Identifying organ dysfunction trajectory-based subphenotypes in critically ill patients with COVID-19

Identification of Parkinson's disease PACE subtypes and repurposing treatments through integrative analyses of multimodal data

Parkinson's disease (PD) is a serious neurodegenerative disorder marked by significant clinical and progression heterogeneity. This study aimed at addressing heterogeneity of PD through integrative analysis of various data modalities. We analyzed clinical progression data (≥5 years) of individuals with de novo PD using machine learning and deep learning, to characterize individuals' phenotypic progression trajectories for PD subtyping. We discovered three pace subtypes of PD exhibiting distinct progression patterns: the Inching Pace subtype (PD-I) with mild baseline severity and mild progression speed; the Moderate Pace subtype (PD-M) with mild baseline severity but advancing at a moderate progression rate; and the Rapid Pace subtype (PD-R) with the most rapid symptom progression rate. We found cerebrospinal fluid P-tau/α-synuclein ratio and atrophy in certain brain regions as potential markers of these subtypes. Analyses of genetic and transcriptomic profiles with network-based approaches identified molecular modules associated with each subtype. For instance, the PD-R-specific module suggested STAT3, FYN, BECN1, APOA1, NEDD4, and GATA2 as potential driver genes of PD-R. It also suggested neuroinflammation, oxidative stress, metabolism, PI3K/AKT, and angiogenesis pathways as potential drivers for rapid PD progression (i.e., PD-R). Moreover, we identified repurposable drug candidates by targeting these subtype-specific molecular modules using network-based approach and cell line drug-gene signature data. We further estimated their treatment effects using two large-scale real-world patient databases; the real-world evidence we gained highlighted the potential of metformin in ameliorating PD progression. In conclusion, this work helps better understand clinical and pathophysiological complexity of PD progression and accelerate precision medicine.

Temporal patterns of organ dysfunction in COVID-19 patients hospitalized in the intensive care unit: A group-based multitrajectory modeling analysis

BACKGROUND

The course of organ dysfunction (OD) in Corona Virus Disease 2019 (COVID-19) patients is unknown. Herein, we analyze the temporal patterns of OD in intensive care unit-admitted COVID-19 patients.

METHODS

Sequential organ failure assessment scores were evaluated daily within 2 weeks of admission to determine the temporal trajectory of OD using group-based multitrajectory modeling (GBMTM).

RESULTS

A total of 392 patients were enrolled with a 28-day mortality rate of 53.6%. GBMTM identified four distinct trajectories. Group 1 (mild OD, n = 64), with a median APACHE II score of 13 (IQR 9-21), had an early resolution of OD and a low mortality rate. Group 2 (moderate OD, n = 140), with a median APACHE II score of 18 (IQR 13-22), had a 28-day mortality rate of 30.0%. Group 3 (severe OD, n = 117), with a median APACHR II score of 20 (IQR 13-27), had a deterioration trend of respiratory dysfunction and a 28-day mortality rate of 69.2%. Group 4 (extremely severe OD, n = 71), with a median APACHE II score of 20 (IQR 17-27), had a significant and sustained OD affecting all organ systems and a 28-day mortality rate of 97.2%.

CONCLUSIONS

Four distinct trajectories of OD were identified, and respiratory dysfunction trajectory could predict nonpulmonary OD trajectories and patient prognosis.

Early trajectories of virological and immunological biomarkers and clinical outcomes in patients admitted to hospital for COVID-19: an international, prospective cohort study

BACKGROUND

Serial measurement of virological and immunological biomarkers in patients admitted to hospital with COVID-19 can give valuable insight into the pathogenic roles of viral replication and immune dysregulation. We aimed to characterise biomarker trajectories and their associations with clinical outcomes.

