RAGE has potential pathogenetic and prognostic value in nonintubated hospitalized patients with COVID-19

Abstract

BackgroundThe value of the soluble receptor for advanced glycation end-products (sRAGE) as a biomarker in COVID-19 is not well understood. We tested the association between plasma sRAGE and illness severity, viral burden, and clinical outcomes in hospitalized patients with COVID-19 who were not mechanically ventilated.MethodsBaseline sRAGE was measured among participants enrolled in the ACTIV-3/TICO trial of bamlanivimab for hospitalized patients with COVID-19. Spearman's rank correlation was used to assess the relationship between sRAGE and other plasma biomarkers, including viral nucleocapsid antigen. Fine-Gray models adjusted for baseline supplemental oxygen requirement, antigen level, positive endogenous anti-nucleocapsid antibody response, sex, age, BMI, diabetes mellitus, renal impairment, corticosteroid treatment, and log2-transformed IL-6 level were used to assess the association between baseline sRAGE and time to sustained recovery. Cox regression adjusted for the same factors was used to assess the association between sRAGE and mortality.ResultsAmong 277 participants, baseline sRAGE was strongly correlated with viral plasma antigen concentration (ρ = 0.57). There was a weaker correlation between sRAGE and biomarkers of systemic inflammation, such as IL-6 (ρ = 0.36) and CRP (ρ = 0.20). Participants with plasma sRAGE in the highest quartile had a significantly lower rate of sustained recovery (adjusted recovery rate ratio, 0.64 [95% CI, 0.43-0.90]) and a higher unadjusted risk of death (HR, 4.70 [95% CI, 2.01-10.99]) compared with participants in the lower quartiles.ConclusionElevated plasma sRAGE in hospitalized, nonventilated patients with COVID-19 was an indicator of both clinical illness severity and plasma viral load. Plasma sRAGE in the highest quartile was associated with a lower likelihood of sustained recovery and higher unadjusted risk of death. These findings, which we believe to be novel, indicate that plasma sRAGE may be a promising biomarker for COVID-19 prognostication and clinical trial enrichment.Trial RegistrationClinicalTrials.gov NCT04501978.FundingNIH (5T32GM008440-24, 18X107CF6, HHSN261201500003I, R35HL140026, and OT2HL156812).

Keywords: Antigen; COVID-19; Clinical practice; Cytokines; Pulmonology.

Conflict of interest statement

Conflict of interest: LS declares support from NIH grant 18X107CF6 through a contract with Leidos Biomedical. JDN declares research support from the NIH and NIAID through a contract with Leidos Biomedical. RLD declares support from NIH grant HHSN261201500003I through a contract with Leidos Biomedical. BTT declares personal fees from Bayer, Novartis Pharmaceuticals, and Genentech and participation in data safety monitoring for the NIH, Canadian Institutes of Health Research, the Department of Defense, and the University of Toronto. MAM declares research grants from Genentech-Roche and Quantum Leap Healthcare collaborative and personal fees from Citius Pharmaceuticals, Novartis Pharmaceuticals, Johnson & Johnson Pharmaceuticals, and Pliant Therapeutics.

Figures

Figure 1. Study design.

Figure 2. Log2-transformed plasma sRAGE concentration across levels of baseline oxygen requirement.

Horizontal lines represent medians and boxes represent upper and lower quartiles. The bottom whiskers represent the lowest values within 1.5 IQR of the lower quartiles, and the top whiskers represent the highest values within 1.5 IQR of upper quartiles.

Figure 3. Cumulative incidence of sustained recovery and death stratified by sRAGE levels of equal to or more than 6800 pg/mL versus those less than 6800 pg/mL.

Cumulative incidence of sustained recovery is represented by dashed lines, and death is represented by solid lines. The P value for sustained recovery is from Gray’s test. The P value for mortality is from the log-rank test from unadjusted Cox proportional hazards model.