Detection of neutralizing anti–severe acute respiratory syndrome coronavirus 2 antibodies in patients with common variable immunodeficiency after immunization with messenger RNA vaccines

Common variable immunodeficiency (CVID) is a rare disorder, occurring in approximately 2 to 4 per 100,000 individuals in the general population. Nevertheless, it is the most common symptomatic primary immunodeficiency disorder in adults and may account for a substantial proportion of patients in specialized immunology clinics. The cause is unknown; diagnosis relies on multiple criteria because no single clinical manifestation or laboratory test can aid in recognizing this entity.¹ Specifically, CVID may be diagnosed in case of a marked decrease of immunoglobulin G (IgG) (at least 2 SD below the mean for age) and a marked decrease in IgM or IgA, in addition to the following criteria: (1) onset of immunodeficiency at greater than 2 years of age; (2) absent isohemagglutinins and poor response to vaccines; and (3) exclusion of other possible causes of hypogammaglobulinemia.¹ Antibody production is always impaired in CVID, as a result of primary B-cell dysfunction or from lack of T cell help for antibody production.¹

The current pandemic of the coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) represents an unprecedented challenge for health care systems across the world,² but the ongoing global mass vaccination against SARS-CoV-2 may help reduce the hospital burden worldwide. Patients with CVID, however, may not be included into the vaccination programs because of the common perception that they are not responsive to vaccination.¹ In countries such as Italy, where general practitioners are in charge of the recruitment of “fragile” patients into vaccination programs, it is paramount to address the issue of the effectiveness of COVID-19 vaccines in patients with primary immunodeficiency, also considering the low awareness on this peculiar group of diseases among physicians not specializing in clinical immunology.³

Therefore, to establish whether SARS-CoV-2 vaccination may be meaningful in individuals with CVID, we investigated whether these patients could generate protective antibodies against SARS-CoV-2 after administration of messenger RNA (mRNA) vaccines⁴ and compared the outcome in healthy subjects from hospital staff undergoing COVID-19 vaccination.

A total of 5 patients (4 females, 1 male; median age, 54 years) with CVID (median age at diagnosis, 35 years) on monthly intravenous immunoglobulin replacement therapy agreed to receive an mRNA vaccine.⁴ Serum SARS-CoV-2 antibodies were measured in all patients immediately before vaccination and the next IgG infusion (because of the possibility of antibodies against SARS-CoV-2 in commercial intravenous immunoglobulin preparations⁵) and 30 days after the second vaccine jab, again just before the next IgG infusion. Adverse effects were recorded for all patient after each vaccine dose administration. Comorbidities and immunologic features were also registered. Patient #4 (Table 1 ) had a mild form of the natural infection with SARS-CoV-2 4 months before the vaccination. At that time, this patient only complained of fever (38°C) for 2 days and cough but it took nearly 4 weeks for his nasopharyngeal swab to turn negative on molecular recognition of SARS-CoV-2. Among the 5 patients subjected to vaccination, he was also the only one to have naturally occurring virus-specific neutralizing antibodies, which greatly increased after completion of the vaccination schedule (Table 1). The remaining 4 patients had no detectable anti–SARS-CoV-2 antibodies before vaccination. Postvaccination, all patients but 1 developed neutralizing antibodies against SARS-CoV-2, with varying degrees of positivity. Of note, the only patient who failed to have a substantial rise in postvaccination titers of anti–SARS-CoV-2 antibodies also had a marked decrease in the frequency of circulating B cells on flow cytometry assessment. Adverse effects were mild and transient and did not differ from those reported in other series.⁶ Mild pain at injection site was reported by nearly all patients. Clinical and laboratory features of all patients are summarized in Table 1. Control subjects (3 males, 7 females; age range, 41-62 years) had variable titers of neutralizing antibodies as well (range, 256-9060 binding arbitrary units per milliliter), with median values close to 1200 binding arbitrary units per milliliter. They had been all vaccinated with the Pfizer-BioNTech mRNA vaccine (COMIRNATY, Pfizer Manufacturing Belgium NV, Puurs, Belgium, and BioNTech Manufacturing GmbH, Mainz, Germany).

Table 1.

