Head-to-head comparison of diagnostic accuracy of TB screening tests: Chest-X-ray, Xpert TB host response, and C-reactive protein.

Background: Accessible, accurate screening tests are necessary to advance tuberculosis (TB) case finding and early detection in high-burden countries. We compared the diagnostic accuracy of available TB triage tests. Methods: We prospectively screened consecutive adults with ≥2 weeks of cough presenting to primary health centers in the Philippines, Vietnam, South Africa, Uganda, and India. All participants received the index tests: chest-X-ray (CXR), venous or capillary Cepheid Xpert TB Host Response (HR) testing, and point-of-care C-reactive protein (CRP) testing (Boditech iChroma II). CXR images were processed using computer-aided detection (CAD) algorithms. We assessed diagnostic accuracy against a microbiologic reference standard (sputum Xpert Ultra, culture). Optimal cut-points were chosen to achieve sensitivity ≥90% and maximize specificity. Two-test screening algorithms were considered, using two approaches: 1) sequential negative serial screening in which the second screening test is conducted only if the first is negative and positive is defined as positive on either test and 2) sequential positive serial screening, in which the second screening test is conducted only if the first is positive and positive is defined as positive on both tests. Results: Between July 2021 and August 2022, 1,392 participants with presumptive TB had valid results on index tests and the reference standard, and 303 (22%) had confirmed TB. In head-to-head comparisons, CAD4TB v7 showed the highest specificity when using a cut-point that achieves 90% sensitivity (70.3% vs. 65.1% for Xpert HR, difference 95% CI 1.6 to 8.9; 49.7% for CRP, difference 95% CI 17.0 to 24.3). Among the possible two-test screening algorithms, three met WHO target product profile (TPP) minimum accuracy thresholds and had higher accuracy than any test alone. At 90% sensitivity, the specificity was 79.6% for Xpert HR-CAD4TB [sequential negative], 75.9% for CRP-CAD4TB [sequential negative], and 73.7% for Xpert HR-CAD4TB [sequential positive]. Conclusions: CAD4TB achieves TPP targets and outperforms Xpert HR and CRP. Combining screening tests further increased accuracy. Cost and feasibility of two-test screening algorithms should be explored. Registration: NCT04923958.

An independent, multi-country head-to-head accuracy comparison of automated chest x-ray algorithms for the triage of pulmonary tuberculosis.

Background: Computer-aided detection (CAD) algorithms for automated chest X-ray (CXR) reading have been endorsed by the World Health Organization for tuberculosis (TB) triage, but independent, multi-country assessment and comparison of current products are needed to guide implementation. Methods: We conducted a head-to-head evaluation of five CAD algorithms for TB triage across seven countries. We included CXRs from adults who presented to outpatient facilities with at least two weeks of cough in India, Madagascar, the Philippines, South Africa, Tanzania, Uganda, and Vietnam. The participants completed a standard evaluation for pulmonary TB, including sputum collection for Xpert MTB/RIF Ultra and culture. Against a microbiological reference standard, we calculated and compared the accuracy overall, by country and key groups for five CAD algorithms: CAD4TB (Delft Imaging), INSIGHT CXR (Lunit), DrAid (Vinbrain), Genki (Deeptek), and qXR (qure.AI). We determined the area under the ROC curve (AUC) and if any CAD product could achieve the minimum target accuracy for a TB triage test (≥90% sensitivity and ≥70% specificity). We then applied country- and population-specific thresholds and recalculated accuracy to assess any improvement in performance. Results: Of 3,927 individuals included, the median age was 41 years (IQR 29-54), 12.9% were people living with HIV (PLWH), 8.2% living with diabetes, and 21.2% had a prior history of TB. The overall AUC ranged from 0.774-0.819, and specificity ranged from 64.8-73.8% at 90% sensitivity. CAD4TB had the highest overall accuracy (73.8% specific, 95% CI 72.2-75.4, at 90% sensitivity), although qXR and INSIGHT CXR also achieved the target 70% specificity. There was heterogeneity in accuracy by country, and females and PLWH had lower sensitivity while males and people with a history of TB had lower specificity. The performance remained stable regardless of diabetes status. When country- and population-specific thresholds were applied, at least one CAD product could achieve or approach the target accuracy for each country and sub-group, except for PLWH and those with a history of TB. Conclusions: Multiple CAD algorithms can achieve or exceed the minimum target accuracy for a TB triage test, with improvement when using setting- or population-specific thresholds. Further efforts are needed to integrate CAD into routine TB case detection programs in high-burden communities.

