ABSTRACT
We present a genome polymorphisms/machine learning approach for severe COVID-19 prognosis. Ninety-six Brazilian severe COVID-19 patients and controls were genotyped for 296 innate immunity loci. Our model used a feature selection algorithm, namely recursive feature elimination coupled with a support vector machine, to find the optimal loci classification subset, followed by a support vector machine with the linear kernel (SVM-LK) to classify patients into the severe COVID-19 group. The best features that were selected by the SVM-RFE method included 12 SNPs in 12 genes: PD-L1, PD-L2, IL10RA, JAK2, STAT1, IFIT1, IFIH1, DC-SIGNR, IFNB1, IRAK4, IRF1, and IL10. During the COVID-19 prognosis step by SVM-LK, the metrics were: 85% accuracy, 80% sensitivity, and 90% specificity. In comparison, univariate analysis under the 12 selected SNPs showed some highlights for individual variant alleles that represented risk (PD-L1 and IFIT1) or protection (JAK2 and IFIH1). Variant genotypes carrying risk effects were represented by PD-L2 and IFIT1 genes. The proposed complex classification method can be used to identify individuals who are at a high risk of developing severe COVID-19 outcomes even in uninfected conditions, which is a disruptive concept in COVID-19 prognosis. Our results suggest that the genetic context is an important factor in the development of severe COVID-19.
Subject(s)
COVID-19 , Genome, Human , Humans , B7-H1 Antigen , Interferon-Induced Helicase, IFIH1 , Brazil/epidemiology , COVID-19/diagnosis , COVID-19/genetics , Artificial Intelligence , Algorithms , GenomicsABSTRACT
We aimed to determine whether neck circumference predicts mortality among hospitalized COVID-19 patients with respiratory failure. We performed a prospective multicenter (Italy and Brasil) study carried out from March to December 2020 on 440 hospitalized COVID-19 patients with respiratory failure. Baseline neck circumference was measured. The study outcome was 30- and 60-days mortality. Female and male participants were classified as "large neck" when exceeding fourth-quartile. Patients had a median age of 65 years (IQR 54-76), 68% were male. One-quarter of patients presented with grade-1 or higher obesity. The median neck circumference was 40 cm (IQR 38-43): 38 cm (IQR 36-40) for female and 41 cm (IQR 39-44) for male subjects. "Large neck" patients had a significantly higher prevalence of hypertension (63 vs. 48%), diabetes (33 vs. 19%), obesity (26 vs. 14%), and elevated C-reactive protein (CRP) (98 vs. 88%). The cumulative mortality rate was 13.1% (n = 52) and 15.9% (n = 63) at 30 and 60 days, respectively. After adjusting for age, BMI, relevant comorbidities, and high C-reactive protein to albumin ratio, "large neck" patients showed a significantly increased risk of death at 30- (adjusted HR 2.50; 95% CI 1.18-5.29; p = 0.017) and 60-days (adjusted HR 2.26; 95% CI 1.14-4.46; p = 0.019). Neck circumference is easy to collect and provides additional prognostic information to BMI. Among hospitalized COVID-19 patients with respiratory failure, those with large neck phenotype had a more than double risk of death at 30 and 60 days.