Review Reports - Effect of Oat Flakes on Glycemic Variability, Dyslipidemia, and Pancreatic Duodenum Homeobox-1 (PDX-1) Level Among Adolescents with Type 1 Diabetes: A Randomized Crossover Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript is interesting, but there are several concerns that should be addresssed in the manuscript.

What is the reason for dividing into Group A and Group B?
In Figure 3, is it significantly increased in Group A after 6 months in PDX?
It appears that the changes in PDX and the changes in TG, cholesterol, and LDL-C are opposite in both Group A and B. Is there any correlation among these?

Author Response

Thanks a lot for your positive feedback.

Comment: What is the reason for dividing into Group A and Group B?

Comment: In Figure 3, is it significantly increased in Group A after 6 months in PDX?

Response: Thanks for the comment, the PDX1 was significantly increased at 3 months then decreased at 6 months (p < 0.001). This was added to the figure.

			Group A					Test value	P-value
			Baseline	At 3 month		At 6 month
PDX1 (baseline) ng/dl	Mean ± SD		135.94 ± 65.49	957.05 ± 476.44		237.72 ± 97.01		104.32•1	<0.001
	Range		50.76 – 286.2	316.1 – 1474		57.3 – 371.8
Post hoc analysis
		Baseline Vs At 3 month			Baseline Vs At 6 month		At 3 month Vs At 6 month
PDX1 ng/dl		<0.001			<0.001		<0.001

Comment: It appears that the changes in PDX and the changes in TG, cholesterol, and LDL-C are opposite in both Group A and B. Is there any correlation among these?

Response: Thanks for the comment. Although PDX1, TG, cholesterol and LDL increased during B-glucan supplementation, no significant correlation was found between the percent change of PDX1 and percent change of TG, cholesterol and LDL.

Correlation between change in PDX1 level of the studied patients and percentage of change of the studied parameters

	PDX1 ng/dl
	r	p-value
Weight (kg) on z score	0.121	0.526
Height (cm) on z score	-0.107	0.573
BMI on z score	0.399*	0.029
Total daily dose (unit/ kg/ day)	0.142	0.455
HbA1c %	0.219	0.092
TG mg/dl	-0.038	0.773
Cholesterol mg/dl	0.183	0.161
LDL mg/dl	-0.073	0.580
HDL mg/dl	0.233	0.073
Time in range(TIR) 70-180 mg/dl (%)	-0.315*	0.014
Time below range (TBR) < 54 mg /dl( %)	0.287	0.080
Time below range (TBR) 54-69 mg /dl( %)	-0.266*	0.047
Time above range (TAR) 180 -250mg /dl( %)	0.082	0.545
Time above range (TAR) >250 mg /dl ( %)	0.243	0.074

Spearman correlation coefficients

Reviewer 2 Report

Comments and Suggestions for Authors

In this manuscript, the authors describe a crossover-clinical design study examining the effect of beta-glucan supplementation in adolescents with type 1 diabetes. They examine the effect of beta-glucan on several parameters relevant to T1D progression including hemoglobin-A1C, blood lipids and measures from continuous glucose monitoring.

The authors data indicate that providing 45g of oat flakes containing 2g of beta-glucan at each meal resulted in improvements in measures of glucose control, increasing time in range, and decreasing both time above and below range. The main endpoint of glycemic variability was significantly reduced under supplementation in both treatment groups.

The manuscript is generally well written, but has a few major flaws that must be addressed.

My three major criticisms are as follows,

The use of serum PDX-1 as a marker of pancreatic reserve is problematic. PDX-1 is a transcription factor found in beta cells, and to my knowledge not secreted in any circumstance. The fact that the authors present this changing as a result of their treatment is extremely confusing, and the fact that they detect this in plasma calls into question the validity of their assay. To my knowledge the studies that measure PDX-1 as a marker of pancreatic reserve/differentiation are done in isolated pancreases or beta cell cultures. Given this I do not think any data related to PDX-1 in this paper can be considered valid, and it must be removed.
The authors either misrepresent their treatment throughout their paper, or significantly more detail must be added for their isolation of beta-glucan. As presented the title, abstract, results, and discussion imply an isolated beta-glucan supplement, however their true treatment as described in the methods appears to be 3 servings (45g each) of an oatmeal-like product at each meal. There is a dramatic difference between providing a whole food source of an assumed bioactive substance and an isolated supplement, especially when this quantity of food can have dramatic effects on satiety and other factors related to glycemic control. The authors must adjust their description throughout the paper to better reflect what this trial is providing. If the authors are indeed giving a 2g supplement of beta-glucan the details on its isolation or the product that provides this must be updated.
The statistical analysis and presentation of the data used here are both confusing and under described. Given this is a crossover trial, It is my understanding that the authors should be able to examine the treatment and placebo effects of the combined trial arms, (Combining data from Baseline to 3 Months of Group A and 3 Month to 6 month of Group B as a “treatment effect”; similarly 3 month to 6 month of Group A and 3-month to 6-month of Group B as placebo), especially if there is no assumed carryover effect following the washout. As the data are presented, the table is nearly incomprehensible and makes it harder to interpret the major effects.

While I appreciate the presentation of the individual arms, I feel this would be better shown in supplemental tables, and I would recommend the authors show the main mean differences for placebo versus treatment using combined arms for the main figure of the paper. This would also streamline the description of the major endpoints and permit more specific descriptions in the results of the major findings. As written, no specific data are described in the results text, it would be beneficial to at least highlight the specific main results in text form.

Further the graphs in figure 2 do not substantially add to the overall narrative as they simply repeat the data shown in table 1 without measures of variability or significance and should be removed.

It is unclear why the authors do not report if there are alterations in BMI and/or height following these interventions.

It appears that many references are not properly aligned for the statements they are meant to support.

