A Targeted Metabolomic Assessment of Oral Glutathione Bioavailability and Safety in Humans: A Randomized Crossover Clinical Trial

Round 1

Reviewer 1 Report

• In Materials and Methods, the reviewer requests clearer reporting of sample size (n) for each study arm at the start of each relevant subsection (pharmacokinetics; safety/tolerability), rather than later in the text.
• The reviewer raises a concern regarding Table 2 half-life values (very long; shortest reported 85 hours), suggesting that the stated one-week washout may be insufficient based on typical washout considerations (5–7 half-lives), and requests author commentary in light of the molecule’s characteristics.

• The reviewer’s primary substantive critique is the simultaneous presentation of raw (non–dose-corrected) and dose-corrected outcomes, arguing that the manuscript does not adequately justify including non–dose-corrected data and recommending that results be limited to dose-corrected analyses to improve concision and focus.

 

This is an interesting study on the pharmacokinetics of glutathione. The introduction clearly demonstrates that the authors are aware of the challenges associated with studying the pharmacokinetics of an endogenous molecule. It provides a solid background and appropriately outlines the context of the topic under investigation.

 

When reaching the Materials and Methods section, this reviewer believes that the sample size for each study arm (n) should be stated more clearly from the outset. Although this information is provided later, it would be straightforward to include it in the first paragraph of each study subsection (pharmacokinetics / safety and tolerability).

 

I understand that the choice of different doses for the reference formulation likely reflects available commercial presentations. Including a placebo or control group might have been beneficial; however, since the objective is to compare formulations, this may not be strictly necessary. Nevertheless, administering two capsules in the reference group could potentially compromise the study blinding.

 

Regarding the analytical method, in line 286 I believe there is a typographical error: “from 2% A to 65% B in 1.5 minutes.” My understanding is that the gradient of solvent B changes from 2% to 65% over 1.5 minutes. This should be clarified.

 

The sample size is relatively small and could potentially limit the robustness of the statistical analysis. However, the authors report significant findings despite the limited sample size, which reinforces the conclusions of the study.

 

A minor concern relates to the results presented in Table 2. The reported half-lives are very long, with the shortest being 85 hours. This would imply a washout period of approximately five to seven half-lives, suggesting that a one-week washout period may not be sufficient. The authors may wish to comment on this point, particularly considering the characteristics of the molecule under investigation.

 

I also noted the presence of Asian script characters in lines 488 and 503.

 

Finally, my main criticism concerns the presentation of both raw and dose-corrected results. In this reviewer’s opinion, there is insufficient justification for presenting the non–dose-corrected data. If the intention is to demonstrate that the LipoMicel formulation performs better even at a lower dose, this is already evident from the dose-corrected results. Presenting both sets of data adds length to the manuscript without providing substantial additional value. I would recommend limiting the presentation to dose-corrected results.

 

In conclusion, I would like to congratulate the authors for designing a study that clearly demonstrates improved outcomes with the LipoMicel formulation without altering the redox balance. My recommendation is acceptance for publication in Antioxidants.

 

 

Author Response

Please see attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

NA

I have the following comments

1. For Figure 4, the time-concentration profile of GSH appears to still be in the absorption or distribution phase. It is unclear how the authors calculated the terminal half-life. Please clarify the method used for this calculation.

 

2. If the half-life reported in Table 2 is accurate, a one-week washout period is insufficient. At least 5 half-lives will be needed to ensure complete elimination. The authors should conduct statistical analyses excluding potential dose-accumulation effects. Furthermore, I recommend providing individual PK profiles, preferably showing the profiles of 3 drugs in a figure, so readers can easily determine whether there is dose accumulation.

 

3. The standard deviation (SD) reported for LMG is excessively large. For example, Table 2, half-life = 7 ± 252.9. The authors should present this data using the median and range(or interquartile range) to provide a more accurate representation.

Author Response

Dear Reviewer,

Thank you very much for your careful review of our manuscript and for the constructive comments. We have revised the manuscript accordingly and believe the changes have strengthened the clarity and rigor of the work. Key revisions include: (i) clarification of baseline-adjusted (incremental) PK analyses for GSH, an endogenous analyte, (ii) streamlined presentation of dose-normalized results given unequal administered doses, and (iii) edits throughout the Results and Discussion to improve consistency and interpretation.

Below, we respond to each comment point-by-point.

