Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors,

The manuscript entitled “Microbiological Profiling of Menstrual Blood Aspirated from Uterus in Patients Undergoing Frozen Embryo Transfer” investigates the microbiota isolated from menstrual blood in patients undergoing embryo transfer. This is an interesting and valuable study with potential implications for maternal and fetal health, and it presents findings that may be of interest to the readership. However, several points should be addressed before considering the manuscript for publication:
1. When writing bacterial names, the genus should be written in italics, followed by the abbreviated species, where the abbreviation of the species should be written in regular font. Please revise accordingly (check the pdf file).
2. Higher taxonomic ranks, such as phylum, class, order and family, should be written in regular (non-italic) font. The manuscript should be carefully revised to ensure consistency with this rule.
3. Although DNA was extracted, the primers used for amplification were not specified. Based on the results, sequences appear to include both bacterial and fungal taxa, suggesting the use of at least two primer sets (e.g., 16S rRNA for bacteria and ITS/18S for fungi). This should be clearly described in the methodology section.
4. The results would be strengthened by presenting the relative abundance of the microbiota, either at the phylum or genus and species level.
5. In the Results section, microbial diversity is presented by mixing taxa at different taxonomic levels (e.g., genus and family), which reduces clarity. A more consistent and scientifically accurate presentation is recommended.

Kindly, check the comments in the pdf file.

Comments for author File: Comments.pdf

Author Response

Esteemed Reviewer,

We are grateful for the time and effort dedicated to providing valuable comments on our manuscript. Thank you for highlighting mistakes directly in the PDF file; this has helped us tremendously. We have done our best to revise the manuscript and include all the suggestions provided. Additionally, we have corrected typos and grammar throughout the text. In the revised manuscript, the changes are highlighted using MS Word tools.

Below you may find a point-by-point response to the received comments.

Comment 1: When writing bacterial names, the genus should be written in italics, followed by the abbreviated species, where the abbreviation of the species should be written in regular font.

Response 1: Thank you for pointing this out. We have carefully checked the manuscript to correct these mistakes in both the main text and figures.

 

Comment 2: Higher taxonomic ranks, such as phylum, class, order and family, should be written in regular (non-italic) font. The manuscript should be carefully revised to ensure consistency with this rule.

Response 2: Thank you for pointing this out. We have carefully checked the text to correct these mistakes in both main text and figures.

 

Comment 3: Although DNA was extracted, the primers used for amplification were not specified. Based on the results, sequences appear to include both bacterial and fungal taxa, suggesting the use of at least two primer sets (e.g., 16S rRNA for bacteria and ITS/18S for fungi). This should be clearly described in the methodology section.

Response 3: We are grateful for this valuable suggestion. As it is stated in the methodology section we used pre-validated commercial multiplex assays for qPCR, namely “Femoflor 16”, “TNC Complex”, “Herpes Multiplex” (their detailed description is available at the manufacturer’s website). The manufacturer does not disclose details regarding the sequences of primers and probes for their assays. It is worth mentioning that these assays are not designed for sequencing. These reagents are actively used in clinical practice and were used in many studies published to date (for example, DOI: 10.1080/19396368.2023.2195040; 10.3390/biomedicines11051284; 10.17816/JOWD66457-67).

To improve the clarity of this methodological aspect, we have revised the statement in the study limitations section: “…To begin, microbiological profiling was based on a multiplex assay for real-time PCR, which does not provide such a comprehensive characterization of microbial communities as sequencing-based approaches. Thus, our results should not be directly compared to findings of studies that utilized various sequencing techniques. The rationale for selecting multiplex real-time PCR instead of next-generation sequencing was based on two characteristics of the former. It is inexpensive and allows the analysis of the sample within a single day, whereas the turnaround time for microbiome analysis using sequencing is typically 1-3 weeks, which limits its potential to provide prognostically relevant data prior to embryo defrosting and hormonal stimulation [41–43]…”.

 

Comment 4: The results would be strengthened by presenting the relative abundance of the microbiota, either at the phylum or genus and species level.