METHODS

In this international, prospective cohort study, patients admitted to hospital with COVID-19 and enrolled in the Therapeutics for Inpatients with COVID-19 platform trial within the Accelerating COVID-19 Therapeutic Interventions and Vaccines programme between Aug 5, 2020 and Sept 30, 2021 were included. Participants were included from 108 sites in Denmark, Greece, Poland, Singapore, Spain, Switzerland, Uganda, the UK, and the USA, and randomised to placebo or one of four neutralising monoclonal antibodies: bamlanivimab (Aug 5 to Oct 13, 2020), sotrovimab (Dec 16, 2020, to March 1, 2021), amubarvimab-romlusevimab (Dec 16, 2020, to March 1, 2021), and tixagevimab-cilgavimab (Feb 10 to Sept 30, 2021). This trial included an analysis of 2149 participants with plasma nucleocapsid antigen, anti-nucleocapsid antibody, C-reactive protein (CRP), IL-6, and D-dimer measured at baseline and day 1, day 3, and day 5 of enrolment. Day-90 follow-up status was available for 1790 participants. Biomarker trajectories were evaluated for associations with baseline characteristics, a 7-day pulmonary ordinal outcome, 90-day mortality, and 90-day rate of sustained recovery.

FINDINGS

The study included 2149 participants. Participant median age was 57 years (IQR 46-68), 1246 (58·0%) of 2149 participants were male and 903 (42·0%) were female; 1792 (83·4%) had at least one comorbidity, and 1764 (82·1%) were unvaccinated. Mortality to day 90 was 172 (8·0%) of 2149 and 189 (8·8%) participants had sustained recovery. A pattern of less favourable trajectories of low anti-nucleocapsid antibody, high plasma nucleocapsid antigen, and high inflammatory markers over the first 5 days was observed for high-risk baseline clinical characteristics or factors related to SARS-CoV-2 infection. For example, participants with chronic kidney disease demonstrated plasma nucleocapsid antigen 424% higher (95% CI 319-559), CRP 174% higher (150-202), IL-6 173% higher (144-208), D-dimer 149% higher (134-165), and anti-nucleocapsid antibody 39% lower (60-18) to day 5 than those without chronic kidney disease. Participants in the highest quartile for plasma nucleocapsid antigen, CRP, and IL-6 at baseline and day 5 had worse clinical outcomes, including 90-day all-cause mortality (plasma nucleocapsid antigen hazard ratio (HR) 4·50 (95% CI 3·29-6·15), CRP HR 3·37 (2·30-4·94), and IL-6 HR 5·67 (4·12-7·80). This risk persisted for plasma nucleocapsid antigen and CRP after adjustment for baseline biomarker values and other baseline factors.

INTERPRETATION

Patients admitted to hospital with less favourable 5-day biomarker trajectories had worse prognosis, suggesting that persistent viral burden might drive inflammation in the pathogenesis of COVID-19, identifying patients that might benefit from escalation of antiviral or anti-inflammatory treatment.

FUNDING

US National Institutes of Health.

COVID-19 subphenotypes at hospital admission are associated with mortality: a cross-sectional study

BACKGROUND

We have an incomplete understanding of COVID-19 characteristics at hospital presentation and whether underlying subphenotypes are associated with clinical outcomes and therapeutic responses.

METHODS

For this cross-sectional study, we extracted electronic health data from adults hospitalized between 1 March and 30 August 2020 with a PCR-confirmed diagnosis of COVID-19 at five New York City Hospitals. We obtained clinical and laboratory data from the first 24 h of the patient's hospitalization. Treatment with tocilizumab and convalescent plasma was assessed over hospitalization. The primary outcome was mortality; secondary outcomes included intubation, intensive care unit (ICU) admission and length of stay (LOS). First, we employed latent class analysis (LCA) to identify COVID-19 subphenotypes on admission without consideration of outcomes and assigned each patient to a subphenotype. We then performed robust Poisson regression to examine associations between COVID-19 subphenotype assignment and outcome. We explored whether the COVID-19 subphenotypes had a differential response to tocilizumab and convalescent plasma therapies.