Summary of Clinical and Laboratory Features of Patients With Common Variable Immunodeficiency and Outcome of Vaccination

CVID patient number	Sex	Age	Years since diagnosis	Lymphocyte counts	Lymphocyte subsets	Comorbidities	Latest serum immunoglobulin assessmenta	Type of vaccine	Preimmunization anti–SARS-CoV-2 (antispike) antibodies	Postimmunization anti–SARS-CoV-2 (antispike) antibodies	Adverse effects after first vaccine dose	Adverse effects after second vaccine dose
1	F	58	9	Normal	Normal	Obesity, hypertension, thyroid nodule	IgG, 832 mg/dL IgA, 6 mg/dL IgM, 9 mg/dL	Pfizer-BioNTech	Absent	8.68 BAU/mL	Pain at injection site, diffuse myalgias, headache, fatigue	Pain at injection site
2	F	54	19	Normal	Normal	None	IgG, 777 mg/dL IgA, 2 mg/dL IgM, 2 mg/dL	Moderna	Absent	812.90 BAU/mL	Pain at injection site	Low-grade fever, myalgias, headache, fatigue
3	F	63	1	Mild lymphopenia (1060/μL)	Normal	Celiac disease, osteoporosis with vertebral fractures	IgG, 665 mg/dL IgA, 14 mg/dL IgM, 6 mg/dL	Pfizer-BioNTech	Absent	212.7 BAU/mL	Pain at injection site	Pain at injection site
4	M	45	12	Normal	Normal	None	IgG, 721 mg/dL IgA, 3 mg/dL IgM, 2 mg/dL	Pfizer-BioNTech	160 BAU/mL	10185 BAU/mL	Headache	Pain at injection site
5	F	37	6	Normal	CD19 B cells <1%b	Bronchiectasis, liver nodular lymphoid hyperplasia	IgG, 515 mg/dL IgA, 2 mg/dL IgM, 1 mg/dL	Pfizer-BioNTech	Absent	Absent	Pain at injection site, headache, myalgias, chills	Pain at injection site

Abbreviations: BAU, binding arbitrary unit, according to the World Health Organization international standard (<1, negative; >1, positive); CVID, common variable immunodeficiency; ECLIA, electrochemiluminescence immunoassay; F, female; Ig, immunoglobulin; M, male.

NOTE. Antibodies were detected by ECLIA.

^aNormal values for serum immunoglobulin isotypes were as follows: IgG, 700 to 1600 mg/dL; IgA, 70 to 400 mg/dL; IgM, 40 to 230 mg/dL.

^bNormal B-cell frequencies: 6% to 20%.

Overall, our experience reveals that SARS-CoV-2 vaccination is safe and effective even in patients with primary antibody deficiencies. Nevertheless, as inferred by the only patient who did not have an antibody response to vaccination, production of neutralizing immunoglobulins may at least require preservation of the B lymphocyte population. Whether the serum levels measured in patients with CVID are able to fully protect these subjects from mild or severe manifestations of COVID-19 is currently unknown, because there are not yet ad hoc studies. Further complicating things, patients with CVID are known to be able to even produce antibodies at normal titers, but these antibodies may have poor avidity with impaired opsonophagocytic function.⁷ Likewise, comparing serum levels of anti–SARS-CoV-2 antibodies between patients with CVID and immunocompetent subjects to evaluate the magnitude of vaccination efficacy may as well be meaningless, because the level of neutralizing antibodies required to confer protection has not been definitively established.⁸ Indeed, serum levels of anti–SARS-CoV-2 antibodies seemed to be overall more elevated in healthy hospital personnel than in patients with CVID, but, again, the clinical significance of this difference is currently unknown.⁸ Interestingly, the only patient (#4) with previous, naturally acquired immunization against SARS-CoV-2 had an impressive rise in the titers of neutralizing antibodies after vaccination; thus, if previous immunization is a prerequisite for remarkable antibody responses after vaccination, it may theoretically be anticipated that a further vaccine jab would probably substantially enhance the protective antibody response in the remaining patients as well. Our data are fully consistent with those recently reported by 2 other studies of mRNA COVID-19 vaccine immunogenicity in patients with primary immunodeficiency.^{9 , 10} In both reports, all patients with CVID but 2 (19 total, 6 and 13 in the 2 studies,^{9 , 10} respectively) were able to mount specific antibody responses after vaccination. Accordingly, patients with absence of B cells (ie, patients with X-linked agammaglobulinemia) failed to have production of neutralizing antibodies against SARS-CoV-2.^{9 , 10} Nevertheless, these patients did produce robust antiviral cellular responses,¹⁰ which still may justify vaccination. Likewise, adverse effects to vaccination were mild in both studies. Taken together, the results of our and these 2 other studies strongly support the notion that patients with CVID must be included in COVID-19 vaccination programs because of the ability of mRNA vaccines to safely induce production of neutralizing antibodies in this category of patients.