Fecal Microbiota Transplantation (FMT) as a Prophylaxis of Necrotizing Enterocolitis (NEC)—Protocol for a Safety Study

1. Summary

Thank you very much for taking the time to review this manuscript, and for your positive feedback.

We Have found your comments and suggestions both interesting and useful.

Please find the detailed responses below and the corresponding revisions highlighted in the re-submitted files.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: What measures will be taken or statistical adjustments made to prevent the study results from being confounded by differences in the microbiome between babies born vaginally and those born by cesarean section?

Response 1: As the mode of delivery does not influence patient inclusion in the study or control groups, randomisation should offset the impact of this variable. Nevertheless, we have taken this into account, which is why the initial examination of the patient's gut microbiota composition is conducted prior to the administration of FMT. This allows us to include this factor in post-hoc analyses, which will be employed to develop a protocol for testing the efficacy of the preparation on larger patient groups.

Comments 2: Will fecal microbiome analyses be performed on the study and control groups? If so, when will they be performed?

Response 2: As described in detail in point 4.5, microbiome analyses will be performed at the following time points: 0, 7, 14, 30, 60 and 120 days after the FMT2 time point for the experimental group, or after the 'matching FMT' time point for the control group.

Comments 3: How and according to which criteria will it be determined whether the infants in the study and control groups have dysbiosis?

Response 3: The presence or absence of dysbiosis is not a criterion for inclusion or exclusion from the study; therefore, it will not be assessed prior to patient enrolment. Instead, it will be evaluated retrospectively and correlated with clinical outcomes from previously collected biological samples.

Comments 4: If dysbiosis analysis will not be performed before FMT, how will the study results be compared in terms of the improvement of dysbiosis before and after FMT?

Response 4: Improvement of dysbiosis will be assessed as a secondary outcome in a retrospective analysis. The primary outcome will be the clinical effect, specifically whether or not serious adverse events related to the treatment occur.

Comments 5: Why will the FMT performed on infants be taken from other women rather than from their own mothers' fecal samples?

Response 5: Using a healthy, tested donor for the preparation affects positively its quality and safety. Firstly, as premature birth is an emergency by nature, it is impossible to properly test the mother e.g. for multidrug-resistant strains and prepare the preparation in advance. Secondly, according to the literature, maternal dysbiosis may contribute to premature birth (Baldassarre, M.E.; Di Mauro, A.; Capozza, M.; Rizzo, V.; Schettini, F.; Panza, R.; Laforgia, N. Dysbiosis and Prematurity: Is There a Role for Probiotics? Nutrients 2019, 11, 1273. ) Therefore, we do not want to administer microbiota to a newborn that is more likely to be dysbiotic.  

Comments 6: Microorganisms in the flora can become pathogenic when they move to other parts of the body. Therefore, theoretically, placing an adult's fecal flora into an infant's colon is a practice that increases the risk of pathogenicity of microorganisms in the flora. What measures will be taken to reduce this risk?

Response 6: Firstly, obtaining the material for the preparation from a tested donor involves a set of tests that each donor undergoes in accordance with the manufacturer's procedure (https://human-biome.com/en/fmt-preparations/). Secondly, the material for this study is obtained from a pregnant woman, whose microbiota differs from that of a non-pregnant adult, and furthermore colonises the newborn during natural childbirth.

Comments 7:

60g of raw donor's stool is suspended in 200 ml 0.9% NaCL, and aliquoted adding 10% glycerol. There is no assessment of relative bacteria abundance in every sample and it is a standard practice. High standards in donor qualification protocol is the strongest quality assurance. Microbiota composition is addressed scientificly after the clinical experiment closing, for the scientific purpose.

Comments 8 is the sane as comment 7.

Comments 9: The FMT procedure to be performed is theoretically a method that could cause sepsis risk. It is necessary to clearly explain why the FMT method is preferred over probiotic application and its advantages and disadvantages in terms of biosafety.

Response 9: A relevant explanation is provided in the introduction, lines 73–90. To further highlight the issue, we have also added a relevant paragraph in the discussion section.

Comments 10: Probiotics must be used at a specific dose and for a specific period of time for flora restoration. In this study, the hypothesis regarding the correction of presumed dysbiosis with FMT should be written in detail, along with the scientific rationale. How can it be predicted that presumed dysbiosis can be improved or corrected with a single FMT?

Response 10: At this stage of the study, its primary objective is not to correct dysbiosis, which is why we do not describe the specific method of testing for it.  The hypothesis regarding the potential benefit of FMT in the prevention of NEC is derived directly from the aforementioned experimental studies. Furthermore, the correction of dysbiosis after a single dose of FMT is described in other disease entities where the efficacy of FMT has been more extensively studied, e.g. IBD or Clostridioides difficile infection (Lahoud, C.; Habib, T.; Kalta, D.; Dimachkie, R.; El Sayegh, S.; Deeb, L. Intestinal Microbiota and Fecal Transplantation in Patients with Inflammatory Bowel Disease and Clostridioides difficile: An Updated Literature Review. J. Clin. Med. 2025, 14, 5260. https://doi.org/10.3390/jcm14155260).

