Title: Humoral response and safety of the <scp>BNT162b2</scp> and <scp>mRNA</scp>‐1273 <scp>COVID</scp>‐19 vaccines in patients with haematological diseases treated with <scp>anti‐CD20</scp> antibodies: An observational study
Abstract: In clinical trials, mRNA vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have shown remarkable efficacy and safety in healthy participants.1, 2 However, patients receiving anti-CD20 antibodies, such as rituximab or obinutuzumab (R/Obi), have been shown to have a poor humoral response to mRNA vaccines, although the response improves gradually if more time has elapsed.3, 4 R/Obi exerts an immunosuppressive effect through B-cell depletion, which could explain the impaired response to mRNA vaccines. The factors related to vaccine ineffectiveness in patients receiving R/Obi have not been fully investigated. Therefore, this study aimed to assess the effectiveness and safety of COVID-19 mRNA vaccines in patients who had received R/Obi, and the factors associated with an impaired humoral response. Patients and healthy volunteers were recruited in June and July 2021. The Institutional Review Board of Kobe City Medical Center General Hospital approved the study. Informed consent was obtained from all participants. Blood samples were collected 14–90 days after the second dose of vaccine. The primary end-point was the proportion of participants who acquired IgG antibodies to the receptor-binding domain in the S1 subunit of the SARS-CoV-2 spike protein (anti-S1 IgG antibodies). We evaluated humoral responses in participants who received R/Obi within less than 6, 6–9, 9–12, or more than 12 months prior to vaccination.4 To identify factors contributing to higher anti-S1 antibody titres, we investigated lymphocyte subsets including the CD19-positive B-cell fraction. The antibody response in patients whose B-cell fractions of total lymphocytes were 0%, 0%–3% or more than 3% was assessed to estimate the minimal proportion of B cells required to induce a humoral response following B-cell reconstitution.5 The inclusion and exclusion criteria, details of the data, kit used for antibody detection, and statistical issues are described in the Supporting Information. A total of 151 patients with haematological diseases who had received R/Obi, and 38 healthy volunteers were included in the study. Three patients were excluded from the final analysis (Table 1). A positive antibody response to the vaccine was observed in 93/148 patients (64%) and in all 38 healthy controls. The median anti-S1 IgG titre was 98 [interquartile range (IQR) 0–576] and 499 (IQR 311–826) BAU/ml in the patients and controls, respectively. (Figure 1). Among the seropositive patients, 40 (43%) were weak responders (anti-S1 IgG titres < 260 BAU/ml). Based on the results of subgroup analysis (Figures S1 and S2), we set the optimal cut-off at nine months from the last administration of R/Obi and 3% B cells. According to this cut-off, 91.7% of patients receiving R/Obi earlier than in the previous nine months achieved seropositivity, whereas only 11.5% of patients receiving R/Obi within nine months did so. Similarly, 90 patients (89.8%) with more than 3% B cells achieved a seropositive response, while 12% of those with 3% or fewer B cells were seropositive. The combined humoral response to time and B-cell fraction is shown in Figure 1. Of note, patients receiving the last R/Obi within nine months did not exhibit a humoral response even with more than 3% B cells and, conversely, those receiving R/Obi more than nine months earlier but still in a B-cell-depleted state displayed a poor response (Figure 1). Hence, both time from the last R/Obi treatment to vaccination and B-cell reconstitution were important for a humoral response. Details of adverse events and other subgroup analyses are available in the supplementary material (Figures S3 and S4). During the observation period, four cases of symptomatic COVID-19 (one severe and two mild cases in weak and non-responders, and one mild case in a responder) were recorded. Some guidelines and several prospective studies recommend providing vaccination no earlier than six months after the last administration of B-cell-depleting therapy due to expected impaired humoral response.6 The underlying causes of poor humoral response to vaccination in patients with B-cell malignancies are attributed to both disease-related immune dysregulation and therapy-related immunosuppression.7 In patients treated with R/Obi, the depletion of B cells, which can last for six months, plays an essential role because B cells drive serologic responses.5, 7 B-cell reconstitution resumes gradually after approximately 6–9 months, in parallel with the onset of seroconversion.5 Our study identified a threshold of 3% B cells as a predictor of the humoral response. Two findings should be noted. First, patients failed to show a humoral response if the last R/Obi treatment was within the previous nine months, despite detectable B cells. Second, patients who received their last R/Obi more than nine months earlier, failed to show a humoral response if they did not possess sufficient B cells. Therefore, to accurately estimate vaccine effectiveness in patients receiving R/Obi, a combination of time from the last R/Obi administration and peripheral B-cell fraction (>3%) should be considered. Administration of a third dose of mRNA vaccine to immunodeficient patients is being promoted in developed countries. However, a single-centre prospective study revealed a low seroconversion rate of only 18.2% in non-responders who had been exposed to rituximab and received a third dose.8 In this study, the median interval between the second and third vaccinations was 69 days, which is too short for adequate B-cell recovery.8 Therefore, allowing a period of six or nine months before the third dose while implementing strict protective measures, such as wearing masks and social distancing, may benefit the humoral response for non-responders and weak responders who received R/Obi. The use of pre-exposure prophylaxis to the monoclonal tixagevimab–cilgavimab antibody combination was also promising in this group.9 This study has several limitations. First, the number of patients was relatively small and regimens used in combination with R/Obi were heterogeneous. Therefore, the potential impact of medications other than R/Obi on the humoral response should be considered. Second, although we excluded participants with reported COVID-19, some participants may have had asymptomatic infections, which might have interfered with the results.10 Third, we did not assess the cellular immune response. In summary, we confirmed the safety and the lower effectiveness of mRNA vaccines among patients who received anti-CD20 therapy. A combination of time from the last R/Obi treatment and peripheral B-cell fraction (>3%) can help predict a humoral response to mRNA vaccines. The authors would like to thank the nurses at the Haematology Department and laboratory staff at the Clinical Laboratory Department and LSI Medience Laboratory of Kobe City Medical Center General Hospital. The authors have no financial or proprietary interest in any material discussed in this article. Masashi Nishikubo and Yoshimitsu Shimomura initiated the trial, designed the study, and analysed the data. Masashi Nishikubo wrote the manuscript and Yoshimitsu Shimomura supervised the statistical analysis and manuscript writing, editing, and review. Hayato Maruoka, Seiko Nasu, Tomomi Nishioka and Kenji Sakizono performed, managed, and reported the measurements of anti-SARS-CoV-2 antibodies and other related laboratory data. Naoki Okada, Daishi Nakagawa, Kimimori Kamijo, Hiroharu Imoto, Ryusuke Yamamoto, Yuya Nagai, Nobuhiro Hiramoto, Satoshi Yoshioka, Noboru Yonetani, Akiko Matsushita and Takayuki Ishikawa collected data and provided patient information. Yoshimitsu Shimomura and Takayuki Ishikawa coordinated the project and edited the manuscript. Chisato Miyakoshi provided comments on statistical analysis. Asako Doi provided comments on epidemiology and infection control. All authors have read and approved the final manuscript. This work was supported by a grant from the “Kasahara Cancer Research Fund”, Japan. This study was registered with the University Hospital Medical Information Network (UMIN) in Japan (UMIN000046043). Written informed consent was obtained from all study participants. Figures S1–S4 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.