AOAC SPIFAN Review of GB Methods (March 12, 2020)

AOAC Stakeholder Panel on Infant Formula and Adult Nutritionals NUTRIENTS EXPERT REVIEW PANEL (ERP)

Meeting at the Gaithersburg Marriott Washingtonian Center 9751 Washingtonian Boulevard, Gaithersburg MD 20878, USA GB Methods Review

Thursday, March 12, 2020

AOAC INTERNATIONAL 2275 Research Blvd., Suite 300 Rockville, MD, 20850

UNITED STATES dboyd@aoac.org 301.924.7077 x126

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method GB 29989-2013

Title: Determination of L-carnitine in foods for infants and young children, milk and milk products Author: National Standard of the People´s Republic of China Reviewer Name: Reviewer 1 Summary of Method: The sample is extracted with water and the protein is precipitated with perchloric acid and then filtered. After the filtrate is saponified with alkali, the bound L-carnitine in the solution is released. L-carnitine reacts with acetyl-CoA under the catalysis of acetylcarnitine transferase to form acetylcarnitine and free coenzyme A. Free coenzyme A and 2-nitrobenzoic acid react to form a yellow substance which is proportional to the content of free coenzyme A. Free coenzyme A has an equimolar relationship with L-carnitine and can be used to indirectly calculate the content of L-carnitine in the sample.

Method Scope/Applicability: The method is applicable to determination of L-carnitine in infant foods and young children, milk and milk products.

General comments about the method: The method take the absorbance measure of the sample treatment solution and then calculate the concentration of the test solution by comparing to the calibration curve .

Method Clarity: The method is clear and easy to follow.

Pros/Strengths: •

Cons/Weaknesses •

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Supporting Data • General Comment:

Method Optimization ------------

• Performance Characteristics:

Analytical Range: No data

LOQ: SMPR: LOQ: 2mg/100g

Accuracy/Recovery: No data

Precision (RSD r ): SMPR: The absolute difference between two independent determination results obtained under repeatable conditions should not exceed 10% of the arithmetic mean.

Reproducibility (RSD R ): -----

• System suitability:

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

1. Is the Validation Study Report in a format acceptable to AOAC? No validation Study Report presented.

2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? No data presented

3. Are the figures and tables sufficiently explanatory without the need to refer to the text? No data presented

4. Are all the figures and tables pertinent? No data presented

5. Could some be omitted and covered by a simple statement? -----------

6. Are the references complete and correctly annotated? No references presented

7. Does the method contain adequate safety precaution reference and/or statements? No mention about safety precaution

Recommendation: Is necessary to get information about the validation procedure and data of performance characteristics to have more tools to evaluate the method.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method GB29989-2013

Title: National Food Safety Standards- Determination of L-carnitine in infant foods and dairy products Author: GB29989-2013 Reviewer Name: Reviewer 2 Summary of Method: The sample was extracted with water, and the protein was precipitated with perchloric acid and then filtered. After the filtrate is saponified with alkali, the bound L-carnitine in the solution is released. L- carnitine reacts with acetyl-CoA under the catalysis of acetyl-carnitine transferase to form acetyl-carnitine and free coenzyme A. Free Coenzyme A and 2-nitrobenzoic acid react to form a yellow substance, the color of which is proportional to the content of free coenzyme A. Free coenzyme A has an equimolar relationship with L-carnitine, and can be used to indirectly calculate the content of L-carnitine in the sample. Method Scope/Applicability: This standard specifies the method for the determination of L-carnitine in foods for infants and young children, milk and milk products.

This standard is applicable to the determination of L-carnitine in foods for infants and young children, milk and milk products.

General comments about the method: Primary concerns center on the fact that colorimetric methods are less selective and prone to matrix- impact. This has been observed in our experience.

Methods of this type can be used as a QC tool but are not generally recommended for Official or Dispute Resolution purposes.

Given the method specificity for L-Carnitine, method provides ability to confirm presence of L-Carnitine.

Method Clarity: The method is well written and easy to follow execution details.

Reagent quality – Additional detail recommended/needed “Unless otherwise specified, the reagents used in the method are the ones of analytically pure, and the water used is the Level 3 Water regulated in GB/T 6682.”

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Solutions – Storage detail and expiry is provided in sufficient detail.

Standard(s) - Sourcing detail not provided, purity and storage detail is provided.

Equipment needed to execute method with tolerance detail is provided.

