This is an interesting manuscript but it feels like it has been rushed to submission and is really lacking clear descriptions and presentation of the data. The focus on Kynurenine lacks merit based on how the manuscript is written and the data presented.

I have suggested Major revisions based on the following issues.

1.Baseline characteristics should be presented in a table. It would be good to see a comprehensive breakdown of the T-reg data in a graph as the placebo group has notably lower levs, how do the median values compare?
2. The authors' state "We also
showed a non-significant decrease of Treg in placebo group (mean decrease of 14 ±25 %, p=0·25) over the same period. "
I dont see the point of even mentioning this and it just detracts from the argument unless Significant or very close to significant changes are observed
3.Figure 1 A is far too small to see the data effectively
4. The authors state- A total of 186 metabolites were detected in the blood of the 24 subjects.
Please Show the traces and list all the metabolites detected.
5. The authors state
“Baseline metabolome profiles showed no difference between groups T and P after a permutation test performed on the sPLS-DA model (data not shown) “
This data should be shown
6. The authors state
“. Metabolites retained for the model or which were significantly different between the two groups with raw p-values (none were using the FDR method) included Phosphatidylcholine (PC) aa C32:3, C12:1, lysoPC a C26:1, PC ae 42:0, PC aa C32:2, PC aa C34:4, PC aa C36:4 and kynurenine.
Does included mean there were more than described above. Again a table should be given of all changes if additional to above.

1. The authors state
   “Contrary to the findings of group P, the paired sPLSDA-model was able to discriminate the three times points within group T, as observed in the scores plot, with a correct separation of the metabolomes at the different time points (Figure 1B). The model was significant with a modest accuracy (average error rate of 42%). The most important metabolites for the model were kynurenine, PC aa C36:5, PC aa C38:5, alanine, SM (OH) C14:1 and citrulline.
   Again, please define what important for the model mean? This is not clear
2. The authors state
   Among the important metabolites, kynurenine displayed an interesting pattern, with an increase of its level between D1 and D64 followed by a reduction between D64 and D85 in group T (Figure 1C)
   Again, what does interesting mean scientifically for kynurenine, the language used does not help here. Please be specific. Especially as it looks like kynurenine did not reach significance in terms of changes and dos not correlate with T-reg number? This is an important point based on the title of the paper. Why is it more interesting for ALS vs PC or alanine or SM (OH) aside from IL-2 inducing IDO-1 activity?
3. The authors state
   “Based on the metabolome response between D1 and D64 in group T, we noted 16 metabolites with a raw p-value below 0·05 and two (kynurenine and histidine) with a p-value below 0·1 after FDR correction (but not <0·05).”
   Please provide details of the 16 metabolites
4. The prediction model data is poorly described and discussed. ADMA is not defined anywhere, some discussion is given to methionine but nothing to the rest.

In my opinion overall the authors should alter the focus away from kynurenine and change the title. There is some interesting overall findings including the models built but the way the paper is written and the lack of focus in some areas produces a disjointed manuscript, which lacks discussion of the data presented and is not transparent enough in how the data I presented. The figures should be bigger with titles and better descriptive legends.