Predicting COVID-19 Severity Integrating RNA-Seq Data Using Machine Learning Techniques

A fundamental challenge in the fight against COVID-19 is the development of reliable and accurate tools to predict disease progression in a patient. This information can be extremely useful in distinguishing hospitalized patients at higher risk for needing UCI from patients with low severity. How SARS-CoV-2 infection will evolve is still unclear. Method(s): A novel pipeline was developed that can integrate RNA-Seq data from different databases to obtain a genetic biomarker COVID-19 severity index using an artificial intelligence algorithm. Our pipeline ensures robustness through multiple cross-validation processes in different steps. Result(s): CD93, RPS24, PSCA, and CD300E were identified as COVID-19 severity gene signatures. Furthermore, using the obtained gene signature, an effective multi-class classifier capable of discrimi-nating between control, outpatient, inpatient, and ICU COVID-19 patients was optimized, achieving an accuracy of 97.5%. Conclusion(s): In summary, during this research, a new intelligent pipeline was implemented to develop a specific gene signature that can detect the severity of patients suffering COVID-19. Our approach to clinical decision support systems achieved excellent results, even when processing unseen samples. Our system can be of great clinical utility for the strategy of planning, organizing and managing human and material resources, as well as for automatically classifying the severity of patients affected by COVID-19.Copyright © 2023 Bentham Science Publishers.

Disease-Modifying Therapies, Outcomes, Risk Factors and Severity of COVID-19 in Multiple Sclerosis: A MENACTRIMS Registry Based Study

Background: Multiple sclerosis (MS) patients have been considered a higher-risk population for COVID-19 due to the high prevalence of disability and disease-modifying therapy use;however, there is little data in our Middle East and North Africa region (MENA) identifying clinical characteristics of MS associated with worse COVID-19 outcomes. Material(s) and Method(s): This a nationwide, multicenter, retrospective cohort study conducted between March 2020 and February 2021 and included MS patients with a suspected or confirmed COVID-19. Using data collected from the MENACTRIMS registry and local COVID-19 registries, the association of patient demographics, MS disease characteristics, and use of disease-modifying therapies with outcomes and severity of COVID-19 illness were evaluated by multivariate logistic models. Result(s): A total of 600 MS patients with suspected (n=58) or confirmed (n=542) COVID-19 (mean age: 36.4 +/- 10.16 years;414 (69%) females;mean disease duration: 8.3+/- 6.6 years) were analyzed. Seventy-three patients (12.2%) had a COVID-19 severity score of 3 or more, and 15 patients (2.5%) died of COVID-19. The median EDSS was 2.0 (range, 0-9.5), and 559 patients (93.2%) were receiving disease-modifying therapy (DMT). There was a higher proportion of patients with a COVID-19 severity score of 3 or more among patients treated with DMTs relative to untreated patients (82.9% vs 17.1%;P < .001), from whom the majority (n=117;19.7%) were maintained on anti-CD20 therapies such as ocrelizumab and rituximab. Comorbidities mainly hypertension and cardiovascular diseases, progressive MS, disease duration, and EDSS were associated with severe or worse COVID-19 disease outcome. Multivariate logistic regression analysis showed that older age (odds ratio per 10 years, 1.5 [95%CI, 1.1-2.0]), male gender (OR, 2.1 [95%CI. 1.2-3.8]), obesity (OR, 2.8 [95%CI, 1.3-5.8]), and treatment ocrelizumab/rituximab (OR for ocrelizumab, 4.6 [95%CI. 1.2-17.7], OR for rituximab, 14.1 [95%CI, 4.8-41.3]) or off-label immunosuppressive medications such as azathioprine or mycophenolate mofetil (OR, 8.8 [95%CI. 1.7-44.0]) were risk factors for moderate to severe COVID-19 requiring hospitalization. Surprisingly, smoking and diabetes were not identified as risk factors for severe COVID-19 disease in our cohort. Conclusion(s): In this registry-based cohort study of patients with MS, age, sex, EDSS, obesity, progressive MS were independent risk factors for severe COVID-19. Moreover, there was an association found between exposure to anti-CD20 DMTs and COVID-19 severity. Knowledge of these risk factors may help improve the clinical management of MS patients with COVID-19 infection.Copyright © 2022

COVID-19 Severity Index (SI): is it a sensitive predictor score for disease outcome in hospitalized patients?