One example is reference 22 which has no mention of PPARs anywhere in the paper. Please ensure proper alignment of references without this, I cannot properly assess the validity of the statements made in the discussion.

Within the results, the authors directly contradict themselves citing a study 1 week in duration (line 228) with beneficial effects, then later citing a study with null results, the explanation of which is due to its short duration of 2 weeks (line 238). Please align your statements.

Author Response

Reviewer 2:

The manuscript is generally well written, but has a few major flaws that must be addressed.

Thanks a lot for your positive feedback.

Comment: The use of serum PDX-1 as a marker of pancreatic reserve is problematic. PDX-1 is a transcription factor found in beta cells, and to my knowledge not secreted in any circumstance. The fact that the authors present this changing as a result of their treatment is extremely confusing, and the fact that they detect this in plasma calls into question the validity of their assay. To my knowledge the studies that measure PDX-1 as a marker of pancreatic reserve/differentiation are done in isolated pancreases or beta cell cultures. Given this I do not think any data related to PDX-1 in this paper can be considered valid, and it must be removed.

Response: Thanks for your comment. I agree with you that PDX-1 is a nuclear protein that is primarily known for its role in pancreatic development and ß-cell maturation, rather than as a hormone or traditional protein biomarker that circulates at high levels in the serum of healthy individuals. However, research indicates that PDX-1 can be detected in human sera of people with diabetes, people with pancreatic cancer, and healthy populations which they attributed to hormonal changes, PDX1 secretion in extracellular vesicles (e.g., exosomes) under certain pathological stress or conditions (such as diabetes, pregnancy and cancer) and, in cases of cellular stress, nuclear proteins like PDX1 could leak into the extracellular space and subsequently enter the bloodstream. Diabetes is a special condition with stress and inflammation, during which PDX1 may present in serum. I added this to the discussion. " In the present study, serum PDX-1, a master regulator protein in β-cell maturation and identity preservation, was found to be detectable in sera of adolescents with T1D. PDX-1 is a nuclear protein that is primarily known for its role in pancreatic development and ß-cell maturation, rather than as a hormone or traditional protein biomarker that circulates at high levels in the serum of healthy individuals (30). However, research indicates that PDX-1 can be detected in human sera of people with diabetes, people with pancreatic cancer, and healthy populations (31, 32). The presence of PDX-1 in serum could be attributed to hormonal changes; secretion of PDX-1 in extracellular vesicles (e.g., exosomes) under pathological stress (such as diabetes and cancer) and cellular stress and inflammation, which could cause leakage of nuclear proteins like PDX-1 into the extracellular space and subsequent entry into the bloodstream (32). Interestingly, a significant increase of serum PDX-1 level was observed during β-glucan supplementation, reflecting an increase in the pancreatic reserve. This goes in line with Zhang and colleagues, who reported that increased serum PDX1 in early pregnancy is associated with decreased risks of gestational diabetes mellitus in a prospective study including 231 pregnant women (32)."

References:

Zhang Q, Zhang QQ, Dong SQ, Liu X, Wei J, Li K, Lu Y. PDX1 in early pregnancy is associated with decreased risks of gestational diabetes mellitus and adverse pregnancy outcomes. Front Endocrinol (Lausanne). 2025 May 8;16:1486197. doi: 10.3389/fendo.2025.1486197. PMID: 40405977; PMCID: PMC12094967.

Duarte-Medrano G, Lopez-Méndez I, Ramírez-Luna M, Valdovinos-Andraca F, Cruz-Martínez R, Medina-Vera I, et al. Analysis of circulating blood and tissue biopsy pdx1 and msx2 gene expression in patients with pancreatic cancer: A case-control experimental study. Medicine. (2019) 98:e15954. doi: 10.1097/md.0000000000015954

Comment: The authors either misrepresent their treatment throughout their paper, or significantly more detail must be added for their isolation of beta-glucan. As presented the title, abstract, results, and discussion imply an isolated beta-glucan supplement, however their true treatment as described in the methods appears to be 3 servings (45g each) of an oatmeal-like product at each meal. There is a dramatic difference between providing a whole food source of an assumed bioactive substance and an isolated supplement, especially when this quantity of food can have dramatic effects on satiety and other factors related to glycemic control. The authors must adjust their description throughout the paper to better reflect what this trial is providing. If the authors are indeed giving a 2g supplement of beta-glucan the details on its isolation or the product that provides this must be updated.

Response: Thanks for your valuable comment. Indeed we used oat flakes as a natural source for B glucan. We adjusted this in the title, abstract, results and disciussion and added its possible effect to the limitations " Another limitation is the use of a whole food source of β-glucan, as the quantity of oat flakes can affect satiety and other factors related to glycemic control."

Comment: The statistical analysis and presentation of the data used here are both confusing and under described. Given this is a crossover trial, It is my understanding that the authors should be able to examine the treatment and placebo effects of the combined trial arms, (Combining data from Baseline to 3 Months of Group A and 3 Month to 6 month of Group B as a “treatment effect”; similarly 3 month to 6 month of Group A and 3-month to 6-month of Group B as placebo), especially if there is no assumed carryover effect following the washout. As the data are presented, the table is nearly incomprehensible and makes it harder to interpret the major effects.

Response: Thanks for the nice idea. This was adjusted in tables 2 and 3. Indeed the results are much more clear.

Comment: While I appreciate the presentation of the individual arms, I feel this would be better shown in supplemental tables, and I would recommend the authors show the main mean differences for placebo versus treatment using combined arms for the main figure of the paper. This would also streamline the description of the major endpoints and permit more specific descriptions in the results of the major findings. As written, no specific data are described in the results text, it would be beneficial to at least highlight the specific main results in text form.

Response: Thanks for the comment. Table 2 was shifted to supplementary appendix 5. Comment: Further the graphs in figure 2 do not substantially add to the overall narrative as they simply repeat the data shown in table 1 without measures of variability or significance and should be removed.

