan says this is highly complementary to QbD. “Continuous processing can be more easily achieved after QbD is implemented, for the simple fact that it is nearly impossible to automate or make a process continuous if it is poorly understood or in fact not fully controlled.

The first requirement to make a process continuous is to have a proper understanding of the product requirements, and then the critical unit operations. After that point, whether to make a process continuous is a business decision as to whether it fits with the product volume and the facility constraints.”

QbD is being introduced slowly, and shouldn’t be used for products already on the market. Alpasan says: “I know of companies that try to implement QbD in products not far from their expected regulatory filing date. To implement QbD at this point is very difficult, since much that is required had not been planned for and quite honestly, much of the good science that is necessary for this approach is not typically done.”

The principles of QbD are suited to the manufacture of biologics, and most biologics manufacturers are likely to adopt QbD principles in the design of their processes from the outset. According to Alpasan, QbD can be applied to any of the product/process operations. One needs to understand what is ultimately required beyond what the regulators are asking for in the final documentation. This can essentially be said about any of the regulations and the regulated industry. His experience is that most individuals do not think through the basics of the science involved in the development of their product or process.

The regulations are a minimum and a means to ask the right questions to assure that everyone at least considers the basic requirements in the development of a product. “I have seen individuals either overcomplicate a process or product, or underestimate the science and just ask ‘just tell me what to do’. All this stems from a lack of understanding of not only the basic sciences but the practical application of those principles on a case by case basis.”

Pilots and progress

Pharma is now beginning to implement QbD, with most of the big companies now having at least one manufacturing leader with responsibility for QbD, and many are now running pilots.

One notable example is Novartis’ colllaboration with the Massachussetts Institute of Technology (MIT) on ‘blue sky’ innovation in manufacturing, which has been running since 2007.

James Evans, associate director of the Novartis/MIT Center for Continuous Manufacturing, recently presented prog-ress of their work at the AAPS 2010 meeting in November. He claimed that Novartis-MIT’s continuous paradigm (encompassing QbD) has the potential to produce astonishing reductions in production times – from a typical current range of 200-300 days, to getting the product to the patient in less than ten days.

The collaborators say they will have a full scale process running this year, with a more advanced version expected by 2015.

Many other companies are working on similar projects which are set to produce initial results in the next few years; pharma manufacturing QbD has been slow in coming, but its effects will nevertheless be revolutionary.

Huw Kidwell is an industry writer and journalist with Dolffin