key: cord-1040505-urqjfhnw
authors: Fernandez Salinas, A.; Piano Mortari, E.; Terreri, S.; Quintarelli, C.; Pulvirenti, F.; Di Cecca, S.; Guercio, M.; Milito, C.; Bonanni, L.; Auria, S.; Romaggioli, L.; Cusano, G.; Albano, C.; Zaffina, S.; Perno, C. F.; Spadaro, G.; Locatelli, F.; Carsetti, R.; Quinti, I.
title: SARS-CoV-2 Vaccine Induced Atypical Immune Responses in Antibody Defects: everybody does their best
date: 2021-06-28
journal: nan
DOI: 10.1101/2021.06.24.21259130
sha: f06a2cfe8c98431fb2baeb106b1e925bb71c7b21
doc_id: 1040505
cord_uid: urqjfhnw

Background. Patients with Primary Antibody Deficiencies (PAD) represent a potential at-risk group in the current COVID-19 pandemic. However, unexpectedly low cumulative incidence, low infection-fatality rate, and mild COVID-19 or asymptomatic SARS-CoV-2 infections were frequently reported in PAD. The discrepancy between clinical evidence and impaired antibody production requires in-depth studies on patients immune responses. Methods. Forty-one patients with Common Variable Immune Deficiencies (CVID), 6 patients with X-linked Agammaglobulinemia (XLA), and 28 healthy age-matched controls (HD) were analyzed for anti-Spike and anti-RBD antibody production, generation of low and high affinity Spike-specific memory B-cells, Spike-specific T-cells before and one week after the second dose of BNT162b2 vaccine. Results. HD produced antibodies, and generated memory B-cells with high affinity for Trimeric Spike. In CVID, the vaccine induced poor Spike-specific antibodies, and atypical B-cells with low affinity for Trimeric Spike, possibly by extra-follicular reactions or incomplete germinal center reactions. In HD, among Spike positive memory B-cells, we identified receptor-binding-domain-specific cells that were undetectable in CVID, indicating the incapability to generate this new specificity. Specific T-cell responses toward Spike-protein were evident in HD and defective in CVID. Due to the absence of B-cells, patients with XLA responded to immunization by specific T-cell responses only. Conclusions. We present detailed data on early non-canonical immune responses in PAD to a vaccine against an antigen never encountered before by humans. From our data, we expect that after BNT162b2 immunization, XLA patients might be protected by specific T-cells, while CVID patients might not be protected by immunization. Key words: Primary Antibody Deficiencies, Common Variable Immune Deficiencies, X-linked Agammaglobulinemia, COVID-19, SARS-CoV-2, BNT162b2 vaccine, memory cells, affinity, Trimeric Spike, receptor-binding-domain.

and one week after the second dose of the BNT162b2 vaccine. Results showed lack of antibody responses in two thirds of patients with Common Variable Immune Deficiencies (CVID), and in all patients with X-linked Agammaglobulinemia (XLA) who have no B cells but produce specific T-cell responses at the same extent of HD.

CVID patients generated non-canonical B cell responses, as well as a variable response to the vaccination in terms of Spike-specific T-cells. This is the first study on the early B-cells response in PAD to a vaccine against an antigen never encountered before by humans.

Study Design and Patients. We studied patients regularly followed by the Italian Care Centers for adult patients with primary immune deficiencies in Rome and Naples. The study was carried out in 47 patients with PAD who agree to undergo SARS-CoV-2 immunization. We included also 28 immunized age-matched health care workers of the Bambino Gesù Children Hospital as healthy controls (HD). Eligible patients were informed on the study, including its safety profile and supply procedures, and signed the informed consents for vaccination and for the immunological study. The BNT162b2 vaccine was administered as prescribed, in two doses, 21 days apart. Two blood samples were obtained from each participant for serological and cellular immunity assessment at time 0 (T0), before the first dose, and seven days after the second dose (T1). During the study, patients were allowed to continue their therapies, and were monitored for their clinical status. The study was approved the Ethical Committee of the Sapienza University of Rome (Prot. 0521/2020, July 13, 2020). The study was performed in accordance with the Good Clinical Practice guidelines, the International Conference on Harmonization guidelines, and the most recent version of the Declaration of Helsinki. Methods for identify anti-Spike and anti-Receptor univariate analysis assessed the impact of variable of interest. Values were compared by the non-parametric Kruskal-Wallis test and, if not significant, the Wilcoxon matched pair signed-rank test or the two-tailed Mann-Whitney U-test were used. Differences were deemed significant when P < 0.05. Statistical Package for Social Sciences version 15 (SPSS Inc., 233 South Wacker Drive, 11th Floor, Chicago) will be used for the analysis.

