Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents an epidemiological and microbiological investigation of NDM-1–producing carbapenem-resistant Acinetobacter baumannii isolates collected in Costa Rica. The topic is relevant, as the emergence and dissemination of NDM-producing A. baumannii remain a significant global concern, particularly in regions where genomic data are still limited. The study provides antimicrobial susceptibility data for a collection of clinical isolates and includes whole-genome sequencing of two selected strains.

Major comments

A major concern is the lack of justification for selecting only two isolates for genomic analysis. The manuscript does not explain whether the two sequenced strains are representative of the population, or whether they were selected based on distinct resistance profiles, temporal distribution, clinical origin, or genetic diversity. Without a clear rationale, it is difficult to determine whether the genomic findings are meaningful or generalizable.

2. The manuscript presents antimicrobial susceptibility data but does not sufficiently integrate these findings into an epidemiological framework. It is unclear whether the isolates represent sporadic cases or clonal spread within healthcare settings. The relationship between isolates is not explored (e.g., clustering or transmission patterns). 

3. The figure legends are insufficiently informative and do not allow the figures to be interpreted independently of the main text. In particular: key methodological details are missing, abbreviations are not consistently defined, and the meaning of colors, symbols, or annotations is not clearly explained.

Minor comments

1. The introduction could be more focused on the regional importance of NDM-producing A. baumannii and the knowledge gap this study aims to address.

2. The Methods section should provide more detail on antimicrobial susceptibility testing standards (CLSI vs EUCAST), quality control procedures, sequencing and assembly pipelines.

3. The manuscript would benefit from a table summarizing key characteristics of the sequenced isolates.

4. Terminology related to resistance (e.g., carbapenem-resistant, MDR, XDR) should be used consistently.

5. Minor language and formatting issues are present and should be corrected.

Author Response

PLEASE CHECK THE "CHANGES CONTROL" FILE!

Comment 1

This manuscript presents an epidemiological and microbiological investigation of NDM-1–producing carbapenem-resistant Acinetobacter baumannii isolates collected in Costa Rica. The topic is relevant, as the emergence and dissemination of NDM-producing A. baumannii remain a significant global concern, particularly in regions where genomic data are still limited. The study provides antimicrobial susceptibility data for a collection of clinical isolates and includes whole-genome sequencing of two selected strains.

 

RESPONSE 1: Thank you very much for the critical comments. They were very helpful to improve our manuscript. We have worked on a new version of the paper including the point of concern.

 

Comment 2

Major comments

A major concern is the lack of justification for selecting only two isolates for genomic analysis. The manuscript does not explain whether the two sequenced strains are representative of the population, or whether they were selected based on distinct resistance profiles, temporal distribution, clinical origin, or genetic diversity. Without a clear rationale, it is difficult to determine whether the genomic findings are meaningful or generalizable.

 

RESPONSE 2:

Thank you for pointing this out.

 

We have addressed this point to further emphasize the importance of conducting this study within the context of Costa Rica and Latin America, particularly considering the current limitations in infrastructure and installed capacity for genomic surveillance, which remain highly challenging. In the revised version, we provide a more detailed discussion of this perspective. We kindly ask you to review the updated manuscript, where—from the title to the abstract, conclusions, and main text—we have adopted a more exploratory framing, clearly indicating that these are initial analyses.

Additionally, we have justified how our context limits the ability to substantially expand the number of sequenced isolates. The following text provide a more detailed explanation of these constraints:

