ABSTRACT
Background and aim. The SAVE-MORE study showed that the early start of treatment with the IL-1alpha/beta inhibitor anakinra, guided by suPAR (Soluble Urokinase-Type Plasminogen Activator Receptor) >=6ng/mL, in patients with moderate or severe COVID-19, significantly reduced the risk of worse clinical outcome at day 28. With press release 665 of 28/09/2021, AIFA has approved the inclusion of anakinra in the 648/96 list for the treatment of hospitalized adults with COVID-19 and suPAR >=6ng/mL. However, suPAR methods are not widely available, which hinders the prescription and clinical use of anakinra. Aim of this study was to identify a panel of biochemical tests as a surrogate marker of suPAR positivity (>=6ng/mL). Methods. The study included 456 (median (IQR) age: 75 ys (60-83);M: F 54:46%) hospitalized patients in the Infectious Disease Unit (n=124) and Medical ICU (n=332) of the Maggiore Policlinico Hospital of Milan with molecular diagnosis of COVID-19. suPAR was measured at admission by suPARnostic TurbiLatex kit (Vendor: ViroGates A/S, Denmark;Italian distributor: B.S.N. Srl) on Roche Cobas c702. Results. Median suPAR was 7.6ng/mL (4.8-10.8), with 63% of patients displaying suPAR >=6ng/mL. At the univariate logistic regression analysis, suPAR was found to be associated with age (p<0.001), WBC (p=0.002), #NE (p<0.001), Hb (p<0.001), CREA (p<0.001), LDH (p=0.005), FERR (p=0.006), CRP (p<0.001), Fib (p=0.005), DD (p<0.001), but not with sex (p=0.943), #LY (p=0.444), #MO (p=0.233), PLT (p=0.064), ALT (p=0.238), TBIL (p=0.534), TnT (p=0.153) and TSH (p=0.970). However, at the multivariate analysis, only age (p<0.001), Hb (p=0.043), CREA (p<0.001), LDH (p=0.021), CRP (p=0.005) and DD (p=0.004) were found as independent predictors of suPAR positivity. Percentage of correct classification (< vs >= 6ng/mL) and AUC of the multivariate model were 75.1% and 0.83 (95%CI 0.80-0.87). Conclusions. suPAR is independently associated with age, Hb, CREA, LDH, CRP and DD. Due to the moderate % of correct classification of the multivariate model (75%), we conclude that this combination of blood markers cannot be used as a surrogate of suPAR for anakinra prescription. Further clinical validation is needed to assess a possible role of the model in predicting COVID-19 severity and mortality.