BCG-Induced DNA Methylation Changes Improve Coronavirus Disease 2019 Vaccine Immunity Without Decreasing the Risk for Severe Acute Respiratory Syndrome Coronavirus 2 Infection

Abstract

Background: The BCG vaccine induces trained immunity, an epigenetic-mediated increase in innate immune responsiveness. Therefore, this clinical trial evaluated if BCG-induced trained immunity could decrease coronavirus disease 2019 (COVID-19)-related frequency or severity.

Methods: A double-blind, placebo-controlled clinical trial of healthcare workers randomized participants to vaccination with BCG TICE or placebo (saline). Enrollment included 529 healthcare workers randomized to receive BCG or placebo. Primary analysis evaluated COVID-19 disease frequency, while secondary analysis evaluated coronavirus immunity in a subset of participants. Study enrollment ceased early in December 2020 following introduction of COVID-19-specific vaccines.

Results: Study enrollment was halted early, prior to reaching the targeted recruitment, and was not powered to detect a decrease in COVID-19 frequency. Symptomatic COVID-19 occurred in 21 of 263 and 10 of 266 participants in the BCG and placebo arms, respectively (P = .50, Fisher exact test). Participants vaccinated with BCG, but uninfected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), demonstrated increased coronavirus vaccine immunity (increase spike-inducible levels of tumor necrosis factor, interleukin 6, and interleukin 1β) 12 months after BCG vaccination compared to participants receiving placebo. Immune responsiveness to SARS-CoV-2 antigens correlated with BCG-induced DNA methylation changes.

Conclusions: Due to early study closure, the study was not powered to evaluate COVID-19 frequency. Secondary analysis demonstrated that 12 months following vaccination, BCG increased coronavirus vaccine immunity compared to those who did not receive BCG. This increase in COVID-19 vaccine immunity correlated with BCG-induced DNA methylation changes.

Keywords: BCG vaccine; COVID-19; DNA methylation; epigenetics; innate training.

Figure 1.

Consolidated Standards of Reporting Trials (CONSORT) diagram of the clinical study design and analyses. ¹Of 2023 healthcare workers who applied online, 1300 were eligible, and 529 were enrolled. Participants included those working in hospitals or medical centers/clinics, as well as first responders (paramedics, firefighters, law enforcement). ²People with human immunodeficiency virus, known immunodeficiency, neutropenia, or participants on immunotherapy, oral or intravenous steroids, or immunosuppressants. ³Placebo: 0.9% sodium chloride solution, 0.1 mL, 1-time administration intradermally in the deltoid area. ⁴BCG-vaccinated group: TICE BCG live strain (Merck), US Food and Drug Administration approved, 0.1 mL (∼2 × 10⁵ colony-forming units), 1-time administration intradermally in the deltoid area. ⁵Participants were followed up to assess for symptoms and adverse events. Optional blood donations were collected for the isolation of peripheral blood mononuclear cells and for serological analysis of severe acute respiratory syndrome coronavirus 2 antibodies using dried blood spot samples. Additionally, participants’ messenger RNA vaccine history was documented. Abbreviations: CRP, C-reactive protein; ELISA, enzyme-linked immunosorbent assay.

Figure 2.

BCG augments severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike immunity. Peripheral blood mononuclear cells (PBMCs) were collected 12 months after BCG vaccination (n = 31; red) or placebo vaccination (n = 30; light blue). The cells were frozen immediately after collection and later thawed for batch immune evaluation. PBMCs were left unstimulated or stimulated overnight with overlapping peptide pools of the spike protein found specifically in SARS-CoV-2 with the supernatant evaluated by multiplex enzyme-linked immunosorbent assay (LegendPlex, BioLegend). Responses are expressed as log₂ fold change of stimulated over unstimulated with the center line representing the median. Estimation effect measured by Mann-Whitney test comparing cytokine response in both groups with Holm-Sidak correction. Abbreviations: IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; MCP-1, Monocyte Chemotactic protein 1; PBMCs, peripheral blood mononuclear cells; TNF, tumor necrosis factor.

Figure 3.

BCG augments coronavirus nucleocapsid immunity. Peripheral blood mononuclear cells (PBMCs) were collected 12 months after BCG (n = 31) or placebo (n = 30) vaccination, frozen, and later thawed for batch immune evaluation. PBMCs were left unstimulated or stimulated overnight with overlapping peptide pools of the nucleocapsid found in severe acute respiratory syndrome coronavirus 2 and other coronaviruses with the supernatant evaluated by multiplex enzyme-linked immunosorbent assay (LegendPlex, BioLegend). Responses are expressed as log₂ fold change of stimulated over unstimulated with the center line representing the median. Estimation effect measured by Mann-Whitney test comparing cytokine response in both groups with Holm-Sidak correction. Abbreviations: IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; MCP-1, Monocyte Chemotactic protein 1; PBMCs, peripheral blood mononuclear cells; TNF, tumor necrosis factor.

Figure 4.

BCG-induced DNA methylation changes. A, 12 months after BCG or placebo vaccination, 32 participants donated additional blood for DNA methylation. B, Location of BCG-induced hyper- and hypomethylation. C, Volcano plot demonstrating select hyper- and hypomethylated genes 12 months after BCG vaccination (a complete list of differential methylated genes is shown in Supplementary Table 2). D, Overrepresentation analysis depicting the enriched hyper- and hypomethylated hallmark gene pathways. E, Select pathways and genes hypomethylated 12 months after BCG vaccination (a complete list of all enriched pathways is shown in Supplementary Table 2). Abbreviations: CpG, The majority of DNA methylation occurs on cytosines that precede a guanine nucleotide; IFN, interferon; IL, interleukin; mTOR, mammalian target of rapamycin; ROS, reactive oxygen species; TNF, tumor necrosis factor; UV, ultraviolet.

Figure 5.

BCG-induced DNA methylation changes correlate with coronavirus disease 2019 and lipopolysaccharide-induced immunity. A, The DNA methylation summed z-score for each pathway, with center line representing the median z-score: estimation effect measured by 2-way analysis of variance comparing placebo vs BCG. B, Heatmap showing Pearson correlations between the DNA methylation summed z-scores for selected pathways (columns) and host immunity markers (rows) induced by lipopolysaccharide, spike, or nucleocapsid peptides. Only significant correlations are displayed (P < .05). Red indicates positive correlations, blue indicates negative correlations, and white represents nonsignificant correlations. Abbreviations: BMI, body mass index; IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; MCP-1, Monocyte chemtotactic protein 1; mTOR, mammalian target of rapamycin; N, nucleocapsid; S, spike; TNF, tumor necrosis factor.