Response: We appreciate this suggestion. Given the nature of the multiplex assays for qPCR used in this study, the only viable option to present the abundance of taxa analyzed was to divide the concentration of target amplicons for each taxon by the total bacterial load (the manufacturer does not disclose the primers and probes used to quantify the total bacterial load, but according to the information provided, this assay targets a highly conservative sequence of the 16S rRNA gene which enables detection of 95%+ of bacteria). In the methodology section, we state that “The results of microbiological profiling for each taxon analyzed are presented as “abundance” (% of total bacterial load)”.

 

Comment 5: In the Results section, microbial diversity is presented by mixing taxa at different taxonomic levels (e.g., genus and family), which reduces clarity.

Response: Thank you for this suggestion. We have mixed taxa at different taxonomic levels because of the peculiarities of the multiplex qPCR assays used. The manufacturer decided that the most effective way to design these assays was to make some primers and probes to target family-specific DNA sequences, whereas others – to target genus-specific DNA sequences. By its nature, qPCR does not allow to include too many targets. On the one hand, this is a definitive drawback (which is stated in the limitations section, please see response to comment 3). On the other hand, this allowed us to significantly reduce the price of this study. Moreover, these reagents and widely used in clinical practice; thus, translating our results from research to practice could be significantly easier. In the methodology section, we refer to our previously published pilot study, where the methodology is described more comprehensively (we did not duplicate the text as it is usually suggested by editors).

We have added a new sentence to the end of the respective methodology subsection to make this limitation more obvious: “Due to the limitations of the multiplex real-time PCR assays used, the data are presented at the family level for some microorganisms and at the genus level for others”.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors address an interesting and clinically relevant topic, namely the role of endometrial microbiota in ART outcomes and the potential use of menstrual blood aspirated directly from the uterus as a minimally invasive sampling method. The proposed approach is novel and could have important implications if validated. However, in its current form, the manuscript requires substantial revision to improve clarity, methodological transparency, and the interpretation of findings.

The introduction provides a broad overview of the human microbiota and its role in health, but it would benefit from a more focused and concise presentation. The early paragraphs are overly general and should be streamlined to emphasize the specific relevance of endometrial microbiota in reproductive medicine. While relevant references are included, the critical appraisal of existing literature is limited. The authors are encouraged to better define the current gaps in knowledge and clearly articulate how their study addresses these gaps. In particular, the novelty of using menstrual blood as a surrogate for endometrial sampling should be framed more explicitly as a hypothesis-driven approach.

Regarding the study design, the prospective nature of the work is appropriate; however, the absence of defined inclusion and exclusion criteria raises concerns about heterogeneity and potential confounding. While the intention to reflect a real-world ART population is understandable, the lack of control over important variables (e.g., hormonal regimens, prior treatments, antibiotic exposure) may influence the results and should be acknowledged more explicitly. Additionally, the sample size is relatively limited, especially considering the low detection rates of several microbial taxa, which reduces statistical power and limits the robustness of subgroup analyses.

The methods section is generally adequate but would benefit from further detail and clarification. Key methodological aspects are referenced from a previous publication, but essential elements should be summarized within the manuscript to ensure it is self-contained. The choice of multiplex real-time PCR instead of sequencing-based approaches should be better justified, including a discussion of its limitations in capturing microbial diversity. Furthermore, the decision not to apply corrections for multiple comparisons is a significant limitation given the number of taxa analyzed and should be more thoroughly justified and discussed. Additional clarification on contamination control measures and the rationale behind the selected detection thresholds would also strengthen the methodological rigor.

The results section presents the data comprehensively but lacks clarity and focus. The narrative is dense and, at times, difficult to follow. The authors are encouraged to reorganize the results to distinguish clearly between primary outcomes and exploratory findings. The main finding of no significant association between most microbiological parameters and embryo transfer outcomes should be emphasized more clearly. Statistically significant observations, such as the presence of Candida spp. and Enterobacteriaceae in patients with negative outcomes, are based on very small numbers and should be interpreted with caution and explicitly described as preliminary or hypothesis-generating. Similarly, the analyses involving Shannon diversity indices and Lactobacillus dominance would benefit from clearer explanation and contextualization.

