Search

Background: Estimation of prevalence and diagnostic test accuracy in tuberculosis (TB) prevalence surveys suffer from reference standard and verification biases. The former is attributed to the imperfect reference test used to bacteriologically confirm TB disease. The latter occurs when only the participants screening positive for any TB-compatible symptom or chest X-ray abnormality are selected for bacteriological testing (verification). Bayesian latent class analysis (LCA) alleviates the reference standard bias but suffers verification bias in TB prevalence surveys. This work aims to identify best-practice approaches to simultaneously alleviate the reference standard and verification biases in the estimates of pulmonary TB prevalence and diagnostic test performance in TB prevalence surveys. Methods: We performed a secondary analysis of 9869 participants aged ≥15 years from a community-based multimorbidity screening study in a rural district of KwaZulu-Natal, South Africa (Vukuzazi study). Participants were eligible for bacteriological testing using Xpert Ultra and culture if they reported any cardinal TB symptom or had an abnormal chest X-ray finding. We conducted Bayesian LCA in five ways to handle the unverified individuals: (i) complete-case analysis, (ii) analysis assuming the unverified individuals would be negative if bacteriologically tested, (iii) analysis of multiply-imputed datasets with imputation of the missing bacteriological test results for the unverified individuals using multivariate imputation via chained equations (MICE), and simultaneous imputation of the missing bacteriological test results in the analysis model assuming the missing bacteriological test results were (iv) missing at random (MAR), and (v) missing not at random (MNAR). We compared the results of (i)-(iii) to the analysis based on a composite reference standard (CRS) of Xpert Ultra and culture. Through simulation with an overall true prevalence of 2.0%, we evaluated the ability of the models to alleviate both biases simultaneously. Results: Based on simulation, Bayesian LCA with simultaneous imputation of the missing bacteriological test results under the assumption that the missing data are MAR and MNAR alleviate the reference standard and verification biases. CRS-based analysis and Bayesian LCA assuming the unverified are negative for TB alleviate the biases only when the true overall prevalence is <3.0%. Complete-case analysis produced biased estimates. In the Vukuzazi study, Bayesian LCA with simultaneous imputation of the missing bacteriological test results under the MAR and MNAR assumptions produced overall PTB prevalence of 0.9% (95% Credible Interval (CrI): 0.6–1.9) and 0.7% (95% CrI: 0.5–1.1) respectively alongside realistic estimates of overall diagnostic test sensitivity and specificity with substantially overlapping 95% CrI. The CRS-based analysis and Bayesian LCA assuming the unverified were negative for TB produced 0.7% (95% CrI: 0.5–0.9) and 0.7% (95% CrI: 0.5–1.2) overall PTB prevalence respectively with realistic estimates of overall diagnostic test sensitivity and specificity. Unlike CRS-based analysis, Bayesian LCA of multiply-imputed data using MICE mitigates both biases. Conclusion: The findings demonstrate the efficacy of these advanced techniques in alleviating the reference standard and verification biases, enhancing the robustness of community-based screening programs. Imputing missing values as negative for bacteriological tests is plausible under realistic assumptions. [ABSTRACT FROM AUTHOR]

Because TB control is still hampered by the limitations of diagnostic tools, diagnostic uncertainty is common. The decision to offer treatment is based on clinical decision-making. The therapeutic threshold, test threshold and test-treatment threshold can guide in making these decisions. This review summarizes the literature on methods to estimate the therapeutic threshold that have been applied for TB. Only five studies estimated the threshold for the diagnosis of TB. The therapeutic threshold can be estimated by prescriptive methods, based on calculations, and by descriptive methods, deriving the threshold from observing clinical practice. Test and test-treatment thresholds can be calculated using the therapeutic threshold and the characteristics of an available diagnostic test. Estimates of the therapeutic threshold for pulmonary TB from intuitive descriptive approaches (20%–50%) are higher than theoretical prescriptive calculations (2%–3%). In conclusion, estimates of the therapeutic threshold for pulmonary TB depend on the method used. Other methods exist within the field of decision-making that have yet to be implemented or adapted as tools to estimate the TB therapeutic threshold. Because clinical decision-making is a core element of TB management, it is necessary to find a new, clinician-friendly way to unbiasedly estimate context-specific, agreed upon therapeutic thresholds. [ABSTRACT FROM AUTHOR]

