COVID-19 varies from a mild illness in some people to fatal disease in others. Patients with severe disease tend to be older and have underlying medical problems. People with kidney failure have a particularly high risk of developing severe or fatal COVID-19.

Patients with severe COVID-19 have high levels of inflammation, causing damage to tissues around the body. Many drugs that target inflammation have already been developed for other diseases. Therefore, to repurpose existing drugs or design new treatments, it is important to determine which proteins drive inflammation in COVID-19.

Here, Gisby, Clarke, Medjeral-Thomas et al. measured 436 proteins in the blood of patients with kidney failure and compared the levels between patients who had COVID-19 to those who did not. This revealed that patients with COVID-19 had increased levels of hundreds of proteins involved in inflammation and tissue injury. Using a combination of statistical and machine learning analyses, Gisby et al. probed the data for proteins that might predict a more severe disease progression. In total, over 200 proteins were linked to disease severity, and 69 with increased risk of death. Tracking how levels of blood proteins changed over time revealed further differences between mild and severe disease. Comparing this data with a similar study of COVID-19 in people without kidney failure showed many similarities. This suggests that the findings may apply to COVID-19 patients more generally.

Identifying the proteins that are a cause of severe COVID-19 – rather than just correlated with it – is an important next step that could help to select new drugs for severe COVID-19.

Coronavirus disease (COVID-19), caused by the SARS-CoV-2 virus, displays wide clinical heterogeneity from asymptomatic to fatal disease. Patients with severe disease exhibit marked inflammatory responses and immunopathology. The mechanisms underlying this remain incompletely characterised, and the key molecular mediators are yet to be determined. The first treatment shown to reduce mortality from COVID-19 in randomised trials was dexamethasone (Horby et al., 2020), a corticosteroid that has broad non-specific effects on the immune system. Even with corticosteroid treatment, mortality in severe COVID-19 remains significant. There is a wide armamentarium of existing drugs that target inflammation more selectively, providing potential repurposing opportunities for the treatment of COVID-19. Recently, the REMAP-CAP trial has demonstrated efficacy of anti-IL6 receptor blockade in patients admitted to intensive care units with severe disease (Gordon et al., 2021). In order to select the most promising agents for future trials, we urgently need to better understand the molecular drivers of severe disease. Proteins are the effector molecules of biology and the targets of most drugs. Therefore, proteomic profiling to identify the key mediators of severe disease provides a valuable tool for identifying and prioritising potential drug targets (Suhre et al., 2021).

Risk factors for severe or fatal COVID-19 include age, male sex, non-European ancestry, obesity, diabetes mellitus, cardiovascular disease, and immunosuppression (Williamson et al., 2020). End-stage kidney disease (ESKD) is one of the strongest risk factors for severe COVID-19 (estimated hazard ratio for death 3.69) (Williamson et al., 2020), and ESKD patients hospitalised with COVID-19 have a mortality of approximately 30% (Docherty et al., 2020; Corbett et al., 2020; Ng et al., 2020; Valeri et al., 2020). ESKD patients have a high prevalence of vascular and cardiometabolic disease (e.g. hypertension, ischaemic heart disease, diabetes), either as a result of the underlying cause of their renal disease or as a consequence of renal failure. In addition, ESKD results in both relative immunosuppression and chronic low-grade inflammation, which may impact viral defence and the host inflammatory response.

Here we performed proteomic profiling of serial blood samples of ESKD patients with COVID-19, leveraging the unique opportunity for longitudinal sampling in both the outpatient and inpatient settings afforded by a large multi-ethnic haemodialysis cohort (Figure 1a). These data revealed 221 proteins that are dysregulated in COVID-19 versus matched non-infected ESKD patients. Using linear mixed models, joint models, and machine learning, we identified proteins that are markers of COVID-19 severity and risk of death. Finally, we characterised the temporal dynamics of the blood proteomic response during COVID-19 infection in ESKD patients, uncovering 32 proteins that display altered trajectories in patients with severe versus non-severe disease.

Figure 1 with 1 supplement see all

Download asset Open asset

We recruited 55 ESKD patients with COVID-19 (subcohort A; Table 1). All patients were receiving haemodialysis prior to acquiring COVID-19. Blood samples were taken as soon as feasible following COVID-19 diagnosis. At time of initial sample, 30 patients were outpatients attending haemodialysis sessions and 25 were hospitalised inpatients (see Materials and methods, Figure 1). Following the initial blood sample, serial sampling was performed for 51/55 patients. We also recruited 51 non-infected haemodialysis patients as ESKD controls, mirroring the age, sex, and ethnicity distribution of the COVID-19 cases (Figure 1—figure supplement 1a–c). We used the Olink proteomics platform to measure 436 proteins (Supplementary file 1a) in 256 plasma samples from the COVID-19 patients and the 51 control samples. The proteins measured consisted of five multiplex ‘panels’ focussed on proteins relevant to immuno-inflammation, cardiovascular, and cardiometabolic disease. The 436 proteins assayed showed strong enrichment for immune-related proteins (Supplementary file 1b).

In addition, we performed the Olink proteomic assays in 52 serum samples from a separate set of 46 COVID-19-positive ESKD patients (subcohort B) and 11 serum samples from ESKD COVID-19-negative controls (a subset of the controls described above). For the large majority of patients in subcohort B, only a single timepoint was available. A higher proportion of these patients (41/46, 89%) were hospitalised and had severe disease (Table 2) than in subcohort A (Figure 1, Table 1).

Proteomic differences between COVID-19-positive and -negative ESKD patients

Principal component analysis (PCA) of proteomic data from subcohort A demonstrated differences between samples from COVID-19-positive cases and controls, although the two groups did not separate into discrete clusters (Figure 2a,b). To examine the effects of COVID-19 on the plasma proteome, we performed a differential expression analysis in subcohort A between COVID-19 cases (n = 256 samples passing quality control [QC] from 55 patients) and non-infected ESKD controls (n = 51) using linear mixed models, which account for serial samples from the same individual (see Materials and methods). This revealed 221 proteins associated with COVID-19 (5% false discovery rate, FDR); the vast majority were upregulated, with only 40 downregulated (Figure 3a, Supplementary file 1c). In order to provide a succinct and standardised nomenclature, we report proteins by the symbols of the genes encoding them (see Supplementary file 1a for a mapping of symbols to full protein names). The most strongly upregulated proteins (in terms of fold change) were DDX58, CCL7, IL6, CXCL11, KRT19, and CXCL10, and the most strongly downregulated were SERPINA5, CCL16, FABP2, PON3, ITGA11, and MMP12 (Figure 3—figure supplement 1). Notably, many of the upregulated proteins were chemotaxins.

Figure 2 with 2 supplements see all

Download asset Open asset

Figure 3 with 4 supplements see all

Download asset Open asset

We observed that a high proportion of the measured proteins were associated with COVID-19. Given the highly targeted nature of the Olink panels that we used (enriched for immune and inflammation-related proteins), this was not surprising. Nevertheless, to ensure that the Benjamini–Hochberg adjustment of p-values was controlling the FDR at the 5% level, we performed two additional analyses (see Materials and methods). First, we estimated the FDR using an alternative method (the plug-in procedure ; Hastie et al., 2001); this confirmed appropriate FDR control. Second, we used permutation to estimate the distribution of the number of proteins expected to be declared significant under the null hypothesis of no association between any proteins and COVID-19. This showed that the probability of observing the number of differentially abundant proteins we identified was highly unlikely under the null (empirical p<1×10⁻⁵; Figure 3—figure supplement 2).