Response: Figure 2 was removed.

Comment: It is unclear why the authors do not report if there are alterations in BMI and/or height following these interventions.

Response: Interestingly there was a significant decrease in weight and BMI z score on B-glucan supplementation as shown in table 2.

	B-Glucan group (n=60)					Ordinary diet group (n=60)
	Baseline		Post treatment	Test value	P-value	Baseline	Post treatment	Test value	P-value
Weight z score	Median (IQR)	0.1 (-0.15 - 0.27)	0.08 (-0.18- 0.23)	-3.728≠	<0.001	0.1 (-0.15 - 0.27	-0.03 (-0.29 - 0.11)	-0.862≠	0.389
	Range	-1.58 – 1.11	-1.82 – 0.99			-1.58 – 1.11	-1.05 – 0.95
Height z score	Median (IQR)	-0.34 (-0.88 -0.13)	-0.34 (-0.88 -0.13)	-1.342≠	0.180	-0.34 (-0.88 -0.13)	-0.25 (-0.68 - -0.09)	-2.375≠	0.018
	Range	-2.51 – 0.65	-2.51 – 0.65			-2.51 – 0.65	-2.84 0.13
BMI z score	Median (IQR)	0.38 (-0.04 -0.62)	0.31 (-0.04 - 0.48)	-2.936≠	0.003	0.38 (-0.04 -0.62)	0.2 (-0.02 - 0.57)	-0.921≠	0.357
	Range	-0.31 – 1.35	-0.31 – 1.11			-0.31 – 1.35	-0.83 – 1.42
Total daily dose (U/ kg/ day)	Mean±SD	0.98 ± 0.24	0.73 ± 0.27	7.726•	<0.001	0.98 ± 0.24	1.07 ± 0.26	1.290•	0.207
	Range	0.63 – 1.58	0.22 – 1.28			0.63 – 1.58	0.55 – 1.67
HbA1c (%)	Mean±SD	7.89 ± 1.56	6.50 ± 0.76	10.389•	<0.001	7.89 ± 1.56	7.62 ± 1.39	0.270•	0.788
	Range	5.6 – 12	5.2 – 8.3			5.6 – 12	5.6 – 12
Triglycerides (mg/dl)	Mean±SD	94.50 ± 30.22	87.70 ± 27.11	4.285•	<0.001	94.50 ± 30.22	93.20 ± 26.44	-2.402•	0.019
	Range	58 – 202	54 – 175			58 – 202	52 – 165
Cholesterol (mg/dl)	Mean±SD	149.20 ± 35.36	134.07 ± 26.20	9.045•	<0.001	149.20 ± 35.36	162.73 ± 18.61	-3.862•	<0.001
	Range	111 – 236	95 – 209			111 – 236	128 – 201
LDL (mg/dl)	Mean±SD	70.33 ± 35.76	54.83 ± 25.48	9.049•	<0.001	70.33 ± 35.76	88.07 ± 22.53	-5.144•	<0.001
	Range	25 – 168	21 – 116			25 – 168	48 – 143
HDL (mg/dl)	Mean±SD	60.87 ± 12.25	65.27 ± 7.24	-4.357•	<0.001	60.87 ± 12.25	61.03 ± 9.09	-0.325•	0.747
	Range	33 – 92	49 – 84			33 – 92	40 – 82
TIR (%)	Mean±SD	60.22 ± 13.90	77.37 ± 11.30	-12.936•	<0.001	60.22 ± 13.90	59.17 ± 10.41	2.903•	0.005
	Range	20 – 90	44 – 100			20 – 90	39 – 81
TBR < 54 mg/dl (%)	Mean±SD	1.42 ± 2.08	0.50 ± 0.79	3.328•	0.002	1.42 ± 2.08	1.75 ± 1.74	-3.418•	0.001
	Range	0 – 9	0 – 3			0 – 9	0 – 7
TBR 54-69 mg/dl (%)	Mean±SD	6.80 ± 5.80	4.37 ± 3.56	3.326•	0.002	6.80 ± 5.80	7.08 ± 5.32	-1.640•	0.106
	Range	0 – 28	0 – 15			0 – 28	0 – 27
TAR 180 - 250 mg/dl (%)	Mean±SD	22.47 ± 12.76	14.03 ± 7.69	5.440•	<0.001	22.47 ± 12.76	20.63 ± 9.60	0.043•	0.966
	Range	0 – 80	0 – 29			0 – 80	4 – 37
TAR > 250 mg/dl (%)	Mean±SD	9.10 ± 6.21	3.73 ± 4.33	5.673•	<0.001	9.10 ± 6.21	11.37 ± 7.30	-2.533•	0.014
	Range	0 – 25	0 – 22			0 – 25	0 – 35
GMI %	Mean±SD	7.00 ± 0.73	6.63 ± 0.64	4.822•	<0.001	7.00 ± 0.73	7.17 ± 0.81	-2.723•	0.008
	Range	5.6 – 8.1	5.3 – 7.7			5.6 – 8.1	5.6 – 8.8
CV %	Mean±SD	38.61 ± 7.61	32.72 ± 7.41	5.725•	<0.001	38.61 ± 7.61	39.67 ± 6.47	-3.167•	0.002
	Range	18.4 – 48.2	17.9 – 46.1			18.4 – 48.2	26.3 – 55.3
PDX1 (ng/dl)	Mean±SD	164.20 ± 43.69	1016.54 ± 401.50	-16.369•	<0.001	164.20 ± 43.69	183.31 ± 77.70	-0.465•	0.644
	Range	69.8 – 286.2	316.1 – 1474			69.8 – 286.2	57.3 – 371.8

Comment: It appears that many references are not properly aligned for the statements they are meant to support.

Response: Thanks for your notice. Reference 23 was forgotten by mistake. I added it and revised the rest of the references.