Pros/Strengths: • Specificity for L-Carnitine

• Common, inexpensive equipment used for method. • Relatively easy to implement and routinely use. • Can be a good QC tool

Cons/Weaknesses • General

o Colorimetric methods tend to be susceptible to matrix impact.  Impact of hydrolyzed-protein formula on method performance has been noted. o QC detail is thin and could impact lab performance and inter-lab method performance consistency. o LOQ does not meet SMPR 2012.010 and is higher than current generation UHPLC-MS/MS methods; OMA 2015.10 • Specific Comments o Acetylcarnitine transferase (CAT) Enzyme solution: It can be difficult to discard the supernatant per procedure - a small volume of CAT solution can remain at bottom of Eppendorf centrifuge tube which can result in a cloudy final solution and can impact detection (and result). o Enzyme reaction time specified by method (10-min) may not be sufficient. o Only one enzyme supplier with consistently sufficient quality for method. o Enzyme is expensive and the shelf life is 6 months – usually translates to ~3 - 4 months shelf life at laboratory.

Supporting Data • General Comment: None presented

Method Optimization: None presented

• Performance Characteristics:

Analytical Range: Not presented

Limit of Quantitation (LOQ): 2 mg/100g. Per SMPR 2012.010, LOQ is 0.16 mg/100g. LOQ of this Standard does not meet SMPR.

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Limit of Detection (LOD): 0.6 mg/100g. LOD not defined in SMPR 2012.010

Accuracy/Recovery: Not presented

Precision (RSD r ): Not presented

Reproducibility (RSD R ): Not presented

• System suitability: Calibration curve parameters are not specified.

The reported final result is the arithmetic mean of the two independent determination results. Final results must be accurate to the first decimal place.

The absolute difference of the two independent determination results obtained under the repetitive condition shall not exceed 10% arithmetic mean.

1. Is the Validation Study Report in a format acceptable to AOAC? a. NAP 2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? a. Yes 3. Are the figures and tables sufficiently explanatory without the need to refer to the text? a. NAP 4. Are all the figures and tables pertinent? a. NAP 5. Could some be omitted and covered by a simple statement? a. NAP 6. Are the references complete and correctly annotated? a. References are not provided within Standard. 7. Does the method contain adequate safety precaution reference and/or statements? a. Safety statements not provided within Standard. Recommendations: A formal SLV and MLT using SPIFAN kit samples is recommended to better characterize method performance.

Performance of this Standard should be bench-marked to the current-generation AOAC Final Action Method, OMA 2015.10.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method __GB 29989-2013_____________

Title: National food safety standard - Determination of L-carnitine in foods for infants and young children, milk and milk products Author: People’s Republic of China Reviewer Name: Reviewer 3 Summary of Method: L-carnitine is determined indirectly from free co-enzyme A concentration. Samples are extracted with water, and proteins are precipitated with perchloric acid. The resulting filtrate is saponified with alkali to free any bound L-carnitine. L-carnitine reacts with acetyl-CoA to form acetylcarnitine and free coenzyme A. Free coenzyme A reacts with nitrobenzoic acid to form a yellow compound that is detected spectrophotometrically at 412 nm. Both standards are samples are treated with acetyl-CoA and nitrobenzoic acid. External standard curves are prepared and used to calculate l- carnitine in samples. Method Scope/Applicability: Foods for infants and young children, milk and milk products

General comments about the method: Similar methods have demonstrated acceptable precision and accuracy for determination of l-carnitine in infant formulas.

Method Clarity: Overall method is clearly written.

Pros/Strengths: • Method is simple. • Method uses equipment available in most laboratories. • Method can differentiate L- and D-carnitine.

Cons/Weaknesses • Method does not determine L-carnitine directly.

• Sample preparation procedure labor intensive. Only 8-10 samples can be prepared in two days. • Method may not work well with products containing hydrolyzed proteins or products with yellow or cloudy filtrates.

• Can be difficult to obtain a clear CAT enzyme solution. • LOQ much higher than newer HPLC/UPLC/MS methods.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Supporting Data: No supporting data provided.

• General Comment:

Method Optimization:

• Performance Characteristics:

Analytical Range:

LOQ:

Accuracy/Recovery:

Precision (RSD r ):

Reproducibility (RSD R ):

• System suitability:

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

1. Is the Validation Study Report in a format acceptable to AOAC? Not applicable 2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? Yes. 3. Are the figures and tables sufficiently explanatory without the need to refer to the text? Not applicable 4. Are all the figures and tables pertinent? Not applicable 5. Could some be omitted and covered by a simple statement? Not applicable 6. Are the references complete and correctly annotated? Not applicable 7. Does the method contain adequate safety precaution reference and/or statements? Method does not contain safely section.