Although our COVID-19-negative controls were well matched in terms of age, sex, and ethnicity (Figure 1—figure supplement 1a–c), perfect matching of comorbidities was not feasible in the context of the healthcare emergency at the time of patient recruitment. There was a higher prevalence of diabetes in the COVID-19 cases compared to the controls (61.8% versus 47.1%, respectively; Table 1). To evaluate whether differing rates of diabetes had impacted the proteins identified as differentially abundant between cases and controls, we performed a sensitivity analysis adding diabetes as an additional covariate in the linear mixed model. This did not materially affect our findings; estimated effect sizes and –log10 p-values from models with and without the inclusion of diabetes were highly correlated (Pearson r > 0.99, and r = 0.95, respectively; Figure 3—figure supplement 3a,b). Full results from both models are shown in Supplementary file 1c. Similarly, there were also differences in the underlying cause of ESKD in cases compared to controls (Table 1). We therefore performed a further sensitivity analysis adjusting for underlying cause of renal failure. This did not make any meaningful difference to our results (Figure 3—figure supplement 3c,d, Supplementary file 1c).

We also considered the possibility that timing of haemodialysis might affect the plasma proteome. To minimise the impact of this, all samples were taken prior to haemodialysis. For the large majority (86.6%) of samples, the most recent haemodialysis was between 48 and 72 hr prior to blood draw. This consistency in timing of blood sampling reduces the potential for impact of this issue. Nevertheless, to evaluate whether timing of haemodialysis might have impacted our results, we performed a sensitivity analysis including time from last haemodialysis as a covariate. Our results were not materially affected by this, with −log10 p-values and estimated effect sizes very highly correlated with those obtained without inclusion of this covariate (Pearson r > 0.99 for effect size estimates and for −log10 p-values; Figure 3—figure supplement 4a,b, Supplementary file 1c).

We used the smaller subcohort B (n = 52 serum samples from 46 patients with COVID-19; see Materials and methods) for validation. We first projected the data from subcohort B into the PCA space of subcohort A to examine how well the separation of cases and controls in the PCA space replicated (see Materials and methods). This revealed clearer separation of infected and non-infected patients than in subcohort A (Figure 2c,d), perhaps reflecting the higher proportion of hospitalised patients (41 of 46 patients) in subcohort B (Table 2). We next performed differential abundance analysis in subcohort B and found 201 proteins that were dysregulated in cases versus controls (5% FDR) (Supplementary file 1c). Of the 221 differentially abundant proteins from subcohort A, 150 (69.7%) were also identified in subcohort B at 5% FDR (Figure 4a). Effect sizes in each dataset showed a strong correlation (r = 0.80, Figure 4b). This demonstrates that our findings are highly reproducible despite differences in sample sizes and blood materials (plasma versus serum in subcohorts A and B, respectively).

Figure 4 with 1 supplement see all

Download asset Open asset

Proteins associated with COVID-19 severity

Examination of the principal components plot labelling samples by clinical severity at the time of sampling (defined by WHO severity scores, graded as mild, moderate, severe, or critical) demonstrated a gradient of COVID-19 severity, best captured by principal components 1 and 3 (Figure 2—figure supplement 1a). To determine the proteomic effects of COVID-19 severity, we tested for associations between proteins and WHO severity score at the time of blood sampling, using linear mixed models with severity encoded as an ordinal predictor (see Materials and methods). This analysis revealed 203 proteins associated with severity (Figure 3b, Supplementary file 1d). The majority of these were upregulated in more severe disease, with only 42 downregulated. A sensitivity analysis adjusting for time since last haemodialysis made no significant impact on our results (Figure 3—figure supplement 4c,d, Supplementary file 1d). Consistent with previous reports, we found that severe COVID-19 was characterised by elevated IL6. In addition, we observed a signature of upregulated monocyte chemokines (e.g. CCL2, CCL7, CXCL10), neutrophil activation and degranulation (e.g. PRTN3, MPO), and epithelial injury (e.g. KRT19, AREG, PSIP1, GRN). (Figures 3b,c and 5). SERPINA5 and leptin showed the greatest downregulation as COVID-19 severity increased (Figure 3b,c).

Figure 5

Download asset Open asset

We next asked how does the COVID-19 severity protein signature relate to the proteins that are differentially abundant between cases and controls? The majority (140/203; 69%) of severity-associated proteins were also identified as differentially abundant in the COVID-19-positive versus -negative analysis (Figure 6a). Log fold changes for proteins in COVID-19 versus non-infected patients were correlated with effect sizes in the severity analysis, such that the proteins most upregulated in cases versus controls also tended to show the greatest upregulation in severe disease (Figure 6b). However, there were some notable exceptions (e.g. CCL20, IL17C, OSM) that were strongly associated with severity, but not differentially expressed in infected versus non-infected patients (Figure 6c).

Figure 6

Download asset Open asset

Supervised learning to predict COVID-19 severity

PCA revealed that some samples from patients who had mild or moderate disease at the time of sampling clustered with samples from patients with severe disease (Figure 2—figure supplement 1a). Examination of the same PCA plot labelling samples according to the patient’s overall clinical course (measured by peak WHO severity score over the duration of the illness) (Figure 2—figure supplement 1b) revealed that these samples came from individuals who subsequently developed severe or critical disease. This suggested that molecular changes may predate clinical deterioration. To evaluate this further, we used supervised learning approaches to test whether the proteomic signature of the first blood sample for each patient in our dataset could identify whether the patient either had clinically severe COVID-19 at the time of sampling or would develop severe disease in the future. Whereas differential expression analyses consider each protein marker separately, machine-learning techniques allow examination of all proteins concurrently, thus capturing non-linear relationships in the dataset. Using Random Forests, we trained a classifier on the first sample for each COVID-19 patient to predict the overall clinical course, defined by peak WHO severity. For the purposes of this analysis, we binarised clinical course into either WHO mild/moderate or severe/critical.

The Random Forests method achieved 71% accuracy in predicting peak severity. By contrast, using only clinically available predictors (demographics, comorbidities, and clinical laboratory results), the Random Forests method achieved 66% accuracy in predicting peak severity. Combining clinical parameters plus proteins did not improve accuracy (71%) compared to using proteomic predictors alone, suggesting that the information contained in the clinical predictors is captured at the proteomic level. While we do not believe that proteomic profiling is likely to enter clinical practice for risk stratification during this pandemic, the features selected by the classifier can highlight proteins of biological importance. We therefore interrogated the model to identify key proteins by calculating feature importance metrics (see Materials and methods, Supplementary file 1e). The most important proteins for indicating the presence of current or future severe disease were IL18BP, CTSD (Cathepsin D), GDF15, KRT19, TNFSF11, and IL1RL1 (ST2) (Figure 7a). It is notable that through this distinct analytical approach, KRT19 again emerged as a key biomarker of severe disease.

Figure 7 with 1 supplement see all

Download asset Open asset

Associations with clinical laboratory tests

A number of routine clinical laboratory tests have well-characterised associations with COVID-19 (e.g. elevated inflammatory markers, d-dimer, and reduced lymphocyte count) (Guan et al., 2020). We therefore compared our proteomic data from COVID-19 patients at each timepoint to contemporaneous clinical laboratory measurements using linear mixed models (see Materials and methods). We found associations between plasma proteins and all clinical laboratory measurements except troponin (Figure 8, Supplementary file 1g). Many of these proteins were also markers of severity (e.g. IL6, KRT19, IFN-gamma, and CXCL10 were strongly associated with raised CRP and ferritin and reduced lymphocyte counts). Of note, CCL7, a monocyte chemokine that was also identified as an important marker of severity by the Random Forests classifier, was associated with lower monocyte count and raised inflammatory markers. Elevated neutrophil count was associated with Oncostatin-M, which regulates IL6, GCSF, and GMCSF production, and with the proteases MMP9 and defensin.