Patients. In the study on Spike-specific antibody responses, we included 41 patients with CVID, 6 patients with XLA, and 28 healthy age-matched controls who received two doses of BNT162b2 vaccine. Demographic, clinical and immunological data at baseline of 47 patients enrolled and individual data are shown in Table S1 .

Immunoglobulin replacement treatment was continued in all PAD patients. Thirty-four CVID patients did not experience a previous SARS-CoV-2 infection, as shown by negativity at the periodical nasopharyngeal swab by PCR testing [7] . Seven CVID patients were vaccinated at least 3 months after recovering from a mild COVID-19.

None PAD patients was infected with SARS-CoV-2 after completing the two doses vaccination, with the exception of one CVID patient who was infected two months after completing the vaccination, when his Spike-specific IgG level was 2.5 AU. He was treated with monoclonal antibodies within 24 hours from the onset of a mild fever and since then he did not show any additional COVID-19 symptoms. Twenty-six CVID, and 6 XLA patients, and 28 HD were included in the study of the specific B-and T-cell responses.

Common Variable Immunodeficiency. CVID is the most prevalent symptomatic PAD [14] with reduced or absent antibody response to infections and immunization, paucity of switched memory B-cells and dysregulated T-cell responses. Spike-specific IgG and IgA, and RBD-specific IgM were evaluated at T0 and T1. Antibodies to SARS-CoV-2 antigens in CVID patients and in HD are shown in Fig.1 . In HD, anti-Spike IgG and IgA significantly increased in post-immunization samples, while anti-RBD antibodies of IgM isotype were already detectable at T0, reflecting the presence of natural or cross-reactive antibodies [15] . IgM increased at T1, with a wide variability between HD (Fig.1) .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 28, 2021. ; https://doi.org/10.1101/2021.06.24.21259130 doi: medRxiv preprint

In 41 patients with CVID, we observed a significant increase in the median values of anti-Spike IgG (from T0: 0.12 AU (IQR 0.07-0.24) to T1: 0.39 AU (IQR 0.13-5.25), p<0.0001), but not of IgA (from T0: 0.05 AU (IQR 0.04-0.11) to T1: 0.07 (IQR 0.04-0.20), p=0.070). Anti-RBD IgM also increased (from T0: 0 µg/mL (IQR 0-0.42) to T1: 0.04 µg/mL (IQR 0.13-5.25), p=0.006) in CVID patients (Fig. 1) . Post immunization antibodies against Spike and RBD were lower in CVID patients than in HD (IgG: p<0.0001, IgA: p<0.0001 and IgM: p=0.017) ( Fig. 1 ). X-linked Agammaglobulinemia. As expected, due to the lack of B cells and the consequent lack of serum antibody in XLA [16] , anti-Spike and anti-RBD specific antibodies were not generated after immunization (Fig. 1 ).