Comment: Within the results, the authors directly contradict themselves citing a study 1 week in duration (line 228) with beneficial effects, then later citing a study with null results, the explanation of which is due to its short duration of 2 weeks (line 238). Please align your statements.

Response: Thanks for the notice. I removed the short duration section to align the statements. " In contrast, a double-armed β-glucan placebo crossover study showed that 2 weeks of β-glucan didn't have a significant effect on glycemic variability in adults with T1D compared to controls (23). The lack of a significant effect in this study could be attributed to the low dose of β-glucan used and the lack of control for a range of factors."

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

After reviewing this study, I have several suggestion:

The rationale of this study is unclear. Similar study design has been conducted in 2017 (10.3390/nu9091004), whereas the dosage and treatment period of β-glucan are insufficient, which is discussed in Line 235-239. Therefore, the key novelty of this study is to demonstrate that high β-glucan meal can support T1D patients in daily glycemic management. However, the authors did not emphasize this argument in the Introduction section.
Please explain the reason of choosing crossover design instead of parallel design. If the reason is to expand sample size, please explain the reason of avoiding wash-out.
Is this study blinded to participants? If not, please explain how to prevent or offset placebo effect.
Please verify the initial time of follow-up. Does that initiate after 6-month of study dosing?
What is the reason of measuring serum PDX-1? If the authors intend to assess the β-cell remodeling, I suggest the authors measuring serum insulin content. That would be a more straightforward evidence.
The authors did not describe the adverse events, body weight, and fasting glucose contents. Since serum PDX-1 significantly increase during study dosing, the insulin sensitivity may be altered.

Author Response

After reviewing this study, I have several suggestion:

Comment: The rationale of this study is unclear. Similar study design has been conducted in 2017 (10.3390/nu9091004), whereas the dosage and treatment period of β-glucan are insufficient, which is discussed in Line 235-239. Therefore, the key novelty of this study is to demonstrate that high β-glucan meal can support T1D patients in daily glycemic management. However, the authors did not emphasize this argument in the Introduction section.

Response: Thanks for the comment. This was emphasized " Thus, β-glucan could be considered to support T1D patients in daily glycemic management. However, results from studies assessing the effect of dietary fiber intake on glycemic metrics in people with T1D are contradictory (8-10), and the long-term effect of high β-glucan oat flakes on glycemic metrics, pancreatic reserve and lipids profile are still obscure. Hence, this study aimed to investigate the effect of long-term oat flakes β-glucan use on pancreatic β-cell reserve, glycemic metrics, and lipid profile among adolescents with T1D.". I hope it is more clear now.

Comment: Please explain the reason of choosing crossover design instead of parallel design. If the reason is to expand sample size, please explain the reason of avoiding wash-out.

Response: Thanks for the comment. A crossover design was used in the study to allow optimal comparison (each participant acts as their own control), eliminate inter-subject variability, and improve the statistical power, allowing for smaller sample sizes while maintaining statistical precision and reducing the costs. A 2-week washout period was used in the current study to eliminate carry-over effect, similar to a previous study assessing the effect of high fiber diet for 6 weeks with a 2- to 4-week washout period between diets. In addition, standardized diet was used during the wash out period to remove the effects of the previous dietary sequence.

Zhang M, Juraschek SP, Appel LJ, Pasricha PJ, Miller ER 3rd, Mueller NT. Effects of High-Fiber Diets and Macronutrient Substitution on Bloating: Findings From the OmniHeart Trial. Clin Transl Gastroenterol. 2020 Jan;11(1):e00122. doi: 10.14309/ctg.0000000000000122. PMID: 31972610; PMCID: PMC7056053.

Comment: Is this study blinded to participants? If not, please explain how to prevent or offset placebo effect.

Response: No, the study was an open label study. This was clarified in the methodology. To offset the placebo effect, the B-glucan was compared against standard care nutritional therapy without using placebo. In addition, standardized nutritional education was given to all participants by the same personnel. Informed consent was given by the same investigator who used neutral language when discussing the study and expected outcomes, objective analysis was used using CGM, serum lipids, HbA1c and PDX-1 and sequential cross over design was used. This was added " Written informed consent and assent were obtained from the patients and their legal guardians before enrollment after a thorough explanation of the study design and expected outcomes using neutral language by the same investigator to all participants."

Comment: Please verify the initial time of follow-up. Does that initiate after 6-month of study dosing?

Response: The initial follow up was after 3 months from the baseline, this was followed by two weeks wash out period then the second follow up (6 months follow up) was after 3 months of the start of the second phase. This was clarified in the text " In the first three months, group A received 6 g of oat flakes β-glucan daily in addition to their ordinary diet, while group B received their ordinary diet alone. This was followed by crossing over both arms for another 3 months after a two-week washout period between the interventions."

Comment: What is the reason of measuring serum PDX-1? If the authors intend to assess the β-cell remodeling, I suggest the authors measuring serum insulin content. That would be a more straightforward evidence.

Response: As you know, all these adolescents have established T1D and are on exogenous insulin, that's why serum insulin assay is not accurate. Hence, we used PDX-1 as a marker of B-cells remodeling.

Comment: The authors did not describe the adverse events, body weight, and fasting glucose contents. Since serum PDX-1 significantly increase during study dosing, the insulin sensitivity may be altered.

Response: Thanks for the comment. Interestingly there was a significant decrease in weight and BMI z score on B-glucan supplementation as shown in table 2.