• Genomic surveillance of pathogens in Latin America, and in particular in Costa Rica, has historically been unequal compared with the Global North, as shown by various studies. I would like to point out this one: https://www.nature.com/articles/s42003-023-05745-7 in which no Latin American country (and most of countries in the global south) has reported >100 Carbapenem-resistant coli (please check Figure 1)
• In this context for LAtin America, among the recurring problems are insufficient infrastructure, lack of investment and funding, shortage of trained personnel, weak teamwork and networking, and limited advances that, when they exist, tend to be anecdotal and focused on specific pathogens and cases, as reported (https://www.neb.com/en/nebinspired-blog/genomes-without-borders-bridging-gaps-in-health-equity, https://www.bgi.com/global/news/new-who-report-exposes-major-equity-gaps-in-global-genomics-research and https://www.sciencedirect.com/science/article/pii/S0959437X2030174X ).
• In the case of Costa Rica, as we have reported in this and previous studies, prevalence and frequency data are practically nonexistent outside the internal records of health care facilities for CRAB and carbapenem-resistant bacteria.
• To help reduce this gap, we have developed a Project among hospitals, reference laboratories, and the University of Costa Rica (https://pathogensportal.ucr.ac.cr/dashboards/ipat-project/), including the establishment of the PAthogen Portal Node Costa Rica https://www.pathogensportal.org/pathogens-portal-nodes, the first one (out of two) to be developed in Latin America with the support of the Swiss Bioinformatics Institute and the EMBL-European Bioinformatics Institute.
• Thus, this study constitutes one of the first reports shared with the scientific community on CRAB and sequencing in the region. While in other contexts sequencing two genomes and analyzing a few samples may be considered basic and outdated, in Costa Rica this initiative and its results represent important milestones at the local level, which are even highlighted in news articles and broadcasts as innovations (example in english: https://thecostaricanews.com/ucr-redesigns-test-to-detect-antibiotic-resistant-bacteria/). Our project at the University of Costa Rica is, by its nature, one of the first in Central America to sequence more than 200 carbapenem-resistant isolates. Of that collection, only seven isolates are CRAB; unfortunately, five of these are no longer available due to documented deficiencies in storage infrastructure and logistics.
• We note that the genomes analyzed here are not representative of all cases in Costa Rica and are reported as specific cases based on data availability; this limits broader interpretation. Nevertheless, we emphasize that these data constitute an initial insight into the diversity and some of the cases circulating in the country.
• Again, we know that the limited availability of samples does not necessarily reflect a low burden of the problem, but rather the logistical and infrastructural difficulties that still need to be overcome. Precisely for this reason we are redefining the manuscript to emphasize this bias and the local importance of the project. At the national level, sequencing has been crucial for bringing together the clinical community, researchers, and epidemiological surveillance teams, who have begun to work in a coordinated way on these initial studies. Although the number of isolates analyzed may seem small in an international context, for epidemiological surveillance in Costa Rica these results are highly relevant: they have enabled improvements in local diagnostic tests.
• For example, in a previous study published in 2025 from this same project we optimized a PCR diagnostic test on an automated platform (GeneXpert) that was not correctly detecting certain resistance genes https://www.mdpi.com/2079-6382/14/8/772. This finding illustrates how automated platforms designed with strains circulating in Europe or North America may fail to detect local variants that were not considered in their design, producing false negatives. Thanks to our sequencing, we were able to identify these limitations and optimize the diagnostic conditions used in Costa Rica, with direct impact on patient care, since these adjustments are now considered in Costa Rica’s molecular diagnostic laboratories and we have already received requests from countries such as Colombia, Peru, and Brazil to use them in their countries.
• In the previous version of the manuscript we did not emphasize this scenario because we did not want to give a sense of victimization, but it is truly a matter of relevance and awareness for democratization. However, it is not easy. Last week I attended the global conference “Antimicrobial Resistance – Genomes, Big Data and Emerging Technologies” and, surprisingly, not a single presentation or researcher was presenting or representing information from Latin America; please check the program here: https://coursesandconferences.wellcomeconnectingscience.org/wp-content/uploads/2025/04/AMR-2026-Final-Programme.pdf. This global effort, again, shows the bias in how “global” research networks aiming for fair and equitable genomic surveillance remain unfair and inequitable, with the particular exclusion of representatives from Latin America.

As part of this critique, in response to your observations and to make the importance of Latin America’s underrepresentation clearer, I have added text to the Introduction and mainly to the Discussion addressing this situation. Title, abstract and the general view has been modified to cover this.

 

 

Comment 3

2. The manuscript presents antimicrobial susceptibility data but does not sufficiently integrate these findings into an epidemiological framework. It is unclear whether the isolates represent sporadic cases or clonal spread within healthcare settings. The relationship between isolates is not explored (e.g., clustering or transmission patterns). 

 

RESPONSE 3:

We revised the manuscript to clarify that the two isolates analyzed are individual, sporadic cases and do not constitute part of a recognized outbreak nor are they epidemiologically linked. Because of the very small number of cases and the lack of epidemiological connections, it is not possible to perform a robust transmission to infer nosocomial spread.

We expanded the phylogenetic results to show that, although both isolates fall within the same broader phylogenetic group when comparing world-wide CRAB genomes, they have distinct MLST profiles and belong to different clonal complexes/sequence types. This indicates possible independent evolutionary histories rather than recent clonal transmission between the two cases. We also note that these isolates differ from the single prior report from Costa Rica, suggesting circulation of diverse variants locally.

Finally, (as previously considered) we explicitly discuss a key limitation: there is very limited genomic representation from Latin America in public databases, which constrains comparisons and makes it premature to classify these isolates as part of any established epidemiologically important clonal lineage. We have added these clarifications and caveats to the Results and Discussion.

 

Comment 4

3. The figure legends are insufficiently informative and do not allow the figures to be interpreted independently of the main text. In particular: key methodological details are missing, abbreviations are not consistently defined, and the meaning of colors, symbols, or annotations is not clearly explained.

 

RESPONSE 4: The Figure legends were completely extended to explain the figures in detail, including abbreviation, symbols, colors and methods. Please see new text.

 

Comment 5

Minor comments

1. The introduction could be more focused on the regional importance of NDM-producing A. baumanniiand the knowledge gap this study aims to address.
2. The Methods section should provide more detail on antimicrobial susceptibility testing standards (CLSI vs EUCAST), quality control procedures, sequencing and assembly pipelines.
3. The manuscript would benefit from a table summarizing key characteristics of the sequenced isolates.
4. Terminology related to resistance (e.g., carbapenem-resistant, MDR, XDR) should be used consistently.
5. Minor language and formatting issues are present and should be corrected.

