Early Immunological Biomarkers for Personalized Treatment Selection in Severe COVID-19: Post Hoc Machine Learning Analysis of a Randomized Clinical Trial

1

CoV-2-STs combined with standard-of-care improved recovery rates and overall survival in severe COVID-19 patients compared to standard-of-care alone.
2

Machine learning models can enhance prediction of COVID-19 treatment outcomes and identify patients unlikely to benefit from standard therapies.
3

The study utilized data from a randomized trial assessing CoV-2-STs' safety and efficacy in hospitalized adults with severe COVID-19.
4

Key biomarkers such as CRP, LDH, ferritin, and D-Dimers were monitored to stratify patient risk and guide treatment decisions.
5

A novel computational tool was developed to predict recovery from severe COVID-19, facilitating tailored immunotherapy strategies.

Jmir Medical Informatics

by Symeon Savvopoulos, Anastasia Papadopoulou, Georgios Karavalakis, Ioanna Sakellari, Grigorios Georgolopoulos, Christos Argyropoulos, Evangelia Yannaki, Haralampos Hatzikirou
June 4, 2026

1