1. Assessment of three new parasite lactate dehydrogenase (pan-pLDH) tests for diagnosis of uncomplicated malaria Assessment of three new parasite lactate dehydrogenase (pan-pLDH) tests for diagnosis of uncomplicated malaria Fogg CL 1, Nabasumba C 1, Twesigye R 1, Batwala V 2, Piola P 1, Kiguli J 1, Mutebi F 1, Hook C 3, Guillerm M 4, Moody A 4, Guthmann J-P 1 1 Epicentre, 2 Mbarara University of Science and Technology, 3 MSF - Malaria Working Group, 4 MSF – Access Campaign

2. Background  Microscopy = recommended method for malaria diagnosis: Trained staff Quality equipment Supervision  Rapid diagnostic tests (RDT) are useful for malaria diagnosis where high standard microscopy is not available: Easy to use Simple to interpret

3. Study rationale  Most common RDT detect Histidine Rich Protein 2 (HRP2), i.e. Paracheck ®  HRP2 tests have 2 major drawbacks: Only detect P. falciparum Remain positive for weeks after treatment  New RDTs detect parasite Lactate DeHydrogenase (pLDH): Identify all plasmodium species Become negative quickly after treatment  The study assessed 3 new pLDH tests

4. Objectives  Primary objective Measure validity of 3 pLDH tests and that of Paracheck ®  Secondary objectives Measure % of positive tests after effective treatment Ease of use Measure inter-reader reliability

5. Methods I: Inclusion  Screening at OPD of Mbarara hospital  Inclusion: symptoms of simple malaria signed informed consent  Two age groups: “under 5” and “5 and above”  Sample size: 400 patients (200 blood smear [+] and 200 blood smear [-])  All tests were double read, blinded and compared to microscopy  Reading of blood smears controlled in Thailand

6. Methods II: Tests performed 1234

7. Methods III: Follow-Up  Patients: Positive BS on D0 OR At least 1 positive RDT on D0  Positive BS on D0 treated with Coartem®  Pregnant women were excluded  Procedures on D3 & D14: Clinical examination Blood smear All tested RDTs

8. Methods IV: Outcomes  Validity: Sensitivity: % of true positives Specificity: % of true negatives  Percentage of RDTs remaining positive at D3 & D14  Inter-reader reliability: kappa coefficient (above 0.80 was considered « Good Agreement»)  Ease of use determined by a score

9. Results (I): Validity Sensitivity (N=248) Specificity (N=212) Paracheck ® 94% [90.2 - 96.6] 87.3% [82.0 – 91.4] Vistapan ® 91.9% [87.8 – 95] 89.6% [84.7 – 93.4] Carestart ® 95.6% [ 92.2 – 97.8] 91.5% [86.9 – 94.9] Parabank ® 84.7 % [79.6 – 88.9] 94.3% [90.3 – 97.0]

10. Results (II): Sensitivity by group  Parasite density: Sensitivity decreased with parasitemia Parasitaemia <100 parasites/ µL (from 41.9% [Parabank] to 67.7% [Carestart])  Age group: Sensitivity increased in under 5’s From 95.3% [Parabank] to 97.7% [Vistapan/Carestart])

11. Results (III): Positive tests after treatment Day 3Day 14 pLDH Parabank ® 17.8% [12.5 – 23.1] 4.6% [1.7 – 7.5] Vistapan ® 36.1% [29.7 – 42.5] 8.9% [5.1 – 12.7] Carestart ® 42.5% [36.1 – 48.9] 9.5% [5.6 – 13.4] HRP2Paracheck ® 86.2% [81.7 – 90.7] 69.7% [63.1 – 75.7]

12. Results (IV): Reliability  Kappa above 0.9 for inter-reader reliability for all tests

13. Results (V): Ease of use CHARACTERISTICParacheckVistapanCarestartParabank I Performance21301821 II Safety7.5 157.5 III Result stability5555 IV Interpretation3021 30 V Storage20 12 TOTAL (/100)83.5 7975.5

14. Conclusions  Carestart and Vistapan best tests: Parabank (poorly sensitive) Paracheck (high % of positives at D14)  Carestart and Vistapan could replace Paracheck for malaria diagnosis in sub- Saharan Africa  However, pLDH tests have slightly higher costs ($0.6 to $0.7 vs $0.45 for Paracheck)

15. Acknowledgements Thanks to:  Mbarara University of Science & Technology  Study team and patients in Mbarara.  Malaria Working Group, MSF.  Access to Essential Medicines Campaign, MSF.  The test manufacturers.  Shoklo Malaria Research Unit.

17. Validity details Sensitivity (N=248) Specificity (N=212) PARACHECK94% (n=233) [90.2 - 96.6] 87.3% (n=185) [82.0 – 91.4] VISTAPAN91.9% (n=228) [87.8 – 95] 89.6%(n=190) [84.7 – 93.4] CARESTART95.6% (n=237) [92.2 – 97.8] 91.5% (n=184) [86.9 – 94.9] PARABANK84.7 % (n=210) [79.6 – 88.9] 94.3% (n=200) [90.3 – 97.0]

18. Positive Tests D3/D14 details Day 3Day 14 pLDH Parabank 17.8% [12.5 – 23.1] (36/202) 4.6% [1.7 – 7.5] (9/196) Vistapan 36.1% [29.7 – 42.5] (79/219) 8.9% [5.1 – 12.7] (19/213) Carestart 42.5% [36.1 – 48.9] (97/228) 9.5% [5.6 – 13.4] (21/221) HRP2Paracheck 86.2% [81.7 – 90.7] (193/224) 69.7% [63.1 – 75.7] (152/218)

19. PPV & NPV PARACHECKVISTAPANCARESTARTPARABANK Sensitivity94%*91.9%*95.6%*84.7 % [90.2 - 96.6][87.8 – 95][92.2 – 97.8][79.6 – 88.9] Specificity87.3%89.6%91.5%*94.3%* [82.0 – 91.4][84.7 – 93.4][86.9 – 94.9][90.3 – 97.0] PPV89.6%91.2%*92.9%*94.6%* [85.3 – 93][87 – 94.4][89.1 – 95.8][90.7 – 97.2] NPV92.5%*90.5%*94.6%*84.0% [87.9 – 95.7][85.7 – 94.1][90.6 – 97.3][78.7 – 88.4]

20. Methods V: External Quality Control  External quality control performed at Shoklo Malaria Research Unit, Thailand  Randomly selected 145 positive slides and 145 negative Day 0 slides.

21. Results (VI): Quality control of slides  Number of discordances: 7 out of 145 (<5%)