Low Lactose ERP - Review Book

3. Is there information demonstrating that the

No, there is very little data supporting the applicability of the method to a broad range of matrices which are lactose-free or low lactose.

method performs within the SMPR Method Performance REquirements table specifications for all analytes in the SMPR applicability statement? If not, please specify what is missing and whether or not the method's applicability should be modified.

The assessment of LoD / LoQ has been perfromed using 2 different approaches (siganl/noise assessement and Assessemnt of SD of a sample containing low lactose concentrations). There is a very big difference between the LoQs assessed by the two approaches which is unusual. Indeed using the signal to noise approach the LoQ would not meet the requirements of the SMPR. Of course the authors selected the approach which gives them the best perfromance, but I am not sure that is appropriate, without argumentation why they rejected the results using the other approach. Suggest the authors try to understand why there is such a large discrepancy between the approaches and try to make a better assessment. They could also supply us with all the data used to make the estmate of LoQ using the approach using the SD of a low lactose sample, as all the raw data are not available for assessment.

The authors have perfromed some spike-recovery experiments on reference materials, but have not used the appropriate calculation to assess recovery.

There is no data demonstrating the applicability of the method to lactose-free infant formula and only one set of data (coop lactose-free milk) indicating that the method may be applicable for the analysis of lactose-free products. The authors need to broaden the number of matrices covered, and perfrom appropriate spike-recovery experiments in those matrices (as well as collecting RSD(r) data)

It is strongly recommended the authors refer to OMA appendix L – SPIFAN guidelines for SLV.

In particular pay attention to the collection of data to estimate RSD(r) and RSD(iR) ( samples analysed in duplicate on 6 days) and trueness (spiked samples or reference samples analysed in duplicate on 3 different days). In general the authors have not collected enough replicates to complete the assessment of either repeatability or trueness. The authors should also note that it is not possible to assess reproducibility (RSD(R)) without a multi-lab study. This should not be referred to when reporting data from a SLV.

IV. General Submission Package 1. Based on the supporting information, were there any additional steps in the evaluation of the method that indicated the need for any additional precautionary statements in the method? 2. Does the method contain system suitability tests or controls as specified by the SMPR? If not, please indicate if there is a need for such tests or controls and which ones.

no

Yes the authors included blank injections and check standards