Circulating cell-free RNA in blood as a host response biomarker for detection of tuberculosis.

Tuberculosis (TB) remains a leading cause of death from an infectious disease worldwide, partly due to a lack of effective strategies to screen and triage individuals with potential TB. Whole blood RNA signatures have been tested as biomarkers for TB, but have failed to meet the World Health Organization's (WHO) optimal target product profiles (TPP). Here, we use RNA sequencing and machine-learning to investigate the utility of plasma cell-free RNA (cfRNA) as a host-response biomarker for TB in cohorts from Uganda, Vietnam and Philippines. We report a 6-gene cfRNA signature, which differentiates TB-positive and TB-negative individuals with AUC = 0.95, 0.92, and 0.95 in test, training and validation, respectively. This signature meets WHO TPPs (sensitivity: 97.1% [95% CI: 80.9-100%], specificity: 85.2% [95% CI: 72.4-100%]) regardless of geographic location, sample collection method and HIV status. Overall, our results identify plasma cfRNA as a promising host response biomarker to diagnose TB.

C-reactive protein-based tuberculosis triage testing: a multi-country diagnostic accuracy study.

Rationale: C-reactive protein (CRP)-based tuberculosis (TB) screening is recommended for people with HIV (PWH). However, its performance among people without HIV and in diverse settings is unknown. Objectives: In a multi-country study, we aimed to determine whether CRP meets the minimum accuracy targets (sensitivity ≥90%, specificity ≥70%) for an effective TB triage test. Methods/Measurements: Consecutive outpatient adults with cough ≥2 weeks from five TB endemic countries in Africa and Asia had baseline blood collected for point-of-care CRP testing and HIV and diabetes screening. Sputum samples were collected for Xpert MTB/RIF Ultra (Xpert) testing and culture. CRP sensitivity and specificity (5 mg/L cut-point) was determined in reference to sputum test results and compared by country, sex, and HIV and diabetes status. Variables affecting CRP performance were identified using a multivariate receiver operating characteristic (ROC) regression model. Results: Among 2904 participants, of whom 613 (21%) had microbiologically-confirmed TB, CRP sensitivity was 84% (95% CI: 81-87%) and specificity was 61% (95% CI: 59-63%). CRP accuracy varied geographically, with higher sensitivity in African countries (≥91%) than Asian countries (64-82%). Sensitivity was higher among men than women (87% vs. 79%, difference +8%, 95% CI: 1-15%) and specificity was higher among people without HIV than PWH (64% vs. 45%, difference +19%, 95% CI: 13-25%). ROC regression identified country and measures of TB disease severity as predictors of CRP performance. Conclusions: Overall, CRP did not achieve the minimum accuracy targets and its performance varied by setting and in some sub-groups, likely reflecting population differences in mycobacterial load.

Solicited Cough Sound Analysis for Tuberculosis Triage Testing: The CODA TB DREAM Challenge Dataset.

Cough is a common and commonly ignored symptom of lung disease. Cough is often perceived as difficult to quantify, frequently self-limiting, and non-specific. However, cough has a central role in the clinical detection of many lung diseases including tuberculosis (TB), which remains the leading infectious disease killer worldwide. TB screening currently relies on self-reported cough which fails to meet the World Health Organization (WHO) accuracy targets for a TB triage test. Artificial intelligence (AI) models based on cough sound have been developed for several respiratory conditions, with limited work being done in TB. To support the development of an accurate, point-of-care cough-based triage tool for TB, we have compiled a large multi-country database of cough sounds from individuals being evaluated for TB. The dataset includes more than 700,000 cough sounds from 2,143 individuals with detailed demographic, clinical and microbiologic diagnostic information. We aim to empower researchers in the development of cough sound analysis models to improve TB diagnosis, where innovative approaches are critically needed to end this long-standing pandemic.

Sex and HIV Differences in Preserved Ratio Impaired Spirometry (PRISm) Among Ugandans Postpneumonia.

