Admin_Adham
A French team has identified biological markers that can predict 3‑month mortality in patients hospitalized for initially mild COVID pneumonia. The researchers developed a predictive score based on age and three markers: two renal and one inflammatory. Pierre‑Louis Tharaux , a nephrologist and research director at Inserm within the Paris Cardiovascular Research Center and a co‑author of the study, explained.
From the First Wave of COVID
The CORIMUNO‑19 study, conducted by researchers at the National Institute of Health and Medical Research (Inserm) and Paris Cité University sponsored by Assistance Publique Hôpitaux de Paris and funded by the Fondation pour la Recherche Médicale and ANRS MIE, was born of a desire to better understand the COVID epidemic from its outset.
“During winter 2019-20, when the outbreak was severe in China and beginning in France and Italy, I contacted Inserm and its REACTing unit (Research and Action targeting emerging infectious diseases), which was working on emerging infectious diseases, and proposed testing anti‑inflammatory drugs in patients with severe COVID. With other physicians, researchers, methodologists, administrators, and logisticians, we were able to set up the CORIMUNO‑19 clinical trial platform — a tremendous collective effort. At the time, we had no vaccines and it was unlikely we would have effective antivirals against this novel virus. When we wrote the therapeutic trial protocol, we proposed obtaining patient consents’ to collect plasma and serum samples to measure certain molecules to better understand the disease, identify patients at risk of severe progression, and possibly find targets for future therapeutic trials,” Tharaux said.
Measuring Circulating Proteins
Samples were collected from hundreds of patients. “For financial reasons, we ultimately focused our study on 196 patients admitted to 15 hospitals with moderate-to-severe pneumonia and included in two clinical trials run by the CORIMUNO‑19 consortium during the first wave of COVID‑19.The advantage was that they received near‑daily follow‑up, which allowed us to gather very precise data,” he said.
The clinical samples made it possible to measure 41 immune mediators and markers of kidney and vascular injury in patients’ blood within 48 hours of hospitalization. “We primarily measured circulating proteins in three categories: proteins involved in inflammation (interleukins [ILs] and cytokines); molecules involved in coagulation and vascular injury because we observed that some patients had thromboses and hemorrhages; and we added markers indicating renal injury, even if subtle,” Tharaux detailed.
Corimuno‑Score
From these assays, the researchers created models combining routine biological parameters and patients’ clinical characteristics and examined what was associated with severe disease progression. They found that age at the start of the study and 14 biological markers, including 11 proteins, were associated with the risk for death within 90 days of measurement. Among these markers they identified two renal markers — Kidney Injury Molecule‑1 and lipocalin‑2 — and a marker classically considered anti‑inflammatory, IL-10.
Sentinel Parameters
Using the plasma concentrations of these three markers combined with patient age, the scientists developed a new severity score called the “Corimuno‑Score” to identify patients at highest risk for fatal complications.
These results were replicated and validated in an independent cohort of 105 individuals.
“We were surprised to discover that renal markers predicted the risk of transfer to intensive care and death, even though majority of patients did not have established kidney failure,” the nephrologist said, calling them “sentinel parameters.”
The prognostic renal injury was often below the thresholds used in international medical definitions. He added, “What is also interesting is that comparing this score with others already in use, such as the 4C Mortality score, which includes many clinical parameters, shows that our very simple score performs at least as well.”
Question of Patient Triage
“We also found another interesting element that is not in the article: measuring these parameters on several days after admission does not add much more information than a single sample taken at admission. So one blood sample can provide prognostic information about what will happen over the next 3 months,” he added.
For him, the study — conducted in patients hospitalized on conventional wards rather than in intensive care — raises the question of patient triage. “Participants included in our study did not necessarily have initially severe pneumonia; yet a number of them died. This highlights the importance of identifying patients at risk early and not delaying transfer to units where they can be better monitored and treated, such as intensive care,” he said.
Recalibration Needs
While the study’s findings add new information, Tharaux cautions that they should be “taken with a grain of salt, because they were established from data collected during the early waves of COVID‑19.” “To use this score now, given the emergence of new SARS‑CoV‑2 variants, it would need recalibration,” he warned. Nevertheless, he believes the renal markers and IL‑10 identified in this study could also be interesting to evaluate in other diseases.
“This observation of unsuspected renal damage in a viral pneumonia draws attention to the role of the kidney as a sentinel organ to explore in other infectious diseases such as influenza, dengue, or chikungunya. That is a line of research worth pursuing,” he concluded.
Biomarkers in Long COVID
“On the biological level, many studies find, in a proportion of patients with long COVID, persistent elevation of inflammatory markers like IL‑6, CRP [C-reactive protein], TNF‑alpha and of certain chemokines/interferons, suggesting low‑grade inflammation and immune dysregulation,” Tharaux said.
A 2023 systematic review identified 113 biomarkers significantly associated with long COVID. Among them, IL‑6, CRP and TNF‑alpha emerge as a recurring proinflammatory “core.” Other studies confirm prolonged elevation in many patients of cytokines and chemokines like IL‑6, TNF‑alpha, CXCL10, type I interferons, acute‑phase proteins (CRP, ferritin); and vascular or neurologic markers such as VEGF, neurofilament light chain, and glial fibrillary acidic protein.
“However, diagnostic limits are linked to the fact that IL‑6 and CRP are nonspecific. Any infection, autoimmune disease, cancer or chronic inflammation can raise them, which prevents using them as a confirmatory test for long COVID,” the specialist said.
Also, “a substantial proportion of patients with long COVID have normal IL‑6 and CRP, so normal values do not exclude the diagnosis. No consensus threshold or combination of biomarkers currently has predictive and diagnostic value robust enough to be recommended for routine outpatient use. The lack of biomarkers specific to long COVID limits diagnostic and treatment precision as well as disease monitoring,” he added.
“Models that integrate clinical data, comorbidities, acute severity and biomarkers allow moderate prediction, especially for objectively measurable sequelae (cardiovascular, renal). Prediction remains more uncertain for fatigue, cognitive problems or sleep disturbances,” Tharaux said.
This story was translated from Medscape's French edition.
