Recognizing the Emergent and Submerged Iceberg of the Celiac Disease: ITAMA Project—Global Strategy Protocol

Round 1

Reviewer 1 Report

A good study looking at the development and feasibility of rapid testing in celiac disease. It would be a significant development to reduce the burden of expensive conventional testing and would also help to test a wider population. 
 
Very well presented protocol for the testing and control measures. It is also good to see the care taken to test and reduce/address the negative predictive value of the POCT compared with conventional testing for Celiac disease.

It would be better to have more write up in the discussion and conclusions section and discuss strategies to improve or propose a future study. 

Author Response

It would be better to have more write up in the discussion and conclusions section and discuss strategies to improve or propose a future study. 

We enriched the discussion and conclusion discussing strategies to improve or propose a future study

Reviewer 2 Report

General Comments:  The investigators discuss their potentially important study that may be partly finished to improve diagnosis of gluten sensitivity through a large health project to screen school children for risk of coeliac disease in Malta and another study in Sicily. The study objective is to evaluate the diagnostic accuracy of a finger prick point-of-care test (POCT) combined with conventional celiac serology supported by an artificial intelligence (AI)-based approach. A main goal is to identify individuals who would benefit from gluten-free diet, including those who lack GI symptoms, serology, and histology and define the spectrum of coeliac disease, perhaps by looking for presence of TGA-IgA deposits in the intestinal mucosa.  The overall plan to undertake screening is intended to contribute structural, procedural, and technological advances to screening. The current manuscript does not include sufficient information to evaluate the originality of this approach.  Although the manuscript has clearly been prepared to some extent by highly trained and scholarly authors, the current report is confusingly written partly due to grammatical issues, needs proof reading, and is disorganized to the point of being incoherent and unreadable. A recent study (Stahl et al, 2021) might be informative for the authors.

Specific comments:

1. Originality: How will/or did the investigators develop a database with “heterogeneous metadata” and produce a test for the diagnosis of AD? Do they mean gluten sensitivity/coeliac disease? How will/ or did they validate diagnostic guidelines and the AI system?

2. The investigators need to provide organized sections for subjects and sites, a methods section that includes procedures, etc., the section on sample size as part of statistical analysis, and protocols for each study, fig 2, fig 3. The results section should include delineation of subject demographics for the sites, and present results of discrete studies as clearly as possible.

3. Clarify when each study began, the number of subjects enrolled at this point and identify the numbers of subjects for each stage of the flow chart for Malta and provide equivalent information for Sicily.

4. The section on Development and Validation of the AI system includes an introduction -should be in the Introduction, the methods used- should be in Methods but shows no results or data to support the conclusions, not even how many subjects had “sufficient information” and were included in the analysis -also not shown.

5. Similarly, the section on analysis of costs should be revised and organized.

6. Since the investigators mention using TGA deposits as an important marker, data from TGA-IgA deposits in the intestinal mucosa need to be presented to assess how this worked out. Provide data for each aim,  such as validation of the POCT, mention any limitations, and then discuss conclusions. 

. 

Author Response

The investigators discuss their potentially important study that may be partly finished to improve diagnosis of gluten sensitivity

We aim at improving diagnosis of celiac disease and not of gluten sensitivity. Even the possible relationship with gluten in presence of anti-transglutaminase antibodies has to be meant as potential celiac disease in that this finding is a marker of forthcoming CD, it is mentioned in the methods section regarding the Study Protocol in Sicily - Celiac Disease Center setting and we added the reference No. 27

Although the manuscript has clearly been prepared to some extent by highly trained and scholarly authors, the current report is confusingly written partly due to grammatical issues, needs proof reading, and is disorganized to the point of being incoherent and unreadable. A recent study (Stahl et al, 2021) might be informative for the authors.

We re-organized the manuscript according to Stahl et al. whose study we cite and comment and a native English-speaking colleague made proof reading, as suggested.

Specific comments:

Originality: How will/or did the investigators develop a database with “heterogeneous metadata” and produce a test for the diagnosis of AD? Do they mean gluten sensitivity/coeliac disease? How will/ or did they validate diagnostic guidelines and the AI system?

The section on "Development and validation of an artificial intelligence-based…", as suggested, has been rewritten and enriched by inserting data relating to the analysis conducted on the ITAMA database and the performance obtained. The AI part of the ITAMA Project has been  described in detail in the manuscript : M. E. Tabacchi, D. Tegolo, D. Cascio, C. Valenti, S. Sorce, V. Gentile , V. Taormina, I. Brusca, G. Magazzù, A. Giuliano and G. Raso: A Fuzzy-based Clinical Decision Support System for coeliac disease (2022). Submitted for publication

We mean celiac disease and not gluten sensitivity.

The investigators need to provide organized sections for subjects and sites, a methods section that includes procedures, etc., the section on sample size as part of statistical analysis, and protocols for each study, fig 2, fig 3. The results section should include delineation of subject demographics for the sites, and present results of discrete studies as clearly as possible.

As suggested, we organized sections for subjects and sites, a method section that includes procedures, according to Stahl et al., the section on sample size as part of statistical analysis and protocols for each study integrating the Figures 1, 2 and 3. We provided delineation of subjects demographics for the sites and present preliminary results, in a new section “Results”

Clarify when each study began, the number of subjects enrolled at this point and identify the numbers of subjects for each stage of the flow chart for Malta and provide equivalent information for Sicily.

We clarified when each study began, the number of subjects enrolled at this point and we added the number in the flow-charts in Figures 1, 2 and 3.

The section on Development and Validation of the AI system includes an introduction -should be in the Introduction, the methods used- should be in Methods but shows no results or data to support the conclusions, not even how many subjects had “sufficient information” and were included in the analysis -also not shown.

In this section, we moved an introduction and Method paragraphs to the appriopriate sections and provided information regarding subjects included in the analysis in the new section “Results”

Similarly, the section on analysis of costs should be revised and organized.

The section on analysis of costs was revised

Since the investigators mention using TGA deposits as an important marker, data from TGA-IgA deposits in the intestinal mucosa need to be presented to assess how this worked out. Provide data for each aim, such as validation of the POCT, mention any limitations, and then discuss conclusions. 

We present now in Figure 4 and 5 the presence and absence of anti-transglutaminase mucosal deposits and provided technical details in the Method section.

As this manuscript is a protocol paper, we present preliminary data only. However for the first aims regarding POCT validation just after we sent this article to Pediatric Reports, two abstract were accepted and will be presented on June as poster and oral presentation at the ESPGHAN Meeting in Copenhagen and at the Royal College of Pediatrics and Child Health Conference in Liverpool, respectively. These abstracts will be published on Journal of Paediatric Gastroenterology and Nutrition and Archives od Diseases in Childhood, respectively, as Proceedings of the Meetings.

We added in the Discussion the limitations of the project.

Reviewer 3 Report

This study presents a protocol for the development of the largest screening project for CD carried out to date in school children aimed at assessing the diagnostic accuracy of a minimally invasive finger prick point-of-care tests (POCT) that combined with conventional celiac serology and the aid of an artificial intelligence-based system may reduce the need of intestinal biopsy. 

Author Response

no comment or change required

Reviewer 4 Report

Very nice paper. No comments to be made. Everything was vary clear and easy to read. Congratulations!

Author Response

no comment or change required

Round 2

Reviewer 2 Report

The manuscript has been extensively revised and  organized. The current paper describes a novel investigation and provides an interesting and important contribution to the field.