CCL3, CCL5, IL-15, IL-1Ra and VEGF compose a reliable algorithm to discriminate classes of adverse events following 17DD-YF primary vaccination according to cause-specific definitions.

In the present study, a range of serum biomarkers were quantified in suspected cases of adverse events following YF immunization (YEL-AEFI) to propose a reliable laboratorial algorithm to discriminate confirmed YEL-AEFI ("A1" class) from cases with other illnesses ("C" class). Our findings demonstrated that increased levels of CXCL8, CCL2, CXCL10, IL-1β, IL-6 and TNF-α were observed in YEL-AEFI ("A1" and "C" classes) as compared to primary vaccines without YEL-AEFI [PV(day 3-28)] and reference range (RR) controls. Notably, increased levels of CCL3, CCL4, CCL2, CCL5, IL-1β, IL-15, IL-1Ra and G-CSF were found in "A1" as compared to "C" class. Venn diagrams analysis allowed the pre-selection of biomarkers for further analysis of performance indices. Data demonstrated that CCL3, CCL5, IL-15 and IL-1Ra presented high global accuracy (AUC = 1.00) to discriminate "A1" from "C". Decision tree was proposed with a reliable algorithm to discriminate YEL-AEFI cases according to cause-specific definitions with outstanding overall accuracy (91%). CCL3, CCL5, IL-15 and IL-1Ra appears as root attributes to identify "A1" followed by VEGF as branch nodes to discriminate Wild Type YFV infection ("C(WT-YFV)") from cases with other illnesses ("C*"). Together, these results demonstrated the applicability of serum biomarker measurements as putative parameters towards the establishment of accurate laboratorial tools for complementary differential diagnosis of YEL-AEFI cases.