Recommendation: Complete single laboratory validation with SPIFAN matrices and compare results to those generated with AOAC final action dispute method.

3

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method _GB 29989-2013

Title: determination of L-carnitine in foods for infants and young children, milk and milk products Author: Reviewer Name: Reviewer 4 Summary of Method: The sample was extracted with water, and the protein was precipitated with perchloric acid and then filtered. After the filtrate is saponified with alkali, the bound L-carnitine in the solution is released. L- carnitine reacts with acetyl-CoA under the catalysis of acetylcarnitine transferase to form acetylcarnitine and free coenzyme A. Free Coenzyme A and 2-nitrobenzoic acid react to form a yellow substance, the color of which is proportional to the content of free coenzyme A (Spectrophotometer at 412 nm). Free coenzyme A has an equimolar relationship with L-carnitine, and can be used to indirectly calculate the content of L-carnitine in the sample. Method Scope/Applicability: This method is applicable to infant foods and dairy products General comments about the method: It is an enzymatic method. It is relatively easy to do. The sample prep step involves protein precipitation and saponification, which is mild comparing to the AOAC 2015.10 Free and total choline and carnitine method, in which the sample is digested with nitric acid at 120oC for 40 min. Method Clarity: In general, it is clear. The grade of some reagents and the activity of enzyme need to be specified. Pros/Strengths: • No expensive equipment is needed Cons/Weaknesses • Might prone to interference since no chromatography is used Supporting Data • General Comment: There is no supporting data.

Method Optimization:

o

Performance Characteristics:

•

Analytical Range:

LOQ:

Accuracy/Recovery:

Precision (RSD r ):

Reproducibility (RSD R ):

System suitability:

•

Page Break Is the Validation Study Report in a format acceptable to AOAC?

1.

2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? 3. Are the figures and tables sufficiently explanatory without the need to refer to the text?

4.

Are all the figures and tables pertinent?

5.

Could some be omitted and covered by a simple statement?

6.

Are the references complete and correctly annotated?

7.

Does the method contain adequate safety precaution reference and/or

statements?

Recommendation: Could be a useful method for labs with limited resources. There is no supporting data to review for any recommendation.

Comments on GB methods Reviewer 5

GB29989-2013 : Determines total L-Carnitine. From our method equivalency work there is no significant difference between results obtained using this GB standard and other similar enzymatic assays based on the following method, ‘Enzymatic determination of free carnitine in milk and infant formulas’, AOAC manuscript, H. Indyk and D. Woollard, J. AOAC Int., 78 (1995) 69-74. This study involved a limited number of formulas – would be good to have data for SPIFAN formula (infant milk) – enzymatic method vs the LC-MS method.

Is there a guidance document on reviewing these GB standards – it would be helpful. I am assuming this is a paper exercise, should we review on its own or compare to AOAC/ISO standards, review key aspects / principle or all aspects, line for line etc….? Maybe there is something already, my apologies if I missed it.

Comments on GB methods Reviewer 6

GB29989-2013 : Determines total L-Carnitine. From our method equivalency work there is no significant difference between results obtained using this GB standard and other similar enzymatic assays based on the following method, ‘Enzymatic determination of free carnitine in milk and infant formulas’, AOAC manuscript, H. Indyk and D. Woollard, J. AOAC Int., 78 (1995) 69-74. This study involved a limited number of formulas – would be good to have data for SPIFAN formula (infant milk) – enzymatic method vs the LC-MS method.

Is there a guidance document on reviewing these GB standards – it would be helpful. I am assuming this is a paper exercise, should we review on its own or compare to AOAC/ISO standards, review key aspects / principle or all aspects, line for line etc….? Maybe there is something already, my apologies if I missed it.

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method GB 5413. 18-2010

Title: Determination of vitamin C in foods for infants and young children, milk and milk products Author: National Standard of the People´s Republic of China

Reviewer Name: Reviewer 1

Summary of Method: Vitamin C (ascorbic acid) is oxidized to be dehydroascorbic acid in presence of activated carbon. The dehydroascorbic acid then reacts with o-phenylenediamine to generate fluorescent substances. The fluorescent spectrophotometer then is applied to determine the fluorescence intensity of the fluorescent substances generated. The fluorescence intensity is proportional to the concentration of ascorbic acid present.

Method Scope/Applicability: This method is applicable to determination of vitamin C in infant foods and young children, milk and milk products.

Method Clarity: The method is clear, descriptive and easy to follow in its current format.

