SARS-CoV-2 Serology Testing in an Asymptomatic, At-Risk Population: Methods, Results, Pitfalls

Round 1

Reviewer 1 Report

The study is a real life demonstration of the limitations of serology testing when using a test with poor analytical performance and in a low prevalence population.  To best represent these findings, I recommend the authors:

1. Re-evaluate the title and manuscript focus to highlight that this approach and using this assay, yields many false positives in a low prevalence/asymptomatic population.
2. Expand the introduction to include information about the importance of the method performance characteristics (sens, spec, PPV and NPV) for the intended application and include any relevant details on the methods used. Also, mention and reference guidelines regarding serology testing in HCW or for screening.
3. Clarify the methods section.
   1. That the IgG/IgM method result IgM and IgG individually
   2. Testing algorithm (consider a diagram) and weekly testing scheme
      1. For example: A person with previous IgM+ would not get tested by PCR next week?
      2. Why a negative test following a positive was considered negative? Was that the instructions by the manufacturer, why not only accept 2/3 concordant results?
   3. Rationale of following up IgM test with IgG test
   4. The appendix was not included in the documents I could see. Is that the method performance characteristics?  Please mention specificity and sensitivity at least. Clarify if the lab validated the assays in-house and the performance.

4. Re-evaluate Figure 1
   • If I understand correctly, this is not seroconversion, because the results were false positives to start and only 2 persons had evidence of COVID-19. I would modify this category and possibly break the display be weeks or groups of weeks.
   • Simplify figure rows. Is follow up Ab status the one performed in 5 minutes?  They ‘all’ repeated the same, with 1 discrepant that was also IgM pos.  I would only display one of these rows and would comment on the discrepancy. 
   • Are any of the 45 individuals with false positive in the subsequent weeks, among the 12 originally positive?
   • Add PCR results to the figure, the pos case(s)
5. Revisit the Discussion
   1. It is important to highlight why this test is not reliable for this application, in the context of the performance characteristics of the test.
   2. And if only 1 test of 298 was IgG positive, discuss the value of using an approach like this in asymptomatic HCW
   3. The study was over a year ago, discuss these findings in light of today’s situation with increased cases in children.
   4. Comment on other studies like this in HCW, even if adults.

Author Response

The study is a real-life demonstration of the limitations of serology testing when using a test with poor analytical performance and in a low prevalence population.  To best represent these findings, I recommend the authors:

1. Re-evaluate the title and manuscript focus to highlight that this approach and using this assay, yields many false positives in a low prevalence/asymptomatic population.

1. Expand the introduction to include information about the importance of the method performance characteristics (sens, spec, PPV and NPV) for the intended application and include any relevant details on the methods used. Also, mention and reference guidelines regarding serology testing in HCW or for screening.

We thank the reviewer for these comments, we have changed the title and added information to the introduction and methods section to address the importance of the method performance characteristics.  Guidelines have also been referenced in the discussion section.

----------------------------

1. Clarify the methods section.
1. That the IgG/IgM method result IgM and IgG individually
2. Testing algorithm (consider a diagram) and weekly testing scheme
1. For example: A person with previous IgM+ would not get tested by PCR next week?
2. Why a negative test following a positive was considered negative? Was that the instructions by the manufacturer, why not only accept 2/3 concordant results?
3. Rationale of following up IgM test with IgG test
4. The appendix was not included in the documents I could see. Is that the method performance characteristics?  Please mention specificity and sensitivity at least. Clarify if the lab validated the assays in-house and the performance.

We apologize that the version you received didn’t have an appendix, it can be found in this version, located prior to the reference section.

Regarding 1 – We have clarified this in the manuscript.

Regarding 2 – At our institution, during this period of the pandemic, the policy for positive IgM antibody results was the removal of the health care worker from shift until PCR test results could be obtained.  PCR results during this period could be delayed up to 5 days.  With the understanding of the limited sensitivity of the test, particularly the IgM results, the recommendation from hospital administration, guided by the infectious disease experts at our institution, was to immediately repeat antibody testing if positive.  We conducted the repeat testing, in order to comply with our institution’s request.

Regarding 3 – The Abbott assay was used as the confirmatory antibody test in this study, it only tests for IgG.  We assessed for current infection using PCR. In this way we were screening for infection and history of infection, but not validating the IgM portion of the test by itself. However, it would be expected that if a true positive, an individual should either test positive via PCR or in the coming weeks continue to test IgM+ or become IgG+ (in which case another Abbott test would be sent).

Regarding 4 – The Abbott assay was validated by our lab prior to use, the Wytcote kit was not validated by our lab prior to use; this has been clarified in the appendix.