RESULTS

A total of 4620 patients were included. LCA identified six subphenotypes, which were distinct by level of inflammation, clinical and laboratory derangements and ranged from a hypoinflammatory subphenotype with the fewest derangements to a hyperinflammatory with multiorgan dysfunction subphenotypes. Multivariable regression analyses found differences in risk for mortality, intubation, ICU admission and LOS, as compared to the hypoinflammatory subphenotype. For example, in multivariable analyses the moderate inflammation with fever subphenotype had 3.29 times the risk of mortality (95% CI 2.05, 5.28), while the hyperinflammatory with multiorgan failure subphenotype had 17.87 times the risk of mortality (95% CI 11.56, 27.63), as compared to the hypoinflammatory subphenotype. Exploratory analyses suggested that subphenotypes may differential respond to convalescent plasma or tocilizumab therapy.

CONCLUSION

COVID-19 subphenotype at hospital admission may predict risk for mortality, ICU admission and intubation and differential response to treatment.KEY MESSAGEThis cross-sectional study of COVID patients admitted to the Mount Sinai Health System, identified six distinct COVID subphenotypes on admission. Subphenotypes correlated with ICU admission, intubation, mortality and differential response to treatment.

Clinical Sepsis Phenotypes in Critically Ill Patients

Sepsis, defined as the life-threatening dysregulated host response to an infection leading to organ dysfunction, is considered as one of the leading causes of mortality worldwide, especially in intensive care units (ICU). Moreover, sepsis remains an enigmatic clinical syndrome, with complex pathophysiology incompletely understood and a great heterogeneity both in terms of clinical expression, patient response to currently available therapeutic interventions and outcomes. This heterogeneity proves to be a major obstacle in our quest to deliver improved treatment in septic critical care patients; thus, identification of clinical phenotypes is absolutely necessary. Although this might be seen as an extremely difficult task, nowadays, artificial intelligence and machine learning techniques can be recruited to quantify similarities between individuals within sepsis population and differentiate them into distinct phenotypes regarding not only temperature, hemodynamics or type of organ dysfunction, but also fluid status/responsiveness, trajectories in ICU and outcome. Hopefully, we will eventually manage to determine both the subgroup of septic patients that will benefit from a therapeutic intervention and the correct timing of applying the intervention during the disease process.

Comparison of time series clustering methods for identifying novel subphenotypes of patients with infection

OBJECTIVE

Severe infection can lead to organ dysfunction and sepsis. Identifying subphenotypes of infected patients is essential for personalized management. It is unknown how different time series clustering algorithms compare in identifying these subphenotypes.

MATERIALS AND METHODS

Patients with suspected infection admitted between 2014 and 2019 to 4 hospitals in Emory healthcare were included, split into separate training and validation cohorts. Dynamic time warping (DTW) was applied to vital signs from the first 8 h of hospitalization, and hierarchical clustering (DTW-HC) and partition around medoids (DTW-PAM) were used to cluster patients into subphenotypes. DTW-HC, DTW-PAM, and a previously published group-based trajectory model (GBTM) were evaluated for agreement in subphenotype clusters, trajectory patterns, and subphenotype associations with clinical outcomes and treatment responses.

RESULTS

There were 12 473 patients in training and 8256 patients in validation cohorts. DTW-HC, DTW-PAM, and GBTM models resulted in 4 consistent vitals trajectory patterns with significant agreement in clustering (71-80% agreement, P < .001): group A was hyperthermic, tachycardic, tachypneic, and hypotensive. Group B was hyperthermic, tachycardic, tachypneic, and hypertensive. Groups C and D had lower temperatures, heart rates, and respiratory rates, with group C normotensive and group D hypotensive. Group A had higher odds ratio of 30-day inpatient mortality (P < .01) and group D had significant mortality benefit from balanced crystalloids compared to saline (P < .01) in all 3 models.

DISCUSSION

DTW- and GBTM-based clustering algorithms applied to vital signs in infected patients identified consistent subphenotypes with distinct clinical outcomes and treatment responses.

CONCLUSION

Time series clustering with distinct computational approaches demonstrate similar performance and significant agreement in the resulting subphenotypes.

