Challenges in childhood tuberculosis

Comparison of gene set scoring methods for reproducible evaluation of tuberculosis gene signatures

BACKGROUND

Blood-based transcriptional gene signatures for tuberculosis (TB) have been developed with potential use to diagnose disease. However, an unresolved issue is whether gene set enrichment analysis of the signature transcripts alone is sufficient for prediction and differentiation or whether it is necessary to use the original model created when the signature was derived. Intra-method comparison is complicated by the unavailability of original training data and missing details about the original trained model. To facilitate the utilization of these signatures in TB research, comparisons between gene set scoring methods cross-data validation of original model implementations are needed.

METHODS

We compared the performance of 19 TB gene signatures across 24 transcriptomic datasets using both rrebuilt original models and gene set scoring methods. Existing gene set scoring methods, including ssGSEA, GSVA, PLAGE, Singscore, and Zscore, were used as alternative approaches to obtain the profile scores. The area under the ROC curve (AUC) value was computed to measure performance. Correlation analysis and Wilcoxon paired tests were used to compare the performance of enrichment methods with the original models.

RESULTS

For many signatures, the predictions from gene set scoring methods were highly correlated and statistically equivalent to the results given by the original models. In some cases, PLAGE outperformed the original models when considering signatures' weighted mean AUC values and the AUC results within individual studies.

CONCLUSION

Gene set enrichment scoring of existing gene sets can distinguish patients with active TB disease from other clinical conditions with equivalent or improved accuracy compared to the original methods and models. These data justify using gene set scoring methods of published TB gene signatures for predicting TB risk and treatment outcomes, especially when original models are difficult to apply or implement.

Pediatric Tuberculosis: A Review of Evidence-Based Best Practices for Clinicians and Health Care Providers

Advances in pediatric TB care are promising, the result of decades of advocacy, operational and clinical trials research, and political will by national and local TB programs in high-burden countries. However, implementation challenges remain in linking policy to practice and scaling up innovations for prevention, diagnosis, and treatment of TB in children, especially in resource-limited settings. There is both need and opportunity to strengthen clinician confidence in making a TB diagnosis and managing the various manifestations of TB in children, which can facilitate the translation of evidence to action and expand access to new tools and strategies to address TB in this population. This review aims to summarize existing guidance and best practices for clinicians and health care providers in low-resource, TB-endemic settings and identify resources with more detailed and actionable information for decision-making along the clinical cascade to prevent, find, and cure TB in children.

Comparison of gene set scoring methods for reproducible evaluation of multiple tuberculosis gene signatures

Rationale

Many blood-based transcriptional gene signatures for tuberculosis (TB) have been developed with potential use to diagnose disease, predict risk of progression from infection to disease, and monitor TB treatment outcomes. However, an unresolved issue is whether gene set enrichment analysis (GSEA) of the signature transcripts alone is sufficient for prediction and differentiation, or whether it is necessary to use the original statistical model created when the signature was derived. Intra-method comparison is complicated by the unavailability of original training data, missing details about the original trained model, and inadequate publicly-available software tools or source code implementing models. To facilitate these signatures' replicability and appropriate utilization in TB research, comprehensive comparisons between gene set scoring methods with cross-data validation of original model implementations are needed.

Objectives

We compared the performance of 19 TB gene signatures across 24 transcriptomic datasets using both re-rebuilt original models and gene set scoring methods to evaluate whether gene set scoring is a reasonable proxy to the performance of the original trained model. We have provided an open-access software implementation of the original models for all 19 signatures for future use.

Methods

We considered existing gene set scoring and machine learning methods, including ssGSEA, GSVA, PLAGE, Singscore, and Zscore, as alternative approaches to profile gene signature performance. The sample-size-weighted mean area under the curve (AUC) value was computed to measure each signature's performance across datasets. Correlation analysis and Wilcoxon paired tests were used to analyze the performance of enrichment methods with the original models.