The discussion is comprehensive and demonstrates familiarity with the literature; however, it would benefit from greater focus and a more balanced interpretation of the findings. Some sections are overly detailed, particularly those addressing contamination issues, while others tend to overinterpret exploratory results. The claim that menstrual blood is a promising biomaterial for clinical application appears premature based on the current data and should be moderated. The limitations section is appropriate but should more clearly emphasize the exploratory nature of the study, the limited statistical power, and the constraints of the analytical methods used.

The conclusions should be revised to better reflect the results presented. While the study introduces an interesting concept, it largely demonstrates a lack of strong association between microbiological profiles and embryo transfer outcomes. Therefore, statements suggesting clinical applicability or utility should be tempered, and greater emphasis should be placed on the need for further validation in larger, well-controlled studies.

Figures and tables are generally informative but would benefit from improved clarity and more detailed legends to ensure they are interpretable independently of the main text. Simplifying the visual presentation to highlight key findings may also enhance readability.

Finally, the English language is generally understandable but would benefit from professional editing. Several sentences are overly long or complex, which affects readability and clarity. Improving the linguistic quality of the manuscript would significantly enhance its overall presentation.

In summary, this study presents a novel and potentially valuable approach, but substantial revisions are required to improve methodological clarity, data interpretation, and overall presentation before it can be considered for publication.

Author Response

Esteemed Reviewer,

We are grateful for the time and effort dedicated to providing valuable comments on our manuscript. We have done our best to revise the manuscript and include all the suggestions provided. In the revised manuscript, the changes are highlighted using MS Word tools.

Below you may find a point-by-point response to the received comments.

 

Comment 1: Several sentences are overly long or complex, which affects readability and clarity. Improving the linguistic quality of the manuscript would significantly enhance its overall presentation.

Response 1: We are grateful for this comment. We have submitted our manuscript for English editing. To our shame, there were so many changes that it is not possible to list them within this response. Please see the revised version of the manuscript. All corrections were applied using the track changes tool.

 

Comment 2: Figures and tables are generally informative but would benefit from improved clarity and more detailed legends to ensure they are interpretable independently of the main text. Simplifying the visual presentation to highlight key findings may also enhance readability.

Response 2: Thank you for pointing this out. We agree that figures should be easily interpretable without reference to the main text. To improve the clarity, we have expanded the figure legends.

 

Figure 1: “…Vertical axis corresponds to samples from study participants, whereas horizontal axis–to the data for each taxon analyzed… The intensity of cell coloration corresponds to the abundance value according to the scale of the heatmap.”

Figure 2: “…Bacterial abundances were calculated as % of total bacterial load. “Lactobacillus spp. dominance” refers to categorical variable that was positive when the abundance of Lactobacillus spp. was higher than the abundance of all other taxa combined. P-values for the comparisons of categorical variables were obtained using Fisher’s exact test, whereas for continuous variables–using Mann-Whitney’s U test. * This p-value did not pass the threshold for Benjamini-Hochberg correction for multiple comparisons at a false discovery rate of 0.2.”

 

Comment 3: The introduction provides a broad overview of the human microbiota and its role in health, but it would benefit from a more focused and concise presentation. The early paragraphs are overly general and should be streamlined to emphasize the specific relevance of endometrial microbiota in reproductive medicine. While relevant references are included, the critical appraisal of existing literature is limited. The authors are encouraged to better define the current gaps in knowledge and clearly articulate how their study addresses these gaps. In particular, the novelty of using menstrual blood as a surrogate for endometrial sampling should be framed more explicitly as a hypothesis-driven approach.

Response 3: Thank you for this insightful comment. We have done our best to improve the introduction section in accordance with your comments. Below is the list of additions/corrections.