High specificity and affinity are the most important characteristics of protective memory B-cells (MBCs), generated by the adaptive immune system in response to infection or vaccination [17] . MBCs, atypical memory B cells (ATM), activated MBCs, and plasmablasts (PBs) were identified by flow-cytometry, based on the expression of CD19, CD27, CD24 and CD38 markers (Fig. 2 ). MBCs were identified as CD19+ CD24+ CD27+ CD38-cells; ATMs were identified as CD19+ CD27-CD24-CD38-cells, that also are also CD21 negative [18, 19] . Activated MBCs also lack CD21, but express CD27 [20, 21] . PBs were identified as CD19+CD24-CD38++CD27++ cells. Overall, CVID patients showed lower frequency of MBCs, and higher frequency of ATMs and PBs than HD at all study times (Fig.3 , and Table 1 ). This baseline pattern that distinguishes the B-cell populations of HD from those of CVID patients affects the specific response to vaccination.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All HD responded to immunization by generating a classical B cell response with low and high affinity MBCs (p=0.001 and p<0.0001, respectively) and high affinity activated MBCs (p=0.030). HD generated low affinity Trimeric Spike ATMs (p=0.0001), and low and high affinity Trimeric Spike PBs (p<0.0001) ( Fig. 3 and Table 1 ), demonstrating that these populations are induced by the immune response to vaccination [23].

On the contrary, patients with CVID did not generate classical and activated MBCs, but vaccination induced Trimeric Spike ATMs with low affinity only (p=0.019) ( Fig.3 and Table 1 ). This was observed only in about one third of CVID patients, where the frequency of low affinity ATMs at least doubled seven days after the second dose of vaccine. ATMs originally observed in tonsils and later in peripheral blood, are observed in conditions of chronic antigen stimulation [23] and may be produced by extra-follicular reactions or failed and incomplete germinal center reactions [24,25]. As expected, high affinity PBs for Trimeric Spike were produced in HD at T1, but not in CVID, with the exception of three patients who also produced specific IgG and IgA. Most CVID patients were able to generate low affinity Trimeric Spike PBs (p=0.001) that are possibly short-lived PBs derived from the non-canonical pathway of ATMs. Among total Spike positive MBCs (S+ plus S++), we also identified RBDspecific cells. RBD+ cells represent a minority of the MBCs generated by vaccination and significantly increased at T1 in HD (p<0.0001) ( Table 1 and Figure S1 ). These cells, with high specificity against the virus RBD and thus producing most of the neutralizing antibodies, are undetectable in CVID patients indicating the incapability of CVID B-cells to undergo somatic mutation in the germinal center and thus generate this new specificity.

XLA patients lacks B cells and consequently did not generate any specific B-cells (Table 1) .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 28, 2021. ; https://doi.org/10.1101/2021.06.24.21259130 doi: medRxiv preprint

Interferon gamma (IFN-g) Colony Forming Units that significantly increased (from T0: median 2.5 SFU (IQR 0-0.69) to T1: 115 median SFU (IQR 0.13-5.25), p=0.041). Differently from what observed after influenza vaccination [26], in CVID patients, Spike-specific T-cells producing IFN-g did not increase (from T0: median 0 SFU (IQR 0-9) to T1: median 7 SFU (IQR 0-35), p=0.151) and were significantly lower than in HD (p=0.0001).

Nevertheless, XLA patients developed a significant T-cell response to vaccination. Indeed, Spike-specific T-cells were detected in 5 out of 6 tested patients (from T0: median 37.5 SFU (IQR 0-50.5) to T1: median 152.5 SFU (IQR 41-185), p=0.041), with a negligible difference respect to the one established in HD (p=0.937).