A two-gene marker for the two-tiered innate immune response in COVID-19 patients

For coronavirus disease 2019 (COVID-19), a pandemic disease characterized by strong immune dysregulation in severe patients, convenient and efficient monitoring of the host immune response is critical. Human hosts respond to viral and bacterial infections in different ways, the former is characterized by the activation of interferon stimulated genes (ISGs) such as IFI27, while the latter is characterized by the activation of anti-bacterial associated genes (ABGs) such as S100A12. This two-tiered innate immune response has not been examined in COVID-19. In this study, the activation patterns of this two-tiered innate immune response represented by IFI27 and S100A12 were explored based on 1421 samples from 17 transcriptome datasets derived from the blood of COVID-19 patients and relevant controls. It was found that IFI27 activation occurred in most of the symptomatic patients and displayed no correlation with disease severity, while S100A12 activation was more restricted to patients under severe and critical conditions with a stepwise activation pattern. In addition, most of the S100A12 activation was accompanied by IFI27 activation. Furthermore, the activation of IFI27 was most pronounced within the first week of symptom onset, but generally waned after 2-3 weeks. On the other hand, the activation of S100A12 displayed no apparent correlation with disease duration and could last for several months in certain patients. These features of the two-tiered innate immune response can further our understanding on the disease mechanism of COVID-19 and may have implications to the clinical triage. Development of a convenient two-gene protocol for the routine serial monitoring of this two-tiered immune response will be a valuable addition to the existing laboratory tests.

Aplicación de la calculadora de probabilidad fenotípica FEN-COVID en pacientes hospitalizados por COVID-19 en una población latinoamericana

Introducción: la variabilidad del comportamiento clínico del COVID-19 puede ser uno de los determinantes que limitan la toma de decisiones terapéuticas. Se busca clasificar a pacientes latinoamericanos hospitalizados mediante la herramienta FEN-COVID para la identificación de fenotipos clínicos y determinar su asociación con mortalidad e ingreso a la unidad de cuidado intensivo (UCI). Métodos: estudio observacional de cohorte retrospectivo, que incluyó adultos hospitalizados en dos centros de tercer nivel de atención con COVID-19 confirmado entre septiembre 2020 y marzo 2021. A cada paciente seleccionado se asignó un fenotipo aplicando la calculadora FEN-COVID. Se llevó a cabo un análisis multivariado para documentar las asociaciones entre el fenotipo, las complicaciones hospitalarias y los desenlaces clínicos. Resultados: se identificaron 126 pacientes hospitalizados por COVID-19, edad promedio de 58 años, 45 mujeres (35.7%), 23% diabéticos, 45% hipertensos y 20% obesos. 108 (85.7%) fueron del fenotipo B y 18 (14.2%) fenotipo C. Aunque en este último los desenlaces fueron peores (requerimiento de UCI 77.8% vs 45.4% y mortalidad 66% vs 22%, OR 1.408, IC95% 3.191-5.243, p <0.007), esta asociación no se mantuvo en el análisis multivariado con OR 1.110 (IC95% 0.780 - 1.581, p de 0.555) Conclusión: los fenotipos identificados a partir de FEN-COVID parecen discriminar un subgrupo de pacientes que ostenta el peor comportamiento clínico, aunque no tuvo representación del fenotipo más leve. El análisis bivariado documentó asociación con la muerte o ingreso a UCI que no se mantuvo en el modelo multivariado.

COVID-19 Acute Respiratory Distress Syndrome: One Pathogen, Multiple Phenotypes

Acute respiratory distress syndrome (ARDS) is a heterogeneous syndrome arising from multiple causes with a range of clinical severity. In recent years, the potential for prognostic and predictive enrichment of clinical trials has been increased with identification of more biologically homogeneous subgroups or phenotypes within ARDS. COVID-19 ARDS also exhibits significant clinical heterogeneity despite a single causative agent. In this review the authors summarize the existing literature on COVID-19 ARDS phenotypes, including physiologic, clinical, and biological subgroups as well as the implications for improving both prognostication and precision therapy.