Background: Preserved ratio impaired spirometry (PRISm), defined as a normal ratio of forced expiratory volume in 1 second (FEV1) to forced vital capacity (≥0.70) with low FEV1 (<80% predicted), has been associated with increased mortality in the general population. Female sex has been associated with increased odds of PRISm in people without HIV. People with HIV (PWH) are at increased risk for lung function abnormalities, but whether HIV modifies the effect of sex on PRISm development is largely unknown. Methods: Adults with and without HIV underwent baseline followed by serial spirometry after completing therapy for pneumonia, predominantly tuberculosis (TB), in Kampala, Uganda. Using generalized estimating equations adjusted for age, body mass index, smoking, biomass fuel exposure, HIV, and TB status, we compared individuals with PRISm with those with normal spirometry. These models were stratified by HIV status. Results: Of 339 baseline participants, 153 (45%) were women; 129 (38%) had HIV, of whom 53% were women. Overall, 105/339 participants (31%) had PRISm at baseline. HIV was associated with lower odds of PRISm (adjusted odds ratio [aOR], 0.38; 95% CI, 0.21-0.68; P = .001). Female sex trended toward increased odds of PRISm among all participants (aOR, 1.65; 95% CI, 0.99-2.75; P = .052). The association between female sex and PRISm tended to be stronger among PWH (aOR, 3.16; 95% CI, 1.14-8.76; P = .03) than among those without HIV (aOR, 1.34; 95% CI, 0.73-2.45; P = .34); this study was underpowered to detect an HIV-sex interaction of this magnitude (P = .30). Conclusions: Among Ugandan adults who recovered from pneumonia, female sex was associated with increased odds and HIV with decreased odds of PRISm, suggesting independent sex and HIV effects on PRISm pathogenesis.

Accuracy of the tuberculosis molecular bacterial load assay to diagnose and monitor response to anti-tuberculosis therapy: a longitudinal comparative study with standard-of-care smear microscopy, Xpert MTB/RIF Ultra, and culture in Uganda.

BACKGROUND: In 2018, the tuberculosis molecular bacterial load assay (TB-MBLA), a ribosomal RNA-based test, was acknowledged by WHO as a molecular assay that could replace smear microscopy and culture for monitoring tuberculosis treatment response. In this study, we evaluated the accuracy of TB-MBLA for diagnosis and monitoring of treatment response in comparison with standard-of-care tests. METHODS: For this longitudinal prospective study, patients aged 18 years or older with presumptive tuberculosis (coughing for at least 2 weeks, night sweats, and weight loss) were enrolled at China-Uganda Friendship Hospital Naguru (Kampala, Uganda). Participants were evaluated for tuberculosis by TB-MBLA in comparison with Xpert MTB/RIF Ultra (Xpert-Ultra) and smear microscopy, with Mycobacteria Growth Indicator Tube (MGIT) culture as a reference test. Participants who were positive on Xpert-Ultra were enrolled on a standard 6-month anti-tuberculosis regimen, and monitored for treatment response at weeks 2, 8, 17, and 26 after initiation of treatment and then 3 months after treatment. FINDINGS: Between Nov 15, 2019, and June 15, 2022, 210 participants (median age 35 years [IQR 27-44]) were enrolled. 135 (64%) participants were male and 72 (34%) were HIV positive. The pretreatment diagnostic sensitivities of TB-MBLA and Xpert-Ultra were similar (both 99% [95% CI 95-100]) but the specificity was higher for TB-MBLA (90% [83-96]) than for Xpert-Ultra (78% [68-86]). Ten participants were Xpert-Ultra trace positive, eight (80%) of whom were negative by TB-MBLA and MGIT culture. Smear microscopy had lower diagnostic sensitivity (75% [65-83]) but higher specificity (98% [93-100]) than TB-MBLA and Xpert-Ultra. Among participants who were smear microscopy negative, the sensitivity of TB-MBLA was 96% (95 CI 80-100) and was 100% (95% CI 86-100) in those who were HIV positive. 129 (61%) participants were identified as tuberculosis positive by Xpert-Ultra and these individuals were enrolled in the treatment group and monitored for treatment response. According to TB-MBLA, 19 of these patients cleared bacillary load to zero by week 2 of treatment and remained negative throughout the 6-month treatment follow-up. Positivity for tuberculosis decreased with treatment as measured by all tests, but the rate was slower with Xpert-Ultra. Consequently, 31 (33%) of 95 participants were still Xpert-Ultra positive at the end of treatment but were clinically well and negative on TB-MBLA and culture at 6 months of treatment. Two patients were still Xpert-Ultra positive with a further 3 months of post-treatment follow-up. The rate of conversion to negative of the DNA-based Xpert-Ultra was 3·3-times slower than that of the rRNA-based TB-MBLA. Consequently for the same patient, it would take 13 weeks and 52 weeks to reach complete tuberculosis negativity by TB-MBLA and Xpert-Ultra, respectively. Participants who were positive on smear microscopy at 8 weeks, who received an extra month of intensive treatment, had a similar TB-MBLA-measured bacillary load at 8 weeks to those who were smear microscopy negative. INTERPRETATION: TB-MBLA has a similar performance to Xpert-Ultra for pretreatment diagnosis of tuberculosis, but is more accurate at detecting and characterising the response to treatment than Xpert-Ultra and standard-of-care smear microscopy. FUNDING: European and Developing Countries Clinical Trials Partnership, Makerere University Research and Innovation Fund, US National Institutes of Health.

