Identification of a multi-omics factor predictive of long COVID in the IMPACC study

Abstract

Following SARS-CoV-2 infection, ~10-35% of COVID-19 patients experience long COVID (LC), in which often debilitating symptoms persist for at least three months. Elucidating the biologic underpinnings of LC could identify therapeutic opportunities. We utilized machine learning methods on biologic analytes and patient reported outcome surveys provided over 12 months after hospital discharge from >500 hospitalized COVID-19 patients in the IMPACC cohort to identify a multi-omics "recovery factor". IMPACC participants who experienced LC had lower recovery factor scores compared to participants without LC. Biologic characterization revealed increased levels of plasma proteins associated with inflammation, elevated transcriptional signatures of heme metabolism, and decreased androgenic steroids in LC patients. The recovery factor was also associated with altered circulating immune cell frequencies. Notably, recovery factor scores were predictive of LC occurrence in patients as early as hospital admission, irrespective of acute disease severity. Thus, the recovery factor identifies patients at risk of LC early after SARS-CoV-2 infection and reveals LC biomarkers and potential treatment targets.

Keywords: COVID-19; Heme metabolism; Long COVID; Machine Learning; PASC; SARS-CoV-2; androgenic steroids; inflammation; multi-omics; patient reported outcomes.

Figure 1.. Multi-omics data overview and generation of a predictive LC factor.

(A) Number of samples used in the multi-omics data integration strategy by assay (rows) and scheduled time of collection (columns). Shading indicates the frequency of samples with data availability at the indicated visit. (B) Patient classification in Patient Reported Outcome (PRO) clusters according to the PRO measure survey scores. (C) Individual assay data were preprocessed and split into Train and Test cohorts by participant in an 80/20 split, maintaining the proportion of PRO cluster participants in each partition. (D) Preprocessed assay data and LC response outcomes for the Train cohort were used to identify multi-omics predictive factors with SPEAR. Factor scores were then calculated for the Test cohort. (E) The performance of the multi-omics predictive factors to classify patients into presence and absence of LC was assessed on the Train cohort via cross-validation and then validated on the Test cohort. The predictive factor scores were confirmed to be associated with LC after correcting for possible confounding variables. In depth analysis of enriched biological pathways and significant analytes relevant for the prediction was performed. Factor scores were computed for the acute infection immune profiles, and association analysis with LC at these early time points was performed. See also Figure S1.

Figure 2.. Identification of a convalescent multi-omics recovery factor that discriminates long COVID.

(A) Predictive performance of a lasso model trained on the MOFA and SPEAR factors to discriminate LC vs MIN at the event level. The mean AUROC of a 10-fold cross-validation on the Train Cohort, for 100 bootstrapped model training repetitions are shown. Significance was calculated by standard normal approximation of bootstrapped differences between models (t-test, ****adj. p-value ≤ 0.0001) (B) Predictive performance of the SPEAR Physical model to discriminate LC vs MIN on the Test cohort. ROC curve of model (solid line), random classifier (dashed line), and AUROC value are shown. TPR: true positive rate, FPR: false positive rate. (C) Recovery factor scores for the Test cohort of the MIN and LC groups at 3 months (Visit 7), 6 months (Visit 8), 9 months (Visit 9) and 12 months (Visit 10) after hospital discharge. (D) Recovery factor scores of the individual PRO clusters by visit for the Test cohort. P-values in C and D show the significance of the recovery factor score association with MIN vs LC and pairwise PRO cluster combinations, respectively (see methods for association details). See also Figure S2, S3, and S4.

Figure 3.. Heme metabolism and androgenic steroid pathways, inflammation-associated serum factors, and altered immune cell composition are associated with the recovery factor during convalescence.