Simple Summary: We hypothesised that change in a magnetic resonance imaging (MRI) biomarker, the apparent diffusion coefficient (ADC) after 14 days of treatment could be a proxy for tumour regression grade (TRG) on pathology. Measurement of the imaging biomarker was standardised across centres. We restricted measurements to liver metastases from colorectal cancer and ensured a standardised chemotherapy approach. We identified and eliminated significant measurement error due to MRI scanner performance. We excluded studies that failed to conform to the imaging protocol or where images contained movement artefact. We ensured stability of the scanners by periodic quality control testing and used a standard, widely used data analysis technique to derive the ADC. Despite these measures, our results showed no significant correlation between ADC and TRG or between ADC and percentage of viable tumour, percentage necrosis, percentage fibrosis or a tumour proliferation index. This may reflect the complex cellular architecture of tumours after treatment. Background: Tumour apparent diffusion coefficient (ADC) from diffusion-weighted magnetic resonance imaging (MRI) is a putative pharmacodynamic/response biomarker but the relationship between drug-induced effects on the ADC and on the underlying pathology has not been adequately defined. Hypothesis: Changes in ADC during early chemotherapy reflect underlying histological markers of tumour response as measured by tumour regression grade (TRG). Methods: Twenty-six patients were enrolled in the study. Baseline, 14 days, and pre-surgery MRI were performed per study protocol. Surgical resection was performed in 23 of the enrolled patients; imaging-pathological correlation was obtained from 39 lesions from 21 patients. Results: There was no evidence of correlation between TRG and ADC changes at day 14 (study primary endpoint), and no significant correlation with other ADC metrics. In scans acquired one week prior to surgery, there was no significant correlation between ADC metrics and percentage of viable tumour, percentage necrosis, percentage fibrosis, or Ki67 index. Conclusions: Our hypothesis was not supported by the data. The lack of meaningful correlation between change in ADC and TRG is a robust finding which is not explained by variability or small sample size. Change in ADC is not a proxy for TRG in metastatic colorectal cancer. [ABSTRACT FROM AUTHOR]

Diagnostic accuracy studies in pulmonary tuberculosis (PTB) are complicated by the lack of a perfect reference standard. This limitation can be handled using latent class analysis (LCA), assuming independence between diagnostic test results conditional on the true unobserved PTB status. Test results could remain dependent, however, e.g. with diagnostic tests based on a similar biological basis. If ignored, this gives misleading inferences. Our secondary analysis of data collected during the first year (May 2018 –May 2019) of a community-based multi-morbidity screening program conducted in the rural uMkhanyakude district of KwaZulu Natal, South Africa, used Bayesian LCA. Residents of the catchment area, aged ≥15 years and eligible for microbiological testing, were analyzed. Probit regression methods for dependent binary data sequentially regressed each binary test outcome on other observed test results, measured covariates and the true unobserved PTB status. Unknown model parameters were assigned Gaussian priors to evaluate overall PTB prevalence and diagnostic accuracy of 6 tests used to screen for PTB: any TB symptom, radiologist conclusion, Computer Aided Detection for TB version 5 (CAD4TBv5≥53), CAD4TBv6≥53, Xpert Ultra (excluding trace) and culture. Before the application of our proposed model, we evaluated its performance using a previously published childhood pulmonary TB (CPTB) dataset. Standard LCA assuming conditional independence yielded an unrealistic prevalence estimate of 18.6% which was not resolved by accounting for conditional dependence among the true PTB cases only. Allowing, also, for conditional dependence among the true non-PTB cases produced a 1.1% plausible prevalence. After incorporating age, sex, and HIV status in the analysis, we obtained 0.9% (95% CrI: 0.6, 1.3) overall prevalence. Males had higher PTB prevalence compared to females (1.2% vs. 0.8%). Similarly, HIV+ had a higher PTB prevalence compared to HIV- (1.3% vs. 0.8%). The overall sensitivity for Xpert Ultra (excluding trace) and culture were 62.2% (95% CrI: 48.7, 74.4) and 75.9% (95% CrI: 61.9, 89.2), respectively. Any chest X-ray abnormality, CAD4TBv5≥53 and CAD4TBv6≥53 had similar overall sensitivity. Up to 73.3% (95% CrI: 61.4, 83.4) of all true PTB cases did not report TB symptoms. Our flexible modelling approach yields plausible, easy-to-interpret estimates of sensitivity, specificity and PTB prevalence under more realistic assumptions. Failure to fully account for diagnostic test dependence can yield misleading inferences. [ABSTRACT FROM AUTHOR]