Measurement and Main Results

For many signatures, the predictions from gene set scoring methods were highly correlated and statistically equivalent to the results given by the original diagnostic models. PLAGE outperformed all other gene scoring methods. In some cases, PLAGE outperformed the original models when considering signatures' weighted mean AUC values and the AUC results within individual studies.

Conclusion

Gene set enrichment scoring of existing blood-based biomarker gene sets can distinguish patients with active TB disease from latent TB infection and other clinical conditions with equivalent or improved accuracy compared to the original methods and models. These data justify using gene set scoring methods of published TB gene signatures for predicting TB risk and treatment outcomes, especially when original models are difficult to apply or implement.

Evidence to Action: Translating Innovations in Management of Child and Adolescent TB into Routine Practice in High-Burden Countries

Child and adolescent tuberculosis (TB) has been long neglected by TB programs but there have been substantive strides in prioritizing TB among these populations in the past two decades. Yet, gaps remain in translating evidence and policy to action at the primary care level, ensuring access to novel tools and approaches to diagnosis, treatment, and prevention for children and adolescents at risk of TB disease. This article describes the progress that has been made and the gaps that remain in addressing TB among children and adolescents while also highlighting pragmatic approaches and the role of multisectoral partnerships in facilitating integration of innovations into routine program practice.

Repurposing Cefazolin-Avibactam for the Treatment of Drug Resistant Mycobacterium tuberculosis

Background: While tuberculosis (TB) is curable and preventable, the most effective first-line antibiotics cannot kill multi-drug resistant (MDR) Mycobacterium tuberculosis (Mtb). Therefore, effective drugs are needed to combat MDR-TB, especially in children. Our objective was to repurpose cefazolin for MDR-TB treatment in children using principles of pharmacokinetic/pharmacodynamics (PK/PD).

Methods: Cefazolin minimum inhibitory concentration (MIC) was identified in 17 clinical Mtb strains, with and without combination of the β-lactamase inhibitor, avibactam. Next, dose-ranging studies were performed using the intracellular hollow fiber model of TB (HFS-TB) to identify the optimal cefazolin exposure. Monte Carlo experiments were then performed in 10,000 children for optimal dose identification based on cumulative fraction of response (CFR) and Mtb susceptibility breakpoint in three age-groups.

Results: Avibactam reduced the cefazolin MICs by five tube dilutions. Cefazolin-avibactam demonstrated maximal kill of 4.85 log₁₀ CFU/mL in the intracellular HFS-TB over 28 days. The % time above MIC associated with maximal effect (EC₈₀) was 46.76% (95% confidence interval: 43.04–50.49%) of dosing interval. For 100 mg/kg once or twice daily, the CFR was 8.46 and 61.39% in children <3 years with disseminated TB, 9.70 and 84.07% for 3–5 years-old children, and 17.20 and 76.13% for 12–15 years-old children. The PK/PD-derived susceptibility breakpoint was dose dependent at 1–2 mg/L.

Conclusion: Cefazolin-avibactam combination demonstrates efficacy against both drug susceptible and MDR-TB clinical strains in the HFS-TB and could potentially be used to treat children with tuberculosis. Clinical studies are warranted to validate our findings.

Multiparameter Responses to Tedizolid Monotherapy and Moxifloxacin Combination Therapy Models of Children With Intracellular Tuberculosis

Background

Children are often neglected during early development of antituberculosis agents, and most receive treatment after it is first tested in adults. However, very young children have tuberculosis that differs in many respects from adult cavitary pneumonia and could have different toxicity profiles to drugs. Linezolid is effective against intracellular tuberculosis, a common manifestation in young children. However, linezolid has considerable toxicity due to inhibition of mitochondrial enzymes. Tedizolid could be a replacement if it shows equal efficacy and reduced toxicity.

Methods

We performed tedizolid dose-effect studies in the hollow fiber system model of intracellular tuberculosis. We measured linezolid concentrations, colony-forming units (CFU), time-to-positivity, and monocyte viability and performed RNA sequencing on infected cells collected from repetitive sampling of each system. We also compared efficacy of tedizolid vs linezolid and vs tedizolid-moxifloxacin combination.

