MCPD-02 Reviewer Forms (July 19, 2018)

AOAC INTERNATIONAL Stakeholder Panel on Infant Formula and Adult Nutritionals (SPIFAN) MCPD EXPERT REVIEW PANEL

MCPD-02 REVIEWER FORMS (July 19, 2018)

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

Reviewer 1 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method: MCPD-02 (SGS Germany)

Title: 2-Monochloropropanedio (2-MCPD), 3-Monochloropropanedio (3-MCPD), and Glycidol in Infant and Adult/Pediatric Nutritional Formula

Author: SGS Germany (Jan Kuhlmann)

Reviewer Name:

Summary of Method:

• Extraction of fat content from liquid and powder formula samples by Heat-Ultrasonic-Pressure- supported-Solvent-Extraction (HUPsSE) • Determination of free and bound analytes in fat extracts using indirect (GC-MS) analysis (alkaline transesterification – based on AOCS Cd 29b-13)

Method Scope/Applicability:

• Applicable to powder and liquid samples, infant and adult formulas • Method encompasses the determination of both free (2- and 3-MCPD) and bound (2- and 3- MCPD esters and glycidyl esters) analytes

General comments about the method:

• Separate extraction procedure for liquid samples (Rose Gottlieb) – additional text in Section A describes modification of samples (either lyophilization of liquid samples or preparation of powdered samples with water) so that a single extraction procedure can be used for all formula samples • SLV includes data comparison of powdered and liquid samples

Method Clarity:

• Procedural and safety details edited/added; procedure written in individual steps for added clarity

Method Safety Concerns:

• None

Pros/Strengths:

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 1

• Use of GC single quad; quantitation of free and bound analytes in powders does not require separate extraction procedures • Additional text describes modification of samples (either lyophilization of liquid samples or preparation of powdered samples with water) so that a single extraction procedure can be used for all formula samples

Cons/Weaknesses

• Likely requires high level of technical expertise

Supporting Data

• General Comment:

• Note in Section C indicates the purity of the labeled internal standards can be assessed by comparison of the internal standard response to its corresponding native analyte response (response factor of 1) - Method Optimization: • Performance Characteristics: Analytical Range: • Included note that SLV and other validation studies using GC-MS systems have shown that the response factors for the target ions listed in the method (for labeled internal standards and corresponding analytes) are close to 1; linearity of data in SLV report supports response factors are concentration independent • Note in Section B indicates response factor testing should be repeated when using a different GC-MS system. (Note also discusses isotopic purity verification) • Quantitation based on one-point internal standard calibration; SLV validation shows acceptable linearity and precision with single point calibration – additional text in Section A indicates external calibration can be used. • Suitable for range specified in SMPR; may also be applied to more concentrated samples by dilution with a suitable blank material • Meets requirements of SMPR • Defined as s/n ≥ 10; free analytes in powder: 5 µg/kg, free analytes in liquid: 1 µg/kg, bound analytes in powder: 10 µg/kg, bound analytes in liquid: 2 µg/kg Accuracy/Recovery: • Formula reference material unavailable – FAPAS QC sample (soy sauce) • Spiked/native recoveries and dilutions of mixed samples: 90-113% Precision (RSD r ): • Determined in powder and liquid formulas • RSDr: 1.2-12% (across bound and free analytes in liquid and powder formulas) LOQ:

Reproducibility (RSD R ):

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 1

• Data not available

• System suitability: • Method describes the preparation of blank and spiked samples for the determination of system suitability – reference samples to be analyzed in each analytical sequence

Recommendation:

• ACCEPT

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 2 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method: MCPD-02

Title: 2-MCPD, 3-MCPD and glycidol in infant and adult/pediatric nutritional formula

Author: Jan Huhlman

Reviewer Name:

Summary of Method: Ester bound forms: Fat extraction Mild catalysed transesterification of 2- & 3-MCPDE and GE to free form Glycidol conversion to 3-MBPD Derivatisation with PBA Analysis by GC-MS

Free forms: Extraction Derivatisation with PBA Analysis by GC-MS

Method Scope/Applicability: Infant formula and adult/pediatric nutritional formula, solid (powdered) and liquid. General comments about the method: The method consists of a fat extraction followed by the AOCS Cd 29 c-13 = ISO 18363-1:2017 Method Clarity: The description of the different steps in the form of a text makes it hard to follow, especially when it should be applied in a lab.

