Review Reports - Molecular Characterization of Persistent SARS-CoV-2 Infections in Immunocompromised Patients

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this work, the authors describe three patients with prolonged SARS-CoV-2 infection due to immunocompromise or immunosuppression. SARS-CoV-2 acquires mutations in all patients, but the patient immunosuppressed with rituximab does not show spike mutations. That patient had severe Covid-19. Although it is difficult to know for sure in the patient with advanced HIV disease because of the presence of other respiratory infections, the other two patients seemed to have mild or no SARS-CoV-2 symptoms.

The most interesting result is the persistence of SARS-CoV-2 for 4 years in the patient with Good’s syndrome. A major weakness is that no isolation and testing of the viruses is presented, even though the authors state on lines 261-263 that, for the second patient, “Infectious SARS-CoV-2 was successfully cultured from the nasopharyngeal swabs collected on day 189. SARS-CoV-2 isolate showed cytopathic effect on Vero E6 cells supporting persistent SARS-CoV-2 infection.” Without some characterization of the viruses for either immune escape, replication, or both, it is difficult to see how these findings add to the field, given that there is already a lot of published data showing prolonged infections in immunocompromised people.

The authors are also not consistent in describing the patients, having detailed information about infection course only for the first participant. There is also confusion in the differentiation of severe Covid-19 from prolonged infection, which are not the same thing.

Comments on the Quality of English Language

Multiple issues.

Author Response

Response: Thank you for this comment. We agree that characterization of viruses is highly informative in studies of persistent SARS-CoV-2 infection. In the revised manuscript, we have clarified and expanded the virological evidence available for each case.

For the patient living with HIV/AIDS (Case 2), infectious SARS-CoV-2 was confirmed by successful viral isolation from a nasopharyngeal swab collected on day 189, with clear cytopathic effect observed in Vero E6 cells. Whole-genome sequences obtained from this viral isolate are now explicitly referenced in the manuscript, with the corresponding GISAID accession numbers, and a new paragraph has been added to document this finding.

For the patient with follicular lymphoma treated with rituximab (Case 1), viral isolation could not be performed at the time of sample collection. However, persistent detection of subgenomic RNA is presented as a surrogate marker of ongoing viral replication. In addition, longitudinal follow up supports sustained replication in vivo.

To our knowledge, Case 3 represents one of the longest documented cases of SARS-CoV-2 persistence reported to date, spanning approximately four years, with continuous molecular evidence consistent with a single B.1 lineage infection in a patient with Good’s syndrome. While persistent SARS-CoV-2 infection has been described in immunocompromised hosts, cases with such duration and longitudinal genomic documentation remain exceedingly rare. The combination of prolonged follow-up and genome-level data provides a unique natural history perspective on long-term SARS-CoV-2 persistence in profound immunodeficiency, even in the absence of in vitro viral characterization for this specific case.

We have revised the manuscript to better emphasize these points and to highlight the strengths and the limitations of the available data in each case.

Response: Thank you for this comment. We acknowledge the need for consistent and structured descriptions of cases and clear conceptual distinctions between severe COVID-19 and persistent SARS-CoV-2 infection.

In the revised manuscript, we expanded the clinical and virological descriptions for patients 2 and 3, including clinical timelines, diagnostic testing, and the timing of antiviral and other treatments for clarity and added the corresponding figures 4 and 8 to clarify this.

Reviewer 2 Report

Comments and Suggestions for Authors

The mutation of the SARS-CoV-2 virus and its impact on transmissibility and pathogenicity are currently critical public health concerns. Particularly in immunocompromised populations, the inability of the host immune system to clear the virus in a short period provides a long-term breeding ground for viral replication. This environment also facilitates continuous genomic mutation of the virus. Therefore, the topic of this study is highly significant. However, the description of results and analysis need to be refined before publication. Especially, this study appears to devote more attention to the description and discussion of patient conditions than to the analysis and discussion of viral molecular variations, as it’s difficult to establish relationship between virus variation and basic disease or immune-status of patients, the value of patients’ descriptions is diminished.

