AOAC OMA 2021.01 Reviewer Forms (July 25, 2023)

AOAC Stakeholder Program on Infant Formula and Adult Nutritionals (SPIFAN)

AOAC OMA# 2021.01 (GOS-03) REVIEWER FORMS

July 25, 2023

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

dboyd@aoac.org 301.924.7077 x126

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

Expert Review Panel for Infant Formula and Adult Nutrition “FINAL ACTION REVIEW” Evaluation of Method 2021.01 (GOS-03)

Title: Determination β -Galactooligosaccharides (GOS) in Infant Formula and Adult Nutritionals: Collaborative Study Author: Denys Cuany, Sean Austin Reviewer Name: Reviewer 1

Did Method Author Consider ERP’s General Recommendations:

• Was the method tested using all the SPIFAN samples for precision and accuracy? The method was tested using MLT sample kit comprised of seven commercial products from different suppliers. Six samples of Infant Formula with with different composition GOS/FOS, GOS/HMO, partially hydrolyzed protein, organic with probiotics and one sample of Adult Nutritional.

• Was data provided from a standard reference material? Yes, data was provided by NIST SRM 1849b.

• Was the analysis conducted using ready-to-feed (reconstituted) infant formula? Yes, three ready-to-feed infant formula samples were analyzed.

• Was rice-based formula available and was it used? No rice-based formula available.

• Was the method tested using properly characterized standards? Yes, the method was tested using laminaritriose as internal standard and with maltotriose as external standard.

• Is there proof of performance through system suitability? Yes, the system suitability is described in the collaborative study as system suitability test (item I) • Is there clarity of calibration procedure? Yes, it is recommended to use bracketed calibration, injecting 3 standards followed by a maximum of 10 samples then 3 standards again. It is used the instrument software to plot a six points standard

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

curve of Instrument response of maltotriose/Instrument response for laminaritriose against the concentration of maltotriose in the standard. • Does the method demonstrate reproducibility? To what extent? The repeatability and reproducibility for six of the samples meet the requeriments of SMPR 2014.003 with values of RSD r between 1.4 and 4.7 % and RSD R between 8.1 and 11.6%. The precision achieved for Adult Nutritional RTF sample does not meet SMPR requirement with an RSDr of 9.9% and RSD R of 12.1%. Did Method Author Consider ERP’s Method Specific Recommendations: Yes, after the approval of the first action method, the team of the study director tested the replacement of sodium cyanoborohydride (Toxic) by 2-methylpyridine borane complex. A comparison of the methods was carried out to determine whether the alternative method is equivalent to the reference method. Five samples were analyzed in duplicate on six different days. The results demonstrate there are no biases between the methods.

Any Revision to Method Scope/Applicability of Method: The First Action method is applicable to SPIFAN II kit, four of the fortified products were found to contains GOS. Five additional matrices were added to increase the number of samples containing GOS.

General comments (i.e. safety) about the method: • Feedback from Users of the Methods since being awarded First Action Official Methods Status Comments are from MLT study: -Laboratory 4 made additional tests by reconstituting with water at 50°C instead of 70°C and stirring for only a few minutes and found no difference in the results, thus they proposed that the reconstitution could be simplified without impacting the results.

- Several labs required guidance during the set-up of the method.

Additional Pros/Strengths: • The method described in AOAC 2021.01acieved good results generally in line with the requirements of AOAC SMPR 2014.003 during both the SLV and MLT. • The replacement of sodium cyanoborohydride (Toxic) by 2-methylpyridine borane complex is a good practice. • The MLT had few outliers from 14 participants.

Additional Cons/Weaknesses • Needed more data on Adult Nutritional samples.

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

Supporting Information

• Performance Characteristics and compared to approved SMPR:

Analytical Range : SMPR 0.2 – 3.0 g/100g 0.23 to 0.86 g/100g For infant formula , the maximum permitted dose of GOS in most countries is 0.8 g/100g LOQ:

SMPR ≤ 0.2 g/100g Method < 0.2 g/100g Accuracy/Recovery: SMPR 90 – 110% Method 91.5 – 103% Precision (RSD r ): SMPR ≤ 6% Infant formula 1.4% to 4.7% Adult Nutritional 9.9% Reproducibility (RSD R ): SMPR ≤ 12% Infant formula 8.1% to 11.6% Adult Nutritional 12.1%

ERP Summary/Recommendation(s):

• Does the method meet the approved SMPR? Yes, for Infant Formula.