Method Safety Concerns:

Pros/Strengths: • “Easier” fat extraction than in the standard method but still effective • No risk of overestimation of GE

Cons/Weaknesses • Use of GC/MS (and not GC-MS/MS) • One-point calibration • Need to apply a transformation factor (3-MCPD to glycidol) which is to be empirically determined for each batch

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 2

Supporting Data • General Comment:

- Method Optimization: • Performance Characteristics:

Analytical Range:

LOQ:

Accuracy/Recovery:

Precision (RSD r

):

Reproducibility (RSD R ):

• System suitability:

Were the recommendations/comments completed as requested by the ERP? • Yes where the authors agree with the reviewers. • Review 9 might still need discussion

Recommendation for First Action Official Methods SM status: Yes

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 3 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method ____MCPD-02___________

Title: 2-Monochloropropanediol (2-MCPD), 3-Monochloropropanediol (3-MCPD), and Glycidol in Infant and Adult/Pediatric Nutritional Formula

Author: SGS (no authors listed)

Reviewer Name:

Summary of Method: Extraction of fat from powders is performed by Heat Ultrasonic Pressure supported Solvent Extraction (HUPsSE). This procedure extracts both free and bound (esterified) 2-MCPD, 3-MCPD, and glycidol. The single extract is partitioned into an organic phase and an aqueous phase. 2-MCPD esters, 3-MCPD esters, and glycidyl esters are determined in the organic phase by the AOCS Cd 29b-13 method. Free 2-MCPD and 3-MCPD are determined in the aqueous phase by a separate procedure. Extraction of fat from liquid samples is performed by the Rose-Gottlieb procedure. 2-MCPD esters, 3-MCPD esters, and glycidyl esters are determined in the extracted fat by the AOCS Cd 29b-13 method. Free 2- MCPD and 3-MCPD are determined in a separate aliquot of liquid sample by a separate procedure.

Method Scope/Applicability:

Liquid and powdered infant formula and pediatric/adult nutritionals

General comments about the method:

Heat Ultrasonic Pressure supported Solvent Extraction (HUPsSE) appears to be a very efficient way to extract both free and bound (esterified) forms of all analytes. However, it’s only applicable to powders. A different extraction procedure (Rose-Gottlieb) needs to be used for liquid samples. Determination of 2- MCPD esters, 3-MCPD esters, and glycidyl esters in the extracted fat (or organic phase from HUPsSE) is performed by an officially recognized method (AOCS Cd 29b-13). The procedure for the separate determination of free 2-MCPD and 3-MCPD has also been properly validated. Overall, the entire proposed method is of high quality, scientifically sound, and properly validated.

Method Clarity:

The method would benefit from proofing for English grammar and syntax.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 3

Method Safety Concerns:

None other than related to the routine handling of a variety of volatile and flammable solvents.

Pros/Strengths: • For powders one single extraction procedure is used for the combined extraction of free and bound (esterified) analytes. This is significant because most samples encountered in routine analysis are powders. • Quantitation is performed by one-point internal calibration. This greatly reduces the analyst’s workload because calibration curves don’t have to be generated with each sample set. • Internal standards are added at the beginning (before extraction), so the most benefit is derived from their use. • A standard GC/MS instrument is used, which is widely available in most laboratories. Cons/Weaknesses • A separate extraction procedure is needed for liquid samples. However, options have been provided to convert liquid samples to powders and vice versa to allow the use of a single extraction protocol. • Quantitation is complicated by the mathematical correction of the result for glycidol and the necessity to determine the transformation factor of 3-MCPD to glycidol with each sample set.

Supporting Data

• General Comment: the supporting data is sufficient.

Validation has been done to exceptional standards. Original and ingenuous matrix-matching scheme has been devised and implemented. Novel and valuable data is being presented on the crossover of free and bound analyte forms between their respective assays.

- Method Optimization: the method is presented in its final, optimized form.