Figure 8

Download asset Open asset

Longitudinal analysis reveals proteins with distinct temporal profiles according to severity

The immune response to infection is dynamic, and therefore snapshot measurements provide only partial insights. Leveraging the dense serial sampling in our dataset (Figure 1), we modelled the temporal trajectory of each protein and asked whether or not any protein trajectories differed in patients with a severe/critical versus mild/moderate overall clinical course. This was achieved using linear mixed models that included a term for time from first symptoms and a time × severity interaction term (see Materials and methods).

One hundred and seventy-eight proteins displayed a significant association with time from first symptoms (5% FDR), demonstrating the temporal variability in plasma proteins across the disease course (Supplementary file 1h). Moreover, we identified 32 proteins for which there was significant interaction between time and severity, that is, proteins displaying differential temporal trajectories between mild/moderate and severe/critical infections (Supplementary file 1h, Figure 9). Among the proteins with the strongest temporal differences according to clinical course were the integrins ITGA11 and ITGB6, the adhesion molecule ICAM1, TNFRSF10B (a receptor for TRAIL), and PLAUR (the receptor for urokinase plasminogen activator). Most of these proteins exhibited rising profiles in the more severe patients but flat profiles in milder cases. ACE2, the receptor for SARS-CoV-2, also displayed this pattern (Figure 9). In contrast, abundance of ITGA11, which was also identified as reduced in the analysis of infected versus non-infected patients, fell over time in the severe group.

Figure 9 with 2 supplements see all

Download asset Open asset

In this study, we performed plasma proteomic profiling of haemodialysis patients with COVID-19. A strength of our study was that we were able to perform serial blood sampling in both the outpatient and inpatient settings, including longitudinal samples from the same individual before and after hospitalisation. This was possible because haemodialysis patients are unable to fully isolate as they must continue to attend for regular dialysis sessions. Moreover, haemodialysis patients represent an important group since ESKD is one of the strongest risk factors for death from COVID-19 (Williamson et al., 2020; Corbett et al., 2020; Ng et al., 2020; Valeri et al., 2020). Data from the UK Renal Registry shows that 7 and 14 day mortality for COVID-19-infected in-centre haemodialysis patients was 11% and 19%, respectively (COVID-19 Data, 2020). Data from the Scottish Renal Registry estimates 30 day mortality following a positive COVID-19 test as 22%, and as of 31 May 2020, 28.2% of renal replacement therapy patients who had a positive COVID-19 test had died (Scottish Renal Registry COVID-19 report, 2020). In our local population of 1352 in-centre haemodialysis patients, 315 patients had tested positive for COVID-19 by the end of our study period (31 May 2020), of whom 53% required hospitalisation and 85 (27%) died. The OpenSAFELY study (Williamson et al., 2020) examined ~17 million UK primary care records and linked these to the UK COVID-19 mortality register. Patients with estimated glomerular filtration rate (eGFR) < 30 ml/min/1.73 m² had a hazard ratio (HR) for death of 3.56 after adjustment for age and sex.

In part, the high mortality from COVID-19 in ESKD patients likely reflects the fact that these patients are enriched for cardiometabolic traits that predispose to severe COVID-19. However, in multivariable analyses adjusting for these factors, impaired renal function remains an independent risk factor for severe COVID-19 (Williamson et al., 2020). Moreover, there is an inverse relationship between renal function and risk of death from COVID-19 across the spectrum of chronic kidney disease. These observations support the notion that the state of ESKD per se is an important determinant of outcome in COVID-19. ESKD is well-recognised as an immunosuppressed state (Eiselt et al., 2016; Girndt et al., 1999; Sarnak and Jaber, 2000), with defects in both innate and adaptive immunity (Alexiewicz et al., 1991; Massry and Smogorzewski, 2001; Girndt et al., 2001; Meier et al., 2002). Accordingly, ESKD confers increased vulnerability to viral infections including influenza and respiratory syncytial virus (Betjes, 2013; Boattini et al., 2020; Prasad et al., 2020; Usvyat et al., 2012). In addition, ESKD results in a chronic low-grade inflammatory state (Carrero and Stenvinkel, 2010). This tendency to a pro-inflammatory state, combined with reduced ability to respond to viruses, may contribute to the abnormal host response to SARS-CoV-2 infection, producing the immunopathology that leads to severe COVID-19.

Our comparison of COVID-19-positive and -negative haemodialysis patient plasma samples revealed 221 proteins that were differentially abundant in COVID-19. The majority of these were upregulated, with strong representation of viral response proteins (e.g. DDX58, IFNG), cytokines/chemokines (e.g. IL6, CCL7, CXCL10, and CXCL11), and epithelial proteins (e.g. KRT19, PSIP1) (Figure 3a). The COVID-19-negative controls in this analysis were carefully matched to cases in terms of age, sex, and ethnicity. However, complete matching of clinical characteristics was not feasible; there were differences in the prevalence of diabetes and the underlying causes of ESKD between COVID-19-positive cases and controls (Table 1). Sensitivity analyses adjusting for these covariates gave highly consistent results, indicating that our findings are robust. In addition, we validated our findings when we analysed serum samples from a separate subcohort of COVID-19-positive ESKD patients.

ESKD is itself likely to significantly impact the plasma proteome. Previous cross-sectional studies have shown that the levels of many circulating proteins have an inverse relationship with eGFR (Naseeb et al., 2015; Christensson et al., 2018). A longitudinal study using an Olink proteomics panel (although not one used in our study) found that for 74% of the 84 proteins measured, protein levels rose as eGFR fell (Lind et al., 2019). For many proteins, it is unclear whether this inverse relationship with renal function reflects cause or effect. Some proteins may be increased in chronic kidney disease due to reduced renal clearance, some may be elevated secondary to tissue injury or chronic inflammation, and others may be drivers of renal injury. Regardless, this observation of widespread changes in the blood proteome of kidney disease patients emphasises the importance of using COVID-19 ESKD patients rather than healthy individuals as our control group.

Analysis within COVID-19 cases revealed 203 proteins associated with disease severity, the strongest of which was IL6 (Figure 3b). Association of IL6 with severe disease is well-established and has already received considerable attention (Wu et al., 2020; Li et al., 2020). Despite promising initial case reports of IL6R receptor blockade in COVID-19, convincing efficacy was not demonstrated in early randomised trials (Furlow, 2020). More recently, the REMAP-CAP trial has shown the benefit of anti-IL6R therapy when given to critically ill patients on admission to intensive care units (Gordon et al., 2021), indicating that IL6 does contribute to critical illness from COVID-19. Our finding that IL6 was most strongly upregulated in severe disease demonstrates the value of plasma proteomic profiling in identifying putative drug targets.

Members of the CCL and CXCL chemokine families (e.g. CCL2, CCL7, CCL20, and CXCL10) were strongly associated with severity. Likewise, higher levels of CCL2, CCL7, CCL20, and CXCL10 were associated with lower blood lymphocyte count and higher inflammatory markers (CRP and ferritin) (Supplementary file 1g), which are clinical markers of severe disease and poorer outcome in COVID-19 (Gupta et al., 2021). Of note, CCL20 is a chemoattractant for lymphocytes (Schutyser et al., 2003), and its negative association with lymphocyte count may reflect a direct effect on migration of lymphocytes from the blood into the tissues rather than simply marking severe disease. CCL2 (also known as MCP-1) and CCL7 (MCP-3) are both chemokines for monocytes, and CXCL10 has pleiotropic immunological effects including chemotaxis. These chemokines were also negatively correlated with blood monocyte count, suggesting recruitment of these innate immune cells into damaged tissues.

The neutrophil proteases PRTN3 (proteinase-3) and MPO (myeloperoxidase) (Figure 5) and the neutrophil-derived protein AZU1 were associated with severe disease (Supplementary file 1d), indicating that neutrophil activation and degranulation are features of severe COVID-19. Degranulation of neutrophils releasing PRTN3 and MPO could potentially contribute to oxidative damage in the lungs and thus more severe disease.

