Title: Evaluation of host biomarker-based point-of-care tests for targeted screening for active TB (Screen TB)

Introduction: Tuberculosis (TB) places severe pressure on health care services of the developing world. Despite the introduction of the highly sensitive and specific GeneXpert MTB/RIF (GeneXpert) test [1] with a potential turn-around time of two hours, many people in high TB prevalence areas still do not have access to efficient TB diagnostic services due to logistical constraints in these settings. A cost effective, rapid, point-of-care screening test with high sensitivity would identify people with a high likelihood for active TB and would prioritize them for testing with more expensive, technically or logistically demanding assays including GeneXpert or liquid culture, facilitating cost-effective diagnostic work-up in resource-limited settings. A serum cytokine signature for active TB disease, discovered in the AE-TBC project, with a sensitivity of 89% (CI 78 - 95%) and specificity of 76% (CI 68 - 83%), will be optimised and utilized in a point-of-care format (TransDot) to rapidly test for TB disease in symptomatic people.

Hypothesis: The TransDot test will achieve a sensitivity of > 90% for TB disease, in a training set of people suspected of having TB disease, and be validated (achieve similarly high sensitivity) subsequently in a prospective test set of people suspected of having TB disease, when compared to a composite gold standard of sputum culture, smear, GeneXpert, chest X-ray, TB symptoms and TB treatment response.

Objectives: The overall objective of the study is to incorporate a six-marker serum signature into a multiplex UCP-LFA format, referred to as TransDot, for finger-prick blood testing. The end point of the study is the accuracy (sensitivity and specificity) of the UCP-LFA TransDot test on finger-prick blood for active TB and will be prospectively compared against gold standard composite diagnostic criteria (GeneXpert, MGIT culture, TB sputum smear, CXR, TB symptom screen and response to TB treatment).

Primary: The primary outcome of interest will be accuracy, sensitivity and specificity of the TransDot finger-prick test when compared with the composite gold standard tests.

Protocol Summary Title: Evaluation of host biomarker-based point-of-care tests for targeted screening for active TB (Screen TB)

Population: A total of 800 people presenting at primary health care clinics with presumed active pulmonary tuberculosis, aged 18 to 70 years, male or female gender, will be recruited. They should be willing to give informed, written consent, including consent for HIV testing. They should have symptoms that could be compatible with active TB (cough > two weeks, plus at least one of the following: fever, weight loss, haemoptysis and night sweats). Participants should not have been on TB treatment for the past 90 days and should not have received immune suppressive therapy, be known with alcohol of drug abuse, have a haemoglobin level <9g/dl or be pregnant or breastfeeding. HIV co-infection is not an exclusion criterion. Participants will be recruited from primary health care clinics in Cape Town, South Africa, Windhoek in Namibia, Addis Ababa in Ethiopia, Banjul in The Gambia and Kampala in Uganda.

Number of Sites: Five sites

Study Duration: 3 years

Subject Duration: 18 months for TB cases, 2 months for non-TB cases

Objectives:

The overall objective of the study is to incorporate a six-marker serum signature into a multiplex UCP-LFA format, referred to as TransDot, for finger-prick blood testing. The end point of the study is the accuracy (sensitivity and specificity) of the UCP-LFA TransDot test on finger-prick blood for active TB and will be prospectively compared against gold standard composite diagnostic criteria (GeneXpert, MGIT culture, TB sputum smear, CXR, TB symptom screen and response to TB treatment).

Inclusion Criteria:

• Symptoms suggestive of TB disease: cough for more than two weeks with fever, malaise, weight loss, night sweats, haemoptysis, chest pain or loss of appetite.
• Willingness to give consent to take part in the study.
• Willingness to undergo HIV testing or be willing to have their HIV infection status disclosed to the study field workers.
• Eighteen years or older and aged 70 years or younger.

Exclusion Criteria:

• Permanent residence in study area for less than 3 months or with no permanent address.
• Pregnancy or breastfeeding.
• HB<9g/l
• On TB treatment currently or in the last ninety days.
• HIV positive patients currently on INH prophylaxis, or in the last ninety days.
• Known quinolone or aminoglicozide antibiotic use reported in the past 60 days.

Adult participants with suspected active TB disease will be recruited in South Africa, the Gambia, Uganda, Namibia and Ethiopia. Each site will recruit approximately 160 participants with suspected TB, until the desired overall total of about 800 participants is reached. In South Africa, up to 300 participants will be recruited from primary health care clinics (Adriaanse, Elsiesriver, Uitsig, Ravensmead, Fisantekraal, Durbanville and Dunoon) in Cape Town.

Patients presenting to the health care facility with symptomatic pulmonary disease and a high likelihood of having tuberculosis will be enrolled and followed up for outcome classification. Participants who had previous TB, extra-pulmonary TB in addition to pulmonary TB, drug resistance detected on GeneXpert or culture or other concomitant diseases, will not be excluded from enrolment. Both HIV positive and HIV negative individuals will be enrolled.