Paula Brown & Dr. Joseph Betz04.01.10
In the January/February “Quality Focus” column, “A Riddle, Wrapped in an Enigma, Shrouded by Mystery,” the challenges in botanical identity testing were examined and readers were left with the following advice in a cliffhanger closing: “…whatever method you use (TLC, LS/MS, UV-Vis, Near FTIR), the key will be to try to capture data from as many authenticated reference materials as possible. So, what are and where do we get authentic reference materials…next time.”
Although cliffhangers often frustrate readers by offering no resolution, the intent was to impart the Zeigarnik effect with the hopes that this unfinished task would be better remembered. In keeping with that line of thought, botanical reference materials will not be addressed just yet, but rather we will look into another topic that falls within the category of reference materials: the deceptively simple chemical calibration standard.
The Plot
Calling on support from my colleague, Dr. Joseph Betz, this column will delve into the complexity of chemical reference standards, also called chemical calibration standards. In a few short paragraphs we will attempt to explain the following:
• What are they?
• Where can we get them?
• When and how should we use them?
• What role do they play in quality control? Regulatory compliance?
The Cast
According to AOAC Official Method 892.35, a Reference Material (RM) is defined as material or substance, of whose property values are sufficiently homogeneous and well established to be used for calibration of an apparatus, the assessment of a measurement method, or for assigning values to materials. This definition is contextual in that it describes how an RM is used, but not much about what it is.
AOAC provides a second definition: material, sufficiently homogeneous and stable with respect to one or more specified properties, which has been established to be fit for its intended use in a measurement process. So you have to prove the material is suitable for the way you want to use it. This definition also comes with a few footnotes that address the use of an RM: calibration of a measurement system, assessment of a measurement procedure, assigning values to other materials and quality control. Just as important, an RM cannot be used for more than a single purpose in a given measurement (General and Statistical Principles for Certification, ISO Guide 35, 3rd Ed., 2006). Loosely translated—no multitasking allowed! For example, a material can’t be assigned a value based on an analysis, but also be used as the calibrant in the analysis; that is called “self-referencing.”
And for one more opinion, The International Vocabulary of Basic and General Terms in Metrology, (2nd ed., ISO 1993) describes an RM as, “Material or substance one or more of whose property values are sufficiently homogeneous and well established to be used for the calibration of an apparatus, the assessment of a measurement method, or for assigning values to materials (ISO VIM, 6.13, 7).”
In all these definitions the key descriptors are “sufficiently homogenous” and “well established.” The values that must be assigned? Identity and purity.
How do you determine if your material fits the bill? How does one go about qualifying a reference material as defined above?
Guidance can be found in ISO guide 35 (Reference materials-General and statistical principles for certification), which is meant to provide statistical principles to assist in the understanding and development of valid methods to assign values to reference material. It pretty much says that the requirements are that you must take all the steps necessary to demonstrate that the values you assign are correct.
Wow. Thanks guys.
Behind the Scenes
Moving on to the U.S. Dietary Supplement GMPs, it takes a bit of digging but reference materials are mentioned.
In 21 CFR Part 111 Subpart D-Equipment and Utensils, Section 111.25 asks the question “What are the requirements under this subpart for written procedures?” The agency answers that you must establish and follow written procedures for (a)
Calibrating instruments and controls that you use in manufacturing or testing a component or dietary supplement; and (b)
Calibrating, inspecting, and checking automated, mechanical and electronic equipment…” Further, Section 111.27 “What Requirements Apply to the Equipment and Utensils That You Use?” goes on to note that you must calibrate instruments and controls you use in manufacturing or testing a component or dietary supplement. Section 111.35 says you must have written procedures for performing these calibrations and that you are required to keep records of the results of the calibrations.
Basically FDA expects you to calibrate and keep records about your calibrating
activities. What in a facility might need calibration? Just about anything and everything, including drying ovens, refrigerators, freezers, fans, scales, pipets and pH meters. Consider a basic HPLC-UV; the autosampler, oven and lamp all must be calibrated and you haven’t even run a method yet!