Pros/Strengths:

Cons/Weaknesses This method could present interferences.

Supporting Data • General Comment:

Method Optimization: No data presented.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

• Performance Characteristics:

Analytical Range: No data

LOQ: No data

Accuracy/Recovery: No data

Precision (RSD r ): No data

Reproducibility (RSD R ): ----

• System suitability: No data

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

1. Is the Validation Study Report in a format acceptable to AOAC? No validation data report presented.

2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? No data presented

3. Are the figures and tables sufficiently explanatory without the need to refer to the text? No data presented

4. Are all the figures and tables pertinent? No data presented

5. Could some be omitted and covered by a simple statement? -----

6. Are the references complete and correctly annotated? No references presented

7. Does the method contain adequate safety precaution reference and/or statements? No comments about safety.

Recommendation: Is necessary to get information about the validation procedure and data of performance characteristics to have more tools to evaluate the method.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method GB5413.18-2010

Title: National Food Safety Standards- Determination of vitamin C in infant foods and dairy products Author: GB5413.18-2010 Reviewer Name: Reviewer 2 Summary of Method: Vitamin C (Ascorbic acid) is oxidized to dehydro-ascorbic acid in the presence of activated carbon. The dehydroascorbic acid then reacts with o-phenylenediamine to generate fluorescent substances. Fluorescence spectrophotometry is applied to measure the resulting fluorescence intensity of the fluorescent substances generated. The measured fluorescence intensity is proportional to the concentration of ascorbic acid. External calibration is used. Method Scope/Applicability: This standard is applicable to the determination of vitamin C in foods for infants and young children, milk and milk products. Results indicate total vitamin C content defined as both reduced and oxidized vitamin C. General comments about the method: Primary concerns center on the fact that colorimetric methods are less selective and prone to matrix- impact. This has been observed in our experience.

Methods of this type can be used as a QC tool but are not generally recommended for Official or Dispute Resolution purposes.

We have completed an internal assessment of this method on limited matrices, including milk-based IF, extensively hydrolyzed protein IF, partially-hydrolyzed protein FUF and Adult formulations. Relative agreement of the GB method with AOAC 2012.21 (First Action, Back-up method) ranged from 65% - 155%.

Method Clarity: The method document is generally well written with experimental details easy to follow.

Reagent quality – Additional detail recommended/needed “Unless otherwise specified, the reagents used in the method are the ones of analytically pure, and the water used is the Level 3 Water regulated in GB/T 6682.”

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Solutions – Some detail provided (prepared when required). Additional preparation, storage and expiry detail is recommended.

Standard(s) – Additional detail is recommended including required minimum standard purity, storage conditions, purity/quality assessment, expiry, etc.

Recommend adding additional detail describing working standards preparation including defining use of volumetric glassware or pipettes and measure of final volume.

Equipment – Recommend method provide more detailed description of equipment needed to execute the method.

Pros/Strengths: • Common, inexpensive equipment. • Easy to implement and routinely use. • Can be a good QC tool.

Cons/Weaknesses • Fluorometric assay – good selectivity but generally not as good as LC methods o Does not allow for speciation • Additional method detail is needed for needed lab equipment, standards and reagent quality, storage and expiry, preparation of working standards, and general method QC. • QC detail is thin and could impact method/lab performance and inter-lab consistency. • Colorimetric methods tend to be susceptible to matrix impact

Supporting Data • General Comment: None presented Method Optimization:

• Performance Characteristics:

Analytical Range: Not presented

LOQ: Not presented

LOD: The detection limit of this standard is 0.1 mg/100g and exceeds SMPR 2012.012 LOD requirement of 0.3 mg/100g

Accuracy/Recovery: Not presented

Precision (RSD r ): Not presented

Reproducibility (RSD R ): Not presented

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

• System suitability:

Quality checks (purity, stability) on the standard is not discussed.

External calibration is used but no details are provided defining acceptable curve parameters.

The final reported result is the arithmetic mean of two independent determinations. Results are reported to one decimal place.

The absolute difference of the two independent determination results obtained under the repetitive condition shall not exceed 10% arithmetic mean.

1. Is the Validation Study Report in a format acceptable to AOAC? NAP 2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? Yes 3. Are the figures and tables sufficiently explanatory without the need to refer to the text? NAP 4. Are all the figures and tables pertinent? NAP 5. Could some be omitted and covered by a simple statement? NAP 6. Are the references complete and correctly annotated? No references provided 7. Does the method contain adequate safety precaution reference and/or statements?