 

RESPONSE 5:

We have extensively reviewed the minor comments to improve this version. Please find all the modifications in the “Changes control PDF” and the final text.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

First of all, congratulations on your work. While it is a valuable study, one of its main limitations is the very small sample size used for genomic characterization. Of 190 carbapenem-resistant isolates, only seven were identified as Acinetobacter baumannii, and whole-genome sequencing was performed on only two strains. This limited number of genomes significantly restricts the ability to draw robust conclusions about the genomic diversity and epidemiology of CRAB in Costa Rica. The rationale for sequencing only two isolates should be explained, and the reason for excluding additional isolates due to unavailability, unsuitability for sequencing, or other reasons should be stated. If sequencing of the remaining isolates was not possible, this should be clearly justified. Furthermore, the study should be clearly framed as an exploratory or preliminary genomic analysis rather than a comprehensive characterization of circulating CRAB strains in the country. Another methodological concern relates to the estimation and interpretation of prevalence. The article reports carbapenem resistance rates of 9.8% for imipenem and 6.1% for meropenem; However, these values ​​are derived from data collected from only five hospitals. Therefore, it is unclear whether these figures reflect a national prevalence estimate or only the situation in selected institutions. The scope of these estimates should be clarified, and they should be avoided as nationally representative unless supported by broader surveillance data. Furthermore, additional methodological details are needed, including how epidemiological data were collected, how isolates were identified (infection or colonization), and whether duplicate isolates from the same patient were excluded from the analysis. Finally, some conclusions appear to overinterpret the available data. For example, the statement that Costa Rican isolates form a "unique monophyletic clade" may be too strong given that the genomic analysis is based on only two isolates. With such a limited sample, it is difficult to confidently infer the existence of a separate lineage. A more cautious interpretation would be to note that the analyzed isolates clustered near genomes linked to IC9-related lineages, rather than defining them as a single monophyletic clade.

Author Response

PLEASE CHECK THE "CHANGES CONTROL" FILE!

Comment 1

First of all, congratulations on your work. While it is a valuable study, one of its main limitations is the very small sample size used for genomic characterization. Of 190 carbapenem-resistant isolates, only seven were identified as Acinetobacter baumannii, and whole-genome sequencing was performed on only two strains.

 

RESPONSE 1: Thank you very much for the critical comments. They were very helpful to improve our manuscript.

We have worked on a new version of the paper including the point of concern.

 

Comment 2

This limited number of genomes significantly restricts the ability to draw robust conclusions about the genomic diversity and epidemiology of CRAB in Costa Rica. The rationale for sequencing only two isolates should be explained, and the reason for excluding additional isolates due to unavailability, unsuitability for sequencing, or other reasons should be stated. If sequencing of the remaining isolates was not possible, this should be clearly justified. Furthermore, the study should be clearly framed as an exploratory or preliminary genomic analysis rather than a comprehensive characterization of circulating CRAB strains in the country.

 

RESPONSE 2:

Thank you for pointing this out.

 

We have addressed this point to further emphasize the importance of conducting this study within the context of Costa Rica and Latin America, particularly considering the current limitations in infrastructure and installed capacity for genomic surveillance, which remain highly challenging. In the revised version, we provide a more detailed discussion of this perspective. We kindly ask you to review the updated manuscript, where—from the title to the abstract, conclusions, and main text—we have adopted a more exploratory framing, clearly indicating that these are initial analyses.

Additionally, we have justified how our context limits the ability to substantially expand the number of sequenced isolates. The following text provide a more detailed explanation of these constraints:

• Genomic surveillance of pathogens in Latin America, and in particular in Costa Rica, has historically been unequal compared with the Global North, as shown by various studies. I would like to point out this one: https://www.nature.com/articles/s42003-023-05745-7 in which no Latin American country (and most of countries in the global south) has reported >100 Carbapenem-resistant coli (please check Figure 1)
•  

 

• In this context for LAtin America, among the recurring problems are insufficient infrastructure, lack of investment and funding, shortage of trained personnel, weak teamwork and networking, and limited advances that, when they exist, tend to be anecdotal and focused on specific pathogens and cases, as reported (https://www.neb.com/en/nebinspired-blog/genomes-without-borders-bridging-gaps-in-health-equity, https://www.bgi.com/global/news/new-who-report-exposes-major-equity-gaps-in-global-genomics-research and https://www.sciencedirect.com/science/article/pii/S0959437X2030174X ).
• In the case of Costa Rica, as we have reported in this and previous studies, prevalence and frequency data are practically nonexistent outside the internal records of health care facilities for CRAB and carbapenem-resistant bacteria.
• To help reduce this gap, we have developed a Project among hospitals, reference laboratories, and the University of Costa Rica (https://pathogensportal.ucr.ac.cr/dashboards/ipat-project/), including the establishment of the PAthogen Portal Node Costa Rica https://www.pathogensportal.org/pathogens-portal-nodes, the first one (out of two) to be developed in Latin America with the support of the Swiss Bioinformatics Institute and the EMBL-European Bioinformatics Institute.
• Thus, this study constitutes one of the first reports shared with the scientific community on CRAB and sequencing in the region. While in other contexts sequencing two genomes and analyzing a few samples may be considered basic and outdated, in Costa Rica this initiative and its results represent important milestones at the local level, which are even highlighted in news articles and broadcasts as innovations (example in english: https://thecostaricanews.com/ucr-redesigns-test-to-detect-antibiotic-resistant-bacteria/). Our project at the University of Costa Rica is, by its nature, one of the first in Central America to sequence more than 200 carbapenem-resistant isolates. Of that collection, only seven isolates are CRAB; unfortunately, five of these are no longer available due to documented deficiencies in storage infrastructure and logistics.
• We note that the genomes analyzed here are not representative of all cases in Costa Rica and are reported as specific cases based on data availability; this limits broader interpretation. Nevertheless, we emphasize that these data constitute an initial insight into the diversity and some of the cases circulating in the country.
• Again, we know that the limited availability of samples does not necessarily reflect a low burden of the problem, but rather the logistical and infrastructural difficulties that still need to be overcome. Precisely for this reason we are redefining the manuscript to emphasize this bias and the local importance of the project. At the national level, sequencing has been crucial for bringing together the clinical community, researchers, and epidemiological surveillance teams, who have begun to work in a coordinated way on these initial studies. Although the number of isolates analyzed may seem small in an international context, for epidemiological surveillance in Costa Rica these results are highly relevant: they have enabled improvements in local diagnostic tests.
• For example, in a previous study published in 2025 from this same project we optimized a PCR diagnostic test on an automated platform (GeneXpert) that was not correctly detecting certain resistance genes https://www.mdpi.com/2079-6382/14/8/772. This finding illustrates how automated platforms designed with strains circulating in Europe or North America may fail to detect local variants that were not considered in their design, producing false negatives. Thanks to our sequencing, we were able to identify these limitations and optimize the diagnostic conditions used in Costa Rica, with direct impact on patient care, since these adjustments are now considered in Costa Rica’s molecular diagnostic laboratories and we have already received requests from countries such as Colombia, Peru, and Brazil to use them in their countries.
• In the previous version of the manuscript we did not emphasize this scenario because we did not want to give a sense of victimization, but it is truly a matter of relevance and awareness for democratization. However, it is not easy. Last week I attended the global conference “Antimicrobial Resistance – Genomes, Big Data and Emerging Technologies” and, surprisingly, not a single presentation or researcher was presenting or representing information from Latin America; please check the program here: https://coursesandconferences.wellcomeconnectingscience.org/wp-content/uploads/2025/04/AMR-2026-Final-Programme.pdf. This global effort, again, shows the bias in how “global” research networks aiming for fair and equitable genomic surveillance remain unfair and inequitable, with the particular exclusion of representatives from Latin America.

