Darrell R. Abernethy, MD, PhD, is chief science officer, all at the United States Pharmacopeial Convention, 12601 Twinbrook Parkway, Rockville, MD 20852-1790, tel. 301.816.8184, dra@usp.org

Darrell R. Abernethy

The United States Pharmacopeial Convention (USPC) asked the Institute of Medicine (IOM) of the National Academy of Sciences to convene a workshop at which stakeholders, including pharmaceutical manufacturers, could discuss ways to update 

General Chapter <231> "Heavy Metals" (1). IOM responded positively and formed a planning committee composed of national and international experts in metal measurement and toxicity.  

The workshop took place at IOM Aug. 26–27, 2008. In an effort to promote harmonization and at USP's request, IOM invited compendial experts from Europe and Japan to attend. Because 

 <231> is relevant to food ingredients (standards are provided in USP's 

) and dietary supplements (standards are provided in 

's dietary supplement section), IOM also invited stakeholders from those sectors. 

 <231> relies on classical (i.e., wet) chemistry procedures that have been in the compendium for more than 90 years. The test suffers from a lack of instrumental output and is a nonspecific, subjective test based on the precipitation of metal sulfides from a solution with visual comparison to the color of a standard solution (2). Several groups have shown that the method routinely underestimates the levels of the few metals it was designed to detect (2, 3). Certain toxic metals thus may appear in drug products and their ingredients because of their use as catalysts or starting materials. Other metals such as arsenic, cadmium, lead, and mercury are highly toxic and ubiquitous in the environment. Not surprisingly, many of these metals also have been detected during market surveillance of dietary supplements and food ingredients (4). 

Workshop deliberations resulted in a consensus that the current classical chemistry procedure in 

 <231> is no longer suitable. Participants agreed that the current procedure should be replaced by specific and sensitive instrumental procedures that detect and quantify the broad range of metals and elements of interest in pharmaceuticals, dietary supplements, and food ingredients. 

Modern instrumental methods can identify and quantify at low levels many metals that currently are not detected by 

 <231>. Because many metals may not pose a public health risk when consumed for a brief time or at low levels, the challenge for participants was to identify metals of interest, and levels that should be monitored, especially considering the broad range of exposures posed by articles such as food ingredients. 

For certain metals (e.g., arsenic, cadmium, lead, and mercury) analysis was relatively straightforward. These substances have known toxic effects and are potential contaminants. Thus, analytical methods should be validated to detect arsenic, cadmium, lead, and mercury. The methods should detect the metals at toxicologically relevant concentrations; and pharmacopeial monographs for these metals should be revised. 

Workshop participants grappled with the questions of which metals should be monitored and to what level of detection. The European Medicines Agency (EMEA) has released a 

Guideline on the Specification Limits for Residues of Metal Catalysts

 (5). Because the metals specified in the guideline are sometimes used as catalysts or processing agents and can appear as contaminants in drug substances and drug products, EMEA recommends that pharmaceutical manufacturers test for them at levels commensurate with their toxicity.  

Workshop participants identified various reagents that may require monitoring and contain elements such as aluminum, beryllium, boron, chromium, cobalt, copper, gallium, germanium, gold, indium, iron, lithium, magnesium, manganese, nickel, silicon, silver, tin, thallium, titanium, and zinc. 

Speciation (i.e.,  valence state of the metal or its existence as an organic complex) can play an important role in toxicity. The cases of arsenic and mercury are instructive: Inorganic arsenic compounds are typically highly toxic. Dietary supplements that contain kelp or certain other nontoxic materials also contain high concentrations of essentially nontoxic organoarsenic compounds. Similarly, metallic mercury is relatively nontoxic, but inhaled mercury metal, methyl mercury, and similar organomercury compounds that can be concentrated in foods such as fish are highly toxic. Thus, workshop attendees concluded that tests for arsenic or mercury  by themselves may not yield suitable data on which to base specifications. Further, reports of total arsenic or total mercury content may unnecessarily confuse consumers about the safety of the products they consume, leading the public to question the efficacy of regulatory standards. 

 Participants agreed that several related factors must be considered when evaluating the acceptable rate and extent of exposure in vulnerable populations. Humans may experience degrees of environmental exposure to metals.  

Another consideration is the duration of exposure. Long-term exposure to lead is associated with decreased intelligence quotient, and long-term exposure to organic mercury causes neuropathic effects. Concurrent exposure to both metals has not been well studied, but risk-assessment models suggest incremental risk.  

 If analysts do not separate metal forms (i.e., perform speciation) before undertaking metal analysis, the results will report the total of all forms detected (e.g., the method will report both inorganic and organic arsenic). Workshop participants did not reach a consensus about the level of detection required for each metal. A reasonable approach might involve the development of pharmacopeial specifications for the total amount of an individual metal at some limit (worst case), and requiring manufacturers to use speciation to justify a higher limit. 