Longwise Cluster Analysis for the Prediction of COVID-19 Severity within 72 h of Admission: COVID-DATA-SAVE-LIFES Cohort

The use of routine laboratory biomarkers plays a key role in decision making in the clinical practice of COVID-19, allowing the development of clinical screening tools for personalized treatments. This study performed a short-term longitudinal cluster from patients with COVID-19 based on biochemical measurements for the first 72 h after hospitalization. Clinical and biochemical variables from 1039 confirmed COVID-19 patients framed on the “COVID Data Save Lives” were grouped in 24-h blocks to perform a longitudinal k-means clustering algorithm to the trajectories. The final solution of the three clusters showed a strong association with different clinical severity outcomes (OR for death: Cluster A reference, Cluster B 12.83 CI: 6.11−30.54, and Cluster C 14.29 CI: 6.66−34.43; OR for ventilation: Cluster-B 2.22 CI: 1.64−3.01, and Cluster-C 1.71 CI: 1.08−2.76), improving the AUC of the models in terms of age, sex, oxygen concentration, and the Charlson Comorbidities Index (0.810 vs. 0.871 with p < 0.001 and 0.749 vs. 0.807 with p < 0.001, respectively). Patient diagnoses and prognoses remarkably diverged between the three clusters obtained, evidencing that data-driven technologies devised for the screening, analysis, prediction, and tracking of patients play a key role in the application of individualized management of the COVID-19 pandemics.

A multi-task Gaussian process self-attention neural network for real-time prediction of the need for mechanical ventilators in COVID-19 patients

OBJECTIVE

The Coronavirus Disease 2019 (COVID-19) pandemic has overwhelmed the capacity of healthcare resources and posed a challenge for worldwide hospitals. The ability to distinguish potentially deteriorating patients from the rest helps facilitate reasonable allocation of medical resources, such as ventilators, hospital beds, and human resources. The real-time accurate prediction of a patient's risk scores could also help physicians to provide earlier respiratory support for the patient and reduce the risk of mortality.

METHODS

We propose a robust real-time prediction model for the in-hospital COVID-19 patients' probability of requiring mechanical ventilation (MV). The end-to-end neural network model incorporates the Multi-task Gaussian Process to handle the irregular sampling rate in observational data together with a self-attention neural network for the prediction task.

RESULTS

We evaluate our model on a large database with 9,532 nationwide in-hospital patients with COVID-19. The model demonstrates significant robustness and consistency improvements compared to conventional machine learning models. The proposed prediction model also shows performance improvements in terms of area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) compared to various deep learning models, especially at early times after a patient's hospital admission.

CONCLUSION

The availability of large and real-time clinical data calls for new methods to make the best use of them for real-time patient risk prediction. It is not ideal for simplifying the data for traditional methods or for making unrealistic assumptions that deviate from observation's true dynamics. We demonstrate a pilot effort to harmonize cross-sectional and longitudinal information for mechanical ventilation needing prediction.

Outcomes of Cancer Patients with COVID-19 in a Hospital System in the Chicago Metropolitan Area

Patients with a history of malignancy have been shown to be at an increased risk of COVID-19-related morbidity and mortality. Poorer clinical outcomes in that patient population are likely due to the underlying systemic illness, comorbidities, and the cytotoxic and immunosuppressive anti-tumor treatments they are subjected to. We identified 416 cancer patients with SARS-CoV-2 infection being managed for their malignancy at Northwestern Medicine in Chicago, Illinois, between March and July of 2020. Seventy-five (18.0%) patients died due to COVID-related complications. Older age (>60), male gender, and current treatment with immunotherapy were associated with shorter overall survival. Laboratory findings showed that higher platelet counts, ALC, and hemoglobin were protective against critical illness and death from COVID-19. Conversely, elevated inflammatory markers such as ferritin, d-dimer, procalcitonin, CRP, and LDH led to worse clinical outcomes. Our findings suggest that a thorough clinical and laboratory assessment of infected patients with cancer might help identify a more vulnerable population and implement more aggressive proactive strategies.