• Performance Characteristics:

Analytical Range: 2-MCPD esters 10 – 557 µg/kg, 3-MCPD esters 10 – 1330 µg/kg, glycidyl esters 10 – 476 µg/kg, free 2-MCPD and free 3-MCPD 5 – 275 µg/kg (powders)

2-MCPD esters 2 – 32 µg/kg, 3-MCPD esters 2 – 84 µg/kg, glycidyl esters 2 – 41 µg/kg, free 2-MCPD and free 3-MCPD 1 – 120 µg/kg (liquids)

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 3

LOQ: 2-MCPD esters, 3-MCPD esters, and glycidyl esters: 10 µg/kg (powder), 2 µg/kg (liquid) free 2-MCPD, 3-MCPD: 5 µg/kg (powder), 1 µg/kg (liquid)

Accuracy/Recovery: 91 – 113%

Precision (RSD r

): 1.2 – 1.3% (data supplied only for free 2-MCPD and free 3-MCPD)

Reproducibility (RSD R

): 2.3 – 12%

• System suitability: addressed.

Were the recommendations/comments completed as requested by the ERP?

While the below points have not been fully completed as requested, none of them are serious enough to prevent the method from being advanced to first action official status. • Comments 3 and 25: linearity has been demonstrated separately across partial analytical ranges and not across the entire range (LOQ to upper limit of quantitation). This is not proof that the response factors are constant across the entire analytical range. However other supporting data demonstrate good method performance by the one-point internal calibration approach. • Comment 12: homogeneous distribution of the internal standards does not guarantee quantitative recovery of analytes (as they must be released from the matrix). However other supporting data demonstrate quantitative recovery of fat and analytes.

Recommendation for First Action Official Methods SM status:

I recommend this method for first action official method status.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 4 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method __ MCPD-02

Title: 2-Monochloropropanediol (2-MCPD), 3-Monochloropropanediol (3-MCPD) and Glycidol in infant formula and adult/pediatric nutritional formula

Author: Jan Kuhlmann

Reviewer Name:

Summary of Method: Free and bound analytes are extracted from powder samples liquid samples. For powder infant formula, free and ester-bound analytes are extracted by triple Heat-Ultrasonic-Pressure- supported Solvent Extraction using methanol and tert-butyl-Methyl-Ether (t-BME). By application of a subsequent liquid-liquid extraction with saturated soldium sulfate and isohexane/t-BME, the combined extracts are separated into a polar fraction containing the 2- and 3-MCPD and a non-polar fraction containing the ester-bound analytes. For liquid samples, a flocculation followed by washing step is used as a pre-treatment, before involving a liquid-liquid extraction with Diethyl ether to extract free form. Extraction of ester bound analytes is performed in liquid samples by Rose-Gottlieb approach. Free 2-MCPD and 3-MCPD from liquid or powder samples are derivatized using phenylboronic acid (PBA) and measured against their derivatives by GC/MS. Ester bound forms of 2-MCPD, 3-MCPD and GE are analyzed after release of the free form by an overnight alkaline transesterification. Hydrolysis is stopped by use of an acidic solution containing NaBr for stabilization of glycidol through conversion to Monobromopropanediol. Analytes are then derivatized using PBA and measured against their derivatives by GC/MS.

Method Scope/Applicability: Powder and liquid infant formula and adult/pediatric nutritional formula.

General comments about the method: Method

Method Clarity:

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 4

Method clarity has improved with revision.

Method Safety Concerns:

Safety is addressed in the document (cover page for general handling of chemicals, D(1)b for extraction under pressure).

Pros/Strengths: • Answer SMPR requirements • Internal standards added early in extraction steps to control all analytical steps

Cons/Weaknesses •

Supporting Data • General Comment:

Extensive data collection to support SLV

- Method Optimization:

• Performance Characteristics:

All performances characteristics below meet SMPR:

Analytical Range LOQ Accuracy/Recovery Precision (RSD r ) Reproducibility (RSD R )

• System suitability: System suitability has been added according to recommendations.

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 4

Were the recommendations/comments completed as requested by the ERP?

