1. The USP Performance Test Dissolution Systems Suitability Studies Walter W. Hauck, Ph.D. USP Consultant Presentation to Advisory Committee for Pharmaceutical Sciences October 25, 2005

2. Topics  Systems Suitability Studies for Dissolution  Collaborative Study Design and Analysis  Variances: Correction to May 2005 Discussion  Next Steps

3. System Suitability  USP and ICH Q2b: … to ensure that the validity of the analytical method is maintained whenever used. Typical variations are the stability of the analytical solutions, equipment, and analysts. … System suitability tests are based on the concept that the equipment, electronics, analytic operations, and samples to be analyzed constitute an integral system that can be evaluated as such.

4. Types of Dissolution Systems Suitability Studies  Chemical Calibration  The term is a misnomer—it is a periodic Systems Suitability or system verification study  USP calibrator tablets support this type of study  The approach evaluates the system as a whole (apparatus plus operator plus procedures)  Acceptance ranges for dissolution values of the calibrator tablets determined by collaborative studies for each new lot (e.g., new Prednisone Lot P)  Other Approaches  Mechanical Calibration  Engineering Approaches

5. Terminology  Apparatus  One vessel  One stirring element  One position in the assembly  Assembly  One motor  Same temperature controller  Multiple apparatuses (6 or more)  Also called ‘bath’  Experiment  Single units tested in all apparatuses in the assembly (in the collaborative study, n= 6)

6. Systems Suitability: Chemical Calibration  Calibration procedure  For assembly under study, use standard dissolution procedure; test all apparatuses in the assembly  Assembly/apparatus must be mechanically calibrated  Acceptance applied to each position (apparatus) If one position result is unacceptable then entire assembly is also unacceptable Increased risk of failing due to multiple testing  Evaluates the integrated function of the system to prepare a dissolution sample

7. Systems Suitability: Mechanical Calibration  If a mechanical system is properly aligned and conforms to tolerances established for all such systems, then the operation of the system should be suitable  Dissolution test apparatus and assembly are mechanical systems  If mechanical calibration is the only control, no further check on the system is performed and system is assumed to be suitable

8. Systems Suitability: Mechanical Calibration  Deficiency of purely mechanical calibration  Assumes that the integrated system is no more than the sum of its parts  Does not evaluate that standard flow of medium is achieved  Limited understanding of the appropriate range and measurement mode for vibration  Difficulty of determining that vessel is suitable (surface irregularity, deviation from hemisphere and cylinder) List of suggested improved mechanical tolerances in Dissolution Calibration: Recommendations for Reduced Chemical Testing and Enhanced Mechanical Calibration, PhRMA Dissolution Calibration Subcommittee, PF 26(4) [July-Aug. 2000]

9. Topics  Systems Suitability Studies for Dissolution  Collaborative Study Design and Analysis  Variances: Correction to May 2005 Discussion  Next Steps

10. Collaborative Study Design  Standard product (tablet)  Standard procedures employed  Determine range of results from mechanically-calibrated assemblies  Determine bounds for “acceptable” results  25-35 collaborating laboratories, international

11. Design Evolution  Originally, each lab performed one experiment per condition (e.g., basket at 50 rpm)  Beginning with 2003, each lab performs two experiments per condition with separate operator and equipment;  Allows separation of inter-laboratory and intra-laboratory (intermediate) precision

12. Analysis Considerations  Analysis is preceded by a control chart analysis to select labs/experiments for inclusion in determination of acceptance ranges  Follows what had been PhRMA procedure  Statistical analysis determines three variances –  apparatus/tablet/assay  between-experiment, with-in laboratory  between-laboratory

13. Sample Xbar and S Control Charts for Apparatus 2, 30 Minutes

14. Representative CV’s (%) CV total CV between -lab CV within -lab CV between + within- lab CV apparatus/ tablet/ assay Prednisone - basket 4.52.53.73.23.5 Prednisone - paddle 10.08.13.07.06.7 Sal’c acid - basket 4.22.81.83.12.7 Sal’c acid - paddle 7.45.12.85.2

15. Comments  Variability due to the combination of apparatus, tablet, and assay is low, particularly for the basket (Apparatus 1)  Assay variability around 2% CV, so much of the contribution from apparatus/tablet/assay is due to the assay

16. Determination of Acceptance Ranges  Acceptance range based on sum of the three variances  Intended to represent range of values to be expected from a random tablet tested at a random laboratory using good practice

17. Issue for Acceptance Ranges  Ranges determined for a single tablet, but applied to sets of 6 and must pass all 6  Possible solutions  Standard statistical multiple testing adjustment; i.e., widen the intervals to 99% from 95%  Allow retesting  Set acceptance ranges on sample mean and standard deviation as done in collaborative study; i.e., would the data qualify for inclusion in the determination of the acceptance range

18. Topics  Systems Suitability Studies for Dissolution  Collaborative Study Design and Analysis  Variances: Correction to May 2005 Discussion  Next Steps

19. Calibrator to Control Product Variability  Total product variability, σ 2 total = σ 2 product + σ 2 measurement  σ 2 total for calibrator = σ 2 calibrator + σ 2 C*measurement  Since σ 2 calibrator is not known use σ 2 total for calibrator  σ 2 total for calibrator is given in the acceptable ranges  σ 2 total = σ 2 product + calibrator range Argument against USP RS in dissolution system suitability delivered at various venues

20. Restating Concerns About Calibrator Variability  This argument does not hold; maybe σ 2 total < σ 2 product + calibrator range  Calibrator variability does not contribute to product variability

21. Topics  Systems Suitability Studies for Dissolution  Collaborative Study Design and Analysis  Variances: Correction to May 2005 Discussion  Next Steps

22. Next Steps  Continued work to improve chemical calibration, e.g., approach to avoid ‘multiple testing’  Acknowledge that dissolution is ‘case by case’—resolved with ‘flexible monograph’—differences acceptable if bioequivalent  Assess various approaches to Systems Suitability for the dissolution procedure; e.g., develop methodology to measure fluid flow that is applicable in QC environment (engineering approach)  Setting acceptance criteria for marketed dosage forms (not calibrators): tolerance interval approach based on clinical trial batches (see Hauck et al., Pharmaceutical Research 2005)  USP willing to work with all stakeholders on complex science/technical issues for Performance test; ‘Quality by Design’ approaches should be highly valuable.