Results

There was no downregulation of mitochondrial enzyme genes, with a tedizolid 0-24 hour area under the concentration-time curve (AUC0-24) of up to 90 mg*h/L. Instead, high exposures led to increased mitochondrial gene expression and monocyte survival. The AUC0-24 to minimum inhibitory concentration ratio associated with 80% of maximal bacterial kill (EC80) was 184 by CFU/mL (r2 = 0.96) and 189 by time-to-positivity (r2 = 0.99). Tedizolid EC80 killed 4.0 log10 CFU/mL higher than linezolid EC80. The tedizolid-moxifloxacin combination had a bacterial burden elimination rate constant of 0.27 ± 0.05 per day.

Conclusions

Tedizolid demonstrated better efficacy than linezolid, without the mitochondrial toxicity gene or cytotoxicity signatures encountered with linezolid. Tedizolid-moxifloxacin combination had a high bacterial elimination rate.

Suboptimal Antituberculosis Drug Concentrations and Outcomes in Small and HIV-Coinfected Children in India: Recommendations for Dose Modifications

This work aimed to evaluate the once‐daily antituberculosis treatment as recommended by the new Indian pediatric guidelines. Isoniazid, rifampin, and pyrazinamide concentration–time profiles and treatment outcome were obtained from 161 Indian children with drug‐sensitive tuberculosis undergoing thrice‐weekly dosing as per previous Indian pediatric guidelines. The exposure–response relationships were established using a population pharmacokinetic‐pharmacodynamic approach. Rifampin exposure was identified as the unique predictor of treatment outcome. Consequently, children with low body weight (4–7 kg) and/or HIV infection, who displayed the lowest rifampin exposure, were associated with the highest probability of unfavorable treatment (therapy failure, death) outcome (P_unfavorable). Model‐based simulation of optimized (P_unfavorable ≤ 5%) rifampin once‐daily doses were suggested per treatment weight band and HIV coinfection status (33% and 190% dose increase, respectively, from the new Indian guidelines). The established dose‐exposure–response relationship could be pivotal in the development of future pediatric tuberculosis treatment guidelines.

Partnerships to Design Novel Regimens to Treat Childhood Tuberculosis, Sui Generis: The Road Ahead

There has been a recent expansion of preclinical models to predict the efficacy of regimens to treat adults with tuberculosis. Despite increasing global interest in childhood tuberculosis, these same tools have not been employed to develop pediatric regimens. Children differ from adults in bacillary burden, spectrum of disease, the metabolism and distribution of antituberculosis drugs, and the toxicity experienced. The studies documented in this series describe a proof-of-concept approach to pediatric regimen development. We propose a program of investigation that would take this forward into a systematic and comprehensive method to find optimal drug combinations to use in children, ideal exposures, and required dosing. Although the number of possible drug combinations is extensive, a series of principles could be employed to select likely effective regimens. Regimens should avoid drugs with overlapping toxicity or linked mechanisms of resistance and should aim to include drugs with different mechanisms of action and ones that are able to target different subpopulations of mycobacteria. Finally drugs should penetrate into body sites necessary for treating pediatric disease. At an early stage, this body of work would need to engage with regulatory agencies and bodies that formulate guidelines, so that once regimens and dosages are identified, translation into clinical studies and clinical practice can be rapid. The development of child-friendly drug formulations would need to be carried out in parallel so that pharmacokinetic studies can be undertaken as formulations are created. Significant research and development would be required and a wide range of stakeholders would need to be engaged. The time is right to consider a more thoughtful and systematic approach toward identifying, testing, and comparing combinations of drugs for children with tuberculosis.

Drug Concentration Thresholds Predictive of Therapy Failure and Death in Children With Tuberculosis: Bread Crumb Trails in Random Forests

Background. The role of drug concentrations in clinical outcomes in children with tuberculosis is unclear. Target concentrations for dose optimization are unknown.

