In-silico immune cell deconvolution of the airway proteomes of infants with pneumonia reveals a link between reduced airway eosinophils and an increased risk of mortality

Charles J Sande; Jacqueline M Waeni; James M Njunge; Martin N Mutunga; Elijah Gicheru; Nelson K Kibinge; Agnes Gwela

We report on a new method of deconvolving immune cell populations that are resident in the airways of infants and children with different survival outcomes of severe pneumonia. The study mucosal cellular immunity during very severe pneumonia has been hindered by a number of important hurdles including low sample volumes as well as the relatively low abundance of immune cell phenotypes that may be critical in directing the clinical course of pneumonia. In this study, we addressed these problems by using a machine learning approach to identify protein markers that could be used to deconvolve mixed immune cells. We then applied these markers to airway proteome data obtained from children with different survival outcomes of clinical pneumonia. Our results show that protein markers associated with eosinophils are elevated in the airways of survivors and are diminished in children who later died from infection. The airway levels of these eosinophil markers were no different between children who died and well controls, suggesting that the failure to mount an appropriate eosinophil response is a potential mechanism of pneumonia-related mortality, especially in high risk populations such as undernourished children. To validate the findings of the airway proteome analysis, we reviewed the hospitalisation records of >10,000 children who had been admitted to hospital in the previous 10 years with clinical pneumonia. The results of this retrospective analysis confirmed the observations made from analysis of the airway proteome and showed that children who died from pneumonia, had significantly lower blood eosinophil counts relative to those who survived. In contrast, neutrophil levels were not different between survivors and non survivors in both the airway proteome and in the retrospective validation cohort.