Artificial intelligence for COVID-19: battling the pandemic with computational intelligence

The new coronavirus disease 2019 (COVID-19) has become a global pandemic leading to over 180 million confirmed cases and nearly 4 million deaths until June 2021, according to the World Health Organization. Since the initial report in December 2019 , COVID-19 has demonstrated a high transmission rate (with an R₀ > 2), a diverse set of clinical characteristics (e.g., high rate of hospital and intensive care unit admission rates, multi-organ dysfunction for critically ill patients due to hyperinflammation, thrombosis, etc.), and a tremendous burden on health care systems around the world. To understand the serious and complex diseases and develop effective control, treatment, and prevention strategies, researchers from different disciplines have been making significant efforts from different aspects including epidemiology and public health, biology and genomic medicine, as well as clinical care and patient management. In recent years, artificial intelligence (AI) has been introduced into the healthcare field to aid clinical decision-making for disease diagnosis and treatment such as detecting cancer based on medical images, and has achieved superior performance in multiple data-rich application scenarios. In the COVID-19 pandemic, AI techniques have also been used as a powerful tool to overcome the complex diseases. In this context, the goal of this study is to review existing studies on applications of AI techniques in combating the COVID-19 pandemic. Specifically, these efforts can be grouped into the fields of epidemiology, therapeutics, clinical research, social and behavioral studies and are summarized. Potential challenges, directions, and open questions are discussed accordingly, which may provide new insights into addressing the COVID-19 pandemic and would be helpful for researchers to explore more related topics in the post-pandemic era.

Identification of Robust Protein Associations With COVID-19 Disease Based on Five Clinical Studies

Multiple studies have investigated the role of blood circulating proteins in COVID-19 disease using the Olink affinity proteomics platform. However, study inclusion criteria and sample collection conditions varied between studies, leading to sometimes incongruent associations. To identify the most robust protein markers of the disease and the underlying pathways that are relevant under all conditions, it is essential to identify proteins that replicate most widely. Here we combined the Olink proteomics profiles of two newly recruited COVID-19 studies (N=68 and N=98) with those of three previously published COVID-19 studies (N=383, N=83, N=57). For these studies, three Olink panels (Inflammation and Cardiovascular II & III) with 253 unique proteins were compared. Case/control analysis revealed thirteen proteins (CCL16, CCL7, CXCL10, CCL8, LGALS9, CXCL11, IL1RN, CCL2, CD274, IL6, IL18, MERTK, IFNγ, and IL18R1) that were differentially expressed in COVID-19 patients in all five studies. Except CCL16, which was higher in controls, all proteins were overexpressed in COVID-19 patients. Pathway analysis revealed concordant trends across all studies with pathways related to cytokine-cytokine interaction, IL18 signaling, fluid shear stress and rheumatoid arthritis. Our results reaffirm previous findings related to a COVID-19 cytokine storm syndrome. Cross-study robustness of COVID-19 specific protein expression profiles support the utility of affinity proteomics as a tool and for the identification of potential therapeutic targets.

Stratifying individuals into non-alcoholic fatty liver disease risk levels using time series machine learning models

Non-alcoholic fatty liver disease (NAFLD) affects 25% of the population worldwide, and its prevalence is anticipated to increase globally. While most NAFLD patients are asymptomatic, NAFLD may progress to fibrosis, cirrhosis, cardiovascular disease, and diabetes. Research reports, with daunting results, show the challenge that NAFLD's burden causes to global population health. The current process for identifying fibrosis risk levels is inefficient, expensive, does not cover all potential populations, and does not identify the risk in time. Instead of invasive liver biopsies, we implemented a non-invasive fibrosis assessment process calculated from clinical data (accessed via EMRs/EHRs). We stratified patients' risks for fibrosis from 2007 to 2017 by modeling the risk in 5579 individuals. The process involved time-series machine learning models (Hidden Markov Models and Group-Based Trajectory Models) profiled fibrosis risk by modeling patients' latent medical status resulted in three groups. The high-risk group had abnormal lab test values and a higher prevalence of chronic conditions. This study can help overcome the inefficient, traditional process of detecting fibrosis via biopsies (that are also medically unfeasible due to their invasive nature, the medical resources involved, and costs) at early stages. Thus longitudinal risk assessment may be used to make population-specific medical recommendations targeting early detection of high risk patients, to avoid the development of fibrosis disease and its complications as well as decrease healthcare costs.

