Large-Scale Study of Antibody Titer Decay following BNT162b2 mRNA Vaccine or SARS-CoV-2 Infection

Round 1

Reviewer 1 Report

The manuscript by Israel et al., represents an interesting study on the persistence of SARS-CoV-2-spike specific IgGs in two large cohorts of BNT162b2-vaccinated and COVID-19 convalescent individuals not previously infected. The message of a differential decline in antibody titers between both cohorts is relevant. I would like to propose some queries in an attempt to improve/clarify this work.

MAJOR

• Is it a descriptive study of two different populations or is it a comparison between them? If the latter, it is difficult to ascribe the differences only to vaccinated vs. convalescent as there are a lot of variables which are different between both groups. In my opinion, both cohorts cannot be compared between each other’s and therefore, the text (and maybe also the tables) should be structured by describing both populations separately. Also, any comparison should be indicated more carefully (suggest, we propose…).

• Is there a significant linear regression for convalescent individuals in figure 2.b?

MINOR

• Lines 37 and 36: “While convalesced seropositive individuals have approximately 90% protection from SARS-CoV-2 reinfection”. This sentence is stated based on a single (although clearly relevant) reference performed in a specific cohort of health care workers. I believe that this message should be either softened or backed with more references.

• Line 200: “Figure 1a. and 1b.”, relates to Figures 2.

• As you nicely describe the limitations of your study, please, place them in a separated paragraph.

Author Response

Reviewer 1:

Comments and Suggestions for Authors

The manuscript by Israel et al., represents an interesting study on the persistence of SARS-CoV-2-spike specific IgGs in two large cohorts of BNT162b2-vaccinated and COVID-19 convalescent individuals not previously infected. The message of a differential decline in antibody titers between both cohorts is relevant. I would like to propose some queries in an attempt to improve/clarify this work.

MAJOR

Is it a descriptive study of two different populations or is it a comparison between them? If the latter, it is difficult to ascribe the differences only to vaccinated vs. convalescent as there are a lot of variables which are different between both groups. In my opinion, both cohorts cannot be compared between each other’s and therefore, the text (and maybe also the tables) should be structured by describing both populations separately. Also, any comparison should be indicated more carefully (suggest, we propose…).

Response: We are pleased that reviewer 1 has understood our essential point that the populations of a) convalescent individuals and b) vaccinated, never infected individuals are different. As the reviewer seems to understand, public health officials and epidemiologists have repeatedly, since the start of vaccination, made the error of comparing these two populations. Therefore, paradoxically, some additional comparison -- as we have done -- is needed to show on a rigorous immunological basis why this is an error. We take the reviewer’s point that our comparison is imperfect because many of the demographic variables between the populations are unmatched, but the datasets are large and the differences in antibody kinetics are very substantial, so we think that there is still some information to be gained.

More specifically, we do not claim in the manuscript that the convalescent and vaccinated populations are similar and we analyze the two populations in distinct tables and graphs. Our analysis includes multivariable regression that deliberately adjusts for differences in baseline factors. Importantly, the antibody measurements in the two populations are done using the same kit and hence the numerical measurements are directly comparable. Accordingly, we added the following text at the end of the Introduction:

We show that these two populations are different demographically and hence, our analyses treat the vaccinated and convalescent populations separately. We use multivariable regression that largely corrects for demographic and comorbidity differences. Even with this correction, the kinetics of antibody decline in the convalescent and vaccinated populations appear to differ substantially.

and we added the following text in the Discussion:

“Importantly, the antibody measurements in the two populations were done in the same laboratory facility and with the same kits. The two populations have different demographic characteristics and therefore, we analysed the two cohorts separately. Yet, early in the vaccination phase, the most common way to assess whether the vaccination was effective was to compare antibody levels in newly vaccinated individuals to antibody levels in convalescent individuals (see the subsection 4.3 and 4.4 below). To make this comparison possible, we used multivariable regression in order to correct for underlying differences.”

Furthermore, and also in order to fulfill a suggestion from reviewer 2, we have added a new subgroup analysis for patients aged ≥ 60 in both cohorts. In this analysis, the two subcohorts have similar age distributions. New regression analyses have been performed for this subgroup, and the new coefficients are shown in the newly added Table 6. Yet, we still observe a substantial difference in the rate of antibody decline between convalescent and vaccinated individuals. The following text added was added at the end of Results:

As shown in Tables 2 and 3, the populations of convalescent and vaccinated individuals have different demographic characteristics, with the most obvious difference being the age distribution. We have used multivariable regression to attempt to adjust for differences in age, demographics and comorbidities in Tables 4 and Table 5. Because higher age is a risk factor for COVID-19, we also perfomed a subgroup analysis on individuals aged ≥ 60 in both convalescent (n=712) and vaccinated (n=1268) cohorts. In this subgroup analysis, the two groups are quite similar in age (mean age=67.8±6.5 for covalescents, and 69.2±6.5 in vaccinated). Table 6 shows the regression coefficients in this subgroup. For vaccinated individuals aged ≥ 60 the mean decay factor was 0.619 indicating a decrease of about 38% per month; whereas for convalescent individuals aged ≥ 60, the mean decay factor was 0.889, indicating a decrease of about 12% per month. The decline of antibody titers was faster in convalescent individuals aged ≥ 60 than for younger convalescent individuals, and even among individuals aged ≥ 60, we observe a substantial difference in the rate of decay between convalescent and vaccinated individuals.

