The author explains the idea of equivalence and describes how it can facilitate equipment validation.
It is common in the global healthcare industry to have multiple pieces of identical equipment available for the purposes of added capacity and redundancy. These circumstances provide opportunities to streamline qualification and validation activities. When several pieces of equipment are identical in all respects, the qualification effort should seek to demonstrate their basic interchangeability for all uses to reduce the needless repetition of activities. The qualification protocols should identify essential performance criteria for the equipment that each unit must meet to demonstrate its equivalence. The criteria used for this evaluation should be formal, quantitative, reasonably tight, and realistic (the performance of a single piece of equipment will vary over time).* Once equivalence has been demonstrated during the qualification effort, subsequent performance-qualification efforts should be divided between the pieces of equipment to reduce the number of studies that would otherwise be required.
The principles of equivalence can be adapted to several other instances in processing and analysis (e.g., containers, materials, formulations, analytical instruments, and personnel) where basic similarities in performance can be exploited to simplify the overall effort. The principle of equivalence is relevant to even singular instances. The fundamental precept of the US Food and Drug Administration's draft process-validation guidance requires that firms demonstrate the consistency of the process at various scales from development, through scale-up, and continuing into commercial manufacturing (1). At the core of that guidance is an expectation for process equivalence during the course of the initial effort and recommendations for controls that will ensure consistent production over the product's life cycle. The guidance implies the expectation for equivalent performance over scale and time; however it is unfortunately not explicit.
Other citations in the regulatory arena refer to equivalence in the execution of validation. These citations also are somewhat more implicit. In a 1994 Warning Letter, FDA indicated the basic requirements for equivalence:
It is FDA's position, however, that while it is possible to rely on validation data from one chamber to represent that of another, it is only possible to do so for chambers at the same location which are identical in all respects. That means the chambers are of identical construction and installation (i.e., identical plumbing, characteristics, steam supplies, operating environments, etc.) and the product(s) to be sterilized are equivalent in all respects (2).
This letter serves as perhaps the clearest statement with respect to equivalence ever made by FDA. As the letter applies to a sterilization process, one of the more crucial processes requiring validation, industry has every reason to believe that the agency would take a similar view with respect to less crucial process equipment.
FDA's guidance document on revisions to new drug applications (NDAs) and abbreviated new drug applications (ANDAs) come close to touching on the subject of equivalence as it relates to equipment, but focus on the performance of the drug (3). This document and the individual scale-up and postapproval changes guidance documents clearly imply that when individual machines are identical, the implications for process equivalence are clearer and more certain (4).
The cited NDA–ANDA guidance focuses on the potential effect of process changes on the drug product. The following excerpt provides insight into FDA's general expectations but, unfortunately, does not offer quantitative criteria to be satisfied:
When testing is performed, the applicant should usually assess the extent to which the manufacturing change has affected the identity, strength, quality, purity, and potency of the drug product. Typically this is accomplished by comparing test results from pre- and post-change material and determining if the test results are equivalent. Simply stated: Is the drug product made after the change equivalent to the drug product made before the change? An exception to this general approach is that when bioequivalence is re-documented for certain ANDA post-approval changes, FDA recommends that the comparator be the reference listed drug. Equivalence comparisons frequently have a criterion for comparison with calculation of confidence intervals relative to a predetermined equivalence interval. For this, as well as for other reasons, equivalent does not necessarily mean identical. Equivalence may also relate to maintenance of a quality characteristic (e.g., stability) rather than a single performance of a test (3).
Elsewhere in the NDA guidance, equipment elements are discussed in terms of their influence on bioequivalence. The following passage appears later in the same NDA guidance:
Minor Changes (Annual Report)—The following are examples of changes considered to have a minimal potential to have an adverse effect on the identity, strength, quality, purity, or potency of a drug product as these factors may relate to the safety or effectiveness of the drug product.
1. For drug products, changes to equipment of the same design and operating principle and/or changes in scale except as otherwise provided for in this guidance (3).*
This reference is actually more flexible, in that it does not use the word "identical," but only mentions shared design and operating principles. Thus, equivalence in equipment may be more accepted than is ordinarily recognized. It may embrace different vendors, and sometimes encompasses different sizes of equipment.
Additional statements about equipment equivalence appear in various sterilization documents from the International Organization for Standardization. These statements are relevant for medical devices, but pharmaceutical and biotechnology firms seem rarely to consider them (5, 6). Nevertheless, the clear consensus is that demonstrating equivalence as it relates to process equipment is acceptable to regulatory authorities. But current guidances or standards that discuss equivalence lack a clear set of expectations for the demonstration of equivalence between multiple equipment units. The NDA and ANDA guidance document cites bioequivalence, but usually in terms of testing product that is many steps removed from where equivalence is being sought. The effect of essential unit operations such as sterilization, cleaning, and early process steps on bioequivalence is difficult, if not impossible, to discern. For these applications, one must consider how equivalence might be demonstrated independently of bioequivalence. Moreover, as stated consistently in the NDA–ANDA guidance document, bioequivalence is unlikely to be influenced meaningfully by equivalent, if not actually identical, equipment.