- Recommendation 1 : Address system suitability Modification in method answer ERP recommendation

- Recommendation 2: Verify the purity of the labelled standard with every run/batch Note added in section C address the concern raised. Check of purity is enough for each new preparation of standard and not necessarily within each analytical run. - Recommendation 3: Verify the response factors over the analytical range to demonstrate linearity on every system The method still highlights a single point calibration. The note added at the end of section C answer partially the recommendation. Importance of the target ions is also highlighted in the note of section B to ensure similar response factors is obtained between labelled standards and analytes. However, the procedure that is described (end of Section C, check of mixed working solution of isotope-labelled internal standard) should be part of the mandatory control steps after each new preparation of working solution. It should be fully integrated in the section working solution preparation, to ensure same response of isotope labelled internal standard and analyte in the GC-MS. The procedure in the note can be indeed understood as an optional check for laboratories implementing the method. - Recommendation 4a: Written in AOAC format – check for missing steps and information details Typo errors highlighted have been corrected and document gained in clarity. Missing information on evaporation steps and centrifugation speed have been partially added, as well as tolerance on given temperatures. Not all steps have been detailed: e.g. centrifuge conditions are clearly mentioned in D(1)c and D(3)b, but not explicitly given for D(1)e, D(1)f, D(1)g, D(2)b, D(2)c, D(3)c, D(3)e and D(4)b. When not mentioned, condition may refer to Note at the end of D(1). It would be beneficial to either detail conditions, or to place the note adequately in the document. - Recommendation 4b: Written in AOAC format – Add safety Safety advise and risk evaluation has been added in the document (cover page for general handling of chemicals, D(1)b for extraction of sample under pressure in glassware). - Comments: Individual comments from ERP have been addressed and answered either by modification of the manuscript or by giving sound justifications. The reagent section should be expanded to include CAS numbers. Specifically for the target compounds. Although some reagents are common in a laboratory, it would be beneficial for method transfer to include them, especially with grade requirements (e.g. methanol, toluene, tBME).

Recommendation for First Action Official Methods SM status:

Method MCPD-02 described meets SMPR. To accept with additional updates.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 5 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method MCPD-02 (SGS)

Title: 2-Monochloropropanediol (2-MCPD), 3-Monochloropronaediol (3-MCPD) and Glycidol in Infant and Adult/Pediatric Nutritional Formula

Author: Jan Kuhlmann

Reviewer Name:

Summary of Method: A heat and sonicated assisted liquid-liquid extraction with MTBE and methanol is used to extract the bound MCPD and bound glycidol from powdered formula. The organic layer undergoes base catalyzed hydrolysis, is derivatized with PBA, and analyzed by GC-MS/MS. The liquid layer is derivatized with PBA and analyzed by GC-MS. Liquid samples are extracted with the Rose-Gottlieb, with the aqueous layer and organic layers then treated identically to powdered samples.

Method Scope/Applicability: Powdered and liquid infant and adult nutritionals

General comments about the method: Couples a well validated fat extraction method for infant formula to a well validated, collaboratively studied detection method. There is also extensive inter-method comparison with other validated methods, which adds confidence to the accuracy of results.

Method Clarity: The method is clear and provides sufficient detail to ensure it can be reproduced. Also provides complete detail on the equipment conditions and consumable supplies. It might be slightly easier to follow if the steps were individually numbered. Method Safety Concerns: Uses common lab chemicals that require PPE and fume hoods including sodium hydroxide and ethers. Pros/Strengths: • The assay combines the analysis of free and bound in a single extraction. • The use of isotopically labeled standards for all target analytes helps ensure accurate results. • Quantification using widely available GC-MS platform. • Base hydrolysis (as used in AOAC Cd29-13b) is well validated and known not to create or destroy the target analytes. • An extensive inter-laboratory comparison with another validated method adds confidence to the accuracy of results. • Includes a straightforward approach to confirm purity for the isotopically labeled standards. • Includes the possibility of multi-point external calibration.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 5

Cons/Weaknesses • The hydrolysis step takes 15-18 hours, which may be an issue for maintaining production schedules. • Requires a separate analysis for the determination of a transformation factor for every matrix.

Supporting Data • General Comment:

- Method Optimization:

• Performance Characteristics:

Analytical Range: Varies between analytes but are sufficient to ensure that all formula samples could be analyzed without dilution.

LOQ: In powdered formula: Bound 2-MCPD- 10 ug/kg, Bound 3-MCPD- 10 ug/kg, Bound Glycidol- 10 ug/kg, 2-MCPD- 5 ug/kg, 3-MCPD- 5 ug/kg In liquid formula: Bound 2-MCPD- 2 ug/kg, Bound 3-MCPD- 2 ug/kg, Bound Glycidol- 2 ug/kg, 2-MCPD- 1 ug/kg, 3-MCPD- 1 ug/kg

Accuracy/Recovery: Spike recoveries averaged 93-107%

Precision (RSD r

): 1.2 – 12%

Reproducibility (RSD R

): Not reported, but likely not more than 2.4 - 24% based on the RSD r .