Tools and/or instruments are employed to make sure that when specifications are evaluated and measurements made, they be done accurately. Of course you must make sure these tools/instruments are working correctly by performing calibrations. For example, you can calibrate a balance using NIST certified weights. And all volumetric glassware must be calibrated too.
Section 111.35 (b)(3) tells you what you need to include in your calibration documentation. Part 111.35 (b)(3)(iii) states you must “Identify the reference material used including the certification of accuracy of the known reference standard and a history of recertification of accuracy.” And finally, Section 111.315 of Subpart J (Production and Process Control System: Requirements for Laboratory Operations) says that you have to “establish and follow laboratory control processes that are reviewed and approved by quality control personnel, including … (c) Use of criteria for selecting standard RMs used in performing tests and examinations.”
Now we know that 21 CFR Part 111 has a word or five to say about RMs.
Dialogue
The sections within regulations relating to calibration and reference materials encompass some pretty complicated concepts. Starting with 111.315 (c), the QA unit has to justify why they chose the calibration substance(s) they use to perform calibrations. The FDA inspector will expect a QA person to provide a rationale for choosing a particular RM such as: it was the cheapest; best characterized; purest; most readily available; most stable; or some combination of the above. There is no right answer for all situations but there are more appropriate and less appropriate choices for a particular measure.
For example, when measuring anthocyanins in berries, selecting a single aglycone, such as cyanidin, to calibrate your method may be appropriate if the other materials aren’t available and you have another identity method to ensure no adulteration occurs or the materials are available but you have demonstrated the suitability of the surrogate and adulteration is not likely. However, if the glycosides themselves are cheap and readily available, you may have a difficult time justifying your choice.
Be prepared when conducting an indirect (or non-specific) analysis to back up your selection justification with data— chromatographic resolution, extinction coefficient similar to target compound(s) or well-researched correction factor (keeping in mind this tends to be detector specific). The point is you’ve got to have all that information documented or expect some pointed questions from your GMP inspector. The regulations actually tell us what is expected; Part 111.35 (b)(3)(iii) says you need to identify the reference material, including certification of the accuracy of the known reference material and provide a history of recertification of accuracy.
Next time you purchase an RM, look on the label or accompanying certificate of analysis (C of A) for a statement of materials identity and purity, note the certificate date and see if a recertification date has been scheduled. Certainly I have been sold material that was overdue to be re- certified and have also encountered material that was years from a recertification but was completely degraded. The new GMP regulations expect you to justify your choices, including dates, and you should expect your vendor to be able to do the same.
One cautionary note: make sure that you know the effects of water or identity of counter-ion on your material and mass calculations. Berberine hydrochloride and berberine are not the same and cannot be treated as such during calibration. Also important is chirality of the chemical reference material. Selecting the wrong material can impact your analysis; know what isomer is required, and if it is relevant, be sure you can prove you have the right one.
As for purity, 98% pure sounds pretty good, but that means there is 2% other. If that 2% co-elutes with your compound of interest, it becomes a problem. There should also be a statement of the uncertainty in the measurement, such as 98 ± 5%—without it there is no certification of accuracy. According to the GMP, the purity will have to be periodically verified, so stability should definitely be a consideration when choosing a reference material.
The ultimate nightmare would be to discover that your reference material went bad three production runs ago. That sort of thing could trigger a recall. Be forewarned: if the label or C of A tells you through some subtle and circuitous disclaimer that the supplier doesn’t have confidence in the identity and purity of the material, then my advice would be to either not buy it or to have these characteristics independently verified. You may also point them toward ISO Guide 34. General requirements for the competence of reference material producers specifies the general requirements with which a reference material producer has to demonstrate that it operates, if it is to be recognized as competent to carry out the production of reference materials.
Parodies
In analytical chemistry we are taught about primary and secondary reference materials and the differences between them. In the science of measurements, metrology, a primary standard is a standard that is accurate enough that it is not calibrated by or subordinate to other standards (i.e., primary standards are used to calibrate other standards.)