• The first paragraph has been shortened approximately by 20%.
• Shortcomings of the existing studies and potential strengths of the approach presented in our study have been better defined by modifying existing sentences and adding new ones ("…Therefore, researchers have to rely on data obtained in menstrual cycles preceding the one in which IVF will be performed… Moreover, this biomaterial could be collected and rapidly analyzed at the beginning of the respective ET cycle, potentially providing opportunities to reschedule the ART procedure based on the data obtained…").
• In the revised version of the Introduction, we emphasize that menstrual blood might be used as a surrogate for other forms of endometrial sampling as a hypothesis-driven approach by clearly stating that “We hypothesize that…”.

 

Comment 4: Regarding the study design, the prospective nature of the work is appropriate; however, the absence of defined inclusion and exclusion criteria raises concerns about heterogeneity and potential confounding. While the intention to reflect a real-world ART population is understandable, the lack of control over important variables (e.g., hormonal regimens, prior treatments, antibiotic exposure) may influence the results and should be acknowledged more explicitly. Additionally, the sample size is relatively limited, especially considering the low detection rates of several microbial taxa, which reduces statistical power and limits the robustness of subgroup analyses.

Response 4: We are grateful for this comment and agree that our study design had certain shortcomings.

• Indeed, some treatments might affect the microbiota. We did not list the use of antibiotics, antimicrobials, vaginal pre-, pro-, synbiotics as an exclusion criterion, because the use of these medications is not an absolute contraindication for frozen embryo transfer. In this study, we aimed to collect prognostically relevant data suitable for a real-world population. In fact, our questionnaire for study participants included questions regarding the therapy listed above. We were going to use this data for multivariable analysis if needed. As none of the patients had this therapy in the last 3 months prior to the collection of biomaterials, we did not include this information in Table 1 “Relevant demographic and clinical characteristics of the study participants”. In the revised version of the manuscript, we have added the respective rows to the table. (“Therapy in the last 3 months: - antibiotics / antimicrobials; - vaginal pre- / pro- / synbiotics”).
• As for the hormonal regiments, we did provide the information regarding the frequencies of stimulated and natural cycles. In Table 1, it is stated that they were not associated with the outcome of the ART procedure. Some details of hormonal regiments might affect the pregnancy rate, and it has been studied extensively by other authors. In our cohort, it was not the case. Anyway, we have added this information to Table 1 in the form of a note for the row with stimulated cycles (“^# Hormonal regiments did not differ significantly between the groups (p > 0.05)”).
• To emphasize that this study did control potentially important confounding variables, a respective statement in the results section has been changed to read: “…Notably, any attempts to account for the clinical and demographic features of the study participants listed in Table 1 using multivariable logistic regression did not alter the statistical significance of the results presented above…”
• The limitations of the study regarding low sample size are stated in the limitations section, the respective sentence has been changed to read: “…Next, the sample size was limited due to the implementation of a novel methodology regarding the collection of endometrial biomaterials and the exploratory nature of this study, which rendered some statistical analyses underpowered (specifically for taxa with detection rate below 10% at β=0.20 and α=0.05)…”. Moreover, after providing data based on a small sample size, we state in the results that these data must be interpreted with caution, given the low sample size.

 

Comment 5: The methods section is generally adequate but would benefit from further detail and clarification. Key methodological aspects are referenced from a previous publication, but essential elements should be summarized within the manuscript to ensure it is self-contained. The choice of multiplex real-time PCR instead of sequencing-based approaches should be better justified, including a discussion of its limitations in capturing microbial diversity. Furthermore, the decision not to apply corrections for multiple comparisons is a significant limitation given the number of taxa analyzed and should be more thoroughly justified and discussed. Additional clarification on contamination control measures and the rationale behind the selected detection thresholds would also strengthen the methodological rigor.

Response 5: Thank you for such a thorough summary of the things that need to be improved in this section.