• Has the method been performing well in the laboratory? Yes, the method shows good performance • Are there any additional feedback on method performance? No.

• Would you recommend designating this method as a Final Action method to the ERP? Yes, I recommend this method for FINAL Action on Infant Formula. In the case of Adult Nutritional is necessary to collect more data.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

4

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

Expert Review Panel for Infant Formula and Adult Nutrition “FINAL ACTION REVIEW” Evaluation of Method AOAC 2021.01 (GOS-03)

Title: Determination β -Galactooligosaccharides (GOS) in Infant Formula and 1 Adult Nutritionals: Collaborative Study Author: Denis Cuany, Sean Austin Reviewer Name: Reviewer 2

Did Method Author Consider ERP’s General Recommendations:

• Was the method tested using all the SPIFAN samples for precision and accuracy? Samples were prepared by the study director (seven). The kit compromised of four powder infant formulas either intact or hydrolyzed protein including combinations of GOS with FOS and HMO, two RTF Infant formulas in combination with FOS and probiotics and one adult nutritional RTF. All samples were tested on homogeneity. NIST SRM 1849 b was used from SPIFAN sample kit, other samples from SPIFAN sample kit do not contain GOS.

• Was data provided from a standard reference material? YES. NIST SRM 1849b sample with GOS as nutrient was used as practice sample and used to access accuracy

• Was the analysis conducted using ready-to-feed (reconstituted) infant formula? YES. Two Ready to Feed Infant formulas were included in the study and one RTF adult nutritional formula.

• Was rice-based formula available and was it used? No Rice-based formula available with GOS

• Was the method tested using properly characterized standards? Maltotriose is used as standard. Purity and moisture should be checked properly and if necessary used to correct when constructing the calibration curve.

• Is there proof of performance through system suitability? YES system suitability parameters are included in the method protocol.

1

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

• Is there clarity of calibration procedure? Key points addressed in the method 1) Check the purity and moisture content of the maltotriose and correct for those when constructing the calibration curve. (2) Check the integration of the signals in the chromatogram, the chromatograms are not easily integrated using the automated tools built in to chromatographic software (3) Ensure that only peaks with signal/noise ratio of ≥ 10 are included for data processing (including smaller peaks can introduce significant errors) The repeatability and reproducibility for six of the 7 samples meet the requirements of SMPR 2014.003 with values of RSDr between 1.4 and 4.7 % and RSDR between 8.1 and 11.6 %. The precision achieved for one sample, Adult Nutrional RTF (PBK- 9 00007), does not meet completely the SMPR requirement with an RSDr of 9.9% and RSDR of 12.1 %. Horwitz ratio (HorRat) values ranged from 1.9 to 3.0, which is little higher than generally acceptable. However, HorRat is not necessarily applicable to methods with polymeric analytes such as galactooligosaccharides. Did Method Author Consider ERP’s Method Specific Recommendations: • ERP already agreed (after approval to first action and before MLT started) on adaption of the method to remove toxic regent sodium cyanoborohydride replaced by less toxic 2 methylpyridine borane complex. I this MLT report equivalence (no bias) was tested and shown based on comparison on 5 samples (both powder and liquid). Any Revision to Method Scope/Applicability of Method: • First action Method is applicable to infant formula and Adults Nutritionals. At the time of review SLV by ERP, from 10 GOS containing samples precision was established. NINE were infant formulas and ONE adult nutritional. The nine infant formulas had repeatability within requirements and good intermediate reproducibility. Sample 10 is adult nutritional ready to feed with GOS (code N1P - 00019) RSD(r) of 8.3% and RSD(iR) 9.8 %. The ERP recommended to first action for both Infant f ormulas and adult Nutritionals despite repeatability criterium not being met for the only sample on Adult nutritionals. General comments (i.e. safety) about the method: • Feedback from Users of the Methods since being awarded First Action Official Methods Status Thirteen labs contributed to the MLT , a number of the labs needed assistance in identification of the peaks. Two out of thirteen labs used both MS and dextran ladder for identification. The results shown in the MLT report ware based on dextran ladder. Some suggestions were made by labs to simplify the sample treatment of method or data processing. These points can be added as notes in the method. • Does the method demonstrate reproducibility? To what extent?