As part of this critique, in response to your observations and to make the importance of Latin America’s underrepresentation clearer, I have added text to the Introduction and mainly to the Discussion addressing this situation. Title, abstract and the general view has been modified to cover this.

 

 

 

 

Comment 3

Another methodological concern relates to the estimation and interpretation of prevalence. The article reports carbapenem resistance rates of 9.8% for imipenem and 6.1% for meropenem; However, these values ​​are derived from data collected from only five hospitals. Therefore, it is unclear whether these figures reflect a national prevalence estimate or only the situation in selected institutions. The scope of these estimates should be clarified, and they should be avoided as nationally representative unless supported by broader surveillance data. Furthermore, additional methodological details are needed, including how epidemiological data were collected, how isolates were identified (infection or colonization), and whether duplicate isolates from the same patient were excluded from the analysis.

 

RESPONSE 3:

We have incorporated additional information to strengthen the value of these results while also clearly acknowledging that this is not a systematic prevalence study. Accordingly, we have added details in Table 1 and in the Methods section regarding the population covered by each hospital, which together represent more than 90% of Costa Rica’s population.

However, we fully recognize that this is not a controlled or systematic study designed to provide an accurate estimate of prevalence. Given the limited data available for Costa Rica, we consider this to be an important initial contribution. To better reflect this, we have framed the study more explicitly as an exploratory analysis, consistent with the revisions made throughout the manuscript.

We also emphasize the need for systematic studies that account for factors such as the number of infections, laboratory submissions, culture rates, and antimicrobial susceptibility testing, all of which may introduce biases into the observed estimates. While we reiterate the importance of such comprehensive approaches, we present our findings as exploratory prevalence data that provide a useful starting point. We believe this perspective better contextualizes our results while still offering valuable information for epidemiologists and decision-makers regarding carbapenem resistance in the country.

 

 

Comment 4

 

Finally, some conclusions appear to overinterpret the available data. For example, the statement that Costa Rican isolates form a "unique monophyletic clade" may be too strong given that the genomic analysis is based on only two isolates. With such a limited sample, it is difficult to confidently infer the existence of a separate lineage. A more cautious interpretation would be to note that the analyzed isolates clustered near genomes linked to IC9-related lineages, rather than defining them as a single monophyletic clade.

 

RESPONSE 4:

We agree with this point and have revised the manuscript to present the study as an exploratory analysis, removing the reference to a monophyletic group (in all the document) to avoid overinterpretation given the limited data.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript presents WGS results of a pathogen classified as critical by the World Health Organization in a geographic area that is still poorly explored, giving scientific merit to the study despite the extremely small number of isolates that were sequenced (two isolates). The manuscript requires several mandatory corrections.

 

Remarks

Throughout the manuscript, the authors must standardize the nomenclature of genes according to current conventions. For example, the gene “NDM-1” must be corrected to “bla_NDM-1”, in which “bla” must be in italics and “NDM-1” must be in non-italic subscript. The same applies to several other genes (OXA, KPC, VIM, NDM, IMP).

L21 and throughout the manuscript: replace prevalence with frequency. The epidemiological term prevalence is being used incorrectly in the manuscript.

L21: revise the sentence, as “the current prevalence to carbapenems was estimated at 21 9.8% to imipenem and 6.1% to meropenem” does not make sense. Are the authors referring to the frequency of resistant isolates?

L23: confirmed as what? As CRAB?

L27: standardize throughout the manuscript the use of the term “antimicrobial” instead of “antibiotic.” The term antibiotic excludes synthetic or semi-synthetic molecules.

L30: specify the number of genes that constituted the core genome, not only the percentage value.

