Issue link: https://www.e-digitaleditions.com/i/1171020
Pharmaceutical Technology REGULATORY SOURCEBOOK SEPTEMBER 2019 21 be performed if only one laboratory with two work- stations is available. Note that the design in Table VI can be performed by one analyst as analyst was not identified as a risk factor. However, the design in Table V will usually include a different analyst in each laboratory (thus any effect of laboratory and analyst are indistinguishable). Discussion The examples in this article demonstrate how a risk-based approach can be used to design more appropriate IP studies (often as part of a broader holistic precision/reproducibility/ruggedness study) than previously pre-QbD recommended generic studies. The decision on what constitutes an independent analytical run and whether to allocate a factor to repeatability or intermediate precision may not always be clear, as illustrated in Example 2. Careful thought must be given to what it means to repeat the analysis indepen- dently mimicking day-to-day variability. Precision studies are always constrained by the available resource/equipment, so it is imperative to design these experiments thoughtfully to chal- lenge the method as much as possible. Estimating IP as part of a study that includes different labs can often make it easier to incorporate a larger number of levels of IP factors without necessarily increas- ing the size of the design. For most designs, repeat- ability can also be estimated, avoiding the need for a separate study. The final concept paper for ICH Q14, Analyti- cal Procedure Development and revision of Q2(R1) Analytical Validation (27,28) acknowledges that ICH Q14 will provide an opportunity to use en- hanced approaches in line with ICH Q8 and ICH Q11 for analytical procedures. It is anticipated that this will include the use of quality risk man- agement approaches outlined in ICH Q9 (10) in the development and validation analytical proce- dures. The concept paper also highlights the need to harmonize performance criteria for analytical procedures that will also help define how IP stud- ies should be designed and analyzed. A number of recent publications (23,29–32) have illustrated how the concept of the analytical target profile (ATP)—analogous to the quality target product profile (QTPP) as described in ICH Q8 (9)—can be used to define the performance criteria, in- cluding precision requirements, that ensures the measurement of a quality attribute is fit for the intended purpose. Conclusion Generic approaches for studying IP such as the use of the Japanese NIHS design as described in this article have provided industr y in the past with a useful approach for evaluating IP in a pre-QbD world where knowledge associated with the method was not captured systematically and formal risk assessments were not typically performed. Such designs ensured at least five de- grees of freedom were studies across a combina- tion of typical factor combinations (days, analysts, equipment). Since 2007, the principles of QbD have been progressively applied more and more to the development and validation of analytical procedures (13). The use of a risk-based approach aligned with ICH Q9 can aid in the design of more appropriate IP (and broader precision de- signs) designs that focus on the critical few high- risk method factors. The recently agreed concept paper for ICH Q14 and ICH Q2(R2) (27) will pro- vide a golden opportunity to provide guidance