	B-Glucan group (n=60)					Ordinary diet group (n=60)
	Baseline		Post treatment	Test value	P-value	Baseline	Post treatment	Test value	P-value
Weight z score	Median (IQR)	0.1 (-0.15 - 0.27)	0.08 (-0.18- 0.23)	-3.728≠	<0.001	0.1 (-0.15 - 0.27	-0.03 (-0.29 - 0.11)	-0.862≠	0.389
	Range	-1.58 – 1.11	-1.82 – 0.99			-1.58 – 1.11	-1.05 – 0.95
Height z score	Median (IQR)	-0.34 (-0.88 -0.13)	-0.34 (-0.88 -0.13)	-1.342≠	0.180	-0.34 (-0.88 -0.13)	-0.25 (-0.68 - -0.09)	-2.375≠	0.018
	Range	-2.51 – 0.65	-2.51 – 0.65			-2.51 – 0.65	-2.84 0.13
BMI z score	Median (IQR)	0.38 (-0.04 -0.62)	0.31 (-0.04 - 0.48)	-2.936≠	0.003	0.38 (-0.04 -0.62)	0.2 (-0.02 - 0.57)	-0.921≠	0.357
	Range	-0.31 – 1.35	-0.31 – 1.11			-0.31 – 1.35	-0.83 – 1.42
Total daily dose (U/ kg/ day)	Mean±SD	0.98 ± 0.24	0.73 ± 0.27	7.726•	<0.001	0.98 ± 0.24	1.07 ± 0.26	1.290•	0.207
	Range	0.63 – 1.58	0.22 – 1.28			0.63 – 1.58	0.55 – 1.67
HbA1c (%)	Mean±SD	7.89 ± 1.56	6.50 ± 0.76	10.389•	<0.001	7.89 ± 1.56	7.62 ± 1.39	0.270•	0.788
	Range	5.6 – 12	5.2 – 8.3			5.6 – 12	5.6 – 12
Triglycerides (mg/dl)	Mean±SD	94.50 ± 30.22	87.70 ± 27.11	4.285•	<0.001	94.50 ± 30.22	93.20 ± 26.44	-2.402•	0.019
	Range	58 – 202	54 – 175			58 – 202	52 – 165
Cholesterol (mg/dl)	Mean±SD	149.20 ± 35.36	134.07 ± 26.20	9.045•	<0.001	149.20 ± 35.36	162.73 ± 18.61	-3.862•	<0.001
	Range	111 – 236	95 – 209			111 – 236	128 – 201
LDL (mg/dl)	Mean±SD	70.33 ± 35.76	54.83 ± 25.48	9.049•	<0.001	70.33 ± 35.76	88.07 ± 22.53	-5.144•	<0.001
	Range	25 – 168	21 – 116			25 – 168	48 – 143
HDL (mg/dl)	Mean±SD	60.87 ± 12.25	65.27 ± 7.24	-4.357•	<0.001	60.87 ± 12.25	61.03 ± 9.09	-0.325•	0.747
	Range	33 – 92	49 – 84			33 – 92	40 – 82
TIR (%)	Mean±SD	60.22 ± 13.90	77.37 ± 11.30	-12.936•	<0.001	60.22 ± 13.90	59.17 ± 10.41	2.903•	0.005
	Range	20 – 90	44 – 100			20 – 90	39 – 81
TBR < 54 mg/dl (%)	Mean±SD	1.42 ± 2.08	0.50 ± 0.79	3.328•	0.002	1.42 ± 2.08	1.75 ± 1.74	-3.418•	0.001
	Range	0 – 9	0 – 3			0 – 9	0 – 7
TBR 54-69 mg/dl (%)	Mean±SD	6.80 ± 5.80	4.37 ± 3.56	3.326•	0.002	6.80 ± 5.80	7.08 ± 5.32	-1.640•	0.106
	Range	0 – 28	0 – 15			0 – 28	0 – 27
TAR 180 - 250 mg/dl (%)	Mean±SD	22.47 ± 12.76	14.03 ± 7.69	5.440•	<0.001	22.47 ± 12.76	20.63 ± 9.60	0.043•	0.966
	Range	0 – 80	0 – 29			0 – 80	4 – 37
TAR > 250 mg/dl (%)	Mean±SD	9.10 ± 6.21	3.73 ± 4.33	5.673•	<0.001	9.10 ± 6.21	11.37 ± 7.30	-2.533•	0.014
	Range	0 – 25	0 – 22			0 – 25	0 – 35
GMI %	Mean±SD	7.00 ± 0.73	6.63 ± 0.64	4.822•	<0.001	7.00 ± 0.73	7.17 ± 0.81	-2.723•	0.008
	Range	5.6 – 8.1	5.3 – 7.7			5.6 – 8.1	5.6 – 8.8
CV %	Mean±SD	38.61 ± 7.61	32.72 ± 7.41	5.725•	<0.001	38.61 ± 7.61	39.67 ± 6.47	-3.167•	0.002
	Range	18.4 – 48.2	17.9 – 46.1			18.4 – 48.2	26.3 – 55.3
PDX1 (ng/dl)	Mean±SD	164.20 ± 43.69	1016.54 ± 401.50	-16.369•	<0.001	164.20 ± 43.69	183.31 ± 77.70	-0.465•	0.644
	Range	69.8 – 286.2	316.1 – 1474			69.8 – 286.2	57.3 – 371.8

We didn't measure fasting blood glucose and relied on the glycemic metrics in CGM metrics asthey gives a more broad picture of the glycemic status.

The adverse events were added. "As regards the adverse effects of therapy, three patients had GIT manifestations in the form of abdominal distension and flatulence on starting the B-glucan therapy, which were mild and waned by the second week of administration, not necessitating stoppage of the β-glucan oat flakes." We didn't assess insulin sensitivity byt the total daily dose decreased on the β-glucan which is an indirect reflection of insulin sensitivity.

