Fourth mRNA COVID-19 vaccination in immunocompromised patients with haematological malignancies (COBRA KAI): a cohort study

Abstract

Background: Patients with haematological malignancies have impaired antibody responses to SARS-CoV-2 vaccination. We aimed to investigate whether a fourth mRNA COVID-19 vaccination improved antibody quantity and quality.

Methods: In this cohort study, conducted at 5 sites in the Netherlands, we compared antibody concentrations 28 days after 4 mRNA vaccinations (3-dose primary series plus 1 booster vaccination) in SARS-CoV-2 naive, immunocompromised patients with haematological malignancies to those obtained by age-matched, healthy individuals who had received the standard primary 2-dose mRNA vaccination schedule followed by a first booster mRNA vaccination. Prior to and 4 weeks after each vaccination, peripheral blood samples and data on demographic parameters and medical history were collected. Concentrations of antibodies that bind spike 1 (S1) and nucleocapsid (N) protein of SARS-CoV-2 were quantified in binding antibody units (BAU) per mL according to the WHO International Standard for COVID-19 serological tests. Seroconversion was defined as an S1 IgG concentration >10 BAU/mL and a previous SARS-CoV-2 infection as N IgG >14.3 BAU/mL. Antibody neutralising activity was tested using lentiviral-based pseudoviruses expressing spike protein of SARS-CoV-2 wild-type (D614G), Omicron BA.1, and Omicron BA.4/5 variants. This study is registered with EudraCT, number 2021-001072-41.

Findings: Between March 24, 2021 and May 4, 2021, 723 patients with haematological diseases were enrolled, of which 414 fulfilled the inclusion criteria for the current analysis. Although S1 IgG concentrations in patients significantly improved after the fourth dose, they remained significantly lower compared to those obtained by 58 age-matched healthy individuals after their first booster (third) vaccination. The rise in neutralising antibody concentration was most prominent in patients with a recovering B cell compartment, although potent responses were also observed in patients with persistent immunodeficiencies. 19% of patients never seroconverted, despite 4 vaccinations. Patients who received their first 2 vaccinations when they were B cell depleted and the third and fourth vaccination during B cell recovery demonstrated similar antibody induction dynamics as patients with normal B cell numbers during the first 2 vaccinations. However, the neutralising capacity of these antibodies was significantly better than that of patients with normal B cell numbers after two vaccinations.

Interpretation: A fourth mRNA COVID-19 vaccination improved S1 IgG concentrations in the majority of patients with a haematological malignancy. Vaccination during B cell depletion may pave the way for better quality of antibody responses after B cell reconstitution.

Funding: The Netherlands Organisation for Health Research and Development and Amsterdam UMC.

Keywords: Antibody response; Booster vaccination; COVID-19 vaccination; Haematological malignancies; Immunocompromised; SARS-CoV-2.

Fig. 1

Study participants and vaccination schedules. 723 patients were included in this cohort study, of whom 414 met the criteria for fourth vaccination response analyses. Timing of vaccinations was according to the COVID-19 vaccination policy for immunocompromised patients governed by the Dutch Minister of Health (wks: weeks; mo: months). All participants received the primary 3-dose mRNA-1273 vaccination series, followed by a booster (fourth) mRNA vaccination (BNT162b2 in 93% of patients and mRNA-1273 in 7% of participants). Characteristics of patients excluded from analyses are described in Supplementary Table S2. Healthy, age-matched, SARS-CoV-2-unexposed individuals who received a primary 2-dose mRNA-1273 vaccination schedule followed by 1 BNT162b booster vaccination served as a control cohort. ^#Partially overlapping with other exclusion criteria.

Fig. 2

Antibody concentration and virus neutralisation. A. S1 IgG antibody concentration 4 weeks after each vaccination (second, third and fourth) in binding antibody units (BAU) per millilitre, where median and IQR are indicated by black lines. Grey bar indicates IQR of S1 IgG concentration (3219–6820 BAU/mL) in 3-dose vaccinated healthy individuals. B. Correlation between S1 IgG antibody concentration and pseudovirus neutralisation (ID₅₀) of SARS-CoV-2 wild-type and Omicron variants after the third (light blue) and fourth (dark blue) vaccination. C. Antibody maturity, defined as SARS-CoV-2 wild-type and Omicron variant neutralising capacity per antibody after third (light blue) and fourth (dark blue) vaccination. Black lines indicate median values. Ns: Not significant, p > 0.05.

Fig. 3

Heterogeneity between cohorts. A. S1 IgG concentrations 4 weeks after each vaccination (second, third and fourth) in binding antibody units (BAU) per millilitre, for each patient cohort. Orange lines indicate median values. Grey bar indicates IQR (3225–8939 BAU/mL) in 3-dose vaccinated, age-matched, healthy individuals (Fig. 2A). Dotted line indicates threshold for seroconversion (10 BAU/mL). Significances are shown for differences in antibody concentration between second and third vaccination response, and between third and fourth vaccination response. For statistical significance of difference between antibody concentrations after a fourth dose in patients and third dose in healthy individuals, see Table 1. Mo: months; BEAM: Carmustine, etoposide, cytarabine, melphalan; HCT: Haematopoietic cell transplantation; CAR: Chimeric antigen receptor; CLL: Chronic lymphocytic leukaemia; cGvHD: Chronic graft-versus-host disease; IMiD: Immunomodulatory imide drug; HDM: High-dose melphalan; AML: Acute myeloid leukaemia; MDS: Myelodysplastic syndrome; HMA: Hypomethylating agents; MPN: Myeloproliferative neoplasm; CML: Chronic myeloid leukaemia; TKI: Tyrosine kinase inhibitor; ns: p > 0.05; ∗: p ≤ 0.05; ∗∗: p ≤ 0.005; ∗∗∗: p ≤ 0.001. B. S1 IgG concentrations 4 weeks after each vaccination in patients who received cell therapy between the second and third vaccination (grey line), between third and fourth vaccination (black line), or who received a tandem transplantation (one after the second vaccination and one after the third vaccination; dashed line). HDM-autologous HCT: subgroup of multiple myeloma patients who received induction therapy at the time of the first COVID-19 vaccination. Grey bar indicates IQR of S1 IgG concentration (3219–6820 BAU/mL) in 3-dose vaccinated healthy individuals (Fig. 2A). Dotted line indicates threshold for seroconversion (10 BAU/mL). Ns: not significant, p > 0.05.

Fig. 4

Priming effect of vaccinations during B cell depletion. A. S1 IgG concentration after each vaccination for patients with normal absolute B cell numbers (100–500 cells/μL) at the day of the first vaccination (‘never B cell depleted’; n = 119 (Supplementary Table S4)), and for patients who were B cell depleted (0 cells/μL) at the time of the primary 2-dose vaccination and had circulating B cells (≥1 cell/μL) from the third vaccination onwards (‘B cell reconstituting’; n = 22 (Supplementary Table S4)). B. Capacity per antibody to neutralise Omicron BA.1 (ratio of ID₅₀ to the binding S1 IgG concentration) in never B cell depleted and B cell reconstituting patients. N/A: Not applicable.