 

Reviewer comments:

I have the following comments

1. For Figure 4, the time-concentration profile of GSH appears to still be in the absorption or distribution phase. It is unclear how the authors calculated the terminal half-life. Please clarify the method used for this calculation.

We agree with the reviewer. Glutathione (GSH) is an endogenous compound with baseline circulating concentrations and homeostatic regulation. In our 24-hour sampling window, the concentration–time profiles do not consistently demonstrate a clear terminal log-linear decline suitable for robust estimation of the terminal rate constant (λz) and half-life (T½). Accordingly, we have now removed terminal elimination parameters (λz/T½) and MRT from the manuscript and now focus on incremental exposure PK metrics. Specifically, we now report baseline-adjusted incremental exposure (iAUC_0-24) and baseline-adjusted peak response (ΔC_max); please see changes throughout Results and Discussion.

Manuscript changes:

• Removed T½ and MRT from Tables and the associated Results/Discussion text.
• Added baseline-adjusted PKs and adjusted text throughout the paper.
• Added incremental PK parameters for iAUC_0-24, and ΔC_max.

 

1. If the half-life reported in Table 2 is accurate, a one-week washout period is insufficient. At least 5 half-lives will be needed to ensure complete elimination. The authors should conduct statistical analyses excluding potential dose-accumulation effects. Furthermore, I recommend providing individual PK profiles, preferably showing the profiles of 3 drugs in a figure, so readers can easily determine whether there is dose accumulation.

Response:

We agree that the relevant issue is whether there is evidence of carryover/dose accumulation across crossover periods. Because whole-blood GSH and related analytes are endogenous and do not exhibit a consistently identifiable terminal elimination phase within the 24-hour sampling window, conventional “≥5 half-lives” washout criteria are not applicable. We therefore evaluated carryover directly using pre-dose (t=0) concentrations and period/phase comparisons.

First, baseline (t=0) concentrations were statistically comparable across treatments at the start of each period. As shown in Supplementary Materials S2 (baseline comparison), a mixed-effects model with fixed effects for analyte, treatment, and analyte×treatment found no baseline treatment effect (Treatment p = 0.7801; Analyte×Treatment p = 0.6239), and all Tukey-adjusted pairwise comparisons within each analyte were > 0.05. Second, when baseline GSH concentrations were examined by study phase (period) using the randomization schedule to assign the period-specific treatment, pre-dose values were not systematically elevated in later phases (see Supplementary Material S4 - Addendum, Table S19. Pre-dose (t=0) whole-blood GSH by phase (period) and paired phase comparisons; no phase effect).

In addition, as recommended, we provide individual subject-level concentration–time profiles overlaying the available treatment curves for each participant (Supplementary Material S4 - Addendum: Figure S9. Individual whole-blood GSH concentration–time profiles by participant (raw concentrations). Visual inspection does not suggest systematic pre-dose elevation in later periods. While some within-subject baseline variability across periods is evident—expected for an endogenous analyte—this variability is addressed by our primary PK endpoints, which are baseline-adjusted (incremental AUC_0-24 and ΔC_max).

Manuscript changes:

• Removed T½/MRT from Table 2 and related text; added baseline-adjusted PK endpoints (iAUC_0-24, ΔC_max).
• Added Table S2 Supplementary Results section on Carryover/accumulation assessment based on baseline comparability (please see Supplementary Material S4 - Addendum).
• Added baseline GSH by phase (period)- please see Supplementary Materials S2: Baseline Comparisons); Table S9. Baseline (Time 0) concentrations of glutathione and metabolites by treatment.
• Added Supplementary Fig. S1 (individual overlays) for visual assessment of carryover. Please see Supplementary Material S4 - Addendum: Figure S9. Individual whole-blood GSH concentration–time profiles by participant (raw concentrations.

 

1. The standard deviation (SD) reported for LMG is excessively large. For example, Table 2, half-life = 7 ± 252.9. The authors should present this data using the median and range(or interquartile range) to provide a more accurate representation.

We agree that the large dispersion for T½ reflects that the terminal phase was not reliably identifiable for this endogenous analyte under our sampling design, and that the resulting parameter estimates are not meaningful. Consistent with our response to Comment 1, we have removed T½ and MRT from Table 2 and the manuscript. We now report baseline-adjusted endpoints (iAUC_0-24 and ΔC_max), which are well-defined and interpretable for endogenous compounds.

Manuscript changes:

• Removed half-life-related summary values from Table 2.
• Added iAUC_0-24 and ΔC_max

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

All my questions have been addressed and I have no further comments. 

NA