In the Keywords, replace words that already appear in the title with others that better contextualize the study area in order to improve the indexing of the manuscript.

L134: it is necessary to specify the versions of the Clinical and Laboratory Standards Institute guidelines that were followed for establishing breakpoints.

L168: the authors need to include a supplementary file with the accession numbers of all isolates in order to ensure the reproducibility of the study.

Figure 1 is not mentioned in the text and is inserted in the wrong location in the manuscript.

In Table 1, it would be useful to create a column for isolates that were resistant to both antimicrobials, which would be the expected situation. Looking at the table, it is not possible to determine whether the same isolate, when resistant to both antimicrobials, appears in both columns. Therefore, the inclusion of a third column would be interesting to allow the identification of isolates resistant to both molecules and those resistant to only one.

L186: isolates of what? Of the microorganism or of resistant isolates?

The authors need to reformat the tables so that they comply with the format required by the journal.

L225: specify the number of the supplementary material in which the information will be available.

Several figures and tables are presented far from their mentions in the manuscript.

L282: what could explain this lower prevalence? The authors need to present discussions that justify their results, rather than merely providing a literature review.

L292: and how do these pieces of information relate to those presented in this study? Again, I emphasize that the discussion resembles a literature review rather than a discussion, since the authors do not use the presented information to relate it to the results of the study.

L298: revise the sentence.

L334: do the authors not have the phenotypic resistance data for the two isolates that were sequenced? It would be interesting to report the phenotypic resistance profile to evaluate whether it is consistent with the presence of certain genetic elements.

The conclusion needs to be reformulated. In its current version, it consists of a repetition of the main results, resembling an abstract. The authors need to present a conclusion that demonstrates what can actually be concluded from the results of this study.

 

Author Response

PLEASE CHECK THE "CHANGES CONTROL" FILE. 

Comment 1

 

The manuscript presents WGS results of a pathogen classified as critical by the World Health Organization in a geographic area that is still poorly explored, giving scientific merit to the study despite the extremely small number of isolates that were sequenced (two isolates). The manuscript requires several mandatory corrections.

 

RESPONSE 1: Thank you very much for the critical comments. They were very helpful to improve our manuscript

 

Comment 2

Remarks

Throughout the manuscript, the authors must standardize the nomenclature of genes according to current conventions. For example, the gene “NDM-1” must be corrected to “bla_NDM-1”, in which “bla” must be in italics and “NDM-1” must be in non-italic subscript. The same applies to several other genes (OXA, KPC, VIM, NDM, IMP).

 

RESPONSE 2: We have standardized the nomenclature so that all bla genes are now presented using the appropriate format. When referring to the protein or enzyme, the nomenclature was already correct and has been maintained.

 

 

Comments 3 (a general response is provided, see below)

L21 and throughout the manuscript: replace prevalence with frequency. The epidemiological term prevalence is being used incorrectly in the manuscript.

L21: revise the sentence, as “the current prevalence to carbapenems was estimated at 21 9.8% to imipenem and 6.1% to meropenem” does not make sense. Are the authors referring to the frequency of resistant isolates?

L23: confirmed as what? As CRAB?

L27: standardize throughout the manuscript the use of the term “antimicrobial” instead of “antibiotic.” The term antibiotic excludes synthetic or semi-synthetic molecules.

L30: specify the number of genes that constituted the core genome, not only the percentage value.

In the Keywords, replace words that already appear in the title with others that better contextualize the study area in order to improve the indexing of the manuscript.

L134: it is necessary to specify the versions of the Clinical and Laboratory Standards Institute guidelines that were followed for establishing breakpoints.

L168: the authors need to include a supplementary file with the accession numbers of all isolates in order to ensure the reproducibility of the study.

Figure 1 is not mentioned in the text and is inserted in the wrong location in the manuscript.

In Table 1, it would be useful to create a column for isolates that were resistant to both antimicrobials, which would be the expected situation. Looking at the table, it is not possible to determine whether the same isolate, when resistant to both antimicrobials, appears in both columns. Therefore, the inclusion of a third column would be interesting to allow the identification of isolates resistant to both molecules and those resistant to only one.

L186: isolates of what? Of the microorganism or of resistant isolates?

The authors need to reformat the tables so that they comply with the format required by the journal.

L225: specify the number of the supplementary material in which the information will be available.

Several figures and tables are presented far from their mentions in the manuscript.

L282: what could explain this lower prevalence? The authors need to present discussions that justify their results, rather than merely providing a literature review.

L292: and how do these pieces of information relate to those presented in this study? Again, I emphasize that the discussion resembles a literature review rather than a discussion, since the authors do not use the presented information to relate it to the results of the study.

L298: revise the sentence.

L334: do the authors not have the phenotypic resistance data for the two isolates that were sequenced? It would be interesting to report the phenotypic resistance profile to evaluate whether it is consistent with the presence of certain genetic elements.