----------------------------

1. Re-evaluate Figure 1
• If I understand correctly, this is not seroconversion, because the results were false positives to start and only 2 persons had evidence of COVID-19. I would modify this category and possibly break the display be weeks or groups of weeks.
• Simplify figure rows. Is follow up Ab status the one performed in 5 minutes?  They ‘all’ repeated the same, with 1 discrepant that was also IgM pos.  I would only display one of these rows and would comment on the discrepancy. 
• Are any of the 45 individuals with false positive in the subsequent weeks, among the 12 originally positive?
• Add PCR results to the figure, the pos case(s)
•  

We would like to thank the reviewer for these recommendations, we have deleted the word seroconversion from the table. We did not add a breakdown by weeks as we felt it would increase the length of the table without adding information pertinent to the study conclusions. We have added the PCR results and tried to clarify/streamline the table.

Revisit the Discussion

• It is important to highlight why this test is not reliable for this application, in the context of the performance characteristics of the test.

We would like to thank the reviewer for this comment, we have highlighted this in the discussion section.

• And if only 1 test of 298 was IgG positive, discuss the value of using an approach like this in asymptomatic HCW

Please refer to the previous comment, additional language has been added to the discussion section.

• The study was over a year ago, discuss these findings in light of today’s situation with increased cases in children.
• Comment on other studies like this in HCW, even if adults.

We appreciate this comment by the reviewer, we have added additional language regarding both increased cases in children and other studies involving HCW’s to the discussion section.

Reviewer 2 Report

With interest, I have read the ms “Rapid antibody testing for SARS-CoV-2 in pediatric healthcare 2 workers” submitted for publication to Infect. Dis. Rep.

I have just a few minor comments/suggestions.

- Intro is really too short. Something more on the HCWs’ occupational exposure should be written.

- All abbreviations/acronyms should be written out in full on the first appearance in the text.

- Authors might want to specify a little more on the study procedures. In particular, I suggest to add essential information on the statistical analyses performed.

- I’d like to see, in Discussion/Conclusion more on the Public Health impact of the presented study and possible future implications from a policy perspective.

- Garcia-Basteiro, A.L.; Moncunill, G.; Tortajada, M.; Vidal, M.; Guinovart, C.; Jimenez, A.; Santano, R.; Sanz, S.; Méndez, S.; Llupià, A.; et al. Seroprevalence of antibodies against SARS-CoV-2 among health care workers in a large Spanish reference hospital. Nat. Commun. 2020, 11, 3500

Some possible references for context information, and to compare results:

Moscola, J.; Sembajwe, G.; Jarrett, M.; Farber, B.; Chang, T.; McGinn, T.; Davidson, K.W.; Northwell Health COVID-19 Research Consortium. Prevalence of SARS-CoV-2 Antibodies in Health Care Personnel in the New York City Area. JAMA 2020, 324, 893–895

Della Valle, P.; Fabbri, M.; Madotto, F.; Ferrara, P.; Cozzolino, P.; Calabretto, E.; D’Orso, M.I.; Longhi, E.; Polosa, R.; Riva, M.A.; Mazzaglia, G.; Sommese, C.; Mantovani, L.G.; The MUSTANG–OCCUPATION–COVID-19 Study Group. Occupational Exposure in the Lombardy Region (Italy) to SARS-CoV-2 Infection: Results from the MUSTANG–OCCUPATION–COVID-19 Study. Int. J. Environ. Res. Public Health 2021, 18, 2567

Garcia-Basteiro, A.L.; Moncunill, G.; Tortajada, M.; Vidal, M.; Guinovart, C.; Jimenez, A.; Santano, R.; Sanz, S.; Méndez, S.; Llupià, A.; et al. Seroprevalence of antibodies against SARS-CoV-2 among health care workers in a large Spanish reference hospital. Nat. Commun. 2020, 11, 3500

Author Response

With interest, I have read the ms “Rapid antibody testing for SARS-CoV-2 in pediatric healthcare 2 workers” submitted for publication to Infect. Dis. Rep.

 I have just a few minor comments/suggestions.

- Intro is really too short. Something more on the HCWs’ occupational exposure should be written.

We have added a sentence to this effect, placed in the introduction.

----------------------------

- All abbreviations/acronyms should be written out in full on the first appearance in the text.

We have edited the manuscript to include the full spelling of acronyms on first use in the text.

----------------------------

- Authors might want to specify a little more on the study procedures. In particular, I suggest to add essential information on the statistical analyses performed.

We would like to thank the reviewer for this comment. We have modified the methods section for clarity.

----------------------------

- I’d like to see, in Discussion/Conclusion more on the Public Health impact of the presented study and possible future implications from a policy perspective.

We have added a short discussion regarding some of the public health implications of this research.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

None at this moment.  Thanks for addressing the questions and suggestions.