Effective vaccines against SARS-CoV-2 are being administered worldwide with the aim of terminating the COVID-19 pandemic. As for all immunizations, the efficacy has been linked to the production of specific antibodies, which increase in response to all vaccines in use [27, 28] . The majority of patients with PAD show clinical and immunological characteristics implicating a functional impairment of B cell compartment, and a dysregulation of T-cell responses causing hypo-gammaglobulinemia or a-gammaglobulinemia and susceptibility to a wide range of microbial infections [29] . Despite severe impaired antibody responses, when infected with SARS-CoV-2, one fourth of adult PAD patients remained asymptomatic and half of them showed a mild disease [7] . It should be considered that a protective effect against severe COVID-19 could be due to the immunemodulatory effects on innate immunity exerted by immunoglobulin therapy also when administered at replacement dosages [30] . However, the discrepancy between this clinical evidence and the lack of antibodies underlines the limitations of this correlate of protection against the severe forms of COVID-19 and requires in depth studies on SARS-CoV-2 specific B-and T-cells function in PAD. However, data on immunogenicity of SARS-CoV-2 vaccine in patients with inborn errors of immunity are few and limited to anecdotal cases or heterogeneous cohort [31] . Although the natural course of COVID-19 is primarily characterized by the function of the innate immune system, with a secondary involvement of T-and B-cells, vaccines are designed to force the adaptive immune system to generate neutralizing antibodies and memory B-and T-cells that effectively protect from COVID-19.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 28, 2021. Here we showed that while HD produced specific antibodies and generate high affinity MBCs and activated MBCs that significantly increased after immunization, these responses are lacking in all XLA and severely impaired in CVID patients after SARS-CoV-2 immunization, suggesting an incomplete protection. However, in five out of seven CVID patients who were previously infected with SARS-CoV-2, IgG were detectable at the time of vaccination administered at least three months after SARS-CoV-2 infection recovery, suggesting that if IgG were produced, they might persist after the primary infection. Moreover, in these previously infected CVID patients, IgG response was boosted by the subsequent immunization.

Interestingly, in all CVID patients, vaccination induced Trimeric Spike-specific B-cells inside the ATM and PB population, with low affinity for Trimeric Spike, possibly suggesting that the B-cell responses occurred mostly at extra-follicular sites [32] as recently demonstrated [33] . In line with this hypothesis, RBD+ B-cells were undetectable in CVID patients, whereas they represent 20% of the highly specific anti-Trimeric Spike response in HD (unpublished data) able to develop and successfully terminate the germinal center reaction [34] . Thus, CVID patients were able to respond to immunization by two doses of BNT162b2 with non-canonical lineage B-cells induced by a primary exposure to a novel antigen. It has been suggested that atypical B cells are short-lived activated cells, in the process of differentiating into plasmacells [23, 35] . In addition, interesting information was gained by the parallel study of the T-cell responses, showing a robust generation of Spike-specific T-cells in all but one patient with XLA, and in HD. Specific-T-cell responses were induced in a minority of CVID patients with a variable frequency. Based on data on response to influenza virus immunization [11] , we expected a more efficient generation of specific T-cells. However, this was not the case. While after influenza virus immunization T-cells are generated after multiple exposures to viral antigen following infection and immunization, SARS-CoV-2 is a pathogen never encountered before. Then, it is possible that the first antigenic stimulation was not sufficient to induce an early T-cell response.

Based on these data, how can we explain the paucity of symptoms or the mild COVID-19 course in PAD patients, and what might we expect after immunization?

To pathogens for which there is no preexisting immunity, our organism reacts by rapidly engaging the innate immune system with the intent to limit the infection. The adaptive immune response develops slowly and needs . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 28, 2021. ; https://doi.org/10.1101/2021.06.24.21259130 doi: medRxiv preprint two weeks to generate the most specific and effective defensive tools. However, the vast majority of PAD patients infected with SARS-CoV-2 did not show signs of hyper-activation of the innate immunity [3] [4] [5] , an effect possibly due to the immunomodulatory effects on innate immune cells due to the regular use of polyvalent immunoglobulins [30] , and a poor adaptive immunity. This pattern of immune responses resembles what we have already shown in asymptomatic immunocompetent subjects [36] , and further demonstrated that a balanced cytokine production resulting from a functional but not hyper-reactive innate immunity and a poor adaptive immunity are the conditions associated with an early benign COVID-19 course. Virus-neutralizing antibodies are only one of the correlates of protection for COVID-19, as we and others [4] [5] [6] [7] demonstrated in patients with PAD, which, while unable to produce any antibodies, might remain asymptomatic after contracting SARS-CoV-2 infection.

