COVID-19 in Patients Receiving CD20-depleting Immunochemotherapy for B-cell Lymphoma

Abstract

The clinical and immunological impact of B-cell depletion in the context of coronavirus disease 2019 (COVID-19) is unclear. We conducted a prospectively planned analysis of COVID-19 in patients who received B-cell depleting anti-CD20 antibodies and chemotherapy for B-cell lymphomas. The control cohort consisted of age- and sex-matched patients without lymphoma who were hospitalized because of COVID-19. We performed detailed clinical analyses, in-depth cellular and molecular immune profiling, and comprehensive virological studies in 12 patients with available biospecimens. B-cell depleted lymphoma patients had more severe and protracted clinical course (median hospitalization 88 versus 17 d). All patients actively receiving immunochemotherapy (n = 5) required ICU support including long-term mechanical ventilation. Neutrophil recovery following granulocyte colony stimulating factor stimulation coincided with hyperinflammation and clinical deterioration in 4 of the 5 patients. Immune cell profiling and gene expression analysis of peripheral blood mononuclear cells revealed early activation of monocytes/macrophages, neutrophils, and the complement system in B-cell depleted lymphoma patients, with subsequent exacerbation of the inflammatory response and dysfunctional interferon signaling at the time of clinical deterioration of COVID-19. Longitudinal immune cell profiling and functional in vitro assays showed SARS-CoV-2-specific CD8⁺ and CD4⁺ T-effector cell responses. Finally, we observed long-term detection of SARS-CoV-2 in respiratory specimens (median 84 versus 12 d) and an inability to mount lasting SARS-CoV-2 antibody responses in B-cell depleted lymphoma patients. In summary, we identified clinically relevant particularities of COVID-19 in lymphoma patients receiving B-cell depleting immunochemotherapies.

Figure 1.

Severe clinical course of COVID-19 in B-cell depleted lymphoma patients. (A), Schematic overview of clinical course of COVID-19 in B-cell depleted lymphoma (“patients”, top panel) and COVID-19 patients non-B-cell depleted patients without lymphoma (“controls”, bottom panel). *Discharge from hospital at day 140. †Deceased at day 155. (B), Comparison of median durations of hospital stay, ICU stay and mechanical ventilation.

Figure 2.

Hyperinflammation and neutrophil recovery at time of clinical deterioration in B-cell depleted lymphoma patients. Serum levels of (A) C-reactive protein and (B) interleukin-6 (IL-6), (C) blood leukocytes, (D) absolute neutrophil count (ANC), and (E) lymphocyte counts before and at the time of clinical deterioration. (F), Neutrophil-to-lymphocyte ratio (NLR) before and at the time of clinical deterioration.

Figure 3.

B-cell depletion, T-cell and monocyte response in B-cell depleted lymphoma patients and controls. (A), Schematic overview of sampling and timeline. Red mark indicates ICU admission; yellow pins indicate time of PBMC collection. (B), Percentage of B-cells in patients and controls (all time points). (C), Percentage of lymphocytes in PBMCs as indicated in (A). (D), Percentage of lymphocytes and CD8⁺ T-cells in PBMCs as indicated in (A). (E), CD8⁺ T-cell subsets in PBMCs as indicated in (A). (F), Percentage of lymphocytes and CD4⁺ T-cells in PBMCs as indicated in (A). (G), CD4⁺ T-cell subsets as indicated in (A). (H), Percentage of IFNγ-positive CD8⁺ T-cells after ex vivo stimulation with SARS-CoV-2 spike protein. Shown are results of patients with available serial blood samples (n > 3) with interpolation line (black) and 95% confidence interval (gray). (I), Exemplary FACS dot plot for IFNγ-positive CD8⁺ T-cells from patient 3 at indicated time points. (J), Percentage of IFNγ-positive CD4⁺ T-cells after ex vivo stimulation with SARS-CoV-2 spike protein. Shown are results of patients with available serial blood samples (n > 3) with interpolation line (black) and 95% confidence interval (gray). (K), Exemplary FACS dot plot for IFNγ-positive CD4⁺ T-cells from patient 3 at indicated time points. (L), Percentage of monocytes in PBMCs as indicated in (A). (M), Monocyte subsets in PBMCs as indicated in (A). (N), Exemplary FACS contour plot showing monocyte subsets from patient 3 at indicated time points. C = classical; CM = central memory; D = deteriorating; E = early; EM = effector memory; INT = intermediate; L = late; N = naive; NC = non-classical; TEMRA = T-effector memory cells re-expressing CD45RA.

Figure 4.

Dysregulated immune response in B-cell depleted lymphoma patients. (A), Schematic overview of sampling and timeline for analyses shown in (B) thorough (D). Red mark indicates ICU admission; yellow pin indicates time of PBMC collection. (B), Volcano plot for differential gene expression in patients vs controls (early time points). Genes depicted in red have a P value ≤ 0.05 and a log2 ratio higher than 1; genes depicted in blue have a P value ≤ 0.05 and a log2 ratio lower than −1. (C), Direct global significant score analysis for each annotated gene set in patients vs controls (at early time points). (D), Heatmap of differentially expressed cytokines and chemokines in patients and controls (at early time points). (E), Schematic overview of longitudinal sampling in patients 1, 2, and 3 for analyses shown in subfigures (F) though (K). Inferred abundance of (F) macrophages and (G) CD8⁺ T-cells (early versus time of clinical deterioration). Changes in the gene expression (log2 ratio) of (I) IL10, (J) IL1A, and (K) CSF1 over time. (H), GSEA for the hallmark signature “inflammatory response.”

Figure 5.

Delayed viral clearance and impaired SARS-CoV-2 antibody response in B-cell depleted lymphoma patients. (A), Comparison of median duration of detectable SARS-CoV-2 virus in B-cell depleted lymphoma patients and non-B-cell depleted patients. SARS-CoV-2 viral load in respiratory specimens from nasopharyngeal swabs and tracheal aspiration (black triangle, dots, and lines, respectively), as well as detection of virus in blood serum (black squares) and SARS-CoV-2-specific IgG levels in blood serum in (B) B-cell depleted lymphoma patients, and (C) non-B-cell depleted patients without lymphoma. Viral load is indicated as log10 values of viral RNA copies per milliliter over time per days. Dashed lines indicate the limit of detection for viral load (upper panel) and positivity threshold for IgG levels (lower panel), respectively.