Effect of artesunateamodiaquine on Ebola virus disease mortality
Effect of artesunate-amodiaquine on Ebola virus disease mortality Etienne Gignoux, Andrew S. Azman, Martin de Smet, Philippe Azuma, Josephus Bolongei, Moses Massaquoi, Dorian Job, Amanda Tiffany, Roberta Petrucci, Esther Sterk, Julien Potet, Motoi Suzuki, Andreas Kurth, Angela Cannas, Anne Bocquin, Thomas Strecker, Christopher Logue, Thomas Pottage, Constanze Yue, Jean-Clement Cabrol, Micaela Serafini, Iza Ciglenecki
Background • Largest ever Ebola virus disease (EVD) outbreak • Little evidence of therapeutic interventions reducing mortality • Treatment symptomatic and presumptive, including anti-malarial treatment: – Systematic or based on positive malaria diagnosis – In MSF: artemether-lumefantrine Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 2
Foya, Liberia New confirmed EVD admissions, Foya, 2014 Image: Martin Zinggl Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 3
Background Image: Martin Zinggl • Shortage of first-line antimalarial during peak of cases • Artemether-lumefantrine (AL) replaced by artesunateamodiaquine (ASAQ) for a period of 12 days • Amodiaquine shown to have invitro activity against Ebola virus Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 4
Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality
Patient characteristics AL (N=194) ASAQ (N=71) None (N=63) P value Age - no. (%) 0 -4 Years 12 (6%) 3 (4%) 5 (8%) 0. 14 5 -29 Years 94 (49%) 43 (61%) 29 (46%) 30 -59 Years 80 (41%) 18 (25%) 25 (40%) 60+ Years 7 (4%) 7 (10%) 4 (6%) Female sex - no. (%) 96 (49%) 33 (46%) 37 (59%) 0. 34 Care seeking delay (no. days) Median (IQR) 4 (2 -6) 3 (2 -5) 4 (2 -5) 0. 79 Cycle threshold value at admission Median (IQR) 19. 6 (17. 1 -22. 7) 18. 5 (17. 2 -23. 1) 18. 4 (16. 1 -21. 9) 0. 05 Malaria test result no. (%) Positive 33 (18%) 13 (22%) 10 (19%) 0. 86 Missing data 13 (7%) 11 (15%) 9 (14%) IV Fluid no. (%) Received no. (%) 62 (32%) 25 (35%) 20 (32%) 0. 87 Mean no. of in-patients 54 102 50 <0. 001 Case per unique home-village 4. 6 3. 9 2. 9 <0. 001 Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 6
Patient characteristics AL (N=194) ASAQ (N=71) None (N=63) P value 0 -4 Years 12 (6%) 3 (4%) 5 (8%) 0. 14 5 -29 Years 94 (49%) 43 (61%) 29 (46%) 30 -59 Years 80 (41%) 18 (25%) 25 (40%) 60+ Years 7 (4%) 7 (10%) 4 (6%) Female sex - no. (%) 96 (49%) 33 (46%) 37 (59%) 0. 34 Care seeking delay (no. days) Median (IQR) 4 (2 -6) 3 (2 -5) 4 (2 -5) 0. 79 Age - no. (%) Cycle threshold value at admission Median (IQR) 19. 6 (17. 1 -22. 7) 18. 5 (17. 2 -23. 1) 18. 4 (16. 1 -21. 9) 0. 05 Malaria test result no. (%) Positive 33 (18%) 13 (22%) 10 (19%) 0. 86 Missing data 13 (7%) 11 (15%) 9 (14%) IV Fluid no. (%) Received no. (%) 62 (32%) 25 (35%) 20 (32%) 0. 87 Mean no. of in-patients Case per unique home-village 54 4. 6 102 3. 9 50 2. 9 <0. 001 Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 7
Relative risk of mortality by antimalarial treatment prescribed among confirmed Ebola virus disease patients, Lofa, Liberia Treatment Cases Deaths Risk Ratio P-value Adjusted Risk Ratio P-value no. of patients % (95% CI) Reference AL 194 125 64. 4 Reference ASAQ 71 36 50. 79 (0. 61 -1. 01) 0. 062 0. 69 (0. 54 -0. 89) 0. 004 None 63 41 65. 1 1. 01 (0. 82 -1. 25) 0. 925 0. 85 (0. 69 -1. 03) 0. 107 Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 8
Relative risk of mortality by antimalarial treatment prescribed among confirmed Ebola virus disease patients, Lofa, Liberia Treatment Cases Deaths Risk Ratio no. of patients % (95% CI) Reference AL 194 125 64. 4 Reference ASAQ 71 36 50. 79 (0. 61 -1. 01) None 63 41 65. 1 1. 01 (0. 82 -1. 25) P-value Adjusted Risk Ratio P-value 0. 062 0. 69 (0. 54 -0. 89) 0. 004 0. 925 0. 107 0. 85 (0. 69 -1. 03) Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 9
Other risk factors for EVD mortality • Age >60 years – RR 1. 45 (95% CI 1. 14 -1. 84, p=0. 002 ) • Cycle threshold (CT) at admission (Log): – RR 0. 09 (95%CI 0. 05 -0. 14, p<0. 001) • Received IV fluid: – RR 1. 29 (95%CI 1. 09 -1. 52, p = 0. 002) • Number of in-patients on day of admission: – RR 1. 004 (1. 001 -1. 006, p=0. 002) Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 10
Sub-analysis • Effect stronger in patients with negative malaria rapid diagnostic test (RDT): – Malaria RDT+: RR 1. 00 (95% CI 0. 54 to 1. 85), p-value 0. 98 – Malaria RDT-: RR 0. 64 (95% CI 0. 49 -0. 85), p-value 0. 002 • Effect stronger in patients with CT value >20 at admission – CT value > 20: RR 0. 52 (0. 26 to 1. 06), p-value 0. 07 – CT value < 20: RR 0. 89 (0. 70 -1. 13), p-value 0. 37 Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 11
Discussion • ASAQ associated with reduced risk of dying • Many limitations: – – – Observational study Data on drug prescription only Quality of data (esp. «none» ) Additional confounding factors Small sample size • Alternative hypothesis: AL is harmful Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 12
Next steps • Recommend change of first-line antimalarial for EVD patients (risk vs benefit) • Further research needed: – Analysis of any available data: • Impact of mass drug administrations (MDAs) on Ebola transmission • Other centres using ASAQ? – Preclinical (in vitro and animal) and clinical studies Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 13
Acknowledgments • • • Patients and staff in Foya Armand Specher, Michel van Herp, Patricia Kahn, Sophie Masson (MSF) Ministry of Health Liberia EMLab Martin Friede (WHO) Mikhail Martchenko and Joel West (Keck Graduate Institute) Ciglenecki I. Effect of artesunate-amodiaquine on Ebola virus disease mortality 14
- Slides: 14