The immune response to Covid-19 mRNA vaccination among Lymphoma patients receiving anti-CD20 treatment

Abstract

The monoclonal antibody rituximab improves clinical outcome in the treatment of CD20-positive lymphomatous neoplasms, and it is an established drug for treatment of these cancers. Successful mRNA COVID-19 (SARS-CoV-2) vaccination is extremely important for lymphoma patients because they tend to be elderly with comorbidities which leaves them at increased risk of poor outcomes once infected by Coronavirus. Anti-CD20 therapies such as rituximab, deplete B-cell populations and can affect vaccine efficacy. Therefore, a knowledge of the effect of COVID-19 vaccination in this group is critical. We followed a cohort of 28 patients with CD20-positive lymphomatous malignancies treated with rituximab that started prior to their course of COVID-19 vaccination, including boosters. We assayed for vaccine "take" in the humoral (IgG and IgA) and cellular compartment. Here, we show that short-term and long-term development of IgG and IgA antibodies directed toward COVID-19 spike protein are reduced in these patients compared to healthy controls. Conversely, the robustness and breath of underlying T-cell response is equal to healthy controls. This response is not limited to specific parts of the spike protein but spans the spike region, including response to the conserved Receptor Binding Domain (RBD). Our data informs on rational vaccine design and bodes well for future vaccination strategies that require strong induction of T-cell responses in these patients.

Keywords: CD20; COVID - 19; Tcell; lymphoma; mRNA; rituximab; vaccination.

Figure 1

Serum immunoglobulin responses among patients with CD20-positive lymphoid malignancies receiving rituximab post mRNA vaccination for SARS-CoV-2. Serum samples were collected from 28 patients. Each dot represents a given patient sample and lines link patient samples over time. The horizontal dotted line represents the threshold at which an S1 spike specific antibody can be reported as positive by the Covid-19 immunoglobulin assay. The concentration of IgG and IgA S1 specific antibodies elicited post vaccination is similar both between males and females (not shown) and between Pfizer and Moderna vaccinees. The bottom panels delineate out the eight patients taking rituximab during (green) the study and those that stopped Rituximab before vaccination. No difference was observed. The Diamonds represent the Antibody response measured from the controls.

Figure 2

T cell responses to SARS-CoV-2 spike protein post mRNA vaccination among 28 patients with CD20-positive neoplasms receiving rituximab. PBMC were isolated from whole blood samples. Each dot represents a given patient sample and lines link patient samples over time. Total ELISPOT-Interferon gamma responses to the entire spike immunogen based on overlapping peptides that cover SARS-CoV-2 spike protein (Wuhan strain) in the Pfizer (blue) and Moderna (red) mRNA vaccines are illustrated. No difference between the two vaccines was observed. The Diamonds (black) represent healthy control responses and the dotted line represents that average of the healthy control T cell interferon gamma response.

Figure 3

Immune responses to mRNA vaccination among patients with CD20-positive malignancy taking B cell depleting therapy (rituximab). Twenty-eight patients and five healthy controls underwent blood draws at approximately 3, 6- and 9-months post mRNA vaccination (standard dose and regimen Pfizer or Moderna; the healthy controls only had one blood draw). Top: Cartoon illustrating SARS-CoV-2 spike protein segments S1 and S2. The spike protein includes Signal sequence (SS), N-terminal Domain (NTD), Subdomain 1 and 2 (SD1, SD2), Fusion Peptide (FP), Heptad repeats (HR1 and HR2), Central Helix (CH), Connector Domain (CD), and Transmembrane Domain (TM) and cytoplasmic tail (CT). Ten pools of peptides (13-17mer 10aa overlap) were used to map T cell responses to the mRNA vaccines. Bottom: A comparison of total peak ELISPOT -IFNγ responses from all patients and healthy controls to the nine pools of peptide covering the vaccine immunogen. Note that all nine pools provide equal coverage. There is variation in the strength of the response but importantly there are individual preferences as to what regions of spike protein the T cells are targeting. Right: PBMC isolated from patients’ blood drawn at peak response post vaccination. ELISPOT responses to Receptor Binding Domain RBD and SARS-CoV-2 (Wuhan) spike protein (BEI resources and Invivogen) showed no significant difference with healthy control T cell responses. A log2(x+1) transformation was applied to the breadth (total of the 9 pools) to reduce skewness. We fitted data using a mixed effects model with group (patient or control) as a fixed effect and patient ID as random effect. The estimated fixed effect of the group is beta=-1.955 ± 1.488 (p=0.1930). Data are dichotomized by ≥40 or <40. We use two methods to compare each of these nine pools between cases (patients) and controls: (1) Fisher’s exact test and (2) logistic regression that adjusts for gender.