• We have consulted with a statistician, and they recommended us to apply a correction for multiple comparisons as well. We were advised to use the Benjamini-Hochberg procedure, which would allow us to control the false discovery rate. The statistical analysis subsection has been expanded: "Because of the exploratory design of the study, strict correction for multiple comparisons (e.g., Bonferroni correction) was not applied. Instead, we assessed the false discovery rate (FDR) using Benjamini-Hochberg procedure at an FDR of 0.2. P values below 0.05 that did not pass the threshold for this correction for multiple comparisons are marked with an asterisk “*”."
  Text in the results section, table and figures have been modified accordingly, and their respective legends have been expanded as well to clarify the meaning of this symbol. This correction allows to maintain the exploratory nature of the study and avoid complete elimination of significance for some of the potentially valuable observations. In the current form, each reader might decide for themselves how to interpret p-values marked with an asterisk.
• We have significantly expanded the discussion of the rationale behind the selection of multiplex real-time PCR:
  “…To begin, microbiological profiling was based on a multiplex assay for real-time PCR, which does not provide such a comprehensive characterization of microbial communities as sequencing-based approaches. Thus, our results should not be directly compared to findings of studies that utilized various sequencing techniques. The rationale for selecting multiplex real-time PCR instead of next-generation sequencing was based on two characteristics of the former. It is inexpensive and allows the analysis of the sample within a single day, whereas the turnaround time for microbiome analysis using sequencing is typically 1-3 weeks, which limits its potential to provide prognostically relevant data prior to embryo defrosting and hormonal stimulation [41–43]…”.
• We have expanded the description of thresholds for quality and contamination control:
  “…These assays included primers and probes for Homo sapiens DNA to control the quality of biomaterial collection and processing (in accordance with the manufacturer's instructions, the threshold was set to 10³ copies per reaction mixture). The influence of the so-called “kitome” (contamination of the final analytical mixtures by bacteria present on the reagents and consumables utilized at various stages of the sample handling process) was assessed by exposure of negative control sterile 0.9% saline solutions to the above-mentioned procedures at all stages of the microbiological profiling. To diminish the influence of any possible “splashome” and contamination of samples by microbial and viral DNA from the lower parts of the reproductive tract, a conservative threshold of 10³ copies of analyzed target per reaction mixture was applied for the detection calls during real-time PCR analysis. According to the information provided by the manufacturer of the assays, this threshold is characterized by a sensitivity of 97% and a specificity of 97%...”
  
  

Comment 6: The results section presents the data comprehensively but lacks clarity and focus. The narrative is dense and, at times, difficult to follow. The authors are encouraged to reorganize the results to distinguish clearly between primary outcomes and exploratory findings. The main finding of no significant association between most microbiological parameters and embryo transfer outcomes should be emphasized more clearly. Statistically significant observations, such as the presence of Candida spp. and Enterobacteriaceae in patients with negative outcomes, are based on very small numbers and should be interpreted with caution and explicitly described as preliminary or hypothesis-generating. Similarly, the analyses involving Shannon diversity indices and Lactobacillus dominance would benefit from clearer explanation and contextualization.

Response 6: Thank you for pointing out these shortcomings.

• As we review the manuscript, it becomes clear that the data are presented in a dense format. We have divided large paragraphs into several smaller ones in the revised version.
• Additionally, to ease the interpretation of this section, we have introduced subsections (3.1. Sample quality; 3.2. Association with Frozen ET Outcomes; 3.3. Association with Clinical Features).
• Subsection 3.2 now ends with a following statement: “…Therefore, the primary outcome of the study regarding the association of microbiological profiles of menstrual blood with post frozen ET pregnancy rate was negative.”
• All significant observations which were drawn from potentially underpowered comparisons are now followed by a statement that these data should be interpreted with caution due to low sample size with appropriate explanations: “…although these results must be interpreted as preliminary as the sample size for positive cases was quite limited (n = 5)…”; “…although given the overall low detection rate of this microorganism (9.2%) these data must be interpreted with caution”.
• We have added an explanation regarding the Lactobacillus spp. dominance values for Figure 2 (see the response to Comment 2). The text about Shannon’s index and the above-mentioned variable now reads as follows:
  "As comparing the abundances of individual taxa was quite challenging, we evaluated two variables that are more suited for the interpretation of microbiological profiling studies, namely Shannon’s index and the Lactobacillus spp. dominance. The former is a quantitative variable which reflects α-diversity of the microbiological community, whereas the latter is a qualitative variable that is positive when the abundance of Lactobacillus spp. is higher than the abundance of all other taxa combined. Neither Shannon’s index nor the Lactobacillus spp. dominance was associated with the outcome of frozen ET (p > 0.05; Figure 2c,d)."