Revision suggestions:

I suggest that the abstract of the article should be rewritten, as it fails to summarize the core findings of the study, particularly regarding the genetic analysis of the variant strains. Instead, it primarily describes the patient sources of the samples.Similarly, the introduction section of the article mainly emphasizes the significance of this study. It lacks any discussion on the recent advances in variant research, as well as an introduction to previously identified mutation sites or the classification of variant strains.
The title of part 3 is baffling, as the content of part 3 is the whole result.
The X axis of Fig.1 should be labeled with time periods after adopted, not the date. And why only the timeline of patient1 was shown?
Line 398, “might me” should be “might be”.
It is recommended to conduct a systematic comparison of the viral molecular variations among the three patients, along with comparative analysis and predictions referencing previously published literature. Currently, the analysis and discussion in this section are insufficient, as merely presenting mutation information does not enhance the academic value of the paper.
The current visualization of the virus phylogenetic tree is not suitable for publication. It is recommended to optimize the representation of the branching structure.

Author Response

Response: Thank you for this comment. We agree that, in a report focused on prolonged SARS-CoV-2 infection, a key result derives from the longitudinal virological data and their interpretation. We have therefore refined the Results and Discussion to strengthen the analysis of viral molecular variation (by product, with emphasis on Spike) and to better integrate the clinical/immunological context as a framework to interpret the observed within-host evolution.

In parallel, we clarified the clinical narrative and retained clinical details where they inform the virological findings (e.g., timing and type of immunosuppression, serological status when available, and timing of antiviral and/or other relevant treatment). This is supported by the established association between humoral immunodeficiency, including malignancies, AIDS progression, and anti-CD20 therapies, and persistent SARS-CoV-2 infection with within-host evolution.

Finally, we have added figures 6 and 11 to make the longitudinal molecular results easier to follow, and we have revised the Discussion to more explicitly describe the virological contribution of this study relative to the existing literature.

Revision suggestions:

I suggest that the abstract of the article should be rewritten, as it fails to summarize the core findings of the study, particularly regarding the genetic analysis of the variant strains. Instead, it primarily describes the patient sources of the samples. Similarly, the introduction section of the article mainly emphasizes the significance of this study.

Response: The abstract was rewritten with focus in genetic analysis of variant strains.

It lacks any discussion on the recent advances in variant research, as well as an introduction to previously identified mutation sites or the classification of variant strains.

Response: we added in discussion and introduction information of identified mutations in different VOC´s.

The title of part 3 is baffling, as the content of part 3 is the whole result.

Response: Apologies for this editorial mistake. Sections have been clearly labeled and Results are now under section 3.

The X axis of Fig.1 should be labeled with time periods after infection, not the date. And why only the timeline of patient1 was shown?

Response: We appreciate this observation. We modified the X-axis labels in Figure 1 accordingly.

Line 398, “might me” should be “might be”.

Response: The text has been modified accordingly.

It is recommended to conduct a systematic comparison of the viral molecular variations among the three patients, along with comparative analysis and predictions referencing previously published literature. Currently, the analysis and discussion in this section are insufficient, as merely presenting mutation information does not enhance the academic value of the paper.

Response: we added two figures (6 and 11) comparing the mutations detected in patient 2 and 3 respectively with some VOC´s and mutations reported in other publications. Also, we added a paragraph in discussion analyzing and comparing between mutations in Spike protein detected in our patients, and the reported in other cases.

The current visualization of the virus phylogenetic tree is not suitable for publication. It is recommended to optimize the representation of the branching structure.

Response: Thank you for this comment. We have revised the phylogenetic tree figures to improve clarity and suitability for publication. The visualization of the branching structure has been optimized, label size have been adjusted to enhance readability. These changes were implemented to facilitate correct interpretation of the phylogenetic relationships.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript presents three cases of persistent SARS-CoV-2 infections in immunocompromised patients: a rituximab-treated follicular lymphoma patient (persistence ~9 months, BA.5.1.23 lineage), an advanced HIV patient (persistence >10 months, BA.2 lineage with extensive mutations), and a Good's syndrome patient (persistence ~4 years, B.1 lineage).

The authors present three cases with different mechanisms of immune suppression, demonstrating quasispecies evolution in the three subjects.