2

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

Additional Pros/Strengths: • The report demonstrates that the method meets the performance requirements defined in AOAC International SMPR 2014.003 for Infant formula powder and liquid samples. Different GOS containing formulas were analyzed containing either FOS, HMO, probiotics or hydrolyzed protein. • The method is based on proven and robust technology (labeling and HPLC-FLD). • The equipment is commonly available in IF manufacturers, authority and contract labs. • Critical point of the method were well described and already published in the First action of the method (standards, S/N chromatogram integration, use of Dextran ladder and MS) • Adaption of the method to remove toxic regent sodium cyanoborohydride replaced by less toxic 2 methylpyridine borane complex (agreed by ERP In earlier stage) • Use of SRM NIST 1949b sample as a Practice sample for accuracy determination. • Low amount of outliers, almost all results from participants were used to assess the reproducibility • Safety paragraph was added to the method part • System suitability paragraph was added to the method • Additional Cons/Weaknesses • Only one Adult nutritional sample RTF (PKB- 0007) was tested during MLT study and seems to have homogeneity issues as RSD(r) = 9.9% (SMPR 6%) . Although it almost meets RSD R of 12% it clearly raises questions on applicability for method for (RTF) Adult Nutritionals. Question to the authors is this the same sample as the AN sample in the SLV which seemed also to have some homogeneity issues? How was the homogeneity of this sample accessed before including this sample in the MLT sample set? • Guidance needed from the Study director before starting the analysis of the MLT samples. • Outcome of the T-test (table 3) would have been a nice addition as a clear prove of no observed bias between the two reductive amination reagents used. •

Supporting Information • General Comments

• In table 7 and 8 results per sample and laboratory (powder and RTF samples are given including outlier testing for Cochran and Grubbs. These outlier testing was checked. Outlies were well indicated and removed. However in formula 1 lab 10 is given as Cochr an outlier, but this is clearly a Grubbs outlier. In formula 3 lab 11 is given as Grubbs outlier , but this is a Cochran outlier. In formula RTF PBK- 00005 lab 10 is given as Cochran outlier, but this is a Grubbs outlier. The other outliers are well denote d. There is need for editing these outliers in the final publication. This has no consequence for the performance characteristics. • The contribution of GOS to Dietary Fibre is given as additional results, these results have not been reviewed at this moment.

3

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION (DRM)

• Performance Characteristics and compared to approved SMPR: Analytical Range: 0.23 to 0.86 g/100 g (SMPR 0.2 - 3.0 g/100g) The study director was not able to source MLT materials as the maximum permitted dose of GOS is 0.8 g/100g) In the SLV was shown that the method is applicable in the range 0.2 - 3.0 g/100g. LOQ: i s the SLV is given that the method can accurately determine GOS at a concentration of 0.2 g/100 g and meets LOQ of the SMPR Accuracy/Recovery: Determined in SLV. PS-1 (SRM 1849b) results have been compared to (non- certified) reference value. MLT result w as 2,61 g/100g and (non - certified) reference value is 2,575 g/100g. MLT result of PS-1 is in the acceptable range of the reference value Precision (RSD r ): Infant Formula powder and liquids : 1,4- 4,7%, (SMPR <6%) ;Adult nutritional sample: 9,9% Reproducibility (RSD R ): IFT Formula powder and liquids : 8,1- 11,6%, (SMPR <12%) ;Adult Nutritional sample: 12,1%

ERP Summary/Recommendation(s):

• Does the method meet the approved SMPR? YES for infant formula. Not proven for Adult nutritionals

• Has the method been performing well in the laboratory? YES, the laboratory needs to gain knowledge on GOS to understand and interpret the results and then the method shown good performance.

• Are there any additional feedback on method performance?

• Would you recommend designating this method as a Final Action method to the ERP?

Recommend to move this method AOAC 2021.01 to final action for the determination of total GOS in Infant formula (powder and liquid). Regarding Adult Nutritionals, only one sample was tested for precision and reproducibility. This Sample does not meet requirement on repeatability. This is the same situation as in the SLV also one sample was tested and did not meet repeatability requirement. At that time the ERP concluded that the method was fit for purpose also for AN? Final Comments? The collaborative study was organized with 14 laboratories from 8 different countries globally. Thirteen laboratories were able to successfully analyze six infant formula and one adult nutritional showing clarity and robustness of method . This method is also in process to become equivalent ISO method ISO /CD 7102

4