In essence, a primary standard has all the characteristics that might be found in a Certified Reference Material (CRM). Also defined in AOAC Official method 892.35 and with ISO Guide 30, a CRM is characterized by a metrologically valid procedure for one or more specified properties, accompanied by a certificate that provides the value of the specified property, its associated uncertainty and a statement of metrological traceability. These metrologically valid procedures for the production and certification of CRMs are given in ISO Guide 34 and 35, while ISO Guide 31 gives guidance on the contents of certificates. It is also noted that for qualitative attributes, uncertainties may be expressed as probabilities.
It is unlikely that anyone would use these extremely expensive CRMs in day-to-day operations. Not only is it cost-prohibitive, it’s actually unnecessary. One main purpose of a CRM is to qualify the identity, purity and uncertainty of a secondary material. So as long as you have verified your secondary material and have the documentation to show it was done, using secondary materials is an acceptable practice. If there is no CRM available you simply have to provide rationale for your choice of RM.
If you’ve bought chemicals to use as RMs then you are aware that there are different types and grades on the market. These range from “reagent grade” through “analytical grade” and into the stratosphere of CRMs. Prices can be low (caffeine, for example) or quite high, and some substances may not be available at any price or only as reagent grade materials. It is reasonable to use these materials, but know that in order to qualify them as RMs you may be required to independently verify identity, purity and uncertainty. Also be aware that when a chemical’s purity is listed as 98% the method used to determine the purity must be considered. If techniques like TLC or HPLC are used, the purity is actually only in reference to the detection of impurities by the specified method parameters. For example by HPLC with UV detection at 280 nm, any impurities not detected may not be accounted for in the purity determination.
Reference materials do not have to be 100% pure, but you should know the purity so you can correct for that in your quantitation. CRMs from a national metrology institute, such as the SRMs sold by NIST, are pretty much the gold standard. Pharmacopoeial standards (USP, EP, etc.) can provide a reasonable level of confidence; just be sure to carefully read the labels and accompanying documents as these materials are intended to support a Pharmacopoeial monograph and as such their certifications are applicable only within the parameters of that monograph. Many pharmacopoeia publishers have realized there is a market for high quality phytochemicals and are working toward having their materials classified as CRMs. Just be sure to read the catalog descriptions.
The Sequel
Tune in next time to find out whether Paula is brave enough to tackle botanical reference materials. Will she do it? Does she dare? How many botanists will she have to contact before one agrees to help her?
Although cliffhangers often frustrate readers by offering no resolution, the intent was to impart the Zeigarnik effect with the hopes that this unfinished task would be better remembered. In keeping with that line of thought, botanical reference materials will not be addressed just yet, but rather we will look into another topic that falls within the category of reference materials: the deceptively simple chemical calibration standard.
The Plot
Calling on support from my colleague, Dr. Joseph Betz, this column will delve into the complexity of chemical reference standards, also called chemical calibration standards. In a few short paragraphs we will attempt to explain the following:
• What are they?
• Where can we get them?
• When and how should we use them?
• What role do they play in quality control? Regulatory compliance?
The Cast
According to AOAC Official Method 892.35, a Reference Material (RM) is defined as material or substance, of whose property values are sufficiently homogeneous and well established to be used for calibration of an apparatus, the assessment of a measurement method, or for assigning values to materials. This definition is contextual in that it describes how an RM is used, but not much about what it is.
AOAC provides a second definition: material, sufficiently homogeneous and stable with respect to one or more specified properties, which has been established to be fit for its intended use in a measurement process. So you have to prove the material is suitable for the way you want to use it. This definition also comes with a few footnotes that address the use of an RM: calibration of a measurement system, assessment of a measurement procedure, assigning values to other materials and quality control. Just as important, an RM cannot be used for more than a single purpose in a given measurement (General and Statistical Principles for Certification, ISO Guide 35, 3rd Ed., 2006). Loosely translated—no multitasking allowed! For example, a material can’t be assigned a value based on an analysis, but also be used as the calibrant in the analysis; that is called “self-referencing.”
And for one more opinion, The International Vocabulary of Basic and General Terms in Metrology, (2nd ed., ISO 1993) describes an RM as, “Material or substance one or more of whose property values are sufficiently homogeneous and well established to be used for the calibration of an apparatus, the assessment of a measurement method, or for assigning values to materials (ISO VIM, 6.13, 7).”