Workshop participants decided that the following tests are probably suitable for compendial metal analysis: electrothermal atomic absorption spectrometry, inductively coupled plasma–optical emission spectrometry, inductively coupled plasma–mass spectrometry, and any other methods with similar specificity and sensitivity. 

 <231> will rely on reference materials to demonstrate recovery (i.e., precision and accuracy) during sample preparation and analysis. Reference materials also are needed to establish systems suitability and quantitation or detection limits, particularly when several technologies may be used to produce acceptable results. 

Workshop participants advocated the revision of 

 <231>. Further consideration of limits for various metals will be required. Arsenic, cadmium, lead, and mercury were singled out by participants for particular attention. Other metals used in manufacture or present in starting materials will be considered using a risk-based approach. USP will continue to work with stakeholders to update General Chapter <231> using harmonized approaches and a public process of review and comment. 

 is executive vice-president and chief executive officer, 

 is chief science officer, all at the United States Pharmacopeial Convention, 12601 Twinbrook Parkway, Rockville, MD 20852-1790, tel. 301.816.8184, 

. *To whom all correspondence should be addressed. 

 26 General Chapter <231>, "Heavy Metals," pp. 133–134. 

 2.  T. Wang et al., "An Atomic Spectroscopic Method as an Alternative to Both 

 3.  USP Ad Hoc Advisory Panel on Inorganic Impurities and Heavy Metals, "General Chapter on Inorganic Impurities: Heavy Metals," 

 4.  D.R. Abernethy et al., "Adulteration of Drugs and Foods: Compendial Approaches to Lowering Risk," 

Guideline on the Specification Limits for Residues of Metal Catalysts 

Articles

Attendees at a recent workshop endorsed new methods to detect metals in drugs, dietary supplements, and food ingredients.

Inside USP: USP Metals Testing: A Workshop Report

In October 2007, the USFood and Drug Administration released a draft 

Guidance for Industry: Use of Mechanical Calibration of Dissolution Apparatus 1 and 2—Current Good Manufacturing Practice

 and called for public comments (1). This article summarizes the basis for the US Pharmacopeia's (USP) request to withdraw the draft guidance, provided in a detailed response to the public docket (2). 

The intent of the USP performance verification test (PVT), which is used for nonsolution orally administered drug products, is to ensure continuing equivalence between the clinical trial material on which safety and efficacy conclusions were made and the manufactured article following approval. US regulatory approaches do not require postapproval reconsideration of bioequivalence for drug products, barring postapproval change. Thus, reliance on USP becomes a key if not the sole means of ensuring consistency in a drug product's performance.  

 General Chapters ‹701› Disintegration and ‹711› Dissolution, and other chapters provide general procedures with acceptance criteria that can be adapted by manufacturers to specific products (3, 4). Data arising from the adaptation undergo regulatory review and, if approved, become one of the tests in the private drug product specification. 

The dissolution procedure described in ‹711› requires an assembly that allows a kinetic measure of drug release over time. Combining effects from the analyst and analytic procedure, the procedure is usually performed at batch release and also as part of stability studies. The assembly is a complicated mechanical device and includes many factors that can influence results. For this reason, USP, and until recently, FDA have emphasized the importance of a periodic PVT* together with careful mechanical calibration to ensure that the combined experimental study yields consistent results. 

FDA's draft guidance argues that only mechanical calibration is needed. Based on substantial research and development efforts, USP maintains instead that both PVT 

 mechanical calibration are critical to ensure the integrity of the dissolution procedure and that mechanical calibration alone cannot ensure the validity of dissolution results (6–15). The dissolution test system consists of the mechanical apparatus (itself composed of several components), the physical environment in which the apparatus exists, the analytical procedure, and the analyst. Mechanical calibration addresses and controls only some of the components of the apparatus. 

 Fundamental metrology concerns the establishment of measurement units, the realization of measurement standards, and the transfer of traceability from these standards to users in society. Applied to manufacturing, metrology ensures the suitability of measurement instruments, their calibration, and quality control of measurements. In the legal sense of the term, metrology enables regulatory requirements for measurements and measuring instruments for the protection of health, public safety, and the environment, and supports decisions regarding protection of consumers and fair trade. A core concept in metrology is traceability, defined as "the property of the result of a measurement or the value of a standard whereby it can be related to stated references, usually national or international standards, through an unbroken chain of comparisons, all having stated uncertainties" (16). The level of traceability establishes the level of comparability of the measurement between results in any and all laboratories that conduct the procedure. USP's tests and procedures are developed with this objective in mind and form the basis for the legal recognition of USP's standards of strength, quality, and purity in the adulteration provisions of the Federal Food, Drug, and Cosmetic Act (17). USP thus works to ensure that manufacturers, for commercial purposes, and FDA, for regulatory purposes, have access to procedures that achieve a high degree of assurance in test results without regard to time and space.† 