 

RESPONSES 3:

All comments were considered:

• In line with all reviewers’ comments, we have refined the conceptual framework by consistently using the term “frequency” rather than “prevalence,” clarifying that this is not a systematic prevalence study. The results are now presented and discussed as an initial estimate and a preliminary contribution to generating data for Costa Rica, where information remains limited. This clarification has been incorporated throughout the manuscript, including the title, abstract, conclusions, and main text, to emphasize the exploratory nature of the study.
• Although limited in scope, this work represents an important first step toward strengthening genomic surveillance in Costa Rica and Latin America, where resources and capacities are still developing. It reflects the integration of multiple institutions, laboratory workflows, bioinformatics analyses, and regional collaboration. While modest compared to studies from the Global North, it is highly relevant in the local context and helps lay the foundation for future investment and more comprehensive research.
• We have further revised the manuscript to better articulate its scope and limitations, reinforcing its role as a foundational effort aimed at supporting larger-scale studies in the future. All suggested textual revisions have been incorporated, including enhanced discussion and alignment with the literature.
• A new supplementary material was added with all IDs from NCBI for pangenome analysis.
• Regarding Table 1 and resistance to other antibiotics: the addition of a column indicating whether an isolate shows resistance to both antibiotics (an others) was not feasible due to data limitations. The available system provides antibiotic-level results but does not allow direct cross-referencing at the isolate level without manual, case-by-case extraction, which is no longer possible given current access constraints. This limitation has been noted, and the improvement will be considered in future studies.
• Finally, the observed lower frequency in Costa Rica and certain settings may reflect a combination of infrastructure, surveillance capacity, and control measures, as well as ecological and epidemiological factors. In some contexts, resistance patterns for CRAB are clearly identifiable and of greater concern than currently observed locally. We have added a brief discussion noting that, while this remains an open question without definitive evidence, differences in microbial niches and potential competitive dynamics between strains could be contributing factors, though these remain speculative and require further investigation.

 

 

 

Comments 4

The conclusion needs to be reformulated. In its current version, it consists of a repetition of the main results, resembling an abstract. The authors need to present a conclusion that demonstrates what can actually be concluded from the results of this study.

 

RESPONSE 4:

We agree with this recommendation and have fully restructured the conclusions, emphasizing that this is an initial exploratory study that lays the foundation for strengthening genomic surveillance in Costa Rica and Latin America. We also highlight that, although the scope may appear limited, this effort is highly valuable as it integrates multiple institutions and is part of a broader initiative addressing diverse pathogens, as well as other regional efforts. In this work, we focus on Acinetobacter results, but within a wider context of collaboration and capacity building, and we explicitly contextualize the results as exploratory findings. Please refer to the revised conclusions in the main text.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

First of all , I want to thank you for your effort. The concepts of "prevalence" and "frequency" are used inconsistently. The title appears to have been edited multiple times (traces such as "Pepidemiology prevalence" still remain in the text). Although the authors repeatedly discuss the difference between these two concepts throughout the text, they have not been able to establish consistent usage. Given the study design, the term "prevalence" should not be used; data should only be described as "frequency" or "resistance rate," as a controlled prevalence study was not conducted. Example: "Prevalence" is used in the note below the Abstract and Table 1, but the limitations of this concept are explained in the Results section. This contradiction should be corrected. The sample size is insufficient and cannot be generalized. Complete genomic data are available for only two isolates. Of 190 carbapenem-resistant isolates, only seven A. baumannii isolates were identified; only two of these were sequenced. This situation may make the phrase "pangenome analysis" in the title misleading; the pangenome analysis was originally performed using the existing global database, not based on local isolates. Patient information is incomplete. IPAT126–IPAT130 and IPAT155 isolates are listed in Table 2, but the clinical context (inpatient, intensive care, fatal?) is not given. Why are the sequencing data for these isolates not available? This needs to be justified. The data in Table 1 are inconsistent. "National #4" is listed in Table 2 but not in Table 1. Why was this hospital excluded from the epidemiological analysis? This needs to be explained. Long-read sequencing was not used. It was acknowledged that the plasmid architecture could not be fully resolved with short-read (Illumina) sequencing; however, this is a serious limitation for a study requiring precise localization of plasmids carrying NDM-1. The implications of this limitation should be discussed in more detail. There are inconsistencies between figure titles and body text. In particular, Figure 3 appears in two different places and with two different numbers in the text (page 10 and page 11). This appears to be an error in the editorial review process, but it is essential to correct it. The claim of association with IC9 is strongly worded. The phrase "Clustered near IC9 forming a unique monophyletic clade" should be used with caution; the number of isolates does not sufficiently support this phylogenetic inference. A more cautious phrase such as "Consistent with proximity to IC9" should be preferred. The title of Table 3 is incorrect. While the title of Table 3 is "Structural and functional annotation…", its content presents pangenome data (core/accessory gene numbers). This table should be Table 4, and the title should be corrected accordingly. There are serious errors in the reference list: References [28]–[31] (Andrews/FastQC, Bolger/Trimmomatic, Li/MEGAHIT, Wick/Unicycler) are used in the Introduction section with references to "Latin America" ​​or "Costa Rica"; this is clearly an incorrect designation. These are methodological tools and should not be cited in the Introduction.Reference [20] refers to the article Boutzoukas & Doi (2025) and not PubMLST; a separate reference for PubMLST should be added.

Author Response

Reviewer 2

 

Comment 1

First of all , I want to thank you for your effort.

 

RESPONSE 1: Thank you very much for the critical comments. They were very helpful to improve our manuscript. We have worked on a new version of the paper including the point of concern.

 

Comment 2

The concepts of "prevalence" and "frequency" are used inconsistently. The title appears to have been edited multiple times (traces such as "Pepidemiology prevalence" still remain in the text). Although the authors repeatedly discuss the difference between these two concepts throughout the text, they have not been able to establish consistent usage. Given the study design, the term "prevalence" should not be used; data should only be described as "frequency" or "resistance rate," as a controlled prevalence study was not conducted. Example: "Prevalence" is used in the note below the Abstract and Table 1, but the limitations of this concept are explained in the Results section. This contradiction should be corrected.