In all these definitions the key descriptors are “sufficiently homogenous” and “well established.” The values that must be assigned? Identity and purity.
How do you determine if your material fits the bill? How does one go about qualifying a reference material as defined above?
Guidance can be found in ISO guide 35 (Reference materials-General and statistical principles for certification), which is meant to provide statistical principles to assist in the understanding and development of valid methods to assign values to reference material. It pretty much says that the requirements are that you must take all the steps necessary to demonstrate that the values you assign are correct.
Wow. Thanks guys.
Behind the Scenes
Moving on to the U.S. Dietary Supplement GMPs, it takes a bit of digging but reference materials are mentioned.
In 21 CFR Part 111 Subpart D-Equipment and Utensils, Section 111.25 asks the question “What are the requirements under this subpart for written procedures?” The agency answers that you must establish and follow written procedures for (a)
Calibrating instruments and controls that you use in manufacturing or testing a component or dietary supplement; and (b)
Calibrating, inspecting, and checking automated, mechanical and electronic equipment…” Further, Section 111.27 “What Requirements Apply to the Equipment and Utensils That You Use?” goes on to note that you must calibrate instruments and controls you use in manufacturing or testing a component or dietary supplement. Section 111.35 says you must have written procedures for performing these calibrations and that you are required to keep records of the results of the calibrations.
Basically FDA expects you to calibrate and keep records about your calibrating
activities. What in a facility might need calibration? Just about anything and everything, including drying ovens, refrigerators, freezers, fans, scales, pipets and pH meters. Consider a basic HPLC-UV; the autosampler, oven and lamp all must be calibrated and you haven’t even run a method yet!
Tools and/or instruments are employed to make sure that when specifications are evaluated and measurements made, they be done accurately. Of course you must make sure these tools/instruments are working correctly by performing calibrations. For example, you can calibrate a balance using NIST certified weights. And all volumetric glassware must be calibrated too.
Section 111.35 (b)(3) tells you what you need to include in your calibration documentation. Part 111.35 (b)(3)(iii) states you must “Identify the reference material used including the certification of accuracy of the known reference standard and a history of recertification of accuracy.” And finally, Section 111.315 of Subpart J (Production and Process Control System: Requirements for Laboratory Operations) says that you have to “establish and follow laboratory control processes that are reviewed and approved by quality control personnel, including … (c) Use of criteria for selecting standard RMs used in performing tests and examinations.”
Now we know that 21 CFR Part 111 has a word or five to say about RMs.
Dialogue
The sections within regulations relating to calibration and reference materials encompass some pretty complicated concepts. Starting with 111.315 (c), the QA unit has to justify why they chose the calibration substance(s) they use to perform calibrations. The FDA inspector will expect a QA person to provide a rationale for choosing a particular RM such as: it was the cheapest; best characterized; purest; most readily available; most stable; or some combination of the above. There is no right answer for all situations but there are more appropriate and less appropriate choices for a particular measure.
For example, when measuring anthocyanins in berries, selecting a single aglycone, such as cyanidin, to calibrate your method may be appropriate if the other materials aren’t available and you have another identity method to ensure no adulteration occurs or the materials are available but you have demonstrated the suitability of the surrogate and adulteration is not likely. However, if the glycosides themselves are cheap and readily available, you may have a difficult time justifying your choice.
Be prepared when conducting an indirect (or non-specific) analysis to back up your selection justification with data— chromatographic resolution, extinction coefficient similar to target compound(s) or well-researched correction factor (keeping in mind this tends to be detector specific). The point is you’ve got to have all that information documented or expect some pointed questions from your GMP inspector. The regulations actually tell us what is expected; Part 111.35 (b)(3)(iii) says you need to identify the reference material, including certification of the accuracy of the known reference material and provide a history of recertification of accuracy.