The overarching goal of metrological traceability is the transfer of accuracy (trueness) and precision throughout the measurement "ladder." The two fundamental transfer agents of this ladder are procedures with specified methods and reference materials. USP's monograph procedures are frequently grounded in reference materials (e.g., official USP reference standards), with the understanding that the relationship between the procedure for the measurand (the material being examined) is closely linked to the reference material reflective of that measurand. At times, USP uses informally the terms "vertical standards" for the article under test and "horizontal standards" for the monograph's procedures. A key example of horizontal documentary and reference material standards is the dissolution procedure, described in ‹711›. There is a metrologic aspect to the dissolution procedure because a concentration is determined and can be traced to the base SI units of mass (kilogram). USP, however, notes that the kinetic aspects of the dissolution procedure are not well-grounded in metrology (i.e., there is no absolute number from the kinetic study that can be traced to an SI unit). In this regard, the dissolution procedure has the kinetic character of an enzyme assay that yields results in units per mass rather than in SI units alone. This does not mean that such grounding could not be improved, nor does it mean that the procedure is without value. The complex character of an assembly in fact requires careful consideration of horizontal standards to ensure acceptable results.  

USP endorses the concept of mechanical calibration (grounded in metrology), conducted at periodic intervals, to ensure that the mechanical components meet specifications and are in a state of control. These mechanical checks, however, are necessary but not sufficient because they ignore the chemical and kinetic aspects of the dissolution procedure, an oversight that is common in most chemical tests. For example, even the best-calibrated high-performance liquid chromatography equipment still needs systems suitability standards to ensure proper performance within and across laboratories. In one of the most common chemical measurement tests performed, pH, it is not sufficient to ensure only that the electronics of the meter (essentially a voltmeter) are functioning. Rather, there must be assurance that the entire system (the meter, the glass electrode, the counter electrode, and the analyst) is performing adequately through the use of pH buffer standard solutions. Traceability, accuracy, and comparability can then be ensured. USP reference standard tablets used in PVT help ensure that the complex character of a dissolution assembly, coupled with an analyst and analytical procedure, act in concert to yield results within and among laboratories with acceptable accuracy and precision.  

USP has established that mechanical calibration of a dissolution assembly does not address adequately all factors (e.g., vessel symmetry) that influence dissolution results (9, 13). Recent USP research also leads to the conclusion that the current tolerances and specifications set for mechanical calibration are inadequate and must be tightened (e.g., 7, 9). Even if mechanical calibration could be brought to complete realization (i.e., control of all possible impacting factors through conformance to specified tolerances), a PVT would still be needed to test the system as a whole and in actual use. Control of each factor individually might be possible, but that would not mean that all elements working in concert would yield reproducible and comparable results.  

. In the terminology of the International Organization for Standardization (ISO), USP's chemical reference standard tablets used in PVT are elements of a proficiency test whereby a laboratory may compare its results to those from other laboratories (18). USP emphasizes the value of these publicly available reference standards that allow PVT within and among laboratories to ensure that all laboratories achieve similar results in the interests of setting a metrologically sound and publicly derived USP performance test in the drug product specification. 

 Ensuring the quality of data generated by analytical equipment includes an overall approach to equipment quality. Pharmaceutical scientists have long accepted that equipment qualification—installation qualification (IQ), operational qualification (OQ), and performance qualification (PQ)—under good manufacturing practices (GMPs) should be executed to verify acceptable quality of output from multicomponent analytical instrumentation. Although practitioners may debate where in the IQ, OQ, and PQ process individual qualification activities should take place, the fundamental principles are indisputable. Each component of the verification process contributes to overall quality of data generated from multicomponent analytical instrumentation, and omission of any component will compromise the quality assurance of data (19). On the basis of its research, USP concludes that mechanical calibration must be coupled with PVT to satisfy GMP requirements for OQ and PQ, respectively. 

FDA has claimed that the USP prednisone RS tablet exhibits poor quality. The first time this claim came to USP's attention was during presentations by FDA staff at USP's 2004 Annual Science Meeting in Iselin, NJ, and again at meetings of the FDA Advisory Committee for Pharmaceutical Science in May and October 2005. USP refutes the claim based on studies conducted over the past several years (6). USP's Lot P prednisone RS tablet exhibits variability that is less than that of commercially marketed therapeutic drug products but is about the same as that of FDA's NCDA #2 tablet lot (20, 21).  