 

Comment 7: The discussion is comprehensive and demonstrates familiarity with the literature; however, it would benefit from greater focus and a more balanced interpretation of the findings. Some sections are overly detailed, particularly those addressing contamination issues, while others tend to overinterpret exploratory results. The claim that menstrual blood is a promising biomaterial for clinical application appears premature based on the current data and should be moderated. The limitations section is appropriate but should more clearly emphasize the exploratory nature of the study, the limited statistical power, and the constraints of the analytical methods used.

Response 7: We are grateful for this comment.

• The study limitations section has been updated and expanded to better emphasize the constraints of the methodology.
  “This study has certain limitations. To begin, microbiological profiling was based on a multiplex assay for real-time PCR, which does not provide such a comprehensive characterization of microbial communities as sequencing-based approaches. Thus, our results should not be directly compared to findings of studies that utilized various sequencing techniques. The rationale to select multiplex real-time PCR instead of next generation sequencing was based on two characteristics of the former. It is cheap and allows to analyze the sample within a single day, whereas the turnaround time for microbiome analysis using sequencing is typically 1-3 weeks, which limits its potential to provide prognostically relevant data prior to embryo defrosting and hormonal stimulation [41–43]. Next, the sample size was limited due to the implementation of a novel methodology regarding the collection of endometrial biomaterials and the exploratory nature of this study, which rendered some statistical analyses underpowered (specifically for taxa with detection rate below 10% at β=0.80 and α=0.05). Therefore, further validation of these results in an expanded cohort is required. Finally, our methodology did not include cultivation of detected microorganisms, which limits the potential to draw conclusions regarding their role in ER regulation.”
• We have toned down some claims across the discussion section, highlighting the need for further validation in larger studies by using phrases like “…although this observation requires further validation in larger, well-controlled studies.”
• We chose to discuss contamination issues in such a detailed way, because we believe that it is the main problem of all microbiome studies, especially in reproductive medicine.

 

Comment 8: The conclusions should be revised to better reflect the results presented. While the study introduces an interesting concept, it largely demonstrates a lack of strong association between microbiological profiles and embryo transfer outcomes. Therefore, statements suggesting clinical applicability or utility should be tempered, and greater emphasis should be placed on the need for further validation in larger, well-controlled studies.

Response 8: Thanks for the tips on how to improve this section. We have revised in to better highlight the lack of strong association of microbiological profiles with frozen ET outcomes and the need for further validation.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

you have made substantial and thoughtful efforts to address my comments, and the manuscript has improved significantly in terms of clarity, methodological transparency, and overall scientific rigor.

In particular, the revisions to the methods section, including the implementation of multiple comparison correction and the expanded justification of the analytical approach, are highly appreciated. The restructuring of the results and the clearer distinction between primary and exploratory findings have also strengthened the manuscript. Additionally, the discussion and conclusions are now more balanced and better aligned with the data presented.

A few minor points remain, such as the absence of clearly defined inclusion and exclusion criteria (heterogeneity). Although the “real-world population” approach is understandable, this aspect should continue to be interpreted with caution. Nevertheless, overall, the manuscript is now well-structured, transparent in its limitations, and provides a meaningful exploratory contribution to the field.

Author Response

Esteemed Reviewer,

Once again, thank you for reviewing our manuscript. We are glad that you have found our corrections based on your comments substantial and thoughtful. To address the last point you have made regarding the absence of strict inclusion criteria, we have added a new statement to the study limitations section:
"Besides this, we did not apply any strict inclusion and exclusion criteria for the study enrollment; thus, the cohort was quite heterogenous, and it was difficult to control all confounding factors due to the small sample size."