AOAC SPIFAN Nutrients ERP Reviewer Forms (December 7, 2022)

FOR EXPERT REVIEW PANEL USE ONLY NOT FOR DISTRIBUTION

So, it is necessary to adjust the sample preparation to cover the SMPR working ranges. I would thus suggest adding a table with some guides to ensure the diluted sample is within the calibration curve working range. Something like this may work:-

Expected HMO concentration in reconstituted or RTF product (mg/100g)

Suggested aliquot (g)

Dilution volume (mL)

1.5 – 45 5 - 150 20 - 600

5

25 25 25

1.5 0.4

• The above table could then also include the “DF” to be used in the calculation later? • Suggest adding the LC column dimensions etc, rather than than just the part numbers. • Sample reconstitution and further dilutions are carried out at room temperature (except for the sample needing treatment with enzymes, which is heated to 40°C). I would have thought it be preferable to use hot (or warm) water for the reconstitution (however the results do suggest this is not necessary). •

Supporting Data • General Comment:

The supporting data all look very promising. However the design of the experiments describing how those data were obtained is generally missing. Better descriptions of the design of the spike/recovery experiments and the precision experiments could be added.

Method Optimization: This is not really discussed, however the chromatograms demonstrate good separation of the analytes and the trueness / precision results demonstrate the method is working in a range of different matrices and in the presence of potential interferents including GOS, FOS, NeuGc-Lac (from goat milk), probiotics. Looking at figure 1 (and assuming correct signal annotation) it appears that DFL will elute very close to 3FL. Have the authors assessed if the presence of DFL will impact the measurement of 3FL and if so at which relative concentrations will there be an issue? Or do the authors consider that DFL containing products should be excluded from the scope of the method ?

3