In studies cited in this commentary (6–15, 20) and in various internal and external discussions with its Expert Committee members, stakeholders, and others, USP has determined that a PVT performed at periodic intervals, in addition to mechanical calibration, is required to ensure the integrity and reliability of the dissolution procedure when referenced in the USP performance test and in private regulatory drug product specifications. USP believes that additional work is needed to improve its General Chapters that function to ensure the consistency of drug products and expand them to products other than nonsolution orally administered dosage forms. USP also is in accord with an often stated view that new approaches to assess drug product performance 

In the long term, USP envisions an evolved set of approaches to assess drug product performance for all nonsolution dosage forms over time following regulatory approval. Although the use of a PVT with some kind of physical article will be needed in the short term, these newer approaches could offer better public health protection and ease of use by pharmaceutical manufacturers. It is important not to abandon current approaches that have worked well before new opportunities become available. USP hopes that this work can advance jointly with FDA, industry, and other stakeholders. 

 General Chapter ‹711› has been revised, replacing the term "apparatus suitability test" with "performance verification test" (5). Although USP reference standard tablets have been referred to as calibrators, this is a misnomer. The tablets are not, and cannot, be used for calibration.  

 † The argument has been made that a manufacturer can make its own physical artifact (an in-house reference material standard) for a PVT. Although such standards may work locally, this approach is inconsistent with national and international measurement systems, where comparability and consistency across time and space are essential. This approach is also untenable if the goal is to align laboratories among manufacturers and between manufacturers and regulatory laboratories. The task of developing a reliable and useful reference material tablet is not an easy one and would require much duplication of effort. 

, is chair of the USP Biopharmaceutics Expert Committee. 

, is chief reference materials officer, and 

, is senior scientific fellow, all three at USP, 12601 Twinbrook Parkway, Rockville, MD 20852-1790, tel. 301.816.8141, 

*To whom correspondence should be addressed. 

, Office of Regulatory Affairs (Rockville, MD), 2007.  

2.   Public Docket 2007D-0365, available at 

frwebgate.access.gpo.gov/cgi-bin/getpage.cgi?dbname=2007_register&position=all&page=59298

‹701›, US Pharmacopeial Convention (Rockville, MD, 2007), pp. 276–277. 

‹711›, US Pharmacopeial Convention (Rockville, MD, 2007), pp. 277–284. 

5.    USP, "Interim Revision Announcement, Dissolution ‹711›," 

6.    G. Deng et al., "The USP Performance Verification Test, Part I: Quality Attributes and Experimental Variables Contributing to Dissolution Variance," 

www.springerlink.com/content/j8246315n5404j18/?p+8a0886888caf41b4b69722da48bfd8&pi=0

7.    M. Glasgow et al., "The USP Performance Verification Test, Part II: Collaborative Study of USP's Lot P Prednisone Tablets," 

www.springerlink.com/content/b3726195727184gl/?p=8a0886888caf41b4b69722da48bfb5d8&pi=1

8.   P. Nithyanandan et al., "Evaluation of the Sensitivity of USP Prednisone Tablets to Dissolved Gas in the Dissolution Medium Using USP Apparatus 2," 

9.   J. Eaton et al., "Perturbation Study of Dissolution Apparatus Variables—A Design of Experiment Approach," 

10.  M.R. Liddell et al., "Evaluation of Glass Dissolution Vessel Dimensions and Irregularities," 

11.  W.W. Hauck et al., "Proposed Change to Acceptance Criteria for Dissolution Performance Verification Testing," 

12.  W.W. Hauck et al., "Reliability and Reproducibility of Vertical Diffusion Cells for Determining Release Rates from Semisolid Dosage Forms," 

13.   M. Liddell et al., "Dissolution Testing Variability: Effect of Using Vessels from Different Commercial Sources," 

14.  R.G. Manning et al., "Dissolution Testing and Metrological Measurement of Quality for Solid Oral Dosage Forms," 

15.  G. Deng et al., "Frequency-domain Vibration Measurement and Analysis in Dissolution Testing," poster presented at the 2007 Annual AAPSMeeting (San Diego, CA). 

16.  International Bureau of Weights and Measures et al., 

International Vocabulary of Basic and General Terms in Metrology, 

Second Ed., ISO (Geneva, Switzerland, 1993), p. 47. 

Standard 5725 Accuracy (Trueness and Precision) of Measurement Methods and Results

, Parts 1–6, ISO (Geneva, Switzerland, 1994). 

19.  FDA, "ORA Laboratory Procedure. Assuring the Quality of Test Results. ORA-Lab.5.9," 2007, accessed Jan. 29, 2008.  

20.  P. Nithyanandan et al., "Dissolution Variability: Comparison of Commercial Dosage Forms with US Pharmacopeia Lot P Prednisone Reference Standard Tablets," 

21.  Z. Gao et al., "Gauge Repeatability and Reproducibility for Accessing Variability during Dissolution Testing: A Technical Note," 

The United States Pharmacopeia emphasizes mechanical calibration and a performance test to esnure integrity of the dissolution procedure.