A striking finding of our study was the association of disease severity with upregulation of epithelial proteins (e.g. KRT19) and epithelial tissue repair pathways (e.g. PSIP1, AREG, GRN [progranulin]), most likely reflecting lung and vascular damage. KRT19 was notably prominent in our analyses, as well as the study by Filbin et al., 2020; Supplementary file 1j. KRT19 is an intermediate filament protein, important for the structural integrity of epithelial cells (Saha et al., 2017). These data suggest that severe COVID-19 is characterised by destruction of the lung epithelium and vascular endothelium. Vascular injury might thus explain the high level of vascular thrombosis seen in patients in severe disease. In summary, our data reveal that severe COVID-19 is characterised proteomically by a signature of innate immune activation and epithelial injury.

Sixty-nine percent of proteins associated with severity were also differentially abundant in the case versus control analysis (Figure 6a), and for the large majority of proteins the within-case severity analysis, effect size was proportional to the fold change between cases and controls (Figure 6b). This suggests that, in general, the distinction in the plasma proteome between severe and mild COVID-19 is a quantitative difference in the COVID-19 signature, rather than there being an orthogonal signature involving a different set of proteins. Consistent with this concept, examination of PCA plots coloured by severity revealed that while there was a gradient of COVID-19 severity, the samples from severe or critical patients did not form a discrete cluster distinct from those from patients with milder disease (Figure 2—figure supplement 1). However, there were a few exceptions where proteins that were associated with severity were not upregulated in the case–control analysis. These included OSM, IL17C, and CCL20 (Figure 6c). These proteins therefore reflect biological processes specifically of severe disease and may represent therapeutic targets.

Survival analysis identified 44 proteins associated with increased risk of death (Figure 7b). As expected, many of these were also associated with disease severity, high CRP, and lower lymphocyte count (Figure 7—figure supplement 1). In contrast, 25 proteins were associated with reduced risk of death (Figure 7b). One such protein is the multi-functional cytokine TNFSF12 (TWEAK). Although TWEAK can exert pro-inflammatory effects, it also can inhibit the innate immune response (Maecker et al., 2005) and promote tissue repair and endothelial cell proliferation and survival (Burkly et al., 2007), which may be beneficial responses in COVID-19. This illustrates that although proteins associated with inflammation are often thought to be destructive, the inflammatory response also induces programmes for limiting injury and initiating tissue repair. Insufficient activation of such homeostatic mechanisms may contribute to why some individuals get severe COVID-19.

The host immune response to COVID-19 is a dynamic process, and clinical deterioration typically occurs 7–10 days after first symptoms. Temporal information may therefore be important in determining optimum timing of therapeutic intervention (e.g. blockade of a particular cytokine). By taking serial samples and examining their patterns within individuals over time, we were able to model protein trajectories and found that many proteins display temporal variability during COVID-19. Longitudinal measurements also allow molecular comparison of severe versus mild disease trajectories. By modelling the interaction term between time from first symptoms and overall disease course, we found 32 proteins that displayed distinct temporal profiles in severe versus mild disease. These results point to enhanced leucocyte–endothelial cell interactions indicated by upregulation of cell adhesion molecules (e.g. ITGB6, ICAM1) in severe disease. This endothelial activation may contribute to COVID-19-associated thrombosis discussed above. Management of thrombosis in COVID-19 currently consists of anticoagulation. Our results suggest that disrupting leucocyte–endothelial interactions may be a complementary therapeutic strategy.

Several proteins associated with either risk of death or clinical severity lie in pathways targeted by existing drugs. PARP1 was identified as an important marker of current or future severe COVID-19 and also was associated with risk of death. PARP1 is associated with inflammatory and vascular disease (Henning et al., 2018). PARP1 inhibitors are in use for cancer (Rouleau et al., 2010), and our data suggest that re-purposing of PARP1 inhibition in COVID-19 should be explored further. IL33 was associated with both risk of death and clinical severity, and its receptor IL1RL1 (ST2) was associated with clinical severity and identified as an important predictor of severe clinical course. Monoclonal antibodies against IL33 and its receptor are in late-stage development for asthma (Corren, 2019) and could also be explored in COVID-19. As discussed above, MPO was associated with clinical severity. MPO inhibitors (Galijasevic, 2019) might have a role in reducing neutrophil-mediated tissue injury in COVID-19. Finally, inhibitors of monocyte chemokines (e.g. CCL2) and their receptors have been developed (Vergunst et al., 2008; Haringman et al., 2006), although drugging these pathways is made more challenging by molecular cross-talk. An important caveat is that we cannot determine whether the associations we observed are drivers of pathology in COVID-19 or simply reflect the downstream consequences of inflammation and tissue injury. Future studies using Mendelian randomisation analysis will provide a useful tool for assessing causality and prioritising drug targets.

Other groups have studied the plasma or serum proteome in COVID-19 (Shen et al., 2020; Lucas et al., 2020; Arunachalam et al., 2020; Filbin et al., 2020; Rodriguez et al., 2020), using either mass spectrometry or immunoassays including the Olink platform. Mass spectrometry is less sensitive than immunoassays and so it is likely to be unable to detect many of the cytokines measured here. Conversely, it can provide complementary information by measuring many proteins that our immunoassays did not target. A limitation of our study was that we used Olink panels that measured specific proteins selected on their relevance to inflammation, immunity, cardiovascular, and metabolic disease. This bias precluded formal pathway enrichment analysis of differentially abundant proteins. In general, our results had greater similarities to studies that used immunoassays over mass spectrometry (Supplementary file 1i,j). 47.6% of proteins differentially expressed in COVID-19-positive versus -negative ESKD patients in our study were differentially expressed in COVID-19-positive versus -negative acute respiratory distress syndrome patients in the study of Filbin et al., 2020, who used a different Olink proteomics platform. Moreover, we observed consistent effect sizes (Figure 4—figure supplement 1). These similarities are striking given the difference in clinical populations and control groups; in Filbin et al.’s report, the controls included patients with non-COVID-19 respiratory infections, whereas our control group did not have active infection. The concordance in proteins associated with COVID-19 severity within cases was even higher (81.8%). The similarities suggest a similar plasma proteomic signature of COVID-19 across different clinical populations, particularly the signature associated with severity.

In summary, this study reveals proteins associated with COVID-19 infection and severity and demonstrates altered dynamic profiles between patients with severe disease and those with a more indolent course. Our results emphasise the importance of studying and targeting mechanisms that reduce the lung epithelial and endothelial damage to both alleviate the severity of the infection and reduce the chance of long-lasting complications. These data provide a valuable resource for therapeutic target prioritisation.

All data generated during this study are included in the manuscript and supporting files. Underlying source data for all analyses (individual-level proteomic and clinical phenotyping data) are available without restriction as Source Data Files 1-4. In addition, these data have been deposited in the Dryad Digital Repository (https://doi.org/10.5061/dryad.6t1g1jwxj). Code is available in the following GitHub repository: https://github.com/jackgisby/longitudinal_olink_proteomics copy archived at https://archive.softwareheritage.org/swh:1:rev:32f08137859d44707ec4f086eed9af9b9ee91a87/.