New Manual Quantitative Polymerase Chain Reaction Assay Validated on Tongue Swabs Collected and Processed in Uganda Shows Sensitivity That Rivals Sputum-based Molecular Tuberculosis Diagnostics.

BACKGROUND: Sputum-based testing is a barrier to increasing access to molecular diagnostics for tuberculosis (TB). Many people with TB are unable to produce sputum, and sputum processing increases assay complexity and cost. Tongue swabs are emerging as an alternative to sputum, but performance limits are uncertain. METHODS: From June 2022 to July 2023, we enrolled 397 consecutive adults with cough >2 weeks at 2 health centers in Kampala, Uganda. We collected demographic and clinical information, sputum for TB testing (Xpert MTB/RIF Ultra and 2 liquid cultures), and tongue swabs for same-day quantitative polymerase chain reaction (qPCR) testing. We evaluated tongue swab qPCR diagnostic accuracy versus sputum TB test results, quantified TB targets per swab, assessed the impact of serial swabbing, and compared 2 swab types (Copan FLOQSWAB and Steripack spun polyester). RESULTS: Among 397 participants, 43.1% were female, median age was 33 years, 23.5% were diagnosed with human immunodeficiency virus, and 32.0% had confirmed TB. Sputum Xpert Ultra and tongue swab qPCR results were concordant for 98.2% (95% confidence interval [CI]: 96.2-99.1) of participants. Tongue swab qPCR sensitivity was 92.6% (95% CI: 86.5 to 96.0) and specificity was 99.1% (95% CI: 96.9 to 99.8) versus microbiological reference standard. A single tongue swab recovered a 7-log range of TB copies, with a decreasing recovery trend among 4 serial swabs. Swab types performed equivalently. CONCLUSIONS: Tongue swabs are a promising alternative to sputum for molecular diagnosis of TB, with sensitivity approaching sputum-based molecular tests. Our results provide valuable insights for developing successful tongue swab-based TB diagnostics.

The burden of severe asthma in sub-Saharan Africa: Findings from the African Severe Asthma Project.

Background: Severe asthma is associated with high morbidity, mortality, and health care utilization, but its burden in Africa is unknown. Objective: We sought to determine the burden (prevalence, mortality, and activity and work impairment) of severe asthma in 3 countries in East Africa: Uganda, Kenya, and Ethiopia. Methods: Using the American Thoracic Society/European Respiratory Society case definition of severe asthma, we analyzed for the prevalence of severe asthma (requiring Global Initiative for Asthma [GINA] steps 4-5 asthma medications for the previous year to achieve control) and severe refractory asthma (remains uncontrolled despite treatment with GINA steps 4-5 asthma medications) in a cohort of 1086 asthma patients who had been in care for 12 months and had received all GINA-recommended medications. Asthma control was assessed by the asthma control questionnaire (ACQ). Results: Overall, the prevalence of severe asthma and severe refractory asthma was 25.6% (95% confidence interval [CI], 23.1-28.3) and 4.6% (95% CI, 3.5-6.0), respectively. Patients with severe asthma were (nonsevere vs severe vs severe refractory) older (39, 42, 45 years, P = .011), had high skin prick test reactivity (67.1%, 76.0%, 76.0%, P = .004), had lower forced expiratory volume in 1 second percentage (81%, 61%, 55.5%, P < .001), had lower quality of life score (129, 127 vs 121, P < .001), and had higher activity impairment (10%, 30%, 50%, P < .001). Factors independently associated with severe asthma were hypertension comorbidity; adjusted odds ratio 2.21 (1.10-4.47), P = .027, high bronchial hyperresponsiveness questionnaire score; adjusted odds ratio 2.16 (1.01-4.61), P = .047 and higher ACQ score at baseline 2.80 (1.55-5.08), P = .001. Conclusion: The prevalence of severe asthma in Africa is high and is associated with high morbidity and poor quality of life.

Evaluation of the Xpert MTB Host Response assay for the triage of patients with presumed pulmonary tuberculosis: a prospective diagnostic accuracy study in Viet Nam, India, the Philippines, Uganda, and South Africa.

