Advantages
- Accurately classify COVID-19 patients with the highest severity who should be prioritized for treatment.
Technology Overview & Background
Researchers have previously reported that a network of major cytokines in the blood is associated with severity of sepsis and that a highly accurate prognostic marker can be developed by combining these major cytokines. They hypothesized that if there are major proteins that form a network in the blood of COVID-19 patients, it may be possible to develop a similarly accurate prognostic marker by using these proteins.
As a result of their study, four proteins (WFDC2, CHI3L1, GDF15, and KRT19) associated with prognosis and prolonged respiratory failure were identified [*Experimental details in the selection of the four indicator proteins are as follows]. When these four proteins were combined as a compound marker, the sensitivity was 100%, specificity was 66.3%, the AUC (Area Under the Curve) value of the ROC curve was 0.85, and the 95% CI (Confidence Interval) value ranged 0.777–0.926. This result is a highly accurate indicator compared to CRP (AUC = 0.51) and D-Dimer (AUC = 0.68).
Next, the researchers identified new clinical phenotypes using the 4 key proteins by latent cluster analysis. The clinical phenotypes were named as α, β, γ, and δ phenotypes in order of decreasing severity. These phenotypes are visualized via a dimensional compression technique, and it was clearly to see the difference of distributions between the phenotypes. The Kaplan-Meier curves showed significant differences of survival rates between the phenotypes, and the worse two (γ and δ) phenotypes showed lower survival rates compared to the α and β phenotypes. This suggests that γ and δ phenotypes could be therapeutic targets.
Data (Method for the protein selection study)
- First, using plasma proteomic data (public data) from 1463 cases obtained using Olink’s Proximity Extension Assay (PEA) technology as the first analysis cohort, they found 24 proteins that were significantly elevated in the severe disease group when compared to the non-severe disease group.
- Then, in the data of the second analysis cohort of 53 cases (Osaka University Hospital), among the proteins that were significantly elevated in the groups of prolonged respiratory recovery or death compared to the early respiratory recovery group, five proteins that were common to the first analysis cohort were extracted.
- The five proteins were validated by ELISA in the validation cohort of 115 patients, and four proteins (WFDC2, CHI3L1, GDF15, and KRT19) associated with prognosis and prolonged respiratory failure were identified.
Patent(s)
A patent has been applied in Japan, and it has not yet been decided whether a PCT application will be filed.
Researchers & Academic Institution
Hisatake Matsumoto, MD, PhD (Dept. of Traumatology and Acute Critical Medicine, Osaka University Graduate School of Medicine, Japan), et al.
Further Details
CDA (with Osaka University): Available.
Expectations
Tech Manage Corp. is now looking for companies to collaborate with the researchers and develop this technology further under the licensing of patent(s) described above. You can also consider joint research using the invention, providing know-how under a confidentiality agreement, or setting up evaluation or licensing options for a certain period of time.
Project ID: JT-04124