Background: As untreated visceral leishmaniasis (VL) is fatal, reliable diagnostics are pivotal for accurate treatment allocation. The current diagnostic algorithm for VL in Ethiopia, which is based on the rK39 rapid diagnostic test and microscopy of tissue smears, lacks sensitivity. This probably leads to missed cases and patients not receiving treatment. Methodology: We conducted a retrospective study on stored microscopically negative spleen and bone marrow smears from suspected VL patients collected at the Leishmaniasis Research and Treatment Center (LRTC) in Gondar, northern Ethiopia between June 2019 and November 2020. Sociodemographic, clinical and treatment data were collected and samples were tested by real-time PCR targeting kinetoplast DNA. Principle findings: Among the 191 eligible samples (135 spleen and 56 bone marrow) with a microscopically negative and valid PCR result, 119 (62.3%) were positive by PCR, although Ct values for some were high (median 33.0). Approximately three quarters of these undiagnosed primary VL (77.3%) and relapse (69.6%) patients did not receive antileishmanial treatment. Of the 56 microscopically negative bone marrow samples, 46 (82.1%) were PCR positive, which is considerably higher compared to the microscopically negative spleen samples, for which 73 out of 135 (54.1%) were PCR positive. The odds of being PCR positive were significantly higher for bone marrow aspirates and higher when white blood cell values were lower and splenomegaly (in cm) was more pronounced. Conclusions: This study demonstrates that a lot of suspected VL patients remain undiagnosed and untreated. This indicates the urgent need for better diagnostics for VL in the East-African region. The outcomes of PCR positive should be closely monitored and treatment should be provided if the patient deteriorates. In resource limited settings, implementation of PCR on bone marrow aspirate smears of patients with low WBC values and splenomegaly could lead to considerable improvements in patient management. Author summary: As untreated visceral leishmaniasis (VL) is fatal, reliable diagnostics are important for accurate treatment allocation. The current diagnostic algorithm for VL in Ethiopia, which is based on the rK39 rapid diagnostic test and microscopy of tissue smears, lacks sensitivity. This probably leads to missed cases and patients not receiving treatment. To prove this, we conducted a study on stored microscopically negative spleen and bone marrow aspirate smears from suspected VL patients in Gondar, Ethiopia. Clinical and treatment data were collected and samples were tested for Leishmania by PCR. We found that about 60% of these microscopically negative samples were PCR positive. This PCR positivity rate was considerably higher in patients with a microscopically negative bone marrow compared to splenic aspirate. Importantly, more than three quarters of the patients with a PCR positive sample was not treated for VL. Overall, our study demonstrates the gap in the diagnostic algorithm for VL in northern Ethiopia, especially when bone marrow samples are used. In resource limited settings, we advise to challenge the current diagnostic algorithm and implement molecular tools to accurately diagnose patients. This could lead to considerable improvements in patient management in Ethiopia and beyond. [ABSTRACT FROM AUTHOR]