• System suitability: Recoveries, reproducibilities, inter-day performance, and time of use performance verifications all appear to be determined appropriately and are all within acceptable ranges.

Recommendation: Approve

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 7 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD) Evaluation of Method MCPD-2 SGS Proposal for 2-MCPD, 3-MCPD and Glycidol in Infant and Adult/Pediatric Nutritional Formula___

Title: SLV for 2-MCPD, 3-MCPD and Glycidol in Infant and Adult/Pediatric Nutritional Formula

Author: Jan Kuhlmann

Reviewer Name:

Second review comments underlined

Summary of Method: Free and bound analytes are extracted from powder samples by heat ultrasonic pressure supported solvent extraction (HUPsSE) using methanol and t-butyl-methyl ether followed by a liq- liq extraction with isohexane/tBME and aqueous saturated sodium sulfate solution to separate polar and non-polar compounds. Free 2- and 3- MCPD are extracted in the aqueous phase while the non-polar fraction contains the bound 2- and 3-MCPD and glycidol. Free analytes are transferred to an organic phase by liq-liq extraction. Rose-Gottlieb fat extraction method is applied for the preparation of liquid samples for bound forms of the analytes with a second liq-liq extraction after acidic protein denaturation is utilized for the free analytes. For both sample types, the free MCPD analytes are derivatized using phenylboronic acid (PBA), prior to analysis by GC-MS. The bound analytes undergo an alkaline catalyzed transesterification to release the free forms. The glycidol is converted to MBPD by reaction with acidifed sodium bromide solution. Liq-liq extraction followed by derivatization using PBA follows, with final analysis using GC/MS. Procedure is a modification of AOCS Cd29b-13 method. Method Scope/Applicability: Infant and adult nutritionals, both powder and liquid. Adult nutritionals not tested as stated the results/behavior would be similar. General comments about the method: Requires multiple extractions for powders and liquids which will be very time consuming. AOCS Cd29b-13 method known to be successful for analysis of bound analytes. Importantly, method relies on single point calibration based on internal standard response with GC-MS analysis. HUPsSE extraction procedure was complex. Method Clarity: Considering the complexity of the method, the multiple extractions needed and multiple steps which require controlled temperature, the preparation steps could be followed. Suggest that the protocol be broken down further into multiple steps to help comprehension instead of a single paragraph

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 7

containing all steps for a given matrix type. Protocol was revised to include more step by step instructions which was VERY helpful.

Points for clarification:

Page 69: Liquid test sample-need better explanation

Pg72- Solution 4- “acid solution has to be adjusted by diluting or by adding more o-phosphoric acid to solution 2 ” –Should it be solution 4? Clarification of the calculations for the concentrations of the working solutions for esters would be helpful. This was difficult to follow.

Method Safety Concerns: No specific issues.

Pros/Strengths: • Gave ranges for times and temperatures, where applicable. Very much appreciated comment about weekend transesterification. • Sonication extraction may speed process a bit. • Includes steps to determine transformation factor to account for the conversion of 3-MCPD into glycidol.

• Utilizes GC-MS which may be more widely available to laboratories • Calculations are much easier to follow than as stated in AOCS method • Method validation very clearly explained

Cons/Weaknesses • Single point calibration based on response of internal standard. This is a risk. A single missed injection or prep error could result in an invalid/missing calibration point. Difficult to verify ongoing method/instrument performance unless the full linear range is assessed in each batch using a multipoint calibration with a minimum of three calibration levels. Instrument performance can certainly decline on the low end, and this will be difficult to monitor otherwise. Also risk of analyst failing to catch linear range exceedance in routine testing without a calibration curve as reference. • Sonication extraction procedure is complicated with differing temperature requirements, depending on the step. Can be confusing. • Identified potential carry over effects in samples containing high levels of free MCPD but stated not likely a potential issue as high incurred free MCPD is rare in infant formula. • Utilizes GC-MS which can be more prone to false positives. Chromatograms typically very clean so not likely a significant issue.

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 7

• Variability of fat extraction related to matrix composition not addressed. Addressed in revised method.