(A) GSEA identifies heme metabolism and androgenic steroid pathways as significantly associated with the recovery factor, with significance shown per assay, as well as across assays (joint adj. p < 0.05). (B) 26 significant analytes (SPEAR Bayesian posterior selection probability ≥ 0.95) in the recovery factor across different assays (left) were identified using SPEAR factor loadings (middle; coefficient in the factor), and each was tested for association in the test cohort with MIN vs LC groups (right; adj. intercept p-value). (C) Geometric means of analytes from the significantly enriched gene and metabolite sets and/or significant SPEAR analytes are shown per sample at each convalescent visit in the test cohort. The p-values indicate significance of the association with MIN vs LC. (D) Association in the full cohort of whole blood cell counts determined by CyTOF with the recovery factor for parent and child immune cell types. NK: Natural Killer cells, Mono: Monocytes, PMN: polymorphonuclear neutrophils, B: B lymphocytes, CD4: CD4+ T lymphocytes, CD8: CD8+ T lymphocytes. For a full list of the child populations see Table S3. (* adj. p-value < 0.05, ** adj. p-value < 0.01, *** adj. p-value < 0.001). See also Figure S5.

Figure 4.. Associations of clinical measurements with recovery factor scores.

(A) Association of recovery factor sores with clinical features (demographics, comorbidities, complications and baseline lab measurements). Dot plot shows the signed adjusted p-values indicating the clinical feature term significance from a linear mixed-effect model with enrollment site and participant as random effects to explain the convalescent phase recovery factor scores. Sex and discretized age were further adjusted as fixed effects for clinical features other than sex and age. Only significant associations (adj. p-value <0.05) are shown. (B) Associations of recovery factor scores with individual PRO survey scores (PROMIS scale scores, EQ-5D-5L and health score) in the test cohort. Raw and adjusted p-values indicated the PRO score term significance in linear mixed effect models. (C) Associations of recovery factor scores with each indicated symptom group in the test cohort. Numbers are the uncorrected significance (p-values) of the symptom group term in linear mixed effect models. (D) Recovery factor scores per participant in the test cohort, separated into MIN and LC groups by acute phase trajectory groups, stratified by visit. P-values for panels B-D show the endpoint term of a linear mixed effect model with sex, discretized admit age, and trajectory group as fixed effects and enrollment site as random effect. No individual MIN vs. LC comparisons were significant after p-value correction. (* adj. p-value < 0.05, ** adj. adj. p-value < 0.01, *** adj. p-value < 0.001)

Figure 5.. Recovery factor scores in acute phase data associate with eventual LC status.

(A) Recovery factor scores during the acute disease phase for participants in the LC and MIN groups within 72h of hospital admission (Visit 1) and at day 4 (Visit 2), day 7 (Visit 3), day 14 (Visit 4), day 21 (Visit 5), and day 28 (Visit 6) after admission. (B) Recovery factor scores during the acute disease phase for participants in individual PRO clusters. (C) Geometric mean of analytes in enriched gene and metabolic sets and/or significant SPEAR analytes during the acute phase. No individual per-visit comparisons were significant after p-value correction. P-values in top-right box in A-C show the significance of the recovery factor score or geometric mean signature association with MIN vs LC or pairwise PRO cluster combinations. Bars above the boxplots show the pairwise significance across groups in a per-visit comparison (** p<0.01, *p<0.05). (D) Recovery factor scores association with whole blood CyTOF immune cell populations during the acute phase (* adj. p-value < 0.05, ** adj. p-value < 0.01, *** adj. p-value < 0.001). See also Figure S8.

Figure 6.. Recovery factor scores associate with acute disease phase trajectory groups, but identify LC irrespective of acute severity.

(A) Longitudinal analysis of acute recovery factor scores for the full IMPACC cohort stratified by trajectory group (N=1,148 participants). P-value shows the significance of the trajectory group term in a longitudinal model correcting for age and sex as fixed effects and enrollment site and participant ID as random effects. (B) Recovery factor scores in the acute phase by convalescent MIN/LC label, stratified by acute trajectory group and visit number. P-values show significance in distinguishing MIN vs. LC labels in linear mixed models with sex, discretized admit age, and trajectory group as fixed effects and enrollment site and participant ID as random effects, performed separately for each acute visit and corrected across all visits.