 

RESPONSE 2: We have changed the use of “frequency” and “prevalence”, and now “resistance rate” is considered through the text; we agreed your comment. “frequency or prevalence” are only used to refer to other publications or the context that were focused on that; “resistance rate” is used for our data.

However, when you refer to “Pepidemiology prevalence” or inconsistent use of “frequency and prevalence”: Based on this comment, I have the impression that there may have been an error with the “control changes” file, since that word with the reported error was not used in the clean version, as I have verified.

It is possible that this is due to the fact that, in the tracked-changes version, information from both the previous and the current versions coexists, which during editing can create conflicts between them.

This type of conflict can affect both the visible text and, more significantly, the references, as old and new numbering may coexist.

I appreciate your help in reviewing the PDF version with the incorporated changes, and if an inconsistency of this nature persists.

 

TO AVOID THIS AGAIN: In the new clean version, we highlighted the changes in yellow.

 

Comment 3

The sample size is insufficient and cannot be generalized. Complete genomic data are available for only two isolates. Of 190 carbapenem-resistant isolates, only seven A. baumannii isolates were identified; only two of these were sequenced. This situation may make the phrase "pangenome analysis" in the title misleading; the pangenome analysis was originally performed using the existing global database, not based on local isolates.

 

RESPONSE 3: Yes, it is really sad for us. In discussion was already explained the situation of limited setting to overcome this kind of situations. The title and the general text was modified to show this work as a “exploratory and initial approach” to eventually improve the capabilities to run new studies with enough samples.

And yes, when we run a pangenome analysis, we use database info, and the Costa Rican samples are included. In the title, we don’t know how to clarify in more detail to avoid more text, but we think it is implicit the use of databases.

 

Comment 4

Patient information is incomplete. IPAT126–IPAT130 and IPAT155 isolates are listed in Table 2, but the clinical context (inpatient, intensive care, fatal?) is not given. Why are the sequencing data for these isolates not available? This needs to be justified.

 

RESPONSE 4: Patient information was not available due to limitations established with the approval and the nature of the project (epidemiological study). I added a sentence regarding this in Methods.

Regarding sequencing other samples: The isolates were not available for DNA extraction due to logistics limitations between institutions. In discussion was already explained the situation of limited settings to overcome this kind of situations.

 

Comment 5

The data in Table 1 are inconsistent. "National #4" is listed in Table 2 but not in Table 1. Why was this hospital excluded from the epidemiological analysis? This needs to be explained.

 

RESPONSE 5: The Hospital #4 is a private hospital; we changed the info in the table. We had no info (frequency) regarding this hospital.

 

Comment 6

Long-read sequencing was not used. It was acknowledged that the plasmid architecture could not be fully resolved with short-read (Illumina) sequencing; however, this is a serious limitation for a study requiring precise localization of plasmids carrying NDM-1. The implications of this limitation should be discussed in more detail.

 

RESPONSE 6: We agreed and added a discussion regarding the relevance of long-reads. Please check the text.

 

Comment 7

 

There are inconsistencies between figure titles and body text. In particular, Figure 3 appears in two different places and with two different numbers in the text (page 10 and page 11). This appears to be an error in the editorial review process, but it is essential to correct it. The claim of association with IC9 is strongly worded. The phrase "Clustered near IC9 forming a unique monophyletic clade" should be used with caution; the number of isolates does not sufficiently support this phylogenetic inference. A more cautious phrase such as "Consistent with proximity to IC9" should be preferred. The title of Table 3 is incorrect. While the title of Table 3 is "Structural and functional annotation…", its content presents pangenome data (core/accessory gene numbers). This table should be Table 4, and the title should be corrected accordingly.

 

RESPONSE 7:

I have checked and corrected the title, but again:

It was a problem with the “control changes” file, since  table 3 is shown once in the clean version, as I have verified. It is possible that this is due to the fact that, in the tracked-changes version, information from both the previous and the current versions coexists, which during editing can create conflicts between them.

The same with “unique monophyletic clade"” due to this concept and idea was eliminated completely in the revision. It is not present.  

I appreciate your help in reviewing the PDF version with the incorporated changes, and if an inconsistency of this nature persists.

 

TO AVOID THIS AGAIN: In the new clean version, we highlighted the changes in yellow.

 

Comment 8

There are serious errors in the reference list: References [28]–[31] (Andrews/FastQC, Bolger/Trimmomatic, Li/MEGAHIT, Wick/Unicycler) are used in the Introduction section with references to "Latin America" ​​or "Costa Rica"; this is clearly an incorrect designation. These are methodological tools and should not be cited in the Introduction.Reference [20] refers to the article Boutzoukas & Doi (2025) and not PubMLST; a separate reference for PubMLST should be added.

 

RESPONSE 8:Again, the “changes control version” has the previous numeration, but we confirm (you can check) that the references in the clean version were/are correct, for example, 28-31 are references related to Latin America or in the context of PABS

 

[28]         M. Concha-Toloza, L. C. González, A. H. H. Estrella, D. F. Do Porto, R. Campos-Sánchez, and J. A. Molina-Mora, “Genomic, socio-environmental, and sequencing capability patterns in the surveillance of SARS-CoV-2 in Latin America and the Caribbean up to 2023,” Nov. 2024, doi: 10.21203/RS.3.RS-5321558/V1.