Our data are partially in contrast to observation reported in small cohorts of PAD, showing that CVID patients are able to respond to the BNT162b2 vaccine [31] . Differently from previous studies, we described a homogeneous cohort of patients followed in a single clinical center, analyzing separately CVID and XLA and analyzing different time point for each participant in PAD and in HD group.

In summary, a minority of PAD patients showed adaptive, non-canonical immune responses after SARS-CoV-2 immunization. If these responses to vaccination might result in a partial protection from infection or re-infection is now unknown, since we do not know the levels of antibodies or the frequency of specific B-and T-cells required to protect from the infection. It should be remembered here that each PAD patient should be studied as a unique, in terms of cellular and humoral responses due to the variability of their underlying immune deficiency. In our series, antibody response after two doses of BNT162b2 immunization -overlapping that of HD -was found in one patient homozygous for TNFRSF13B mutation, but not in two patients with a heterozygous TNFRSF13B mutation. In a previous study, we demonstrated that CVID patients with biallelic TNFRSF13B mutations responded also to polysaccharide vaccines, while CVID with only one TNFRSF13B mutation showed an impaired response to vaccination [37].

A major limitation of our study is the short time of observation after vaccination. We do not know if the antibody and cellular responses might persist or decline over time, nor if PAD patients might show delayed responses.

However, we do not expect to observe major changes in the immune response to SARS-CoV-2 with time after . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 28, 2021.

immunization. It is difficult to hypothesize an improvement of specific B-cell immunity. On the contrary, data available from the protective T-cell immunity after influenza virus vaccination in PAD [11] 

Authors declare that they have no competing interests.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 28, 2021. test. *p < 0.05, **p < 0.01, ***p < 0.001; ****p< 0.0001. N = 28 HD and N= 32 CVID patients.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 28, 2021. ;

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 28, 2021. ; https://doi.org/10.1101/2021.06.24.21259130 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 28, 2021. ; https://doi.org/10.1101/2021.06.24.21259130 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 28, 2021. ; https://doi.org/10.1101/2021.06.24.21259130 doi: medRxiv preprint

Clinical and immunological features of severe and moderate coronavirus disease 2019

International Union of Immunological Societies: 2017 Primary Immunodeficiency Diseases Committee Report on Inborn Errors of Immunity

Primary Immunodeficiency Diseases in COVID-19 Pandemic: A Predisposing or Protective Factor?

Coronavirus disease 2019 in patients with inborn errors of immunity: An international study

Clinical outcomes and features of COVID-19 in patients with primary immunodeficiencies in New York City

A possible role for B cells in COVID-19? Lesson from patients with agammaglobulinemia

Clinical outcome, incidence, and SARS-CoV-2 infectionfatality rates in Italian patients with inborn errors of immunity

Safety and Efficacy of the BNT162b2 mRNA Covid-19

Epidemiological and evolutionary considerations of SARS-CoV-2 vaccine dosing regimens

Preserved Cellular Immunity Upon Influenza Vaccination in Most Patients with Common Variable Immunodeficiency

Persistence and Evolution of SARS-CoV-2 in an Immunocompromised Host

Health-Related Quality of Life in Common Variable Immunodeficiency Italian Patients Switched to Remote Assistance During the COVID-19 Pandemic

International Consensus Document (ICON): Common Variable Immunodeficiency Disorders

Systems serology detects functionally distinct coronavirus antibody features in children and elderly

Long-term follow-up of 168 patients with X-linked agammaglobulinemia reveals increased morbidity and mortality

Memory B cells: effectors of long-lived immune responses

Expression of the immunoregulatory molecule FcRH4 defines a distinctive tissue-based population of memory B cells

Immunogenicity of Pfizer-BioNTech COVID-19 Vaccine in Patients with Inborn Errors of Immunity

Generation of memory B cells inside and outside germinal centers

B cell subsets as severity-associated signatures in COVID-19 patients

Extrafollicular B cell responses correlate with neutralizing antibodies and morbidity in COVID-19

Maturation and persistence of the anti-SARS-CoV-2 memory B cell response

COVID-19 -pathogenesis and immunological findings across the clinical manifestation spectrum