Analytical Method Development: Beware the Rabbit Hole

By Wayland Rushing, Ph.D.

Senior Scientific Advisor

ABC Laboratories

www.abclabs.com

Developing analytical methods for pharmaceutical analysis can be one of the most time consuming and rewarding experiences in today’s analytical lab. I fondly recall the days (from many years ago) of sitting in front of the HPLC anxiously awaiting the data from the latest run for evaluation.  Either hours/days of hard work were about to be vindicated or hopes dashed and it was back to the drawing boards.  This scenario is still occurs on a daily basis throughout our industry. Unfortunately there is a potential dark side to the developmental process.  It can be very easy for the method developer to get lost in the task, ever trying to develop the “perfect” analytical method.  I have watched many analytical chemists get caught in this web of ever trying to catch this unicorn.  There is never a perfect method, it can always be improved, accuracy tweaked, precision optimized, etc.  So how do we know when enough is enough?

Let’s take the development of an impurities method as an example. The impurity is present at 0.15% and has a specification of 0.5%.  We are going to charge our developer with developing a method to quantitate for the impurity and we turn him lose.  After a couple of weeks we touch base with them and are updated the method isn’t ready, so development continues, couple more weeks pass, then another month, then finally a method is delivered.  The method can see the impurity down to 0.005%, has recovery of >99% with <1% RSD at the quantitation level.  And we stand amazed, not only at the performance of the method (quite impressive), but also finally realizing the time and money spent on a developing a method that far exceeds what it needed to do.  Simply put, for an impurity method, this is overkill, this level of method performance simply isn’t needed.

I have watched this play out in the development arena multiple times over again.  With the main drivers of development being time, quality and cost we must learn how to balance all of them to achieve the goals of development.  The mistake made here is that the development of the method wasn’t guided, there were no expectations set on what the intent of the method should be and what level of performance the method should have.  When initiating any analytical development there should be key questions that are asked and answered before any labwork commences:

• What is the intent of the method?
•  How low does the method need to go down to?
• What level of accuracy is needed?
• How precise does the method need to be?
• What is the range of the method?

So lets re-evaluate our development plan for the above example and answer the above questions first:

• What is the intent of the method?
• Answer: to quantitate the impurity in question for release and stability testing.
• How low does the method need to go down to?
• Answer: the method needs to have an Limit of Quantitation of 0.05%
• What level of accuracy is needed?
• Answer: At the specification the method should have an accuracy between 90 – 110%.
• How precise does the method need to be?
• Answer: At the specification the method should have no greater than 10% RSD.
• What is the range of the method?
• Answer: the method should be linear from 0.05% to 0.75%.

If we apply these questions to the above development we not that the method developer had met these conditions within the first two weeks of the development program.  But his level of understanding of what the method needed to do and the actual needs of the method were not aligned and hence spending an additional couple of months in development, wasting time and money.  This is a common occurrence which is easily preventable, set expectations before you start the work.  Ensure that your internal analyst or the CRO you are working with understands the difference between development a perfect method and developing a method which is suitable for its intended use.  This can often be the difference between having an efficient development program and one that suffers from unexpected delays in time and budget excesses.