1. 1. Gisby J
   2. Clarke CL
   3. Medjeral-Thomas N
   4. Malik TH
   5. Papadaki A
   6. Mortimer PM
   7. Buang NB
   8. Lewis S
   9. Pereira M
   10. Toulza F
   11. Fagnano E
   12. Mawhin M
   13. Dutton EE
   14. Tapeng L
   15. Kirk P
   16. Behmoaras J
   17. Sandhu E
   18. McAdoo SP
   19. Prendecki MF
   20. Pickering MC
   21. Botto M
   22. Willicombe W
   23. Thomas DC
   24. Peters JE

   (2020) Dryad Digital Repository

   Longitudinal proteomic profiling of high-risk patients with COVID-19 reveals markers of severity and predictors of fatal disease.

   https://doi.org/10.5061/dryad.6t1g1jwxj

1. 1. Alexiewicz JM
   2. Smogorzewski M
   3. Fadda GZ
   4. Massry SG

   (1991) Impaired phagocytosis in Dialysis patients: studies on mechanisms

   American Journal of Nephrology 11:102–111.

   https://doi.org/10.1159/000168284

   • PubMed
   • Google Scholar
2. 1. Arunachalam PS
   2. Wimmers F
   3. Mok CKP
   4. Perera R

   (2020) Systems biological assessment of immunity to mild versus severe COVID-19 infection in humans

   Science 6261:1–18.

   https://doi.org/10.1126/science.abc6261

   • Google Scholar
3. 1. Bakdash JZ
   2. Marusich LR

   (2017) Repeated measures correlation

   Frontiers in Psychology 8:1–13.

   https://doi.org/10.3389/fpsyg.2017.00456

   • PubMed
   • Google Scholar
4. 1. Bates D
   2. Mächler M
   3. Bolker B
   4. Walker S

   (2015) Fitting linear Mixed-Effects models using lme4

   Journal of Statistical Software 67:i01.

   https://doi.org/10.18637/jss.v067.i01

   • Google Scholar
5. 1. Betjes MG

   (2013) Immune cell dysfunction and inflammation in end-stage renal disease

   Nature Reviews Nephrology 9:255–265.

   https://doi.org/10.1038/nrneph.2013.44

   • PubMed
   • Google Scholar
6. 1. Boattini M
   2. Almeida A
   3. Christaki E
   4. Cruz L
   5. Antão D
   6. Moreira MI
   7. Bianco G
   8. Iannaccone M
   9. Tsiolakkis G
   10. Khattab E
   11. Kasapi D
   12. Charrier L
   13. Tosatto V
   14. Marques TM
   15. Cavallo R
   16. Costa C

   (2020) Influenza and respiratory syncytial virus infections in the oldest-old continent

   European Journal of Clinical Microbiology & Infectious Diseases 39:2085–2090.

   https://doi.org/10.1007/s10096-020-03959-9

   • Google Scholar
7. 1. Burkly LC
   2. Michaelson JS
   3. Hahm K
   4. Jakubowski A
   5. Zheng TS

   (2007) TWEAKing tissue remodeling by a multifunctional cytokine: role of TWEAK/Fn14 pathway in health and disease

   Cytokine 40:1–16.

   https://doi.org/10.1016/j.cyto.2007.09.007

   • PubMed
   • Google Scholar
8. 1. Carrero JJ
   2. Stenvinkel P

   (2010) Inflammation in end-stage renal disease--what have we learned in 10 years?

   Seminars in Dialysis 23:498–509.

   https://doi.org/10.1111/j.1525-139X.2010.00784.x

   • PubMed
   • Google Scholar
9. 1. Christensson A
   2. Ash JA
   3. DeLisle RK
   4. Gaspar FW
   5. Ostroff R
   6. Grubb A
   7. Lindström V
   8. Bruun L
   9. Williams SA

   (2018) The impact of the glomerular filtration rate on the human plasma proteome

   PROTEOMICS - Clinical Applications 12:1700067.

   https://doi.org/10.1002/prca.201700067

   • Google Scholar
10. 1. Corbett RW
    2. Blakey S
    3. Nitsch D
    4. Loucaidou M
    5. McLean A
    6. Duncan N
    7. Ashby DR
    8. West London Renal and Transplant Centre

    (2020) Epidemiology of COVID-19 in an urban Dialysis center

    Journal of the American Society of Nephrology 31:1815–1823.

    https://doi.org/10.1681/ASN.2020040534

    • PubMed
    • Google Scholar
11. 1. Corren J

    (2019) New targeted therapies for uncontrolled asthma

    The Journal of Allergy and Clinical Immunology: In Practice 7:1394–1403.

    https://doi.org/10.1016/j.jaip.2019.03.022

    • PubMed
    • Google Scholar
12. Website

    1. COVID-19 Data

    (2020) The renal association

    Accessed May 31, 2020.

    https://renal.org/health-professionals/covid-19/covid-19-data
13. Preprint

    1. Docherty AB
    2. Harrison EM
    3. Green CA
    4. Hardwick H
    5. Pius R
    6. Norman L
    7. Holden KA

    (2020) Features of 16,749 hospitalised UK patients with COVID-19 using the ISARIC WHO clinical characterisation protocol

    medRxiv.

    https://doi.org/10.1101/2020.04.23.20076042

    • Google Scholar
14. 1. Eiselt J
    2. Kielberger L
    3. Rajdl D
    4. Racek J
    5. Pazdiora P
    6. Malánová L

    (2016) Previous vaccination and age are more important predictors of immune response to influenza vaccine than inflammation and iron status in Dialysis patients

    Kidney and Blood Pressure Research 41:139–147.

    https://doi.org/10.1159/000443416

    • PubMed
    • Google Scholar
15. Preprint

    1. Filbin MR
    2. Mehta A
    3. Schneider AM
    4. Kays KR
    5. Guess JR
    6. Gentili M
    7. Fenyves BG
    8. Charland NC
    9. Gonye ALK

    (2020) Plasma proteomics reveals tissue-specific cell death and mediators of cell-cell interactions in severe COVID-19 patients

    bioRxiv.

    https://doi.org/10.1101/2020.11.02.365536

    • Google Scholar
16. 1. Furlow B

    (2020) COVACTA trial raises questions about tocilizumab's benefit in COVID-19

    The Lancet Rheumatology 2:e592.

    https://doi.org/10.1016/S2665-9913(20)30313-1

    • PubMed
    • Google Scholar
17. 1. Galijasevic S

    (2019) The development of myeloperoxidase inhibitors

    Bioorganic & Medicinal Chemistry Letters 29:1–7.

    https://doi.org/10.1016/j.bmcl.2018.11.031

    • PubMed
    • Google Scholar
18. 1. Gandolfo LC
    2. Speed TP

    (2018) RLE plots: visualizing unwanted variation in high dimensional data

    PLOS ONE 13:e0191629.

    https://doi.org/10.1371/journal.pone.0191629

    • PubMed
    • Google Scholar
19. 1. Girndt M
    2. Sester U
    3. Sester M
    4. Kaul H
    5. Köhler H

    (1999) Impaired cellular immune function in patients with end-stage renal failure

    Nephrology Dialysis Transplantation 14:2807–2810.

    https://doi.org/10.1093/ndt/14.12.2807

    • Google Scholar
20. 1. Girndt M
    2. Sester M
    3. Sester U
    4. Kaul H
    5. Köhler H

    (2001) Molecular aspects of T- and B-cell function in Uremia

    Kidney International. Supplement 78:S206–S211.

    https://doi.org/10.1046/j.1523-1755.2001.59780206.x

    • PubMed
    • Google Scholar
21. Software

    1. Gisby J

    (2021) longitudinal_olink_proteomics, version swh:1:rev:32f08137859d44707ec4f086eed9af9b9ee91a87

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:306b6a1229754be4c49864b48829f2d68ba4585c;origin=https://github.com/jackgisby/longitudinal_olink_proteomics;visit=swh:1:snp:bbb30cca001138217cf173ace16f1f3577cf2d89;anchor=swh:1:rev:32f08137859d44707ec4f086eed9af9b9ee91a87/
22. Preprint

    1. Gordon AC
    2. Mouncey PR
    3. Al-Beidh F
    4. Rowan KM
    5. Nichol AD
    6. Arabi YM
    7. Annane D
    8. Beane A

