Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents a well-conceived and clinically relevant study on the association between circulating retinoic acid (RA) levels and Type 2 Diabetes Mellitus (T2-DM) in a Saudi Arabian population. The study benefits from a large, well-characterized cohort and a robust methodological approach, including the use of matched controls and a validated surrogate marker for insulin resistance (TyG-IR). The finding of significantly lower RA levels in T2-DM patients and the excellent diagnostic performance of RA (AUC = 0.870) are potentially important contributions to the field of metabolic biomarker discovery.

However, the manuscript in its current form requires major revisions before it can be accepted for publication. The primary issues are related to the quality of the English language, significant internal inconsistencies in the results, and a lack of clarity in data presentation. Addressing these points will substantially improve the manuscript's scientific rigor and readability.

Major Required Revisions:

1. Language and Grammar: The manuscript needs extensive copyediting. There are numerous errors throughout, including subject-verb agreement issues, incorrect prepositions, awkward sentence structures, and typos (e.g., "watched the tenets" instead of "adhered to the tenets", "bioethically approved" is awkward, "teetotalers" is informal). A few examples from the abstract: "persisting global and national health challenge" (repetitive), "suing a specific ELISA" (should be "using"), "The study watched the tenets" (should be "followed"). The entire manuscript requires a thorough line-by-line revision.

2. Major Inconsistencies in Results Section:

   • Complication Numbers: The abstract states complications were observed in 33.9% of subjects (99/292). The text in Table 3 states that complications were present in 101 patients. The text in the "Results" section (line 212) says 33.9% (should be 33.9% of 292 = 99), but Table 3 uses N=101 for the "Yes" group. Please verify the correct number and ensure consistency between the text, abstract, and tables.

   • Lipid Profile Contradiction: In the "Results" (lines 222-225), the text states, "Interestingly, total and LDL-C cholesterol levels did not differ significantly between the two groups (P = 0.964 and P = 0.464, respectively)." However, Table 2 shows the P-value for LDL-C is 0.464, but the mean values are 1.8±0.9 for cases and 2.2±0.3 for controls. This difference of 0.4 mM/L is likely clinically meaningful, and the p-value of 0.464 seems improbable given the standard deviations and sample sizes. Please double-check this statistical analysis. The text's statement about LDL-C not differing seems to contradict the data presented.

3. Clarity of Figures and Tables:

   • Table 1: The range for age is given as 23-99. This is an extremely wide range. Please confirm if 99 is a typo or a real value. If real, comment on its influence.

   • Table 2: The units for RA are ng/mL. The range for controls (2.1 - 25.4) is very wide with a high SD (4.3). Please comment on this heterogeneity in the control group. Are there any outliers?

   • Figure 1: The figure is clear. However, the manuscript text mentions "z statistic = 17.611" for the ROC curve. The standard reporting metric for ROC is the AUC, its CI, and the p-value. The z-statistic is not typically reported and can be omitted.

Minor Required Revisions and Clarifications:

4. Introduction: The introduction is very detailed but overly long (over 75 lines). While comprehensive, it could be streamlined by moving some of the detailed mechanistic descriptions (e.g., lines 96-112) to the discussion section, keeping the introduction focused on establishing the knowledge gap and the study's specific aims.

5. Methods:

   • Sample Size Calculation (lines 158-161): The justification for using a vitamin A deficiency prevalence of 25% is unclear. The study measures RA, not vitamin A. Please clarify this assumption or recalculate the sample size based on a more relevant endpoint (e.g., expected difference in RA levels from a pilot study or prior literature).

   • Treatment Score (line 183): The scoring system (0-3) is introduced but never used in the results. It appears you used a binary (insulin Yes/No) analysis instead. Either use the score in the analysis or remove it from the methods.

   • ELISA Method (line 188): The inter-assay CV (<12%) is acceptable but on the higher side. Please discuss this as a potential limitation.

6. Results:

   • Table 3: The column "Yes (101)" is correct if the total with complications is 101. The earlier text (line 212) needs to be corrected from "99 (33.9%)" to "101 (34.6%)". Also, the row for "Insulin Yes/No" is confusing. It shows "146 (76)" for the 'No' complication group. Does this mean 146 out of 191 patients without complications are on insulin? This seems high. Please clarify the numbers in this row.

   • Correlation Table (Table 4): The correlation between LDL-C and TC (r=0.708, p<0.001) is expected. However, the correlation between LDL-C and RA (r=0.273, p<0.001) is positive, which is interesting because lower RA is associated with worse metabolic health. Typically, LDL-C is higher in poor metabolic health. Please provide a brief interpretation of why lower RA is associated with lower LDL-C (since the correlation is positive, higher RA goes with higher LDL-C). This is a nuanced finding that needs discussion.

7. Discussion:

   • The discussion is strong but overlong. It can be tightened by about 20%.

   • The discussion of the paradoxical findings regarding LDL-C (point #6 above) should be added.

   • The paragraph on clinical translation (lines 345-356) is speculative. While interesting, tone down the therapeutic claims ("the retinoid axis itself represents a therapeutic target") as this is a cross-sectional study, not an interventional trial.

   • The limitations section should also mention the lack of dietary vitamin A intake data and the absence of measurements for retinol or other retinoid metabolites (like 13-cis RA, 9-cis RA), which could provide a more complete picture.

Conclusion on Recommendation:

The core scientific finding is interesting and has potential for publication in Diagnostics. However, the manuscript's presentational flaws and internal inconsistencies significantly undermine its credibility. The authors must carefully address the major inconsistencies in the data (especially regarding complication numbers and lipid profiles) and substantially improve the English language quality. Therefore, my recommendation is to reconsider after the major revision.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper explores the relation between retinoic acid and type 2 diabetes. It is an underexplored field in literature, therefore the research aim is quite interesting and worthing further evaluations in the future. The paper is well structured providing full explanation of the role of retinoic acid and metabolism, sustaining the original hypothesis, providing explanation in the introduction of the role in type 2 diabetes and its complications. It has been conducted s a cross sectional study comparing a group with type 2 diabetes with healthy group, and the statistical analysis applied is coherent. The methodology is appropriate and well designed with clear inclusion criteria, however, the controlled group is smaller(64) comparing with the type 2 dm groups(292) which could provide bias. It may worth in the future to design a similar study with larger control group. The table are well structured and clearly reported; the references are appropriate to the scientific context. The conclusion of this study sustains a relation between retinoic acid and type 2 diabetes, in particular retinoic acid is significantly reduced in type 2 diabetes and inversely correlated with glycaemic control and insulin resistance; this may represent a novel marker in type 2 diabetes,  but needs further exploration with future studies.  

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

It is observed that the authors have made significant improvements based on the feedback and comments provided to them for revision.

Comments on the Quality of English Language

Nothing in particular.