However, patterns differ markedly (ex., minimal spike evolution in cases 1 e 3 vs. convergent mutations in case 2 akin to chronic infections, warranting caution against overinterpreting drivers like immune pressure. moreover deeper quasispecies analysis (e.g., via amplicon sequencing depth metrics), clearance versus evolution is not fully proven. Antiviral timing/outcomes need clearer linkage to mutation emergence

Sample sizes are small (n=3 cases, 7-16 sequences total), limiting generalizability

English requires polishing (ex., "turn SARS-CoV-2 RT-qPCR negative" → "became SARS-CoV-2 RT-qPCR negative"; inconsistent tense in cases/discussion).

Figures/tables are informative but legends need more precision

Given the limitations, the work could be revised and considered as a letter for a less prestigious journal.

Author Response

The authors present three cases with different mechanisms of immune suppression, demonstrating quasispecies evolution in the three subjects.

Response: Thank you for this insightful comment. We agree that the three cases represent distinct immunological contexts and that the observed patterns of viral evolution differ substantially across subjects. We also agree that these differences warrant caution against overinterpretation of specific evolutionary drivers, such as immune pressure, particularly in the absence of a larger sample size and deeper quasispecies analyses.

In the revised manuscript, we have therefore tempered our interpretations to emphasize descriptive longitudinal virus changes rather than any mechanistic inferences associated with the host. We explicitly acknowledge that the evolutionary trajectories observed—ranging from limited Spike diversification in Cases 1 and 3 to more extensive and convergent mutations in Case 2. Such changes likely reflect a complex set of factors, including host HLA and immune status, duration of infection, and therapeutic interventions, rather than a single dominant selective pressure.

To better contextualize these observations, we have expanded the clinical and virological descriptions for all cases, with particular attention to the timing of antiviral and other therapeutic interventions relative to mutation emergence. In addition, we have added schematic figures (6 and 8) summarizing the course of infection, diagnostic findings, and treatments administered, allowing clearer visualization of temporal relationships between viral persistence, evolution, and clinical management.

Finally, we have revised the Discussion to explicitly state the limitations of our dataset, including the lack of deep quasispecies sequencing metrics, and to avoid definitive conclusions regarding viral clearance versus continued within-host evolution. Instead, we frame our findings as consistent with heterogeneous evolutionary dynamics in prolonged SARS-CoV-2 infection across different forms of immunosuppression, contributing longitudinal observational data to a growing body of literature on this topic.

English requires polishing (ex., "turn SARS-CoV-2 RT-qPCR negative" → "became SARS-CoV-2 RT-qPCR negative"; inconsistent tense in cases/discussion).

Response: Thank you for this comment. The manuscript has been carefully revised to improve clarity and English usage throughout, and the suggested corrections (including phrasing such as “became SARS-CoV-2 RT-qPCR negative”) have been implemented. Additional minor language edits were made to ensure consistency and readability.

Figures/tables are informative but legends need more precision

Response: Thank you for this comment. Figures, tables and their legends have been revised to improve clarity and readability.

Given the limitations, the work could be revised and considered as a letter for a less prestigious journal.

Response: While recognizing the limitations discussed, we believe the revised manuscript presents a coherent and substantive contribution as a longitudinal virological study, and we consider it appropriate for consideration within the scope of Viruses.

Reviewer 4 Report

Comments and Suggestions for Authors

This study investigated three patients with persistent COVID-19 infection by performing whole-genome sequencing of the virus. The article is well-structured and offers insights into the pathogenesis and treatment of long COVID patients. However, some minor improvements could be made:

The abstract should summarize the main findings.
The clarity of the figures needs to be improved.
The discussion should include a dedicated section comparing the mutations in the S protein among different immunocompromised patients with long COVID. A review of relevant studies should be included to compare with the findings of this paper.

Author Response

The abstract should summarize the main findings.

Response: The abstract was rewritten with focus in genetic analysis of variant strains.

The clarity of the figures needs to be improved.

Response: we modified the phylogenetic tree, increasing the size of the labels of the sequences in order to optimize the visualization of the samples.

The discussion should include a dedicated section comparing the mutations in the S protein among different immunocompromised patients with long COVID. A review of relevant studies should be included to compare with the findings of this paper.

Response: we added in discussion a comparative between mutations in Spike protein detected in our patients, and the reported in other cases.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I have no further question