Graham, Mary Beth

ImportanceSystemic antibiotic use for patients with a non–Clostridioides difficile infection (CDI) is a major risk factor for recurrent CDI. Increasing use of oral vancomycin for secondary prophylaxis against recurrent CDI in this context has uncertain efficacy.ObjectiveTo evaluate whether oral vancomycin prophylaxis compared with placebo is effective against recurrent CDI during and 8 weeks after the end of study treatment.Design, Setting, and ParticipantsThis phase 2, placebo-controlled, double-blind randomized clinical trial was conducted in 4 large health systems across the upper Midwest US. Adults who had completed treatment for CDI within the past 180 days and were taking a systemic antibiotic for a non-CDI indication were enrolled between May 21, 2018, and March 30, 2023, and followed up for 8 weeks after the end of study treatment.InterventionParticipants were randomized 1:1 to 125 mg of oral vancomycin or placebo once daily during antibiotic use for a non-CDI plus 5 days following cessation of those antibiotics.Main Outcomes and MeasuresThe primary outcome was recurrent CDI incidence during treatment and the 8-week follow-up period. The secondary outcome was vancomycin-resistant Enterococcus carriage in stool.ResultsAmong 81 randomized participants (median age, 59 years [IQR, 50-67 years]), all were included in the primary as-randomized analysis (39 in the vancomycin group; 42 in the placebo group). Sixty patients (74.1%) completed 8-week follow-up and were included in the secondary as–completed treatment analysis (31 in the vancomycin group; 29 in the placebo group). Recurrent CDI occurred in 17 of 39 participants in the oral vancomycin group (43.6%) and 24 of 42 in the placebo group (57.1%; absolute difference in percentage, −13.5% [95% CI, −35.1% to 8.0%]). Adverse events occurred in 27 of 39 participants in the oral vancomycin group (69.2%) and 27 of 42 in the placebo group (64.3%). Vancomycin-resistant Enterococcus carriage was found in 15 of 30 patients in the oral vancomycin group (50.0%) and 6 of 25 in the placebo group (24.0%) (P = .048) 8 weeks after treatment.Conclusions and RelevanceIn this randomized clinical trial, the incidence of recurrent CDI was lower (though did not reach significance) in participants taking oral vancomycin compared with those taking placebo. Because the study was underpowered, it was unable to reveal firm conclusions about the efficacy (or lack thereof) of vancomycin prophylaxis with respect to recurrent CDI.Trial RegistrationClinicalTrials.gov Identifier: NCT03462459

Severe COVID-19 is a multisystem condition with thromboinflammation as a significant driver of morbidity and mortality. Cellular components involved in inflammatory and coagulation responses include platelets, endothelium, monocytes, and neutrophils. Thrombotic manifestations, such as arterial thrombosis, venous thromboembolism, and tissue microthrombosis, occur in at least 20% of ICU COVID-19 patients, possibly more based on postmortem studies. This study investigates the presence of prothrombotic antibodies similar to those found in heparin-induced thrombocytopenia (HIT), a condition caused by platelet-activating antibodies targeting platelet factor 4 (PF4) when complexed with heparin (PF4/H). In a study of samples from 130 hospitalized patients collected on average 3.6 days after COVID-19 diagnosis, 80% were found to have IgG antibodies recognizing PF4/H. This percentage is significantly lower than in 22 HIT patients (100%) but higher than in 51 non-COVID-19 patients with acute respiratory symptoms (ARS) (25%) and 12 healthy subjects (0%). Additionally, 41% of the 130 patients had antibodies inducing PF4-dependent P-selectin expression in CpG-treated normal platelets, a rate significantly lower than in 12 HIT patients (100%) but higher than in 51 non-COVID-19 ARS patients (20%) and 12 healthy subjects (0%). Significantly, platelet-activating antibodies in COVID-19 patients were associated with a specific elevation of platelet α-granule proteins in the plasma and correlated with inflammation and tissue damage markers, suggesting these antibodies' functionality in patients. Significantly, unlike HIT, both PF4/H-reactive and platelet-activating antibodies were found in COVID-19 patients regardless of recent heparin exposure, indicating different antigenic stimuli for these antibodies in COVID-19. Notably, certain antibodies against the receptor-binding domain (RBD) of the spike protein of SARS-CoV-2 possessed sequence motifs, RKH and Y5, in the heavy-chain complementarity-determining region 3 (HCDR3) resembling those in pathogenic HIT antibodies identified in our previous studies, suggesting that these antibodies can activate platelets. Introducing exogenous RBD to COVID-19 plasma or IgG purified from COVID-19 plasma significantly reduced their ability to activate platelets. Furthermore, immunodepleting RBD-reactive antibodies from COVID-19 plasma or IgG purified from COVID-19 plasma using RBD-coated beads significantly reduced their ability to activate platelets. These findings demonstrate that some RBD-specific plasma IgG antibodies can activate platelets. To demonstrate the presence of RBD-specific, platelet-activating antibodies at the clonal level, we produced recombinant antibodies from 42 RBD-binding IgG+ B cells, tested their ability to activate platelets, and identified four that could do so. Interestingly, three antibodies possess HCDR3 sequences resembling RKH or Y5 motifs. Additionally, we identified RKH or Y5 motifs in RBD-specific antibodies reported by other groups, produced 15 such antibodies recombinantly, tested their ability to activate platelets, and identified four more that could do so. IgG+ B cells with these HCDR3 motifs were markedly expanded in severe COVID-19 patients, suggesting an expansion of B cells producing platelet-activating antibodies. The demonstration of functional and structural similarities between certain RBD-specific antibodies in COVID-19 patients and pathogenic HIT antibodies suggests a novel mechanism by which RBD-specific antibodies might contribute to thrombosis in COVID-19.

Abstract Thromboembolic complication is common in severe coronavirus disease 2019 (COVID-19), leading to an investigation into the presence of prothrombotic antibodies akin to those found in heparin-induced thrombocytopenia (HIT). In a study of samples from 130 hospitalized patients, collected 3.6 days after COVID-19 diagnosis, 80% had immunoglobulin G (IgG) antibodies recognizing complexes of heparin and platelet factor 4 (PF4; PF4/H), and 41% had antibodies inducing PF4-dependent P-selectin expression in CpG oligodeoxynucleotide–treated normal platelets. Unlike HIT, both PF4/H-reactive and platelet-activating antibodies were found in patients with COVID-19 regardless of recent heparin exposure. Notably, PF4/H-reactive IgG antibodies correlated with those targeting the receptor-binding domain (RBD) of the severe acute respiratory syndrome coronavirus 2 spike protein. Moreover, introducing exogenous RBD to or removing RBD-reactive IgG from COVID-19 plasma or IgG purified from COVID-19 plasma significantly reduced their ability to activate platelets. RBD-specific antibodies capable of platelet activation were cloned from peripheral blood B cells of patients with COVID-19. These antibodies possessed sequence motifs in the heavy-chain complementarity-determining region 3 (HCDR3), resembling those identified in pathogenic HIT antibodies. Furthermore, IgG+ B cells having these HCDR3 signatures were markedly expanded in patients with severe COVID-19. Importantly, platelet-activating antibodies present in patients with COVID-19 were associated with a specific elevation of platelet α-granule proteins in the plasma and showed a positive correlation with markers for inflammation and tissue damage, suggesting a functionality of these antibodies in patients. The demonstration of functional and structural similarities between certain RBD-specific antibodies in patients with COVID-19 and pathogenic antibodies typical of HIT suggests a novel mechanism by which RBD-specific antibodies might contribute to thrombosis in COVID-19.