Reviewer 4 Report

Comments and Suggestions for Authors

The presented study is interesting both methodologically and clinically. The results suggest promising value of dietary β-glucan supplementation in adolescents with type 1 diabetes.
My comments and reservations include:
1. The results regarding the increase in HDL concentrations would also be worth presenting in Figure 2. How can the continued upward trend in HDL concentrations be explained despite discontinuing supplementation with the additional 6 g of β-glucan?
2. What were the rationale for choosing this specific daily dose of β-glucan?
3. Did the authors observe differences in the effectiveness of dietary supplementation with 6 grams of β-glucan depending on the gender, age, and body weight of the patients?
4. What adverse effects did the children report?
5. What was the reason for excluding 14 patients from the group?
6. It would be worthwhile to expand the discussion with a brief note on which other patient groups, besides type 1 diabetes, would benefit from dietary β-glucan supplementation.

Author Response

Reviewer 4:

The presented study is interesting both methodologically and clinically. The results suggest promising value of dietary β-glucan supplementation in adolescents with type 1 diabetes.

Response: Thanks a lot for your positive feedback.
My comments and reservations include:
1. Comment: The results regarding the increase in HDL concentrations would also be worth presenting in Figure 2. How can the continued upward trend in HDL concentrations be explained despite discontinuing supplementation with the additional 6 g of β-glucan?

Response: Thanks for the interesting notice. Reviewer 2 asked to remove all of figure 2 as it is a repetition of the supplementary table 2 but the data are still present in the table. This is certainly worth discussion. A similar study assessing the effect of B-glucan on serum lipids in men with obesity and hypercholesterolemia revealed that HDL-cholesterol concentrations did not increase immediately with consumption of β-glucan fiber, rather, it took 12 wk, which was 4 wk after the cessation of the β-glucan fiber supplement. The mechanism for the increase in HDL-cholesterol concentrations with β-glucan is unknown. It may be that the amount of β-glucan is the key determinant of whether HDL-cholesterol concentration increases.

This indicates a beneficial carryover effect for the B-glucan on HDL up to 12 weeks after stoppage of the B-glucan supplementation.

This was added to the results " Worth mentioning, a significant increase in HDL was observed in group A during the oat flakes β-glucan supplementation, which persisted even after stoppage of the β-glucan (p=0.018), supplementary appendix 2." and discussion " Interestingly, a continued upward trend in HDL concentration was observed even after the discontinuation of β-glucan supplementation in group A. Robert et al, reported a similar finding in a study assessing the effect of B-glucan on serum lipids in men with obesity and hypercholesterolemia. They found that HDL-cholesterol concentrations did not increase immediately with consumption of β-glucan; rather, it took 12 weeks, which was 4 weeks after the cessation of the β-glucan fiber supplement (32). This suggests that β-glucan might play a role in HDL cholesterol concentration increase, and indicates a beneficial carryover effect for β-glucan on HDL up to 12 weeks after stoppage of the β-glucan supplementation."

Robert Nicolosi, Stacey J Bell, Bruce R Bistrian, Isaac Greenberg, R Armour Forse, George L Blackburn. Plasma lipid changes after supplementation with β-glucan fiber from yeast234. The American Journal of Clinical Nutrition, Volume 70, Issue 2, 1999, Pages 208-212, https://doi.org/10.1093/ajcn.70.2.208.
2. Comment: What were the rationale for choosing this specific daily dose of β-glucan?

Response: The choice of the dose of B-glucan is based on data from a previous study by Bozbulut and colleagues who reported that 6 g/day oat β-glucan have favorable outcomes in glycemic control in adolescents with type 1 diabetes mellitus.

Bozbulut R, Şanlıer N, Döğer E, Bideci A, Çamurdan O, Cinaz P. The effect of beta-glucan supplementation on glycemic control and variability in adolescents with type 1 diabetes mellitus. Diabetes Res Clin Pract. 2020 Nov;169:108464. doi: 10.1016/j.diabres.2020.108464. Epub 2020 Sep 21. PMID: 32971156.
3. Comment: Did the authors observe differences in the effectiveness of dietary supplementation with 6 grams of β-glucan depending on the gender, age, and body weight of the patients?

Response:

Interestingly, gender was not related to the percent change of any parameter except PDX-1, age percent change was correlated with TBR<54mg/dl percent change only and BMI percent change was significantly correlated with triglycerides, cholesterol, GMI and PDX1 percent change.

Relation between gender of the studied patients and percentage of change of the studied parameters

	Males	Females	Test value	P-value	Sig.
	Median (IQR)	Median (IQR)	Test value	P-value	Sig.
Weight (kg) on z score	0.00 (-38.79 - 7.93)	0 (0 - 13.33)	-1.532	0.126	NS
Height (cm) on z score	0.00 (0 - 0)	0 (0 - 0)	-0.343	0.732	NS
BMI on z score	0.00 (-20.28 - 0)	0 (-26.67 - 0)	-0.368	0.713	NS
Total daily dose (unit/ kg/ day)	-19.30 (-34.67 - -14.43)	-20.55 (-38 - -14.42)	-0.042	0.966	NS
HbA1c %	-10.77 (-16.67 - -6.65)	-16.28 (-21.35 - -12.58)	-2.597	0.009	HS
TG mg/dl	-4.12 (-10.36 - -2.96)	-4.67 (-6.3 - -3.74)	-0.332	0.740	NS
Cholesterol mg/dl	-6.87 (-9.68 - -3.85)	-9.25 (-13.5 - -5.57)	-1.540	0.124	NS
LDL mg/dl	-22.43 (-26.19 - -16.67)	-17.65 (-27.34 - -11.67)	-1.291	0.197	NS
HDL mg/dl	7.58 (3.72 - 13.03)	10.22 (4.35 - 15.38)	-0.423	0.673	NS
Time in range(TIR) 70-180 mg/dl (%)	28.83 (9.23 - 45.35)	27.66 (14.29 - 43.77)	-0.196	0.844	NS
Time below range (TBR) < 54 mg /dl( %)	-100.00 (-100 - -79.46)	-100 (-100 - -66.67)	-0.127	0.899	NS
Time below range (TBR) 54-69 mg /dl( %)	-37.50 (-66.67 - 0)	-54.55 (-75 - 0)	-0.686	0.493	NS
Time above range (TAR) 180 -250mg /dl( %)	-40.00 (-65.52 - -9.09)	-22.22 (-50 - 3.57)	-0.771	0.441	NS
Time above range (TAR) >250 mg /dl ( %)	-60.56 (-100 - -35)	-71.43 (-100 - -20)	-0.124	0.901	NS
GMI %	-2.63 (-4.94 - -0.79)	-2.82 (-9.72 - 1.37)	-0.302	0.763	NS
Glucose variability %	-14.41 (-34.49 - -0.26)	-12.84 (-30.12 - -6.13)	-0.377	0.706	NS
PDX1 ng/dl	373.41 (258.31 - 555.4)	268.77 (172.48 - 350.88)	-2.430	0.015	S