BACKGROUND: Non-sputum-based triage tests for tuberculosis are a priority for ending tuberculosis. We aimed to evaluate the diagnostic accuracy of the late-prototype Xpert MTB Host Response (Xpert HR) blood-based assay. METHODS: We conducted a prospective diagnostic accuracy study among outpatients with presumed tuberculosis in outpatient clinics in Viet Nam, India, the Philippines, Uganda, and South Africa. Eligible participants were aged 18 years or older and reported cough lasting at least 2 weeks. We excluded those receiving tuberculosis treatment in the preceding 12 months and those who were unwilling to consent. Xpert HR was performed on capillary or venous blood. Reference standard testing included sputum Xpert MTB/RIF Ultra and mycobacterial culture. We performed receiver operating characteristic (ROC) analysis to identify the optimal cutoff value for the Xpert HR to achieve the target sensitivity of 90% or more while maximising specificity, then calculated diagnostic accuracy using this cutoff value. This study was prospectively registered with ClinicalTrials.gov, NCT04923958. FINDINGS: Between July 13, 2021, and Aug 15, 2022, 2046 adults with at least 2 weeks of cough were identified, of whom 1499 adults (686 [45·8%] females and 813 [54·2%] males) had valid Xpert HR and reference standard results. 329 (21·9%) had microbiologically confirmed tuberculosis. Xpert HR had an area under the ROC curve of 0·89 (95% CI 0·86-0·91). The optimal cutoff value was less than or equal to -1·25, giving a sensitivity of 90·3% (95% CI 86·5-93·3; 297 of 329) and a specificity of 62·6% (95% CI 59·7-65·3; 732 of 1170). Sensitivity was similar across countries, by sex, and by subgroups, although specificity was lower in people living with HIV (45·1%, 95% CI 37·8-52·6) than in those not living with HIV (65·9%, 62·8-68·8; difference of 20·8%, 95% CI 13·0-28·6; p<0·0001). Xpert HR had high negative predictive value (95·8%, 95% CI 94·1-97·1), but positive predictive value was only 40·1% (95% CI 36·8-44·1). Using the Xpert HR as a triage test would have reduced confirmatory sputum testing by 57·3% (95% CI 54·2-60·4). INTERPRETATION: Xpert HR did not meet WHO minimum specificity targets for a non-sputum-based triage test for pulmonary tuberculosis. Despite promise as a rule-out test that could reduce confirmatory sputum testing, further cost-effectiveness modelling and data on acceptability and usability are needed to inform policy recommendations. FUNDING: National Institute of Allergy and Infectious Diseases of the US National Institutes of Health. TRANSLATIONS: For the Vietnamese and Tagalog translations of the abstract see Supplementary Materials section.

Neutrophil-Lymphocyte and Platelet-Lymphocyte Ratios in Distinguishing Lung Cancer in People with HIV.

Objective: The neutrophil-lymphocyte ratio (NLR) and platelet-lymphocyte ratio (PLR) demonstrate good diagnostic accuracy in distinguishing lung cancer patients from healthy individuals, primarily in HIV-negative populations. We determined the sensitivity (Se), specificity (Sp), and area under the curve (AUC) of the NLR and PLR in discriminating between people living with HIV (PLWH) with and without lung cancer. Methods: This is a comparative analysis of secondary data. Cases were PLWH with lung cancer from a retrospective cohort treated at the Uganda Cancer Institute. Controls were unmatched PLWH without lung cancer who were randomly selected from three HIV clinics in Uganda. Se, Sp, and AUC analysis and determination of optimal cutoffs were performed using receiver operating characteristic (ROC) curves. Results: Of 115 PLWH (18 cases and 97 controls), 83 (72.2%) were female, 110 (95.7) were on ART, and the median (IQR) age was 46 (38-51) years. The median (IQR) NLR was higher among cases than controls (3.53 (3.14-7.71) vs. 0.92 (0.67-1.09), p < 0.001). Similarly, the PLR was higher among cases than controls (237.5 (177.8-361.6) vs. 123.6 (100.6-155.4), p=0.001). At a cutoff of 2.44, the respective Se, Sp, and AUC of the NLR were 87.5% (95% CI: 61.7%-98.4%), 100% (95% CI: 96.2%-100%), and 0.94 (95% CI: 0.85-1.00, p < 0.001). Similarly, the respective Se, Sp, and AUC for the PLR were 75% (95% CI: 47.6%-92.7%), 87.2% (95% CI: 78.8%-93.2%), and 0.81 (95% CI: 0.70-0.93, p < 0.001) at a cutoff of 196.3. Conclusion: The NLR and PLR discriminated PLWH with and without lung cancer and could be useful in PLWH with respiratory symptoms in whom lung cancer can easily be misdiagnosed as other lung pathology.