[29]         J. Radeino Ambe, A. Bhan, B. Silaigwana, S. P. Salas, and S. Edwards, “Pathogen access and benefit sharing in a pandemic: working towards fair exchange?,” Lancet Microbe, vol. 0, no. 0, p. 101376, 2026, doi: 10.1016/j.lanmic.2026.101376.

[30]         K. Escandón-Vargas, S. Reyes, S. Gutiérrez, and M. V. Villegas, “The epidemiology of carbapenemases in Latin America and the Caribbean,” Expert Rev. Anti. Infect. Ther., vol. 15, no. 3, pp. 277–297, Mar. 2017, doi: 10.1080/14787210.2017.1268918.

[31]         U. J. Kim, H. K. Kim, J. H. An, S. K. Cho, K.-H. Park, and H.-C. Jang, “ Update on the Epidemiology, Treatment, and Outcomes of Carbapenem-resistant Acinetobacter infections ,” Chonnam Med. J., vol. 50, no. 2, p. 37, 2014, doi: 10.4068/CMJ.2014.50.2.37.

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

National #1 Hospital Anomaly: The dramatic increase in imipenem resistance from 1.1% in 2023 to 23.5% in 2024 is noteworthy. This sudden rise was not addressed in the discussion section. Possible explanations (pandemic, change in testing policy, contamination) must be thoroughly investigated. Plasmid Resolution: The authors already state that short-read Illumina data cannot fully resolve the plasmid architecture. However, the impact of this limitation on the results, particularly in the context of comments regarding the association of blaNDM-1 with the plasmid in IPAT15, should be emphasized more clearly. IC Assignment: The clustering of isolates close to IC9 but not directly assignable is an interesting finding; however, its clinical, epidemiological, and surveillance implications have not been sufficiently discussed.

Author Response

COMMENT 1

National #1 Hospital Anomaly: The dramatic increase in imipenem resistance from 1.1% in 2023 to 23.5% in 2024 is noteworthy. This sudden rise was not addressed in the discussion section. Possible explanations (pandemic, change in testing policy, contamination) must be thoroughly investigated.

 

RESPONSE 1:

The following text was added to the discussion; please find it in light blue (maintext):

“However, unlike other centers, Hospital #1 showed a notable increase in cases from 2023 to 2024 (1.1% to 23.5%), particularly in imipenem resistance. Importantly, no significant changes in molecular diagnostic strategies or tools were reported between these years in this center, suggesting the increase is unlikely to be due to methodological differences. This rise coincides with a reported outbreak driven mainly by Enterobacteriaceae carrying the blaNDM-1 gene, with sequencing data suggesting mobilization via a megaplasmid. Therefore, it cannot be ruled out that the increase in carbapenem resistance in Acinetobacter in 2024 at this hospital is linked to this outbreak (Gutierrez, 2025).”

 

 

COMMENT 2

Plasmid Resolution: The authors already state that short-read Illumina data cannot fully resolve the plasmid architecture. However, the impact of this limitation on the results, particularly in the context of comments regarding the association of blaNDM-1 with the plasmid in IPAT15, should be emphasized more clearly.

RESPONSE 2:

The following text was added to the discussion; please find it in light blue (maintext):

In the IPAT15 strain, the contig containing the blaNDM-1 gene was extremely short, preventing determination of its genomic context within the plasmid. This limitation also created challenges during structural annotation Again, these results highlight the need to employ long-read sequencing technologies to clarify the nature of these elements and their potential relationship to plasmids previously identified in Costa Rica asso-ciated with outbreaks [48], [76].

 

COMMENT 3

IC Assignment: The clustering of isolates close to IC9 but not directly assignable is an interesting finding; however, its clinical, epidemiological, and surveillance implications have not been sufficiently discussed.

RESPONSE 3

The following text was added to the discussion; please find it in light blue (maintext):

“IC of A. baumannii are major drivers of hospital-acquired infections and outbreaks, and are widely distributed across multiple countries (Boutzoukas & Doi, 2025; Castillo-Ramírez et al., 2025). Clinically, these high risk IC are strongly associated with multidrug resistance, including resistance to carbapenems mediated by OXA-type and other carbapenemases (Karah et al., 2025). This profile as IC limits treatment options, often leaving only last-resort drugs such as colistin, and is associated with high morbidity and mortality, especially in critically ill patients (Iovleva et al., 2024).

However, assigning an isolate to an IC (based on MLST profile) is not always straightforward; it depends heavily on the availability and quality of genomic and epidemiological data (Shelenkov et al., 2023). Thus, from an epidemiology perspective, these clones spread efficiently across hospitals and countries, facilitated by patient movement and environmental persistence (Shelenkov et al., 2023). Yet, in regions like Latin America, and particularly Costa Rica, there is still limited genomic surveillance data, making it difficult to determine whether circulating strains truly belong to established IC, even for the cases included in this study that are closet o IC-9. With improved data representation, it may become clearer whether local isolates are part of these global lineages or represent distinct regional variants. Notably, many strains in these regions already harbor carbapenem resistance genes and OXA variants that have been previously linked to globally disseminated, high-risk clones, suggesting potential connections that remain to be fully characterized (Diancourt et al., 2010; Shelenkov et al., 2023).”

 

Author Response File: Author Response.pdf