Methods. Plasma drug concentrations measured in Indian children with tuberculosis were modeled using compartmental pharmacokinetic analyses. The children were followed until end of therapy to ascertain therapy failure or death. An ensemble of artificial intelligence algorithms, including random forests, was used to identify predictors of clinical outcome from among 30 clinical, laboratory, and pharmacokinetic variables.

Results. Among the 143 children with known outcomes, there was high between-child variability of isoniazid, rifampin, and pyrazinamide concentrations: 110 (77%) completed therapy, 24 (17%) failed therapy, and 9 (6%) died. The main predictors of therapy failure or death were a pyrazinamide peak concentration <38.10 mg/L and rifampin peak concentration <3.01 mg/L. The relative risk of these poor outcomes below these peak concentration thresholds was 3.64 (95% confidence interval [CI], 2.28–5.83). Isoniazid had concentration-dependent antagonism with rifampin and pyrazinamide, with an adjusted odds ratio for therapy failure of 3.00 (95% CI, 2.08–4.33) in antagonism concentration range. In regard to death alone as an outcome, the same drug concentrations, plus z scores (indicators of malnutrition), and age <3 years, were highly ranked predictors. In children <3 years old, isoniazid 0- to 24-hour area under the concentration-time curve <11.95 mg/L × hour and/or rifampin peak <3.10 mg/L were the best predictors of therapy failure, with relative risk of 3.43 (95% CI, .99–11.82).

Conclusions. We have identified new antibiotic target concentrations, which are potential biomarkers associated with treatment failure and death in children with tuberculosis.

Examining the quality of childhood tuberculosis diagnosis in Cambodia: a cross-sectional study

BACKGROUND

Cambodia is one of the 22 countries with the highest TB burden. While childhood TB is estimated to make up 10-20% of total TB cases in high-burden settings, this proportion ranges from 1.3 to 39.4% throughout Cambodia's provinces, suggesting potential under- and over-diagnosis of childhood TB, subnationally. The proportion of case notifications classified as extrapulmonary TB out of total TB case notifications in children is 87%, greatly exceeding the expected global range of 20-30%. There is a gap in the literature on how childhood TB is diagnosed in resource-poor settings, and the quality of diagnoses. The study's aim is to quantitatively assess the quality of clinician performance and availability of diagnostic tools, for diagnosing childhood TB in high-burden Operational Districts in Cambodia.

METHODS

Between August and September of 2015, a cross-sectional study was conducted at referral hospitals and villages in 24 high-burden Operational Districts. 40 clinicians, and 104 parents whose child was recently diagnosed with TB were interviewed. Questionnaires assessed availability of diagnostic tools, and clinician knowledge and practice during a clinical examination. Descriptive statistics were calculated to provide cross-sectional data.

RESULTS

Availability of advanced diagnostic tools was low. Only 27.5% of clinicians had Xpert machines available at their facility, and 5% had equipment to perform gastric aspiration. 77.5% of clinicians reported that they had a chest X-ray at their facility, but only 34.6% of parents reported that the clinician conducted a chest X-ray. 72.5% of clinicians could name 5 out of 7 main TB screening criteria; however, parent data suggests that clinicians may not be applying knowledge to practice. The mean number of examinations/tests the clinician conducted during the clinical assessment of the child was 1.64. Of the parents whose child had an enlarged lymph node, 60.22% described lymph node characteristics that were not suggestive of TB.

CONCLUSION

Limited availability of diagnostic tools and suboptimal clinician performance highlight where resources should be allocated to improve quality of diagnoses. Further research needs to be done in low burden Operational Districts to determine the capacity of clinicians and health facilities for diagnosing childhood TB, where cases are likely being missed.

