Q: The USP Validation and Verification Expert Panel recently proposed integrating traditional approaches to method validation, transfer, and verification into the analytical procedure lifecycle process rather than treating them as separate things. What are the main reasons for that change?
A: The USP formed an Expert Panel on Validation and Verification to explore ways to incorporate Quality by Design (QbD) concepts into analytical procedures. Over the past decade, the pharmaceutical industry has been using QbD approaches, as described in ICH Q8, Q9, and Q10, to improve manufacturing processes by using a lifecycle model. If the lifecycle model is applied to analytical procedures, it becomes apparent that method validation, verification, and transfer are not independent activities, but parts of the lifecycle of the procedure.
From an industry perspective, there have been ongoing challenges with method transfer and with out-of-specification (OOS) results, which may be related to method performance and may be improved by adopting a QbD lifecycle approach. That would involve proactively considering the expected performance of the procedure, particularly variability, and the sensitivity of the method to the typical variances in method parameters.
Q: What will that integration mean for lab managers?
A: If the true expectations for method performance are identified and effectively addressed early in the lifecycle of an analytical procedure, then better performance of the method can be expected. The end result will be analytical procedures that, by design, will function well in a wide variety of situations, with fewer transfer issues and fewer OOS results due to analytical variability.
Q: The key to the lifecycle method seems to be the ATP (Analytical Test Profile). What is the ATP, and how does it improve the method validation, transfer, and verification process?
A: QbD has taught us the value of starting with the requirements. The ATP clearly identifies the requirements and expectations of the analytical procedure, which is similar in many ways to the Quality Target Product Profile (QTPP) used in QbD for pharmaceutical manufacturing processes. It typically addresses performance questions about the analytical procedure: What are you trying to measure? Over what concentration range? In the presence of what matrix? What accuracy/ precision/uncertainty is required? With what level of confidence? The ATP may also include some practical constraints, such as requiring that the procedure be performed on the equipment already available in the quality control laboratory or that the analysis be completed within a certain timeframe.
We are likely to find that statistical tools can assist us in attaining these goals, such as using confidence intervals to characterize overall uncertainty instead of the traditional point estimates for accuracy and precision, which can allow unacceptable levels of variability to affect the performance of methods. For method transfer, equivalency protocols may replace the relatively simple comparison of results often employed currently.
While many labs may do this informally now, our experience is that scientists tend to focus on the capabilities of the instruments they are using rather than on the requirements for the reportable results to be able to make good decisions.
Once the requirements for the reportable result have been identified, this naturally leads to selection of an appropriate analytical technique and development of method understanding: well-designed experiments to investigate the method parameters that can impact method success, such as Design of Experiments (DoE) to evaluate optimal sample preparation and chromatographic conditions.
Q: What sort of feedback have you received from lab managers and other lab professionals on the lifecycle approach so far?
A: The feedback has been surprisingly positive, considering that most laboratories have well-established method development and validation procedures. I think this reflects the recognition of managers that the current approaches to method validation are not working as well as they should be.
In fact, USP just held a Workshop on Lifecycle Approach to Validation of Analytical Procedures with Related Statistical Tools on December 8 and 9 at their headquarters in Rockville, MD. Over 140 professionals gathered to discuss current thinking and future plans in this area. The response of the participants was overwhelmingly enthusiastic.
Lab professionals are actively looking for ways to address challenges with method transfer failures and OOS results due to poor method performance, and the use of a proactive, holistic approach is resonating with them. Some chemists are apprehensive about the use of statistics associated with this approach, but I am confident that is a hurdle that can be overcome.
Q: You recently taught a course on the lifecycle approach to analytical methods at the Eastern Analytical Symposium (EAS 2014). How did that go?
A: The course went really well. Eastern Analytical Symposium is a great conference, with lots of local support. I was surprised that the attendees for this course were seasoned professionals, averaging over 15 years’ experience with method development and validation, since EAS often draws those newer to the industry. This again points to the fact that lab professionals are looking for a better way of dealing with method lifecycle, and the issues that result if methods are not meeting performance expectations.
In my courses, I always draw on the experiences of the attendees, since adult learners are more likely to use and apply what they learn if they can relate it to something that is concrete to them. This class had lots of experience to draw upon. When I pointed out that by using the lifecycle approach (with the ATP to articulate the method performance expectations, method understanding to ensure they knew the strengths and weaknesses of the method, appropriate validation acceptance criteria, and ongoing verification of method performance), the attendees could reduce the types of problems they have had in the past, they wanted to learn how they could accomplish the same results.
Q: What were the main concerns of those who attended the course?
A: The attendees were concerned about the failure of methods to perform the way they need to in a GMP environment. If testing results in an OOS result, the impact is huge, and the investigations can take days or weeks. If the investigation concludes that the sample was acceptable, but the results were not initially in specification, then there was wasted effort and lots of anxiety. The bar is even higher when transfer of a method from one laboratory to another is not initially successful. Ultimately, the issues are resolved, but there must be a way to avoid the unnecessary effort and related stress. Using the lifecycle approach has the potential to help the situation significantly.
Q: When is it expected that the changes will be officially adopted? Are lab managers/ researchers beginning to implement them already?
A: The official implementation of this approach is still over a year away for USP, and a planned ICH guidance is not likely before 2017. The good news is that the concepts can start to be applied now, and several companies, including both innovators and generics, have been using them. At the recent USP Workshop on Method Lifecycle, an FDA representative indicated that the FDA had already received and approved applications where the lifecycle approach was used for analytical methods.
The implementation is likely to be evolutionary. The companies that have already adopted this approach generally started with a pilot, but increased the scope when they realized the benefits. While the traditional approaches are still acceptable, when industry finds something that works better, they adopt it. Interestingly, the recent FDA guidance on Analytical Procedures and Methods Validation for Drugs and Biologics contains many of the traditional ideas but leaves the door open for use of the lifecycle concepts.
Q: What changes do you expect to see in the method validation, transfer, and verification process even further down the line?
A: The next five years should be very interesting for laboratory professionals and managers. The USP Validation and Verification Expert Panel expects to publish several more Stimuli Articles and a new General Chapter on Method Lifecycle. PhRMA Analytical Technical Group, EFPIA [European Federation of Pharmaceutical Industries and Associations], BP [British Pharmacopoeia] and EP [European Pharmacopoeia] are actively considering adoption of these principles and an ICH Guidance is in the works. As we see more applications of the lifecycle approach, and if it helps users realize the benefits we anticipate, it should become even more widespread.
For more information, Greg Martin can be reached at firstname.lastname@example.org, or 484-942-3200.