PT - JOURNAL ARTICLE AU - Cecilia Lekpor AU - Felix Botchway AU - Jonathan Stiles AU - Wilson O Nana TI - PO 8356 ANGIOGENIC AND ANGIOSTATIC FACTORS IN THE SALIVA OF MALARIA PATIENTS AID - 10.1136/bmjgh-2019-EDC.75 DP - 2019 Apr 01 TA - BMJ Global Health PG - A29--A29 VI - 4 IP - Suppl 3 4099 - http://gh.bmj.com/content/4/Suppl_3/A29.2.short 4100 - http://gh.bmj.com/content/4/Suppl_3/A29.2.full AB - Background Malaria mortality is associated with exaggerated host responses to inflammatory factors such as C-X-C motif chemokine 10 (CXCL10) and host biomarkers such as angiopoietin 1 (Ang-1) and angiopoietin 2 (Ang-2). The aim of this study was to determine saliva levels of CXCL10, Ang-1 and Ang-2 and compare with plasma levels regarding their potential as biomarkers of malaria, which may be useful for further development of highly efficient non-invasive malaria detection methods.Methods Case control study involving 213 subjects (119 with and 94 without malaria) aged 1–16 years. Haematological determination was done using Haematology Analyser. Plasmodium Lactate Dehydrogenase/Histidine Rich Protein-2 (pLDH/HRP-2) Antigen rapid diagnostic test (RDT) were performed. Plasma and saliva levels of CXCL10, Ang-1 and Ang-2 were measured using Elisa kit. Data was presented as mean ±standard error or median and interquartile range (IQR). A p-value<0.001 was considered statistically significant.Results There was decreased plasma levels of Ang-1 and increased plasma levels of CXCL10 and Ang-2 in individuals with malaria compared to those without malaria (Ang-1, p<0.009; Ang-2, p<0.001; CXCL10 p<0.001). Biomarker levels in both plasma and saliva in subjects with malaria and without malaria were correlated and a significant relationship was found between Ang −2 and CXCL10 which could be used to predict malaria severity (p=0.001 for Ang-2 and p<0.01 for CXCL10). Low Ang-1 and high Ang-2 in both plasma and saliva were significantly associated with increased risk of malaria severity: Ang-1, 2741.04 (1785.85–3582.68), p<0.009; Ang-2, 3508.82 (2139.61–5091.633.9), p<0.001 and Ang-1, 720.27 (439.82–1086.74); 16.98 (10.08–33.26), (p<0.001 for all). Finally, Ang-2 was informative when combined with CXCL10 to predict malaria severity.Conclusion These results provide insight into the use of saliva for a non-invasive diagnostic method and demonstrate that Ang-2 combined with CXCL10 is a promising predictive biomarker of malaria severity.