OUP accepted manuscript

BACKGROUND

A protective antibody response to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is crucial to decrease morbidity and mortality from severe coronavirus disease 2019 (COVID-19) disease. The effects of preexisting anti-human coronavirus (HCoV) antibodies on the SARS-CoV-2-specific immunoglobulin G (IgG) responses and severity of disease are currently unclear.

METHODS

We profiled anti-spike (S), S1, S2, and receptor-binding domain IgG antibodies against SARS-CoV-2 and 6 HCoVs using a multiplex assay (mPLEX-CoV) with serum samples from SARS-CoV-2 infected (n = 155) and pre-COVID-19 (n = 188) cohorts.

RESULTS

COVID-19 subjects showed significantly increased anti-S SARS-CoV-2 IgG levels that were highly correlated with IgG antibodies against OC43 and HKU1 S proteins. However, OC43 and HKU1 anti-S antibodies in pre-COVID-19 era sera did not cross-react with SARS-CoV-2. Unidirectional cross-reactive antibodies elicited by SARS-CoV-2 infection were distinct from the bidirectional cross-reactive antibodies recognizing homologous strains RaTG13 and SARS-CoV-1. High anti-OC43 and anti-S2 antibody levels were associated with both a rapid anti-SARS-CoV-2 antibody response and increased disease severity. Subjects with increased sequential organ failure assessment (SOFA) scores developed a higher ratio of S2- to S1-reactive antibodies.

CONCLUSIONS

Early and rapid emergence of OC43 S- and S2-reactive IgG after SARS-CoV-2 infection correlates with COVID-19 disease severity.

Identification of persistent and resolving subphenotypes of acute hypoxemic respiratory failure in two independent cohorts

BACKGROUND

Acute hypoxemic respiratory failure (HRF) is associated with high morbidity and mortality, but its heterogeneity challenges the identification of effective therapies. Defining subphenotypes with distinct prognoses or biologic features can improve therapeutic trials, but prior work has focused on ARDS, which excludes many acute HRF patients. We aimed to characterize persistent and resolving subphenotypes in the broader HRF population.

METHODS

In this secondary analysis of 2 independent prospective ICU cohorts, we included adults with acute HRF, defined by invasive mechanical ventilation and PaO2-to-FIO2 ratio ≤ 300 on cohort enrollment (n = 768 in the discovery cohort and n = 1715 in the validation cohort). We classified patients as persistent HRF if still requiring mechanical ventilation with PaO2-to-FIO2 ratio ≤ 300 on day 3 following ICU admission, or resolving HRF if otherwise. We estimated relative risk of 28-day hospital mortality associated with persistent HRF, compared to resolving HRF, using generalized linear models. We also estimated fold difference in circulating biomarkers of inflammation and endothelial activation on cohort enrollment among persistent HRF compared to resolving HRF. Finally, we stratified our analyses by ARDS to understand whether this was driving differences between persistent and resolving HRF.

RESULTS

Over 50% developed persistent HRF in both the discovery (n = 386) and validation (n = 1032) cohorts. Persistent HRF was associated with higher risk of death relative to resolving HRF in both the discovery (1.68-fold, 95% CI 1.11, 2.54) and validation cohorts (1.93-fold, 95% CI 1.50, 2.47), after adjustment for age, sex, chronic respiratory illness, and acute illness severity on enrollment (APACHE-III in discovery, APACHE-II in validation). Patients with persistent HRF displayed higher biomarkers of inflammation (interleukin-6, interleukin-8) and endothelial dysfunction (angiopoietin-2) than resolving HRF after adjustment. Only half of persistent HRF patients had ARDS, yet exhibited higher mortality and biomarkers than resolving HRF regardless of whether they qualified for ARDS.

CONCLUSION

Patients with persistent HRF are common and have higher mortality and elevated circulating markers of lung injury compared to resolving HRF, and yet only a subset are captured by ARDS definitions. Persistent HRF may represent a clinically important, inclusive target for future therapeutic trials in HRF.