The use of statistics
For several reasons, this article has mentioned statistics only in passing. First, in applications where the author initially
endeavored to demonstrate equivalence, he did not apply statistics beyond averages because complex tools were simply unavailable.
Second, having demonstrated equivalence successfully without statistics, the author was loath to introduce it later on. Third,
statistical requirements proved tighter than the process's capability in at least two instances. The implied direction of
FDA's draft process-validation guidance leaves little doubt that statistics likely will play a larger role in future equivalence
demonstrations. The author prefers to leave the selection of appropriate statistics to experienced individuals.
Validation is an essential component of operations in the pharmaceutical industry that conform to current good manufacturing
practices, and one can expect it to remain a focus of regulatory attention. Validation has often been considered a costly
and troublesome activity and one badly in need of improvement. Adherence to future regulatory guidance undoubtedly will mandate
new validation studies. Equivalence affords an opportunity to realize a degree of economy in executing those expected new
James Agalloco is president of Agalloco and Associates, PO Box 899, Belle Mead, NJ 08502-0899, tel. 908.874.7558, email@example.com
He also is a member of Pharmaceutical Technology's editorial advisory board.
*Given the material and procedural variation, uncertainties of sampling, and inherent variability in analytical methods, even
a nonvarying piece of equipment will demonstrate performance variation because of limitations in the method of measurement.
*Both statements imply that the production of natural protein-drug substances and drug products require additional caution.
*Bracketing is the practice of interpolating between extremes with respect to variables such as size, dimension, and potency
under the assumption that intermediate situations will provide conforming results. Bracketing is commonly used and accepted
independent of equivalence for the purposes of simplifying validation activities, but is also usable in conjunction with it.
0 is coincidently defined as the equivalent sterilization time related to the temperature of 121 °C and a z value of 10 °C. It thus is uniquely appropriate for equivalence studies for steam-sterilization processes.
1. FDA, Draft Guideline on General Principles of Process Validation (Rockville, MD, Nov. 2008).
2. FDA, Warning Letter to Sterile Recoveries, Nov. 4, 1994.
3. FDA, Guidance for Industry: Changes to an Approved NDA or ANDA (Rockville, MD, Sept. 2004).
4. FDA, SUPAC-MR: Modified Release Solid Oral Dosage Forms—Scale-Up and Postapproval Changes: Chemistry, Manufacturing, and Controls;
In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation (Rockville, MD, Sept. 1997).
5. ISO, "ISO 14160—Liquid Sterilant Validation," (ISO, Geneva, 1998).
6. AAMI–ISO, "DS/ST63/2002-05-01—Sterilization of Health Care Products—Requirements for the Development, Validation, and Routine Control of an Industrial Sterilization Process for Medical Devices—Dry Heat" (AAMI, Arlington, VA, 2002).