Introduction. The COVID-19 pandemic showed the wide ranging of coronavirus disease prognosing, hence clinical identification of patients who are at risk of poor outcomes is a priority. But there is no proven prognostic scoring system yet. COVID-19 SI was developed as a triage tool, that could be used by healthcare personnel to identify highrisk patients. 1 Aim. To estimate whether COVID-19 SI could predict the disease outcome in hospitalized patients with coronavirus disease already on admission? Methods. The study was a single-center retrospective analysis based on data of 632 COVID-19 patients admitted to the City Hospital No 4 (Dnipro) from August to October 2021. The patients' SI on admission and disease outcome were analyzed and statistically processed. Results. Distribution of survivors and nonsurvivors regarding clinical risk in accordance with SI is presented in Table 1. The sensitivity of SI as prognostic score totaled 37 %;specificity - 52,7 %. Conclusions. The study confirmed COVID-19 SI as a good triage tool on admission, but it has low sensitivity and specificity as prognostic score. (Table Presented).

Blood Transcriptomes of Kidney Recipients with Covid-19 Demonstrate Deficient Adaptive Immune Activation

Purpose: Kidney transplant recipients (KTRs) have poor outcomes compared to non-KTRs with acute COVID-19. To provide insight into management of immunosuppression (IS) during COVID-19, we studied immune signatures from the peripheral blood during and after COVID-19 infection from a multicenter KTR cohort. Method(s): Clinical data were collected by chart review. Paxgene blood RNA was polyA-selected and sequenced at enrollment Results: A total of 64 KTRs affected with COVID-19 were enrolled (31 Early cases (<4weeks from a positive SARS-CoV-2 PCR test) and 33 late cases). Out of the 64 patients, eight died and three encountered graft losses during follow-up. Among 31 early cases, we detected differentially expressed genes (nominal p-value < 0.01) in the blood transcriptome that were positively or negatively associated with the COVID-19 severity score (scale of 1 to 7 with increasing severity;Fig 1A). Enrichment analyses showed upregulation of neutrophil and innate immune pathways and downregulation of adaptive immune activation pathways with increasing severity score (Fig 1B). This observation was independent of lymphocyte count, despite reduction in immunosuppression (IS) in 75% of KTRs. Interestingly, compared with early cases, the blood transcriptome in late cases showed "normalization" of these enriched pathways after 4 weeks, suggesting return of adaptive immune system activation despite re-initiation of immunosuppression (Fig 1C). The latter analyses were adjusted for the severity score. Interestingly, similar pathway enrichment with worsening severity of COVID-19 was identifiable from a public dataset of non-KTRs (GSE152418), showing overlapped signatures for acute COVID-19 between KTRs and non-KTRs (overlap P<0.05) (Fig 1D). Conclusion(s): Blood transcriptome of COVID-KTRs shows marked decrease in adaptive immune system activation during acute COVID-19, even during IS reduction, which show recovery after acute illness. (Figure Presented).

Blood Transcriptomes of Covid-19 Infected Kidney Recipients Demonstrate Frequent and Persistent SARS-CoV-2 Virome Gene Expression

Purpose: Kidney transplant recipients (KTRs) have poor outcomes vs non-KTRs with acute COVID-19. To provide insight into management of immunosuppression during acute COVID-19, we studied peripheral blood transcriptomes during and after COVID-19 from a multicenter KTR cohort. Method(s): Clinical data were collected by chart review. Paxgene blood RNA was polyA-selected and sequenced at enrollment. Result(s): A total of 64 KTRs with COVID-19 were enrolled (31 Early cases (<4weeks from a positive SARS-CoV-2 PCR test) and 33 late cases). Out of the 64 patients, eight died and three encountered graft losses during follow-up. Due to presence of mRNA reads in the blood transcriptome unmapped to the human genome, we aligned the mRNA short reads to the SARS-CoV-2 genome. Surprisingly, our strategy detected the SARS-Cov2 mRNA, especially Spike mRNA in 27 (87%) early cases, and 18 (54%) of late cases (Fig 1A and B). We then analyzed the raw reads from a public dataset of non-KTRs with Paxgene RNA (GSE172114). The SARS-CoV-2 Spike mRNA was detected in 2/47 (4.2%) critically ill COVID-19 cases and 0/25 noncritically ill cases in this non-KTR dataset (compared to KTRs, Chi-square P<0.001;Fig 1B). Among our KTRs, the amount of Spike mRNA was associated positively with the COVID-19 severity score (scale of 1 to 7 of increasing severity;Fig 1C) and inversely with time from initial positive PCR (Fig 1D). More interestingly, 7/64 patients had detectable Spike RNA-emia beyond 60 days after COVID-19 diagnosis. Of the 3 graft losses in our cohort, 2 occurred among these 7 patients. Conclusion(s): Blood transcriptome of KTRs with COVID-19 demonstrated a risk for persistent viremia with implications for pathogenesis of COVID-19 disease. This finding also supports using passive immune strategies in COVID-KTRs. (Figure Presented).

