Background Inexpensive rapid screening tests that can be used at the point-of-care (POC) are vital to combat tuberculosis. Particularly, less invasive non-sputum-based biomarker tests for all TB forms can help controlling transmission. Availability of such tests would significantly accelerate and streamline diagnostic approaches, improve cost-efficiency and decrease unnecessary costly GeneXpert referrals. Methods Multi-biomarker test (MBT) devices measuring levels of selections of up to six serum proteins simultaneously on a single lateral flow (LF) strip were produced. The strip contains individual capture lines for a biomarker selection allowing discrimination of TB-patients from other respiratory diseases (ORD). Only biomarkers successfully evaluated with singleplex strips (single biomarker tests) were applied to the MBT device. Quantitative signals are recorded with a low-cost handheld reader compatible with the applied luminescent up-converting particle (UCP) label. Biomarker selection and algorithms used to distinguish potential-TB and ORD are flexible. Results Results obtained with MBT strips containing multiple test lines correlate well with singleplex LF strips. Using LF tests for 5 selected biomarkers a sensitivity of 94% and specificity of 96% could be achieved with a confirmed South African selection of 20 TB and 31 non-TB samples. Patients were designated TB positive when scoring a value above the cut-off threshold for at least 3 out of 5 biomarkers. Serum samples of potential TB patients collected at five medical research institutes (Ethiopia, Namibia, South Africa, The Gambia, Uganda) were tested locally with MBT strips comprised of CRP, SAA, IP-10, Ferritin, ApoA-I and IL-6 and results analysed to obtain an overall pan-Africa applicable signature. Conclusion Evaluated POC applicable UCP-LF devices detecting serum biomarker signatures can help to distinguish active TB from other respiratory diseases and as such can prioritise highest-risk patients for further care. Ongoing prospective studies evaluate the MBT strip with fingerstick blood and do not require a laboratory or trained phlebotomist anymore.
Abstract Background Interferon-γ release assays (IGRA) with Resuscitation promoting factor (Rpf) proteins enhanced tuberculosis (TB) screening and diagnosis in adults but have not been evaluated in children. Children often develop paucibacillary TB and their immune response differs from that of adults, which together affect TB disease diagnostics and immunodiagnostics. We assessed the ability of Rpf to identify infection among household TB-exposed children in The Gambia and investigated their ability to discriminate Mycobacterium tuberculosis complex (MTBC) infection from active TB disease in children. Methods Detailed clinical investigations were done on 93 household TB-exposed Gambian children and a tuberculin skin test (TST) was administered to asymptomatic children. Venous blood was collected for overnight stimulation with ESAT-6/CFP-10-fusion protein (EC), purified protein derivative and RpfA, B, C, D and E. Interferon gamma (IFN-γ) production was measured by ELISA in supernatants and corrected for the background level. Infection status was defined by IGRA with EC and TB disease by mycobacterial confirmation and/or clinical diagnosis. We compared IFN-γ levels between infected and uninfected children and between infected and TB diseased children using a binomial logistic regression model while correcting for age and sex. A Receiver Operating Characteristics analysis was done to find the best cut-off for IFN-γ level and calculate sensitivity and specificity. Results Interferon gamma production was significantly higher in infected (IGRA+, n=45) than in uninfected (IGRA-, n=20) children after stimulation with RpfA, B, C, and D ( P = 0.03; 0.007; 0.03 and 0.003, respectively). Using RpfB and D-specific IFN-γ cut-offs (33.9 pg/mL and 67.0 pg/mL), infection was classified with a sensitivity-specificity combination of 73%-92% and 77%-72% respectively, which was similar to and better than 65%-75% for TST. Moreover, IFN-γ production was higher in infected than in TB diseased children (n=28, 5 bacteriologically confirmed, 23 clinically diagnosed), following RpfB and D stimulation ( P = 0.02 and 0.03, respectively). Conclusion RpfB and RpfD show promising results for childhood MTBC infection screening, and both performed similar to and better than the TST in our study population. Additionally, both antigens appear to discriminate between infection and disease in children and thus warrant further investigation as screening and diagnostic antigens for childhood TB.