    (2021) Interleukin-6 receptor antagonists in critically ill patients with Covid-19 - Preliminary report

    medRxiv.

    https://doi.org/10.1101/2021.01.07.21249390

    • Google Scholar
23. 1. Guan WJ
    2. Ni ZY
    3. Hu Y
    4. Liang WH
    5. Ou CQ
    6. He JX
    7. Liu L
    8. Shan H
    9. Lei CL
    10. Hui DSC
    11. Du B
    12. Li LJ
    13. Zeng G
    14. Yuen KY
    15. Chen RC
    16. Tang CL
    17. Wang T
    18. Chen PY
    19. Xiang J
    20. Li SY
    21. Wang JL
    22. Liang ZJ
    23. Peng YX
    24. Wei L
    25. Liu Y
    26. Hu YH
    27. Peng P
    28. Wang JM
    29. Liu JY
    30. Chen Z
    31. Li G
    32. Zheng ZJ
    33. Qiu SQ
    34. Luo J
    35. Ye CJ
    36. Zhu SY
    37. Zhong NS
    38. China Medical Treatment Expert Group for Covid-19

    (2020) Clinical characteristics of coronavirus disease 2019 in China

    New England Journal of Medicine 382:1708–1720.

    https://doi.org/10.1056/NEJMoa2002032

    • PubMed
    • Google Scholar
24. 1. Gupta RK
    2. Harrison EM
    3. Ho A
    4. Docherty AB
    5. Knight SR

    (2021) Development and validation of the ISARIC 4C deterioration model for adults hospitalised with COVID-19: a prospective cohort study

    Lancet Respiratory Medicine 11:30559-2.

    https://doi.org/10.1016/S2213-2600(20)30559-2

    • Google Scholar
25. 1. Haringman JJ
    2. Gerlag DM
    3. Smeets TJ
    4. Baeten D
    5. van den Bosch F
    6. Bresnihan B
    7. Breedveld FC
    8. Dinant HJ
    9. Legay F
    10. Gram H
    11. Loetscher P
    12. Schmouder R
    13. Woodworth T
    14. Tak PP

    (2006) A randomized controlled trial with an anti-CCL2 (anti-monocyte chemotactic protein 1) monoclonal antibody in patients with rheumatoid arthritis

    Arthritis & Rheumatism 54:2387–2392.

    https://doi.org/10.1002/art.21975

    • PubMed
    • Google Scholar
26. Book

    1. Hastie T
    2. Tibshirani R
    3. Friedman J

    (2001)

    High-Dimensional Problems: p ≫ N

    In: Hastie T, Tibshirani R, Friedman J, editors. The Elements of Statistical Learning. Springer Series in Statistics. pp. 649–698.

    • Google Scholar
27. 1. Henning RJ
    2. Bourgeois M
    3. Harbison RD

    (2018) Poly(ADP-ribose) Polymerase (PARP) and PARP inhibitors: mechanisms of action and role in cardiovascular disorders

    Cardiovascular Toxicology 18:493–506.

    https://doi.org/10.1007/s12012-018-9462-2

    • PubMed
    • Google Scholar
28. 1. Horby P
    2. Lim WS
    3. Emberson JR
    4. Mafham M
    5. Bell JL
    6. Linsell L
    7. Staplin N
    8. RECOVERY Collaborative Group

    (2020)

    Dexamethasone in hospitalized patients with Covid-19 — Preliminary Report

    The New England Journal of Medicine 384:693–704.

    • Google Scholar
29. 1. Ibrahim JG
    2. Chu H
    3. Chen LM

    (2010) Basic concepts and methods for joint models of longitudinal and survival data

    Journal of Clinical Oncology 28:2796–2801.

    https://doi.org/10.1200/JCO.2009.25.0654

    • Google Scholar
30. 1. Kuhn M

    (2008) Building predictive models in R using the caret package

    Journal of Statistical Software 28:i05.

    https://doi.org/10.18637/jss.v028.i05

    • Google Scholar
31. 1. Kuznetsova A
    2. Brockhoff PB
    3. Christensen RHB

    (2017) lmerTest Package: Tests in Linear Mixed Effects Models

    Journal of Statistical Software 82:i13.

    https://doi.org/10.18637/jss.v082.i13

    • Google Scholar
32. 1. Leo B

    (2001) Random forests

    Machine Learning 45:5–32.

    https://doi.org/10.1023/A:1010933404324

    • Google Scholar
33. 1. Li X
    2. Xu S
    3. Yu M
    4. Wang K
    5. Tao Y
    6. Zhou Y
    7. Shi J
    8. Zhou M
    9. Wu B
    10. Yang Z
    11. Zhang C
    12. Yue J
    13. Zhang Z
    14. Renz H
    15. Liu X
    16. Xie J
    17. Xie M
    18. Zhao J

    (2020) Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan

    Journal of Allergy and Clinical Immunology 146:110–118.

    https://doi.org/10.1016/j.jaci.2020.04.006

    • PubMed
    • Google Scholar
34. 1. Lind L
    2. Sundström J
    3. Larsson A
    4. Lampa E
    5. Ärnlöv J
    6. Ingelsson E

    (2019) Longitudinal effects of aging on plasma proteins levels in older adults - associations with kidney function and hemoglobin levels

    PLOS ONE 14:e0212060.

    https://doi.org/10.1371/journal.pone.0212060

    • PubMed
    • Google Scholar
35. 1. Lucas C
    2. Wong P
    3. Klein J
    4. Castro TBR
    5. Silva J
    6. Sundaram M
    7. Ellingson MK
    8. Mao T
    9. Oh JE
    10. Israelow B
    11. Takahashi T
    12. Tokuyama M
    13. Lu P
    14. Venkataraman A
    15. Park A
    16. Mohanty S
    17. Wang H
    18. Wyllie AL
    19. Vogels CBF
    20. Earnest R
    21. Lapidus S
    22. Ott IM
    23. Moore AJ
    24. Muenker MC
    25. Fournier JB
    26. Campbell M
    27. Odio CD
    28. Casanovas-Massana A
    29. Herbst R
    30. Shaw AC
    31. Medzhitov R
    32. Schulz WL
    33. Grubaugh ND
    34. Dela Cruz C
    35. Farhadian S
    36. Ko AI
    37. Omer SB
    38. Iwasaki A
    39. Yale IMPACT Team

    (2020) Longitudinal analyses reveal immunological misfiring in severe COVID-19

    Nature 584:463–469.

    https://doi.org/10.1038/s41586-020-2588-y

    • PubMed
    • Google Scholar
36. 1. Maecker H
    2. Varfolomeev E
    3. Kischkel F
    4. Lawrence D
    5. LeBlanc H
    6. Lee W
    7. Hurst S
    8. Danilenko D
    9. Li J
    10. Filvaroff E
    11. Yang B
    12. Daniel D
    13. Ashkenazi A

    (2005) TWEAK attenuates the transition from innate to adaptive immunity

    Cell 123:931–944.

    https://doi.org/10.1016/j.cell.2005.09.022

    • PubMed
    • Google Scholar
37. 1. Massry S
    2. Smogorzewski M

    (2001) Dysfunction of polymorphonuclear leukocytes in Uremia: role of parathyroid hormone

    Kidney International. Supplement 78:S195–S196.

    https://doi.org/10.1046/j.1523-1755.2001.59780195.x

    • PubMed
    • Google Scholar
38. 1. Meier P
    2. Dayer E
    3. Blanc E
    4. Wauters JP

    (2002)

    Early T cell activation correlates with expression of apoptosis markers in patients with end-stage renal disease

    Journal of the American Society of Nephrology : JASN 13:204–212.

