Abstract 4113573: Long Noncoding RNAs to Predict COVID-19 Neuro-Cardio-Vascular Outcomes

Introduction/Background: Cardiovascular and neurological diseases develop in a significant proportion of COVID-19 patients. Minimally invasive tools to predict outcome after SARS-CoV-2 infection would enable personalized healthcare, potentially easing the disease burden. We showed that blood levels of the long noncoding RNA lymphoid enhancer-binding factor-1 antisense 1 (LEF1-AS1) predict COVID-19 in-hospital mortality. Hypothesis: LEF1-AS1 is associated with long-term clinical outcomes of COVID-19. Aim: Test the capacity of LEF1-AS1 to predict neuro-cardio-vascular outcomes post-SARS-CoV-2 infection. Methods/Approach: We enrolled 104 primo-infected COVID-19 patients aged 18+ recruited from April to December 2020 in the PrediCOVID national cohort for which 12-month follow-up data were available (Ethics Committee approvals 202003/07 and 202310/02-SU-202003/07). Whole blood samples were collected at baseline and expression levels of LEF1-AS1 were assessed by quantitative PCR. Results/Data: Of the 104 patients, 35 had at least one neurological symptom and one cardiovascular symptom at month 12. Levels of LEF1-AS1 at baseline were lower (p=0.019) in patients who developed neurological and cardiovascular symptoms as compared to patients who did not. Lower LEF1-AS1 was associated with symptoms development with an odds ratio of 0.48 (95% CI 0.28-0.83) from logistic regression model adjusted for age, sex, comorbidities and disease severity at baseline. Addition of LEF1-AS1 to a clinical model including age, sex, comorbidities and baseline severity yielded an incremental predictive value as attested by an increased AUC from 0.79 to 0.83 (likelihood ratio test p=0.005), a net reclassification index of 0.54 (p=0.007) and an integrated discrimination improvement of 0.08 (p=0.009). Conclusion: Blood levels of LEF1-AS1 predict 12-month neurological and cardiovascular outcomes of COVID-19 patients. This needs to be validated in larger populations.