Is there a significant linear regression for convalescent individuals in figure 2.b?

Response: Yes, the p-values are shown in Table 4. Because the journal wants the Figures numbered 1, 2, 3, 4, what was Figure 2b was relabeled as Figure 4 in the revised manuscript, but the Figure content is unchanged.

MINOR

Lines 37 and 36: “While convalesced seropositive individuals have approximately 90% protection from SARS-CoV-2 reinfection”. This sentence is stated based on a single (although clearly relevant) reference performed in a specific cohort of health care workers. I believe that this message should be either softened or backed with more references.

Response: We weakened the statement to “While one large study estimated that convalesced seropositive individuals have approximately 90% protection…”. Because we reformatted the manuscript to fit the Vaccines template, line numbers have changed.

Line 200: “Figure 1a. and 1b.”, relates to Figures 2.

Response: This has been corrected. Because we reformatted the manuscript to fit the Vaccines template, line numbers have changed.

As you nicely describe the limitations of your study, please, place them in a separated paragraph.

Response: Done. More generally, we separated the Discussion into Discussion and Conclusions (as permitted by the Vaccines template). We also divided the Discussion into numbered subsections, one of which is entitled “Study Limitations”.

Reviewer 2 Report

I read the manuscript with great interest. I think it's very successful and the results are of great interest.
I would like to add the following points, however.

1. I have recently read a lot of studies or data on the subject of antibody development after vaccination or infection. The specifics of this study should be made clearer.
2. The main problem is the differences between the groups. Those who have recovered are, among other things, significantly younger and have fewer previous illnesses than those who have been vaccinated. I would like to see more subgroup analyzes here. For example how are the courses in the different age groups. Or how it progresses in subjects without previous illnesses.

Author Response

Reviewer 2:

Comments and Suggestions for Authors

I read the manuscript with great interest. I think it's very successful and the results are of great interest.

Response: We thank reviewer 2 for this positive remark.

I would like to add the following points, however.

I have recently read a lot of studies or data on the subject of antibody development after vaccination or infection. The specifics of this study should be made clearer.

The main problem is the differences between the groups. Those who have recovered are, among other things, significantly younger and have fewer previous illnesses than those who have been vaccinated. I would like to see more subgroup analyzes here. For example how are the courses in the different age groups. Or how it progresses in subjects without previous illnesses.

Response:

We thank the reviewer for this excellent suggestion. We have now added a subgroup analysis of individuals of age ≥ 60. In this analysis, the two subcohorts have similar age distributions. New regression analyses have been performed for this subgroup, and the new coefficients are shown in the newly added Table 6. Yet, we still observe a substantial difference in the rate of antibody decline between convalescent and vaccinated individuals. We described this subgroup analysis with the following text added at the end of Results:

As shown in Tables 2 and 3, the populations of convalescent and vaccinated individuals have different demographic characteristics, with the most obvious difference being the age distribution. We have used multivariable regression to attempt to adjust for differences in age, demographics and comorbidities in Tables 4 and Table 5. Because higher age is a risk factor for COVID-19, we also perfomed a subgroup analysis on individuals aged ≥ 60 in both convalescent (n=712) and vaccinated (n=1268) cohorts. In this subgroup analysis, the two groups are quite similar in age (mean age=67.8±6.5 for covalescents, and 69.2±6.5 in vaccinated). Table 6 shows the regression coefficients in this subgroup. For vaccinated individuals aged ≥ 60 the mean decay factor was 0.619 indicating a decrease of about 38% per month; whereas for convalescent individuals aged ≥ 60, the mean decay factor was 0.889, indicating a decrease of about 12% per month. The decline of antibody titers was faster in convalescent individuals aged ≥ 60 than for younger convalescent individuals, and even among individuals aged ≥ 60, we observe a substantial difference in the rate of decay between convalescent and vaccinated individuals.

Also, in response to a related comment from reviewer 1, we have added text to the Introduction and Discussion mentioning the differences between the two cohorts and how we addressed them. We notably added the following text at the end of the Introduction:

We show that these two populations are different demographically and hence, our analyses treat the vaccinated and convalescent populations separately. We use multivariable regression that largely corrects for demographic and comorbidity differences. Even with this correction, the kinetics of antibody decline in the convalescent and vaccinated populations appear to differ substantially.

and the following text in the Discussion:

“Importantly, the antibody measurements in the two populations were done in the same laboratory facility and with the same kits. The two populations have different demographic characteristics and therefore, we analysed the two cohorts separately. Yet, early in the vaccination phase, the most common way to assess whether the vaccination was effective was to compare antibody levels in newly vaccinated individuals to antibody levels in convalescent individuals (see the subsection 4.3 and 4.4 below). To make this comparison possible, we used multivariable regression in order to correct for underlying differences.”