Approaches to bracketing in process validation
Based on the clear support that equivalence seems to have, whether explicit or implicit, one might assume that the concept is in widespread use. Regrettably, industry seems to be reluctant to use it to design validation programs. When industry uses equivalence, little public information about its application and utility is available. This article draws extensively on the author's experience because of the paucity of published information about applying equivalence to the execution of validation studies. The following list of projects used the equivalent performance of multiple pieces of identical equipment to reduce the overall validation effort.
Nine identical ovens. One company used nine ovens to depyrogenate vials of 14 sizes. If equivalence were ignored, the number of studies required to fully validate minimum and maximum loads would be 756 runs (9 × 14 × 3 × 2). Equivalence and bracketing (see sidebar, "Approaches to bracketing in process validation") reduced the number of runs to 126 (9 × 2 × 3 + 12 × 2 × 3). The entire effort comprised the largest and smallest vials in all ovens in both load sizes, plus single-minimum and maximum load-size runs of the other 12 vials in one or more of the other ovens. Equivalence reduced the workload by 83%.*
Paired terminal sterilizers. A company planned to use identical units to sterilize 15 vial sizes filled with 32 formulations. A total of 56 possible product–container combinations needed to be validated. The validation was expected to cover minimum, maximum, and intermediate-sized loads. Without equivalence, the number of required validation runs to be completed would be 1008 (2 × 56 × 3 × 3). Extensive sterilizer-qualification, cycle-development, and water-challenge (covering five different vial sizes) runs were used to demonstrate the sterilizers' equivalence and to support load-size variation across the breadth of containers. The initial product validation of the sterilizers consisted of 54 biochallenge runs selected to broadly confirm the acceptability of the process for several of the formulations. None of the product loads were tested in triplicate. These data and those from the water-challenge runs, in conjunction with D-value determinations, was used to obtain initial facility approval. This information was supplemented postapproval with physical monitoring that reaffirmed the physical parameters for the remaining products, containers, and load sizes. The overall reduction in workload achieved through the use of equivalence was approximately 95%.
Other examples. In addition to the applications described above, the author has participated in the following validation projects in which equivalence reduced the overall effort:
* Terminal sterilization
* Part sterilization
* Clean- and steam-in-place of fixed and portable tanks
* Vial and stopper washers
* Tablet presses
* Coating pans and columns
* Granulation vessels
* Shelf, fluid-bed, and freeze dryers
* Compounding of oral and injectable solutions
* Mills and sieves
* Filling, packaging, labeling, and inspection equipment.
The abovementioned items represent the author's experiences and are not intended to be all-inclusive. One might reasonably expect to demonstrate equivalence for other items of equipment not indicated above. The increased use of automated monitoring and control in contemporary equipment makes it easier to demonstrate equivalence.
The keys to equivalence
Central to the use of equivalence is the establishment of criteria for the evaluation of the equipment's operational performance. This evaluation demonstrates that individual units perform identically. Criteria must be chosen with some restraint; adding too many criteria or nonessential criteria is unlikely to add meaningful value and more likely to result in the conclusion that the equipment is not equivalent. Common sense indicates that the criteria should focus on the crucial aspects of the equipment's performance. For sterilization processes, that most important criterion would customarily be F 0.* Other criteria in this determination for a sterilization process would add no meaningful value because these processes are intended to achieve lethal conditions. Criteria for other processes could be defined by their key performance measures (e.g., content uniformity, potency, fill weight, and residue level). Another possibility is to use a defined and calibrated test set as a standard for assessment. Test sets are ideally suited for attribute, foreign-particle, and label-inspection equipment and systems.
Regardless of what criteria are chosen, it is appropriate to establish in advance an acceptable range for equipment performance. In setting this range, one should consider the following factors:
* The inherent analytical variability or measurement sensitivity used to assess the results
* The normal variation in the underlying process
* The limitations of sampling.
These factors should be first evaluated on a single piece of equipment in replicate studies in which these and other factors are assessed for their influence on the results. A single piece of equipment tested in replicate studies during a brief interval must be considered equivalent to itself. The actual variations between multiple items should be somewhat, but not dramatically, larger. In the absence of an independent assessment, the author suggests a range of not more than ±10% of the target value, which itself is an average of multiple preliminary runs.
Should the firm set overly restrictive criteria or the equipment prove not to be equivalent, nothing is lost. The accumulated data must be augmented with whatever added studies are necessary to complete the overall exercise without relying on the simplifying effect that equivalence can afford.
An aspect of equivalence that has largely been ignored is that of time. It is essential that any process provide consistent results over time, whether the process is performed with a set of equivalent equipment or in a single piece of process equipment. This principle might seem obvious, but FDA codified that expectation in the draft revision of the Guideline on General Principles of Process Validation (1). The expectation that a firm can establish equivalence across multiple items at a single point in time is wholly appropriate for the same firm using a comparable approach to demonstrate the equivalence of the same piece of equipment over several points in time. This is a core expectation of the revised guidance and should raise awareness of equivalence as a regulatory expectation when FDA finalizes that document.
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 validation studies.
James Agalloco is president of Agalloco and Associates, PO Box 899, Belle Mead, NJ 08502-0899, tel. 908.874.7558, firstname.lastname@example.org [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.
*F 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).