Figure to the comment on paper: DOI 10.1371/journal.pmed.1002786
Performance of the three transcriptomic signatures for discriminating between progressors and non-progressors on samples collected within 24 months of TB diagnosis in the test subsets or the full South African and Gambian sets of the GC6-74 cohort. Bar graphs representing area under the ROC curve (AUC) and 95% CI values (error bars) for the Zak16-gene CoR, Sweeney3 or Maertzdorf4 signatures, measured by qRT-PCR, to discriminate between progressors and non-progressors in the 3 test sets of the GC6-74 cohort on the left, as shown in Suliman et al., 2018, or on the full South African and Gambian sets, as shown for the Zak16-gene CoR in Zak et al., 2016. The numbers in brackets below each cohort represent the number of progressors included in the cohort (non-progressors were matched ~4:1 to the progressors). The GC6-74 cohort test sets were comprised of 14 progressors and 83 controls from South Africa, 8 progressors and 56 controls from The Gambia and 12 progressors and 48 controls from Ethiopia. The GC6-74 full cohorts were comprised of 40 progressors and 159 controls from South Africa and 26 progressors and 104 controls from The Gambia. Scores for the Maertzdorf4 and Sweeney3 signatures for the full GC6-74 cohorts were based on cycle threshold data generated by qRT-PCR (Fluidigm, USA) using the classification models published for each signature [5,6]. In cases where the 95% CI crosses 0.5, the error bar is in red, indicating non-significant validation. Areas under the ROC curves show that Sweeney3 now validates in the complete South African cohort, while Maertzdorf4 validates in the complete Gambian cohort.
1. Warsinske H, Vashisht R, Khatri P. Host-response-based gene signatures for tuberculosis diagnosis: A systematic comparison of 16 signatures. Chaisson R, editor. PLoS Med. Public Library of Science; 2019;16: e1002786. doi:10.1371/journal.pmed.1002786
2. Suliman S, Thompson E, Sutherland J, Weiner Rd J, Ota MOC, Shankar S, et al. Four-gene Pan-African Blood Signature Predicts Progression to Tuberculosis. American journal of respiratory and critical care medicine. 2018;197: 1198–1208. doi:10.1164/rccm.201711-2340OC
3. Zak DE, Penn-Nicholson A, Scriba TJ, Thompson E, Suliman S, Amon LM, et al. A blood RNA signature for tuberculosis disease risk: a prospective cohort study. The Lancet. 2016;387: 2312–2322. doi:10.1016/S0140-6736(15)01316-1
4. Sweeney TE, Braviak L, Tato CM, Khatri P. Genome-wide expression for diagnosis of pulmonary tuberculosis: a multicohort analysis. Lancet Respir Med. 2016;4: 213–224. doi:10.1016/S2213-2600(16)00048-5
5. Maertzdorf J, McEwen G, Weiner J, Tian S, Lader E, Schriek U, et al. Concise gene signature for point-of-care classification of tuberculosis. EMBO molecular medicine. EMBO Press; 2016;8: 86–95. doi:10.15252/emmm.201505790
6. Warsinske HC, Rao AM, Moreira FMF, Santos PCP, Liu AB, Scott M, et al. Assessment of Validity of a Blood-Based 3-Gene Signature Score for Progression and Diagnosis of Tuberculosis, Disease Severity, and Treatment Response. JAMA Netw Open. 2018;1: e183779. doi:10.1001/jamanetworkopen.2018.3779