Safety and Immunogenicity of the Recombinant Mycobacterium bovis BCG Vaccine VPM1002 in HIV-Unexposed Newborn Infants in South Africa

Tuberculosis is a global threat to which infants are especially vulnerable. Effective vaccines are required to protect infants from this devastating disease. VPM1002, a novel recombinant Mycobacterium bovis bacillus Calmette-Guérin (BCG) vaccine previously shown to be safe and immunogenic in adults, was evaluated for safety in its intended target population, namely, newborn infants in a region with high prevalence of tuberculosis. A total of 48 newborns were vaccinated intradermally with VPM1002 (n = 36) or BCG Danish strain (n = 12) in a phase II open-labeled, randomized trial with a 6-month follow-up period. Clinical and laboratory measures of safety were evaluated during this time. In addition, vaccine-induced immune responses to mycobacteria were analyzed in whole-blood stimulation and proliferation assays. The safety parameters and immunogenicity were comparable in the two groups. Both vaccines induced interleukin-17 (IL-17) responses; however, VPM1002 vaccination led to an increase of CD8+ IL-17+ T cells at the week 16 and month 6 time points. The incidence of abscess formation was lower for VPM1002 than for BCG. We conclude that VPM1002 is a safe, well-tolerated, and immunogenic vaccine in newborn infants, confirming results from previous trials in adults. These results strongly support further evaluation of the safety and efficacy of this vaccination in larger studies. (This study has been registered at ClinicalTrials.gov under registration no. NCT01479972.).

Tuberculosis

Tuberculosis (TB) is an airborne infectious disease caused by organisms of the Mycobacterium tuberculosis complex. Although primarily a pulmonary pathogen, M. tuberculosis can cause disease in almost any part of the body. Infection with M. tuberculosis can evolve from containment in the host, in which the bacteria are isolated within granulomas (latent TB infection), to a contagious state, in which the patient will show symptoms that can include cough, fever, night sweats and weight loss. Only active pulmonary TB is contagious. In many low-income and middle-income countries, TB continues to be a major cause of morbidity and mortality, and drug-resistant TB is a major concern in many settings. Although several new TB diagnostics have been developed, including rapid molecular tests, there is a need for simpler point-of-care tests. Treatment usually requires a prolonged course of multiple antimicrobials, stimulating efforts to develop shorter drug regimens. Although the Bacillus Calmette-Guérin (BCG) vaccine is used worldwide, mainly to prevent life-threatening TB in infants and young children, it has been ineffective in controlling the global TB epidemic. Thus, efforts are underway to develop newer vaccines with improved efficacy. New tools as well as improved programme implementation and financing are necessary to end the global TB epidemic by 2035.

Childhood Tuberculosis in a Sub-Saharan Tertiary Facility: Epidemiology and Factors Associated with Treatment Outcome

Childhood tuberculosis (TB) is a diagnostic challenge in developing countries, and patient outcome can be influenced by certain factors. We report the disease course, clinical profile and factors associated with treatment outcome in a tertiary facility of Kinshasa. Documentary and analytical studies were conducted using clinical and exploratory data for children aged up to 15 years who were admitted to the University Clinics of Kinshasa for TB. Data are presented as frequencies and averages, and binary and logistic regression analyses were performed. Of 283 children with TB, 82 (29.0%) had smear-negative TB, 40 (14.1%) had smear-positive TB, 159 (56.1%) had extra-pulmonary TB (EPTB), 2 (0.7%) had multidrug-resistant TB (MDR-TB), 167 (59.0%) completed treatment, 30 (10.6%) were cured, 7 (2.5%) failed treatment, 4 (1.4%) died, 55 (19.4%) were transferred to health centers nearest their home, and 20 (7.0%) were defaulters. In the binary analysis, reported TB contacts (p = 0.048), type of TB (p = 0.000), HIV status (p = 0.050), Ziehl-Nielsen test result (p = 0.000), Lowenstein culture (p = 0.004) and chest X-ray (p = 0.057) were associated with outcome. In the logistic regression, none of these factors was a significant predictor of outcome. Tertiary level care facilities must improve the diagnosis and care of patients with childhood TB, which justifies the development of alternative diagnostic techniques and the assessment of other factors that potentially affect outcome.