Supporting Data • General Comment: - Method Optimization: Appears that steps were taken to determine ranges for time critical and temperature critical steps which is very helpful. - Provides example chromatogram and linear range plot, which are helpful. - All data presented met SMPR guidance – no outliers

• Performance Characteristics:

Analytical Range: linearity was determined based on serial dilution of incurred samples for bound analytes and fortifications of free analytes in applicable matrices. Assumed that levels in incurred samples were determined with this procedure as I saw no other mention of how the reference levels were obtained. Concentration ranges tested covered the required analytical range.

Analyte

Range (µg/kg) Powdered IF

Range (µg/kg) Liquid IF

Free 2-MCPD Free 3-MCPD Bound 2-MCPD Bound 3-MCPD Bound Glycidol

5-275

0-120 0-120

0.8 - 275

0-557 0-1330 0-476

0-32 0-84 0-41

LOQ: Determined through S/N assessment of samples containing low levels of the analyte since blank assessment was found to be difficult.

Analyte

LOQ Powdered IF (µg/kg)

LOQ Liquid IF (µg/kg)

Free 2-MCPD Free 3-MCPD Bound 2-MCPD Bound 3-MCPD Total 2-MCPD Total 3-MCPD Bound Glycidol

5 5

1 1 2 2 3 3 2

10 10 15 15 10

Accuracy/Recovery: Determined utilized spiking as well as dilution of incurred samples. Results ranged from 93-109% for all testing across matrices and analytes. Measured values of reference materials within 92-104% of assigned values across analytes.

Precision (RSD r

): Ranged from 2.3-12%, depending on the analyte for powdered infant formula and

1.2 – 11% for liquid formula.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 7

Reproducibility (RSD R

): Not specifically listed in the validation data as RSD R

, but multi-day, multi-

analyst data was provided with RSDs <10% listed across analytes.

• System suitability: results for reference materials were provided; however, there was no specific listing of check standard results at low point and midpoint of ranges. Blank sample checks were included (cold pressed oils). Variability of fat extraction related to matrix composition not thoroughly addressed. Detailed information on the sample compositions and fat content was not provided. Adult nutritional samples not included. Addressed in revised method. Recommendation: Very concerned about proceeding with a method with single point calibration based on internal standards as the only option. Although I realize that the AOCS reference method uses this same calibration procedure and that the method is likely very suitable for routine process monitoring for a specific matrix type, I feel that there is a risk that appropriate controls will not be put in place at all performing laboratories once issued. I would rather see an option to use an external standard calibration with internal standard correction. Method has been shown to work well in our laboratory with external standard calibration curve utilized, so I am OK with moving forward if panel is OK with moving forward with single point calibration as the only listed means of calibration.

4

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 8 Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method: MCPD-02

Title: 2- Monochloro propanediol (2- MCPD), 3- MonoChloropropane diol (3- MCPD) and

Glycidyl in infant and adult/pediatric nutritional formula.

Author: Jan Kuhlmann, SGS Germany Gmbh

Reviewer Name:

Summary of Method : Free and ester bound analytes are extracted from powder, liquid test samples by

Liquid-liquid extraction techniques and free analytes are determined and measured by GCMS.

Method Scope/Applicability: Yes, the method supports the applicability of SMPR as analysis of 2- and 3-

MCPD and glycidyl esters both in bound and free forms were quantified in various forms like powder, liquid

and liquid concentrate of pediatric nutritional formula.

General comments about the method: Method is in accordance with SMPR and it is the detailed approach

for Quantitative analysis of fatty acids.

Method Clarity: Methodology is clearly explained for both powder and liquid samples.

Method Safety Concerns: Safety concerns have been specified.

Pros/Strengths: • Met the requirements of SMPR • Definitions as mentioned in the SMPR were well defined and evaluated

• Extraction procedure is in detail for all forms of Infant and adult pediatric nutritional formulae

• Tables for the method requirements have been well tabulated along with chromatograms

• All the recommendations have been addressed.

Cons/Weaknesses Extraction procedure can be simplified.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 8

Supporting Data • General Comment: - Method Optimization: MRM and method conditions have been mentioned in detail. • Performance Characteristics:

Analytical Range: Yes, covered as per SMPR for 2 & 3- MCPD (25- ≥1000)

LOQ : Has been determined and it is within acceptance limits.

Accuracy/Recovery: Has been determined and it is within acceptance limits.

Precision (RSD r

): Has been determined and it is within acceptance limits.