Background With 2 billion people infected with Mycobacterium tuberculosis (Mtb) worldwide, identification of those most at-risk of progressing to active disease would provide a targeted, cost-effective approach for preventative therapy strategies. The GC6–74 project recruited Mtb-exposed HIV-positive (HIV+) contacts from 5 African countries with the aim of identifying molecular and protein signatures for identification of ‘at-risk’ subjects by comparing those who progressed to active disease (progressors) to those who remained asymptomatic (controls). Methods For this arm of the project, we analysed longitudinal samples from 12 HIV +progressors and 28 HIV +matched controls from Uganda (Makerere University, MAK) and South Africa (Stellenbosch University, SUN). Diluted whole blood was stimulated for 7 days with 7 Mtb-specific antigens plus controls. Supernatant was collected and a 38-plex multiplex assay performed following identification of confirmed progressors and controls. Results The median time to progression to active disease was 510 days for SUN and 425 days for MAK participants. Baseline CD4 counts were 163 cells/µl for progressors and 154 cells/µl for controls. Baseline responses showed significantly lower IL-4 production in progressors following ESAT-6/CFP-10 (EC) stimulation (p=0.0309) and significantly higher macrophage-derived chemokine (MDC) following both Rv3019 and TB10.4 stimulation. For the time-point closest to progression, IL-10 production following EC stimulation and IFN-γ production following Rv3019 stimulation were significantly higher in progressors than controls (p=0.0024 and p=0.0028 respectively). A combination of 12 analytes following EC and TB10.4 stimulation gave 84.4% and 91.1% correct classification respectively. Conclusion We have defined a soluble signature for detecting likely progression to active TB in HIV +subjects 1 year prior to progression. Following validation in other cohorts, this could be exploited for development of a field-friendly test for targeted interventional therapy.
Tuberculosis is one of the leading causes of morbidity and mortality in developing countries. Analysis of the host immune response may help with generating point-of-care tests for personalised monitoring. Thus, the aim of this study was to assess the relationship between immune activation markers: C-reactive protein (CRP), Beta2 microglobulin (B2M) and Neopterin, disease severity prior to treatment and response to therapy in adult pulmonary TB patients. HIV negative adult pulmonary TB index cases (n = 91) were recruited from the TB clinic at MRC, The Gambia. Plasma samples were collected at enrolment and at 2 and 6 months following TB treatment initiation. An enzyme linked immunosorbent assay (ELISA) was performed for evaluation of CRP, B2M and Neopterin levels and correlated with clinical and microbiological parameters including strain of infection. Disease severity was determined using Chest X-ray (CXR), Body Mass Index (BMI) and sputum smear grade. Plasma levels of all three markers were highly elevated in patients at recruitment and declined significantly during TB therapy. No correlation with disease severity was seen at recruitment. CRP showed the most significant decrease by 2 months of treatment (p < 0.0001) whereas levels of B2M and Neopterin showed little change by 2 months but a significant decrease by 6 months of treatment (p = 0.0002 and p < 0.0001 respectively). At recruitment, B2M levels were significantly higher in subjects infected with Mycobacterium africanum (Maf) compared with those infected with Mycobacterium tuberculosis sensu stricto (Mtb) (p = 0.0075). In addition, while CRP and Neopterin showed a highly significant decline post-treatment regardless of strain (p < 0.0001 for all), B2M showed differential decline depending on strain (p = 0.0153 for Mtb and p = 0.0048 for Maf) and levels were still significantly higher at 6 months in Maf compared to Mtb infected subjects (p = 0.0051). Our findings suggest that activation markers, particularly CRP, may have a role in identifying good response to TB therapy regardless of the strain of infection and could be further developed as point-of-care tests. In addition, B2M levels may allow differentiation between Mtb and Maf-infected subjects.