    • PubMed
    • Google Scholar
39. 1. Naseeb U
    2. Axelsson J
    3. Jägerbrink T
    4. Shafqat J
    5. Zarina S
    6. Jörnvall H

    (2015)

    Complementary LC-MS/MS proteomic analysis of uremic plasma proteins

    Journal of the College of Physicians and Surgeons--Pakistan : JCPSP 25:606–609.

    • PubMed
    • Google Scholar
40. 1. Ng JH
    2. Hirsch JS
    3. Wanchoo R
    4. Sachdeva M
    5. Sakhiya V
    6. Hong S
    7. Jhaveri KD
    8. Fishbane S
    9. Northwell COVID-19 Research Consortium and the Northwell Nephrology COVID-19 Research Consortium

    (2020) Outcomes of patients with end-stage kidney disease hospitalized with COVID-19

    Kidney International 98:1530–1539.

    https://doi.org/10.1016/j.kint.2020.07.030

    • PubMed
    • Google Scholar
41. 1. Perperoglou A
    2. Sauerbrei W
    3. Abrahamowicz M
    4. Schmid M

    (2019) A review of spline function procedures in R

    BMC Medical Research Methodology 19:1–16.

    https://doi.org/10.1186/s12874-019-0666-3

    • PubMed
    • Google Scholar
42. 1. Prasad N
    2. Walker TA
    3. Waite B
    4. Wood T
    5. Trenholme AA
    6. Baker MG
    7. McArthur C
    8. Wong CA
    9. Grant CC
    10. Huang QS
    11. Newbern EC

    (2020) Respiratory syncytial Virus–Associated Hospitalizations Among Adults With Chronic Medical Conditions

    Clinical Infectious Diseases 352:ciaa730.

    https://doi.org/10.1093/cid/ciaa730

    • Google Scholar
43. 1. Rizopoulos D

    (2010) JM: an R package for the joint modelling of longitudinal and time-to-event data

    Journal of Statistical Software 35:1–33.

    https://doi.org/10.1111/j.1541-0420.2010.01546.x

    • Google Scholar
44. 1. Rodriguez L
    2. Pekkarinen PT
    3. Lakshmikanth T
    4. Tan Z
    5. Consiglio CR
    6. Pou C
    7. Chen Y
    8. Mugabo CH
    9. Nguyen NA
    10. Nowlan K
    11. Strandin T
    12. Levanov L
    13. Mikes J
    14. Wang J
    15. Kantele A
    16. Hepojoki J
    17. Vapalahti O
    18. Heinonen S
    19. Kekäläinen E
    20. Brodin P

    (2020) Systems-Level immunomonitoring from acute to recovery phase of severe COVID-19

    Cell Reports Medicine 1:100078.

    https://doi.org/10.1016/j.xcrm.2020.100078

    • Google Scholar
45. 1. Rouleau M
    2. Patel A
    3. Hendzel MJ
    4. Kaufmann SH
    5. Poirier GG

    (2010) PARP inhibition: parp1 and beyond

    Nature Reviews Cancer 10:293–301.

    https://doi.org/10.1038/nrc2812

    • PubMed
    • Google Scholar
46. 1. Saha SK
    2. Choi HY
    3. Kim BW
    4. Dayem AA
    5. Yang GM
    6. Kim KS
    7. Yin YF
    8. Cho SG

    (2017) KRT19 directly interacts with β-catenin/RAC1 complex to regulate NUMB-dependent NOTCH signaling pathway and breast Cancer properties

    Oncogene 36:332–349.

    https://doi.org/10.1038/onc.2016.221

    • PubMed
    • Google Scholar
47. 1. Sarnak MJ
    2. Jaber BL

    (2000) Mortality caused by Sepsis in patients with end-stage renal disease compared with the general population

    Kidney International 58:1758–1764.

    https://doi.org/10.1111/j.1523-1755.2000.00337.x

    • PubMed
    • Google Scholar
48. 1. Schutyser E
    2. Struyf S
    3. Van Damme J

    (2003) The CC chemokine CCL20 and its receptor CCR6

    Cytokine & Growth Factor Reviews 14:409–426.

    https://doi.org/10.1016/S1359-6101(03)00049-2

    • PubMed
    • Google Scholar
49. Website

    1. Scottish Renal Registry COVID-19 report

    (2020) Public Health Scotland

    Accessed May 31, 2020.

    https://beta.isdscotland.org/find-publications-and-data/population-health/covid-19/scottish-renal-registry-covid-19-report/
50. 1. Shen B
    2. Yi X
    3. Sun Y
    4. Bi X
    5. Du J
    6. Zhang C
    7. Quan S
    8. Zhang F
    9. Sun R
    10. Qian L
    11. Ge W
    12. Liu W
    13. Liang S
    14. Chen H
    15. Zhang Y
    16. Li J
    17. Xu J
    18. He Z
    19. Chen B
    20. Wang J
    21. Yan H
    22. Zheng Y
    23. Wang D
    24. Zhu J
    25. Kong Z
    26. Kang Z
    27. Liang X
    28. Ding X
    29. Ruan G
    30. Xiang N
    31. Cai X
    32. Gao H
    33. Li L
    34. Li S
    35. Xiao Q
    36. Lu T
    37. Zhu Y
    38. Liu H
    39. Chen H
    40. Guo T

    (2020) Proteomic and metabolomic characterization of COVID-19 patient sera

    Cell 182:59–72.

    https://doi.org/10.1016/j.cell.2020.05.032

    • PubMed
    • Google Scholar
51. 1. Suhre K
    2. McCarthy MI
    3. Schwenk JM

    (2021) Genetics meets proteomics: perspectives for large population-based studies

    Nature Reviews Genetics 22:19–37.

    https://doi.org/10.1038/s41576-020-0268-2

    • PubMed
    • Google Scholar
52. 1. Szklarczyk D
    2. Gable AL
    3. Lyon D
    4. Junge A
    5. Wyder S
    6. Huerta-Cepas J
    7. Simonovic M
    8. Doncheva NT
    9. Morris JH
    10. Bork P
    11. Jensen LJ
    12. Mering CV

    (2019) STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets

    Nucleic Acids Research 47:D607–D613.

    https://doi.org/10.1093/nar/gky1131

    • PubMed
    • Google Scholar
53. 1. Uhlén M
    2. Fagerberg L
    3. Hallström BM
    4. Lindskog C
    5. Oksvold P
    6. Mardinoglu A
    7. Sivertsson Å
    8. Kampf C
    9. Sjöstedt E
    10. Asplund A
    11. Olsson I
    12. Edlund K
    13. Lundberg E
    14. Navani S
    15. Szigyarto CA
    16. Odeberg J
    17. Djureinovic D
    18. Takanen JO
    19. Hober S
    20. Alm T
    21. Edqvist PH
    22. Berling H
    23. Tegel H
    24. Mulder J
    25. Rockberg J
    26. Nilsson P
    27. Schwenk JM
    28. Hamsten M
    29. von Feilitzen K
    30. Forsberg M
    31. Persson L
    32. Johansson F
    33. Zwahlen M
    34. von Heijne G
    35. Nielsen J
    36. Pontén F

    (2015) Proteomics Tissue-based map of the human proteome

    Science 347:1260419.

    https://doi.org/10.1126/science.1260419

    • PubMed
    • Google Scholar
54. 1. Usvyat LA
    2. Carter M
    3. Thijssen S
    4. Kooman JP
    5. van der Sande FM
    6. Zabetakis P
    7. Balter P
    8. Levin NW
    9. Kotanko P

    (2012) Seasonal variations in mortality, clinical, and laboratory parameters in hemodialysis patients: a 5-year cohort study

    Clinical Journal of the American Society of Nephrology 7:108–115.