MACHINE LEARNING ANALYSIS of DNA METHYLATION in COVID-19 DISEASE

Background: SARS-CoV-2 infection has resulted in over 219 million confirmed cases of COVID-19 with 4.5 million fatalities, highlighting the importance of elucidating mechanisms of severe disease. Here we utilized machine learning (ML) technologies to identify DNA methylation footprints of COVID-19 disease from publicly available data. Methods: Genome-wide DNA methylation of SARS-CoV-2 infected and uninfected patients using Illumina HumanMethylationEPIC microarray platform from whole blood was publicly available through NCBI Gene Expression Omnibus. A training cohort (GSE167202) consisting of 460 individuals (164 COVID-19-infected and 296 non-infected) and an external validation dataset (GSE174818) consisting of 128 individuals (102 COVID-19-infected and 26 non-COVID with pneumonia diagnosis) were obtained. COVID-19 severity score (SS) was classified as follows: 0. uninfected;1. released from department to home;2. admitted to in-patient care;3. progressed to ICU;and 4. death. Participants were then dichotomized by SS=0 or SS≥3. Raw data was processed using ChAMP in R 4.1.1, resulting in over 850,000 methylation sites per sample for analysis. Beta values were logit transformed to M values using CpGTools in Python 3.8.8. JADBio AutoML platform was leveraged to analyze these datasets with the goal of identifying a methylation signature indicative of COVID-19 disease. Results: From our training cohort, JADBio utilized LASSO feature selection (penalty=1.5) to identify 4 unique methylation sites capable of carrying the predictive weight of a classification random forest trained on 100 trees with Deviance splitting criterion (minimum leaf size=3). The average area under the curve of receiver operator characteristic (AUC-ROC) of the model was 0.933 (95% confidence interval [0.885, 0.970]), while the average area under the precision-recall curve (AUC-PRC) of 0.965 [0.932, 0.986]. When COVID-19 mild infections (SS = 1 or 2) were returned to the training dataset as an internal control, the model retained its predictive power (AUC-ROC=0.985, AUC-PRC=0.992). When applied to our external validation, this model produced an AUC-ROC of 0.901 with an AUC-PRC of 0.748. Conclusion: We developed a Random Forest Classification model capable of accurately predicting COVID-19 infection leveraging JADBio AutoML platform. These results enhance our understanding of epigenetic mechanisms used by SARS-CoV-2 in disease pathogenesis and identify potential therapeutic targets.

Prognostic impact of CT severity score in childhood cancer with SARS-CoV-2

Background: CT chest severity score (CTSS) is a semi-quantitative measure done to correlate the severity of the pulmonary involvement on the CT with the severity of the disease. The objectives of this study are to describe chest CT criteria and CTSS of the COVID-19 infection in pediatric oncology patients, to find a cut-off value of CTSS that can differentiate mild COVID-19 cases that can be managed at home and moderate to severe cases that need hospital care. A retrospective cohort study was conducted on 64 pediatric oncology patients with confirmed COVID-19 infection between 1 April and 30 November 2020. They were classified clinically into mild, moderate, and severe groups. CT findings were evaluated for lung involvement and CTSS was calculated and range from 0 (clear lung) to 20 (all lung lobes were affected). Results: Overall, 89% of patients had hematological malignancies and 92% were under active oncology treatment. The main CT findings were ground-glass opacity (70%) and consolidation patches (62.5%). In total, 85% of patients had bilateral lung involvement, ROC curve showed that the area under the curve of CTSS for diagnosing severe type was 0.842 (95% CI 0.737–0.948). The CTSS cut-off of 6.5 had 90.9% sensitivity and 69% specificity, with 41.7% positive predictive value (PPV) and 96.9% negative predictive value (NPV). According to the Kaplan–Meier analysis, mortality risk was higher in patients with CT score > 7 than in those with CTSS < 7. Conclusion: Pediatric oncology patients, especially those with hematological malignancies, are more vulnerable to COVID-19 infection. Chest CT severity score > 6.5 (about 35% lung involvement) can be used as a predictor of the need for hospitalization.