Next time you purchase an RM, look on the label or accompanying certificate of analysis (C of A) for a statement of materials identity and purity, note the certificate date and see if a recertification date has been scheduled. Certainly I have been sold material that was overdue to be re- certified and have also encountered material that was years from a recertification but was completely degraded. The new GMP regulations expect you to justify your choices, including dates, and you should expect your vendor to be able to do the same.
One cautionary note: make sure that you know the effects of water or identity of counter-ion on your material and mass calculations. Berberine hydrochloride and berberine are not the same and cannot be treated as such during calibration. Also important is chirality of the chemical reference material. Selecting the wrong material can impact your analysis; know what isomer is required, and if it is relevant, be sure you can prove you have the right one.
As for purity, 98% pure sounds pretty good, but that means there is 2% other. If that 2% co-elutes with your compound of interest, it becomes a problem. There should also be a statement of the uncertainty in the measurement, such as 98 ± 5%—without it there is no certification of accuracy. According to the GMP, the purity will have to be periodically verified, so stability should definitely be a consideration when choosing a reference material.
The ultimate nightmare would be to discover that your reference material went bad three production runs ago. That sort of thing could trigger a recall. Be forewarned: if the label or C of A tells you through some subtle and circuitous disclaimer that the supplier doesn’t have confidence in the identity and purity of the material, then my advice would be to either not buy it or to have these characteristics independently verified. You may also point them toward ISO Guide 34. General requirements for the competence of reference material producers specifies the general requirements with which a reference material producer has to demonstrate that it operates, if it is to be recognized as competent to carry out the production of reference materials.
Parodies
In analytical chemistry we are taught about primary and secondary reference materials and the differences between them. In the science of measurements, metrology, a primary standard is a standard that is accurate enough that it is not calibrated by or subordinate to other standards (i.e., primary standards are used to calibrate other standards.)
In essence, a primary standard has all the characteristics that might be found in a Certified Reference Material (CRM). Also defined in AOAC Official method 892.35 and with ISO Guide 30, a CRM is characterized by a metrologically valid procedure for one or more specified properties, accompanied by a certificate that provides the value of the specified property, its associated uncertainty and a statement of metrological traceability. These metrologically valid procedures for the production and certification of CRMs are given in ISO Guide 34 and 35, while ISO Guide 31 gives guidance on the contents of certificates. It is also noted that for qualitative attributes, uncertainties may be expressed as probabilities.
It is unlikely that anyone would use these extremely expensive CRMs in day-to-day operations. Not only is it cost-prohibitive, it’s actually unnecessary. One main purpose of a CRM is to qualify the identity, purity and uncertainty of a secondary material. So as long as you have verified your secondary material and have the documentation to show it was done, using secondary materials is an acceptable practice. If there is no CRM available you simply have to provide rationale for your choice of RM.
If you’ve bought chemicals to use as RMs then you are aware that there are different types and grades on the market. These range from “reagent grade” through “analytical grade” and into the stratosphere of CRMs. Prices can be low (caffeine, for example) or quite high, and some substances may not be available at any price or only as reagent grade materials. It is reasonable to use these materials, but know that in order to qualify them as RMs you may be required to independently verify identity, purity and uncertainty. Also be aware that when a chemical’s purity is listed as 98% the method used to determine the purity must be considered. If techniques like TLC or HPLC are used, the purity is actually only in reference to the detection of impurities by the specified method parameters. For example by HPLC with UV detection at 280 nm, any impurities not detected may not be accounted for in the purity determination.
Reference materials do not have to be 100% pure, but you should know the purity so you can correct for that in your quantitation. CRMs from a national metrology institute, such as the SRMs sold by NIST, are pretty much the gold standard. Pharmacopoeial standards (USP, EP, etc.) can provide a reasonable level of confidence; just be sure to carefully read the labels and accompanying documents as these materials are intended to support a Pharmacopoeial monograph and as such their certifications are applicable only within the parameters of that monograph. Many pharmacopoeia publishers have realized there is a market for high quality phytochemicals and are working toward having their materials classified as CRMs. Just be sure to read the catalog descriptions.
The Sequel
Tune in next time to find out whether Paula is brave enough to tackle botanical reference materials. Will she do it? Does she dare? How many botanists will she have to contact before one agrees to help her?