Comments 11: Will there be standardization regarding the donors' diets?

Response 11: At this stage no.

Comments 12: Infants undergoing FMT should not have acquired or congenital immunodeficiency

Response 12: Of course. All patients with confirmed or suspected genetic disorders that may affect immunity are excluded from the study. Patients with suspected ongoing infection are also excluded.

Comments 13: How will the pain caused by the Foley catheter used during FMT be assessed? Which analgesic methods will be used?

Response 13: In accordance with the standards of care in the neonatal intensive care unit, the level of pain is constantly monitored using the FLACC scale. If the score is 4 or above, paracetamol is administered.

Comments 14: NEC is a multifactorial disease with a multifactorial etiology that does not develop due to a single cause. Therefore, dysbiosis is not the only factor in the etiopathogenesis of NEC. How will cases of NEC be explained when FMT is performed? How will cases of NEC be explained when FMT is not performed?

Response 14: As you rightly pointed out, NEC is a multifactorial disease, and it seems that correcting dysbiosis is one of the measures that can reduce its incidence, but not eliminate it completely. The next stage of our study is to examine the impact of FMT on the incidence of NEC, which will provide us with information on the preventive efficacy of this procedure.

Other possible factors include, for example, hypoxia due to heart defects.

1. Summary

Thank you very much for taking the time to review this manuscript.

Thank you for all your suggestions and comments, which we found to be very substantive, interesting and important for improving the precision of the test protocol.

Please find the detailed responses below and the corresponding revisions highlighted in the re-submitted files.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: The abstract should clearly state the primary endpoint (safety) and specify how safety comparisons with the control group will be performed.

Response 1: The following part of the abstract: The aim of the study is to investigate the clinical safety of high-viability, high-richness, anaerobic FMT in the infections, sepsis and NEC prophylaxis in premature neonates.

was replaced with: The primary endpoint of the study is safety, defined as the incidence of adverse events (AEs) and serious adverse events (SAEs) occurring from the time of intervention until hospital discharge, classified according to severity and assessed for relatedness to the intervention.

Comments 2: The allocation process is not clearly described. Since the first eligible 20 infants are assigned to the intervention group, there is potential for chronological bias. Please clarify how temporal differences in clinical practice will be controlled.

Response 2: The following part of the Allocation description: Depending on the rate of patient enrolment in the study group, patients in the control group will be enrolled in parallel with the study group or after the study group has been assembled.

was replaced with: Patients will be included in the control and study groups in a 1:1 ratio according to the order of admission to the ward. This means that each subsequent patient included in the study will be included in the opposite group to the previous patient.

Comments 3: The statistical analysis section remains largely descriptive. Please specify which statistical tests will be used to compare safety outcomes between groups.

Response 3: The section on statistical analysis has been expanded.

Comments 4: Given the very small sample size (20 per group), please discuss how the study will interpret differences that are not statistically significant but may still be clinically relevant.

Response 4: Thank you for this important comment. We agree that, given the small sample size (20 participants per group), the study may not be powered to detect small or moderate differences in safety outcomes.

Therefore, the interpretation of results will not rely solely on statistical significance. In addition to p-values, we will emphasize effect size estimates (relative risks) and their 95% confidence intervals. Particular attention will be paid to the magnitude and direction of observed differences, as well as the width of confidence intervals, to assess potential clinical relevance even in the absence of statistical significance. We have clarified this approach in the Statistical Analysis section.

Comments 5: Donor screening procedures are described but would benefit from greater detail. Please clarify whether comprehensive screening for multidrug-resistant organisms (MDROs), viral pathogens (e.g., CMV, EBV), and emerging pathogens will be conducted in accordance with international FMT safety guidelines.

Response 5: All of the aforementioned are included in the donor screening. In the references section, we have added a website address with detailed specifications regarding donor testing.

Comments 6: The sample collection schedule is extensive (up to 120 days). Please clarify feasibility, expected attrition rates, and how missing data will be handled in the analysis.

Response 6: We agree that the extended sample collection schedule requires clarification regarding feasibility and missing data management.

Majority of biological samples are collected during routine hospitalization, which in this population typically extends over several weeks. Therefore, sample collection up to 120 days largely overlaps with standard clinical care and scheduled follow-up visits.

Based on our center’s historical data, we anticipate an attrition rate below 15%  primarily due to early discharge.

Missing data will not be imputed for safety endpoints. Safety analyses will be conducted using all available data (available-case analysis). The number and reasons for missing samples or follow-up data will be reported transparently.

We have clarified these aspects in the Outomes section (5.3 Feasibility and Missing Data Management).

Comments 7: The Discussion appropriately acknowledges limitations related to the open-label design; however, it should more explicitly address the limitations inherent to non-randomized allocation and the small sample size.

Response 7: The second part of the discussion was rewritten and extended.