Recommendation: Additional SLV and MLT data using SPIFAN kit matrices is recommended to further define and assess method performance.

Performance of this Standard should be compared directly to the current-generation AOAC Final Action method, OMA 2012.22, using SPIFAN kit matrices.

3

Comments on GB methods 13 th February 2020 – Reviewer 3

GB 5413.18-2010 : This technique dates back to 1980’s and before (see AOAC 94.26, 964.22), but can be susceptible to interferences. We note that reference methods using some form of chromatographic separation now typical e.g. AOAC 2012.22

Specific comments: 2.2 Insect pest definition - includes birds & rodents. Pests maybe be better heading. (also see 6.4)

3.1.1 editing? contamination risk to food 3.2.4 all pests e.g. mice/rats not just insects 4.1.4 Inspection room - not sure what this is e.g. sub-sampling room? 4.2.4.1 Editing – editing 'timely shut down' ?

6.1.1 Second sentence meaning unclear 6.6.3 “Perfect” condition – not defined 7.1 Last sentence not clear

13.1 Final part of sentence reword “food safety need identified” Appendix A: procedure is not just environmental but also product A 4 (d) “ The sampling points or the monitoring frequency may be properly reduced if the monitoring result is in accordance with the requirements all the time.” Meaning not clear

Comments on GB methods Reviewer 4

GB 5413.18-2010: Technique is old and seems there is no mention so far in GB, of replacing this standard with a HPLC method. From our equivalency work, results generated using this GB Standard are equivalent to the AOAC Official Methods 985.33 and 967.21 (visual titration methods), with small modifications.

Is there a guidance document on reviewing these GB standards – it would be helpful. I am assuming this is a paper exercise, should we review on its own or compare to AOAC/ISO standards, review key aspects / principle or all aspects, line for line etc….? Maybe there is something already, my apologies if I missed it.

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Expert Review Panel for Infant Formula and Adult Nutrition Evaluation of Method __GB 5413.18-2010_____________

Title: National food safety standard - Determination of vitamin C in foods for infants and young children, milk and milk products Author: People’s Republic of China Reviewer Name: Reviewer 5 Summary of Method: In both samples and standards, ascorbic acid is oxidized to dehydroascorbic acid in the presence of activated carbon, and dehydroascorbic acid is converted to a fluorescent compound with o- phenylenediamine. The resulting solutions are analyzed with a fluorometer at an excitation of 350 nm and emission of 430 nm. Blank samples are also prepared and analyzed with samples and standards. External standard curves are prepared and used to calculate total vitamin C in samples.

Method Scope/Applicability: Foods for infants and young children, milk and milk products

General comments about the method: Method similar to AOAC methods 967.22, Vitamin C (Ascorbic Acid) in Vitamin Preparations, and 984.26, Vitamin C (Total) in Food. Method uses derivatization with fluorescence detection. Fluorescence detection typically more specific than UV detection. Sample preparation procedure more labor intensive than newer HPLC/UPLC based vitamin C methods.

Method Clarity: Overall method is clearly written. There are a few sections where the translation to English could be improved.

Pros/Strengths: • Method uses a specific detection technique, fluorescence detection after derivatization. • Method uses equipment available in most laboratories.

Cons/Weaknesses • Method cannot be used to quantitate ascorbic acid and dehydroascorbic acid separately. • Dehydroascorbic acid is less stable than ascorbic acid. Care must be taken to minimize degradation of dehydroascorbic acid during sample preparation and analysis. GB reaction times are approximately twice those of the AOAC methods. • Sample preparation procedure labor intensive

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Supporting Data: No supporting data provided.

• General Comment:

Method Optimization:

• Performance Characteristics:

Analytical Range:

LOQ:

Accuracy/Recovery:

Precision (RSD r ):

Reproducibility (RSD R ):

• System suitability:

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

1. Is the Validation Study Report in a format acceptable to AOAC? Not applicable 2. Is the method described in sufficient detail so that it is relatively easy to understand, including equations and procedures for calculation of results (are all terms explained)? Yes, with the exception of a few sections where translation to English could be improved. 3. Are the figures and tables sufficiently explanatory without the need to refer to the text? Not applicable 4. Are all the figures and tables pertinent? Not applicable 5. Could some be omitted and covered by a simple statement? Not applicable 6. Are the references complete and correctly annotated? Not applicable 7. Does the method contain adequate safety precaution reference and/or statements? Method does not contain safely section.

Recommendation: Complete single laboratory validation with SPIFAN matrices and compare results to those generated with AOAC final action dispute method.

3