P>0.05: Non significant (NS); P <0.05: Significant (S); P <0.01: Highly significant

: Independent t-test; *: Chi-square test; ≠: Mann-Whitney test

Correlation between age of the studied patients and percentage of change of the studied parameters

	Age (years)
	r	p-value
Weight (kg) on z score	0.259	0.167
Height (cm) on z score	0.011	0.955
BMI on z score	0.166	0.382
Total daily dose (unit/ kg/ day)	-0.231	0.218
HbA1c %	0.104	0.583
TG mg/dl	0.062	0.743
Cholesterol mg/dl	-0.049	0.796
LDL mg/dl	-0.028	0.883
HDL mg/dl	0.269	0.151
Time in range(TIR) 70-180 mg/dl (%)	-0.006	0.975
Time below range (TBR) < 54 mg /dl( %)	0.485*	0.030
Time below range (TBR) 54-69 mg /dl( %)	-0.262	0.187
Time above range (TAR) 180 -250mg /dl( %)	-0.027	0.893
Time above range (TAR) >250 mg /dl ( %)	0.319	0.098
GMI %	0.138	0.466
Glucose variability %	-0.121	0.525
PDX1 ng/dl	0.156	0.411

Spearman correlation coefficients

Correlation between change in BMI z score of the studied patients and percentage of change of the studied parameters

	BMI on z score
	r	p-value
Weight (kg) on z score	0.122	0.520
Height (cm) on z score	0.002	0.991
Total daily dose (unit/ kg/ day)	-0.007	0.969
HbA1c %	0.056	0.771
TG mg/dl	0.067	0.724
Cholesterol mg/dl	0.502**	0.005
LDL mg/dl	0.374*	0.042
HDL mg/dl	0.303	0.104
Time in range(TIR) 70-180 mg/dl (%)	0.098	0.608
Time below range (TBR) < 54 mg /dl( %)	0.186	0.432
Time below range (TBR) 54-69 mg /dl( %)	-0.048	0.814
Time above range (TAR) 180 -250mg /dl( %)	-0.120	0.552
Time above range (TAR) >250 mg /dl ( %)	0.025	0.899
GMI %	0.366*	0.046
Glucose variability %	0.142	0.456
PDX1 ng/dl	0.399*	0.029

Spearman correlation coefficients

4. Comment: What adverse effects did the children report?

Response: Thanks for your comment. The B-glucan was well tolerated. Only 3 children had abdominal distension and flatulence with the initiation of B-glucan which waned later on. This was added "As regards the adverse effects of therapy, three patients had GIT manifestations in the form of abdominal distension and flatulence on starting the B-glucan therapy, which were mild and waned by the second week of administration, not necessitating stoppage of the β-glucan oat flakes. "
5. Comment: What was the reason for excluding 14 patients from the group?

Response: Thanks for the comment. They were excluded because they had exclusion criteria i.e coexisting autoimmune diseases affecting glycemic control (e.g., celiac, autoimmune thyroiditis), diseases that require dietary restriction (e.g., celiac disease, food allergy, eating disorders), or chronic diabetes complications (e.g., gastroparesis).
6. Comment: It would be worthwhile to expand the discussion with a brief note on which other patient groups, besides type 1 diabetes, would benefit from dietary β-glucan supplementation.

Response:

Thanks for the valuable comment. This was added " Oat flakes β-glucan has been shown to demonstrate several biological activities, including prebiotic, anti-diabetic, cholesterol-lowering,and immunomodulatory effects. Zalecińska and colleagues reported a beneficial effect of β-glucan in two cases with ulcerative colitis (22). Similarly, Oczkowski and coworkers reported that Oat β-glucan dietary intervention has beneficial antioxidant and anti-inflammatory effects in the testes of rats with TNBS-Induced Colitis (23)."

Zalecińska A, Harasym J, Dziendzikowska K, Sikorska K, Gromadzka-Ostrowska J. Clinical Outcomes of Oat Beta-Glucan Nutritional Intervention in Ulcerative Colitis: Case Reports of a Female and a Male Patient. Nutrients. 2025 Dec 5;17(24):3812. doi: 10.3390/nu17243812. PMID: 41470757; PMCID: PMC12735812.

Oczkowski, M.; Dziendzikowska, K.; Pasternak-Winiarska, A.; Jarmołowicz, K.; Gromadzka-Ostrowska, J. Oat Beta-Glucan Dietary Intervention on Antioxidant Defense Parameters, Inflammatory Response and Angiotensin Signaling in the Testes of Rats with TNBS-Induced Colitis. Nutrients 2024, 16, 2546. https://doi.org/10.3390/nu16152546

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I have no further comments.

Author Response

Comment:

I have no further comments.

Response:

Thanks for your valuable comments that ceratinly added a lot to the manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

1.The explanation and citations offered for the inclusion of serum PDX-1 as a viable biomarker of pancreatic reserve are not convincing for a number of reasons:

For the studies cited, the levels detected in Zhang et. al. are more than an order of magnitude lower than what is detected in the present study at both early and mid pregnancy (Standardized to the current study the highest value is 13ng/dL which is 5X lower than the lowest observed here, with averages between 10 and 100X lower). Despite these lower values, most individuals showed generally normal HOMA-β values, and fasting insulin back-calculated from HOMA-IR and fasting glucose places the average in the range of 8uU/mL, indicating reasonable pancreatic reserve in an insulin resistant state of gestational diabetes.