    https://doi.org/10.2215/CJN.03880411

    • PubMed
    • Google Scholar
55. 1. Valeri AM
    2. Robbins-Juarez SY
    3. Stevens JS
    4. Ahn W
    5. Rao MK
    6. Radhakrishnan J
    7. Gharavi AG
    8. Mohan S
    9. Husain SA

    (2020) Presentation and outcomes of patients with ESKD and COVID-19

    Journal of the American Society of Nephrology 31:1409–1415.

    https://doi.org/10.1681/ASN.2020040470

    • PubMed
    • Google Scholar
56. 1. Vergunst CE
    2. Gerlag DM
    3. Lopatinskaya L
    4. Klareskog L
    5. Smith MD
    6. van den Bosch F
    7. Dinant HJ
    8. Lee Y
    9. Wyant T
    10. Jacobson EW
    11. Baeten D
    12. Tak PP

    (2008) Modulation of CCR2 in rheumatoid arthritis: a double-blind, randomized, placebo-controlled clinical trial

    Arthritis & Rheumatism 58:1931–1939.

    https://doi.org/10.1002/art.23591

    • PubMed
    • Google Scholar
57. 1. Williamson EJ
    2. Walker AJ
    3. Bhaskaran K
    4. Bacon S
    5. Bates C
    6. Morton CE
    7. Curtis HJ
    8. Mehrkar A
    9. Evans D
    10. Inglesby P
    11. Cockburn J
    12. McDonald HI
    13. MacKenna B
    14. Tomlinson L
    15. Douglas IJ
    16. Rentsch CT
    17. Mathur R
    18. Wong AYS
    19. Grieve R
    20. Harrison D
    21. Forbes H
    22. Schultze A
    23. Croker R
    24. Parry J
    25. Hester F
    26. Harper S
    27. Perera R
    28. Evans SJW
    29. Smeeth L
    30. Goldacre B

    (2020) Factors associated with COVID-19-related death using OpenSAFELY

    Nature 584:430–436.

    https://doi.org/10.1038/s41586-020-2521-4

    • PubMed
    • Google Scholar
58. 1. Wu C
    2. Chen X
    3. Cai Y
    4. Xia J
    5. Zhou X
    6. Xu S
    7. Huang H
    8. Zhang L
    9. Zhou X
    10. Du C
    11. Zhang Y
    12. Song J
    13. Wang S
    14. Chao Y
    15. Yang Z
    16. Xu J
    17. Zhou X
    18. Chen D
    19. Xiong W
    20. Xu L
    21. Zhou F
    22. Jiang J
    23. Bai C
    24. Zheng J
    25. Song Y

    (2020) Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China

    JAMA Internal Medicine 180:934–943.

    https://doi.org/10.1001/jamainternmed.2020.0994

    • PubMed
    • Google Scholar

Author details

1. Jack Gisby

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Data curation, Formal analysis, Investigation, Visualization, Methodology, Writing - original draft

   Contributed equally with

   Candice L Clarke and Nicholas Medjeral-Thomas

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-0511-8123

2. Candice L Clarke

   1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
   2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

   Contribution

   Data curation, Investigation, Patient recruitment and sample collection, Clinical phenotyping

   Contributed equally with

   Jack Gisby and Nicholas Medjeral-Thomas

   Competing interests

   No competing interests declared

3. Nicholas Medjeral-Thomas

   1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
   2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

   Contribution

   Data curation, Investigation, Patient recruitment and sample collection, Clinical phenotyping

   Contributed equally with

   Jack Gisby and Candice L Clarke

   Competing interests

   No competing interests declared

4. Talat H Malik

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Investigation, Sample processing

   Competing interests

   No competing interests declared

5. Artemis Papadaki

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

6. Paige M Mortimer

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Investigation, Sample processing

   Competing interests

   No competing interests declared

7. Norzawani B Buang

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Investigation, Sample processing

   Competing interests

   No competing interests declared

8. Shanice Lewis

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Data curation, Investigation, Sample processing, Clinical phenotyping

   Competing interests

   No competing interests declared

9. Marie Pereira

   Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

   Contribution

   Investigation, Sample processing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-0711-3385

10. Frederic Toulza

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Investigation, Sample processing

    Competing interests

    No competing interests declared

11. Ester Fagnano

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Investigation, Sample processing

    Competing interests

    No competing interests declared

12. Marie-Anne Mawhin

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Investigation, Sample processing

    Competing interests

    No competing interests declared

13. Emma E Dutton

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Investigation, Sample processing

    Competing interests

    No competing interests declared

14. Lunnathaya Tapeng

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Investigation, Sample processing

    Competing interests

    No competing interests declared

15. Arianne C Richard

    1. Cambridge Institute for Medical Research, University of Cambridge, Cambridge, United Kingdom
    2. CRUK Cambridge Institute, University of Cambridge, Cambridge, United Kingdom

    Contribution

    Visualization, Methodology, Writing - review and editing, Results interpretation

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0002-8708-9997

16. Paul DW Kirk

    1. MRC Biostatistics Unit, Forvie Way, University of Cambridge, Cambridge, United Kingdom
    2. Cambridge Institute of Therapeutic Immunology & Infectious Disease, University of Cambridge, Cambridge, United Kingdom

    Contribution

    Methodology, Statistical methodology

    Competing interests

    No competing interests declared

17. Jacques Behmoaras

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Writing - review and editing, Results interpretation

    Competing interests

    No competing interests declared

18. Eleanor Sandhu

    1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
    2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

    Contribution

    Investigation, Patient recruitment and sample collection, Clinical phenotyping

    Competing interests

    No competing interests declared

19. Stephen P McAdoo

    1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
    2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

    Contribution

    Investigation, Patient recruitment and sample collection

    Competing interests

    reports personal fees from Celltrion, Rigel, GSK and Cello, outside the submitted work.

20. Maria F Prendecki

    1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
    2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

    Contribution

    Data curation, Investigation, Patient recruitment and sample collection, Clinical phenotyping

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0001-7048-7457

21. Matthew C Pickering

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Conceptualization, Data curation, Funding acquisition, Investigation, Project administration, Writing - review and editing, Conceived and designed the study, Clinical phenotyping, Results interpretation, Funding acquisition

    Competing interests

    No competing interests declared

22. Marina Botto

    Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom

    Contribution

    Conceptualization, Resources, Funding acquisition, Investigation, Project administration, Writing - review and editing, Conceived and designed the study, Sample processing, Results interpretation, Funding acquisition

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0002-1458-3791

23. Michelle Willicombe

    1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
    2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

    Contribution

    Funding acquisition, Investigation, Project administration, Conceived and designed the study, Patient recruitment and sample collection, Led and coordinated patient recruitment.

    Contributed equally with

    David C Thomas and James E Peters

    Competing interests

    No competing interests declared

24. David C Thomas

    1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
    2. Renal and Transplant Centre, Hammersmith Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom

    Contribution

    Conceptualization, Funding acquisition, Investigation, Project administration, Writing - review and editing, Conceived and designed the study. Patient recruitment and sample collection. Results interpretation

    Contributed equally with

    Michelle Willicombe and James E Peters

    Competing interests

    No competing interests declared

25. James E Peters

    1. Centre for Inflammatory Disease, Department of Immunology and Inflammation, Imperial College London, London, United Kingdom
    2. Health Data Research UK, London, United Kingdom

    Contribution

    Conceptualization, Data curation, Supervision, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Conceived and designed the study, Devised the analysis plan and supervised the analysis, Curated clinical phenotyping data, Funding acquisition

    Contributed equally with

    Michelle Willicombe and David C Thomas

    For correspondence

    j.peters@imperial.ac.uk

    Competing interests

    has received travel and accommodation expenses and hospitality from Olink to speak at Olink-sponsored academic meetings.

    "This ORCID iD identifies the author of this article:" 0000-0002-9415-3440

• 4,124

  Page views

• 717

  Downloads

• 42

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.