Exploring the Prognostic Performance of MECKI Score in Heart Failure Patients with Non-Valvular Atrial Fibrillation Treated with Edoxaban

Round 1

Reviewer 1 Report

The manuscript/review by “PET-Derived Radiomics and Artificial Intelligence in Multiple 2 Myeloma: State-of-Art” by Manco et al., is well written with updated information on summary of the current status of radiomics and AI in different clinical context of multiple melanoma. One of the main drawbacks f the study was they often included relatively small patient cohort.

Here are a few minor comments.

1.       Introduction is not sufficient, some broader introduction about MM and its diagnosis and uses various methods. Radiomic, AI etc.…

2.       What were the main reasons for choosing R-Software and Python as the preferred open-source software/programming languages for endpoint prediction. How was the internal validation conducted in the given studies, particularly those using a fold-cross validation procedure not clear. Any notable findings or trends observed during the internal validation process should be provided with proper references.

3.       Include the key findings and performance metrics in the three different models developed by Ni et al. [13] to predict prognosis in newly diagnosed MM patients using PET/CT-derived radiomics features and clinical data.

Moderate Editing is needed. 

Author Response

Manuscript title: Exploring the prognostic performance of MECKI score in heart failure patients with non-valvular atrial fibrillation treated with edoxaban

Reviewer 1

Dear Reviewer, regrettably, due to a technical oversight not attributed to the authors, a copy of a different manuscript, distinct from our intended submission, was erroneously included among our files. Please note that the PDF provided (that is the real manuscript by Mapelli et al.) does not align with the content of the Word document. Consequently, the questions you posed refer to a work that is not ours.

We will bring this matter to the attention of the editorial office to ensure that your valuable review is directed to the correct authors. We sincerely apologize for any confusion this may have caused.

Thank you for your understanding.

Reviewer 2 Report

The authors evaluated the reliability of the MECKI score in HFrEF patients treated with edoxaban for NVAF. The MECKI score allows an evaluation of the risk of heart failure patients using the assessment of exercise capacity, blood samples and echocardiography. However, the conditions of this study were relatively complex, with more variables affecting this evaluation system. There are still several questions that need to be addressed to support the hypotheses/conclusions.

1. The authors enrolled patients with NYHA cardiac functional class I-IV, but whether there are differences in the prognostic assessment of heart failure patients with different classes/stages, especially those with very severe heart failure. 2. The definition/criteria of atrial fibrillation in this study should be clearly stated in the inclusion criteria. The NVAF status at the time of physical examination needs to be stated. 3. Risk factors (CHA2DS2-VASc, HAS-BLED score) and concomitant medications should be clarified. Whether there are patients with suspected adverse drug reactions of edoxaban at the study endpoint. 4. Sex differences research has important implications and should be considered when designing and analyzing basic and clinical research.

Author Response

Manuscript title: Exploring the prognostic performance of MECKI score in heart failure patients with non-valvular atrial fibrillation treated with edoxaban

Reviewer 2

The authors evaluated the reliability of the MECKI score in HFrEF patients treated with edoxaban for NVAF. The MECKI score allows an evaluation of the risk of heart failure patients using the assessment of exercise capacity, blood samples and echocardiography. However, the conditions of this study were relatively complex, with more variables affecting this evaluation system. There are still several questions that need to be addressed to support the hypotheses/conclusions.

1. The authors enrolled patients with NYHA cardiac functional class I-IV, but whether there are differences in the prognostic assessment of heart failure patients with different classes/stages, especially those with very severe heart failure.

We appreciate your thorough review and the relevant question regarding the stratification of patients based on NYHA functional class. In our investigation, we enrolled patients with NYHA cardiac functional class I-IV, and we acknowledge the importance of assessing prognostic differences among different classes, particularly in cases of severe heart failure.

We would like to emphasize that our study population was carefully selected and matched with the "MECKI score" population by several variables, including peak VO2, MDRD, VE/VCO2 slope, LVEF, hemoglobin, sodium, etiology, gender, age, and BMI. We believe that these variables, taken together, provide a comprehensive characterization of the severity of the patients in our study cohort.

However, we recognize that the limited size of our population in this preliminary study may not allow further stratification of subjects based on NYHA. Despite this limitation, we are aware of the importance of exploring prognostic differences among various NYHA functional classes and intend to delve into this aspect in future research.

2. The definition/criteria of atrial fibrillation in this study should be clearly stated in the inclusion criteria. The NVAF status at the time of physical examination needs to be stated.

Thank you for your comment. We acknowledge the importance of clearly defining the criteria for atrial fibrillation (AF) in the inclusion criteria. Unfortunately, the precise AF status (persisting, permanent, paroxysmal) at the time of enrollment is a limitation we encountered. Regrettably, we do not have detailed records of the AF status of patients at the initiation of the study, in particular for the retrospective MECKI group.

It's noteworthy that our data suggests a likelihood of a higher prevalence of paroxysmal AF in the Edoxaban-treated group, given the higher usage of Amiodaron, a medication often used to maintain sinus rhythm. Conversely, the VKA-treated group may have had a higher representation of patients with other forms of AF, as indicated by a lower usage of Digoxin, a medication typically employed to control heart rate in persistent AF.

We understand the importance of providing a clear and comprehensive understanding of the AF status at baseline, and we regret any ambiguity in this regard. A limitation paragraph has been added to text.

3. Risk factors (CHA2DS2-VASc, HAS-BLED score) and concomitant medications should be clarified. Whether there are patients with suspected adverse drug reactions of edoxaban at the study endpoint.

Thank you for your valuable feedback on our manuscript. We appreciate your thorough review.

Regarding the inquiry about risk factors, specifically the CHA2DS2-VASc and HAS-BLED scores, we regret to inform you that some of the required information for the calculation of these scores is not available in the current dataset, especially for the retrospective cohort. This omission is due to the fact that the medication prescription was initiated prior to participants' enrollment in the study as our study's primary objective was to assess the effectiveness of the MECKI score in a population already undergoing treatment.

Concomitant medications are reported in table 1.

Regarding adverse reactions to edoxaban, patients were regularly followed up according to clinical standards.

4. Sex differences research has important implications and should be considered when designing and analyzing basic and clinical research.

Unfortunately, the small sample size of our population has only 10 women, making gender analyses unreliable. This limitation has been added in the text. Thank you for the important observation.