The study by Duarte-Medrano et. al. did not measure PDX protein but rather mRNA expression in whole blood, and therefore likely reflects expression in peripheral blood mononuclear cells not exosomes from pancreatic tissue. However it is impossible to determined what the expression level of these transcripts are as only relative expression is reported.

Thus, my concern with circulating PDX-1 as a surrogate for beta-cell reserve remains, and it should be removed

2.Despite the addition of Oat flakes to each mention of beta-glucan, this is still interpretable as beta-glucan being the primary exposure of this study, which is untrue. The exposure is “oat flakes”, not “oat flakes beta-glucan” as that implies that beta-glucan was isolated from the oat flakes. Other studies using “Oat beta-glucan”, such as Reference 22, uses specifically an extract for their treatment in rats, as does reference 23 in humans. Reference 32 uses an isolated beta-glucan from yeast, and describes it as such.

Oats are still a whole food with may other substances present, and your exposure needs to presented as such. The contribution of beta-glucan can be discussed and included in the introduction as a major bioactive component, but as presented, there are issues with the assumed treatment.

3.Thank you for adjusting the tables and analysis.

4.The supplemental table was not submitted or not viewable, and to my understanding it is the only supplemental table so it should be labeled as supplemental table 1. The other supplemental information (appendix 1) can remain as such.

5.Acknowledged.

6.Thank you for including BMI, it is worth mentioning in the discussion as potential modulator of the beneficial effects of the oat flakes.

7.References are still out of sync.

For example the text “rather than as a hormone or traditional protein biomarker that circulates at high levels in the serum of healthy individuals (33). However, research indicates that PDX-1 can be detected in human sera of people with diabetes, people with pancreatic cancer, and healthy populations (34, 35).”

33 references a review on oat-flakes and beta-glucan.

34 is a review of PDX which does not mention its circulation.

35 is the Duarte-Medrano study mentioned above were they only measured mRNA from whole blood and to represent that as the presence of PDX-1 in serum is misrepresentation of the findings in the study, especially when it is not comparable to anything presented in the current study, or by any other measure given it is all relative expression.

I realize these are staggered by 1 reference, as the prior paragraph mentions a 12 week trial for 32. Regardless it is frustrating that a major request of the first review was to fix reference incongruencies.

Regardless, my point stands about making the statement that prior studies have detected PDX-1 in high concentrations.

8.Acknowledged.

Author Response

Comment:

1. The explanation and citations offered for the inclusion of serum PDX-1 as a viable biomarker of pancreatic reserve are not convincing for a number of reasons:

Thus, my concern with circulating PDX-1 as a surrogate for beta-cell reserve remains, and it should be removed

Response:

Thanks for your comment. It was removed from the whole manuscript.

Comment:

Response:

Thanks for the comment. The term was formulated as advised to oat flakes with discussion of the role of β-glucan as a major component of oat flakes.

Comment:

3.Thank you for adjusting the tables and analysis.

Response:

Thanks for your positive feedback.

Comment:

Response:

They were adjusted as supplementary appendix 1 and supplementary table 1.

Comment:

5.Acknowledged.

Response:

Thanks for the acknowledgment.

Comment:

6.Thank you for including BMI, it is worth mentioning in the discussion as potential modulator of the beneficial effects of the oat flakes.

Response:

Thanks for the comment. This was highlighted in the results and discussion.

Results:

3.1. Oat flakes and BMI:

As shown in Tables 2 and 3, oat flakes administration resulted in a significant reduction in BMI z-score. In addition, BMI z score was significantly correlated with total cholesterol (p= 0.005), LDL-C (p= 0.042), and GMI% (p= 0.046).

Discussion:

In addition to its beneficial effects on glycemia and lipid profile, oat flakes supplementation was found to decrease BMI in the studied adolescents with T1D. This is in concordance with Damsgaard and colleagues, who reported a significant association between whole-grain oat intake and lower fat mass index in healthy children (33). Similarly, an observational study using NHANES data reported an inverse association between total whole-grain intake and BMI in 6- to 18-year-old children and adolescents (35). In addition, two randomized controlled studies including people with T2D showed a significant reduction in BMI on dietary oat supplementation during a follow-up period of three to four weeks (36, 37). This reduction in BMI could be a potential modulator of the beneficial effects of oat flakes on glycemia, insulin resistance, and lipid profile.

Comment:

7.References are still out of sync.

33 references a review on oat-flakes and beta-glucan.

34 is a review of PDX which does not mention its circulation.

Regardless, my point stands about making the statement that prior studies have detected PDX-1 in high concentrations.

Response:

Thanks for your valuable comment. All references were reassessed and hope they are satisfactory now. In addition, the whole part about PDX-1 was removed entirely from the manuscript.

Comment:

8.Acknowledged.

Response:

Thanks for the acknowledgment.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

Thank you for responding my concerns. Regarding the followed-up and wash-out procedure, the authors shall describe them in the study procedure which is unavailable recently.

Author Response

Dear Reviewer,

Comment:

Regarding the followed-up and wash-out procedure, the authors shall describe them in the study procedure which is unavailable recently.

Response:

Thanks for your very valuable comments, which certainly added to the manuscript. The follow-up and washout are now described in the study procedures. "In this open-label, crossover study, participants were randomly assigned to two equally matched groups. In the first three months, group A received 6 g of oat flakes β-glucan daily in addition to their ordinary diet, while group B received their ordinary diet alone. This was followed by crossing over both arms for another 3 months after a two-week washout period between the interventions."