Reproducibility (RSD R

): Has been determined and it is within acceptance limits.

• System suitability: It has been performed as per the SMPR .

Were the recommendations/comments completed as requested by the ERP? Yes, All the recommendations were addressed except for few comments as given below: Recommendation for First Action Official Methods SM status: The method is good candidate to be proposed as AOAC official method of first action

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 9

Expert Review Panel for Infant Formula and Adult Nutrition (MCPD)

Evaluation of Method MCPD-02

Title: MCPD-02 – 2-Monochloropropanediol (2-MCPD), 3-Monochloropropanediol (3-MCPD) and Glycidol in Infantand Adult/Pediatric Nutritional Formula - Gas Chromatographic /Mass spectrometric (GC/MS) Method

Author: Jan Kuhlmann, SGS Germany GmbH

Reviewer Name:

Summary of Method: The purpose of this report is to document a single-laboratory validation performed for an AOAC Submitted Method: 2-Monochloropropanediol (2-MCPD), ), 3-Monochloropropanediol (3- MCPD) and Glycidol in Infant and Adult/Pediatric Nutritional Formula. Purchasable samples containing the analytes from processing as well as spiked samples served for validation. Further infant and toddler formula test materials originated from a monitoring project on behalf of the German Ministry of Agriculture and Food. Other single samples came from a method validation study organized by the German Federal Institute of Risk Assessment (BfR), from the US Food and Drug Administration (FDA), Center for Food Safety and Applied Nutrition (CFSAN) through the course of an interlaboratory comparison and from the Stakeholder Panel on Infant Formula and Adult Nutritionals (SPIFAN) test kit II for nutrient testing. The selected test samples were used to determine such parameters as linearity, limit of quantification (LOQ), specificity, precision, analytical range and accuracy. Analytes tested were free 2-MCPD, ester-bound (bound) 2-MCPD, free 3-MCPD, bound 3-MCPD, and bound glycidol. Because powdered and liquid samples are within the scope of the method both classes of matrices were tested for all cited validation parameters.

Method Scope/Applicability: Applicable for determination of the overall contents of free and ester-bound 2-MCPD, free and ester-bound 3-MCPD and ester-bound glycidol in powdered and liquid infant formula and adult nutritionals, in accordance with AOAC SMPR® 2017.017 - Standard Method Performance Requirements (SMPRs®) for Determination of 2- and 3-MCPD, 2- and 3-MCPD Esters, and Glycidyl Esters in Infant and Adult/Pediatric Nutritional Formula.

General comments about the method: This method appears well written. The scope and applicability are expressed; presented results support final conclusions.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 9

Method Clarity: The manuscript is clearly written; instructions and procedures are expressed with clarity.

Method Safety Concerns: No safety concerns.

Pros/Strengths: 

The manuscript is clearly written.

Cons/Weaknesses  No comments.

Supporting Data  General Comment: the mentioned data support the method as written.

Method Optimization: No comments.

–

Performance Characteristics:



Analytical Range: Variable, depending on the peculiar analyte and the matrix type (Table 6). In accordance with AOAC SMPR® 2017.017.

LOQ: Free 2-MCPD in powder and liquid samples: 5 and 1  g/kg respectively Free 3-MCPD in powder and liquid samples: 5 and 1  g/kg respectively Bound 2-MCPD in powder and liquid samples: 10 and 2  g/kg respectively Bound 3-MCPD in powder and liquid samples: 10 and 2  g/kg respectively Total 2-MCPD in powder and liquid samples: 15 and 3  g/kg respectively Total 3-MCPD in powder and liquid samples: 15 and 3  g/kg respectively

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

Reviewer 9

Bound Glycidol in powder and liquid samples: 10 and 2  g/kg respectively

Accuracy/Recovery: Variable, in the range 72-125 %

Precision (RSD r

): 1.2 to 12 % (< 22 %)

Reproducibility (RSD R

): no comments.

System suitability: No concerns.



Were the recommendations/comments completed as requested by the ERP? Yes, although there is no agreement with concern to ERP recommendations No 15 and 24 (linearity has not been explicitly demonstrated for bound analytes) and No 18 (with relation to the impact of matrix types). However, the Author has provided two detailed answers.

Recommendation for First Action Official Methods SM status: I recommend that the ERP adopt this method for First Action Official Method SM status.

3