Author’s response to reviews Title: Serological Biomarker Testing Helps Avoiding Unnecessary Endoscopies in Obese Patients Before Bariatric Surgery Authors:

Rewrite and structure as suggested above. It is no meaning in doublewriting that the biomarker panel HS and NHS groups were classified according to biomarker profile as this is obvious and also not surprising that the groups profiles therefore were significantly different. Response Changes to abstract made Methods Add more on how the biopanel classification was performed by the software. Heading add Histologi to EGDS heading Result text Structure comparisons of biomarker systematically a) panel results with EGDS and then b) panel with histology results with tables accordingly. Comment on the biomarker panel misclassification of panatrophy vs histology superficial gastritis (table 2). Results tables Table 1 move "n(%) from column to to table characteristics were relevant e.g. females, concomitatnd disease et cetera. Table 3 consider to present as supplementary table as software classification considers several biomarkers?? Change to "." instead of "," in p-value table. Response Table 1 corrected, Table 3 also corrected. Table 4 Partially overlapping results with present table 2? Can all histology results be presented in table 2. Separate table on EDGS (as one of the golden standards) and biomarker panel as suggested above. Response Table 4 was changed, histology was removed from Table 4 and focus was placed on the endoscopic finding vs gastropanel. Gil Faria (Reviewer 2): This is a very interesting clinical question and the methods are adequate to study it. However, there are some minor issues I would like the authors to address. 1 The authors state that the serological panel is cheaper than EGDS. Please provide more information about it and include a cost-effectiveness analysis. From the data presented by the authors, for the serological panel to be cost-effective, the panel would need to cost about 1/3 of the EGDS. What would be the number of EGDS avoided for the panel to be cost-effective? Response Cost-effectiveness is, of course, also country specific, as the prices for EGDS + patohistolocical evaluation can differ to a large degree. For example, the cost per patient for GP in Finland is 125 EUR versus about 600 (average) EUR for EGDS with patohistological evaluation. In Estonia GP is not routinely available on the market and its cost-effectiveness is difficult to evaluate. We agree that for the population cluster studied, the GP price should be about 1/3 of EGDS complex. Information about the GP price was added to the introduction, lines 78-80, page 4. 2 The serological panel detects gastritits and HP infection. But what about other lesions like GIST or NET? In the sample, there were no patients with either pathology, but some reports state that GIST prevalence might be as high as 20%. Would the panel detect those lesions?