Background: The control of Ebola virus disease (EVD) outbreaks relies on rapid diagnosis and prompt action, a daunting task in limited-resource contexts. This study develops prediction scores that can help healthcare workers improve their decision-making at the triage-point of EVD suspect-cases during EVD outbreaks. Methods: We computed accuracy measurements of EVD predictors to assess their diagnosing ability compared with the reference standard GeneXpert® results, during the eastern DRC EVD outbreak. We developed predictive scores using the Spiegelhalter-Knill-Jones approach and constructed a clinical prediction score (CPS) and an extended clinical prediction score (ECPS). We plotted the receiver operating characteristic curves (ROCs), estimated the area under the ROC (AUROC) to assess the performance of scores, and computed net benefits (NB) to assess the clinical utility (decision-making ability) of the scores at a given cut-off. We performed decision curve analysis (DCA) to compare, at a range of threshold probabilities, prediction scores' decision-making ability and to quantify the number of unnecessary isolation. Results: The analysis was done on data from 10432 subjects, including 651 EVD cases. The EVD prevalence was 6.2% in the whole dataset, 14.8% in the subgroup of suspects who fitted the WHO Ebola case definition, and 3.2% for the set of suspects who did not fit this case definition. The WHO clinical definition yielded 61.6% sensitivity and 76.4% specificity. Fatigue, difficulty in swallowing, red eyes, gingival bleeding, hematemesis, confusion, hemoptysis, and a history of contact with an EVD case were predictors of EVD. The AUROC for ECPS was 0.88 (95%CI: 0.86–0.89), significantly greater than this for CPS, 0.71 (95%CI: 0.69–0.73) (p < 0.0001). At -1 point of score, the CPS yielded a sensitivity of 85.4% and specificity of 42.3%, and the ECPS yielded sensitivity of 78.8% and specificity of 81.4%. The diagnostic performance of the scores varied in the three disease contexts (the whole, fitting or not fitting the WHO case definition data sets). At 10% of threshold probability, e.g. in disease-adverse context, ECPS gave an NB of 0.033 and a net reduction of unnecessary isolation of 67.1%. Using ECPS as a joint approach to isolate EVD suspects reduces the number of unnecessary isolations by 65.7%. Conclusion: The scores developed in our study showed a good performance as EVD case predictors since their use improved the net benefit, i.e., their clinical utility. These rapid and low-cost tools can help in decision-making to isolate EVD-suspicious cases at the triage point during an outbreak. However, these tools still require external validation and cost-effectiveness evaluation before being used on a large scale. [ABSTRACT FROM AUTHOR]

Introduction: Despite HIV viral load (VL) monitoring being serial, most studies use a cross-sectional design to evaluate the virological status of a cohort. The objective of our study was to use a simplified approach to calculate viraemic-time: the proportion of follow-up time with unsuppressed VL above the limit of detection. We estimated risk factors for higher viraemic-time and whether viraemic-time predicted mortality in a second-line antiretroviral treatment (ART) cohort in Myanmar. Methods: We conducted a retrospective cohort analysis of people living with HIV (PLHIV) who received second-line ART for a period >6 months and who had at least two HIV VL test results between 01 January 2014 and 30 April 2018. Fractional logistic regression assessed risk factors for having higher viraemic-time and Cox proportional hazards regression assessed the association between viraemic-time and mortality. Kaplan-Meier curves were plotted to illustrate survival probability for different viraemic-time categories. Results: Among 1,352 participants, 815 (60.3%) never experienced viraemia, and 172 (12.7%), 214 (15.8%), and 80 (5.9%) participants were viraemic <20%, 20–49%, and 50–79% of their total follow-up time, respectively. Few (71; 5.3%) participants were ≥80% of their total follow-up time viraemic. The odds for having higher viraemic-time were higher among people with a history of injecting drug use (aOR 2.01, 95% CI 1.30–3.10, p = 0.002), sex workers (aOR 2.10, 95% CI 1.11–4.00, p = 0.02) and patients treated with lopinavir/ritonavir (vs. atazanavir; aOR 1.53, 95% CI 1.12–2.10, p = 0.008). Viraemic-time was strongly associated with mortality hazard among those with 50–79% and ≥80% viraemic-time (aHR 2.92, 95% CI 1.21–7.10, p = 0.02 and aHR 2.71, 95% CI 1.22–6.01, p = 0.01). This association was not observed in those with viraemic-time <50%. Conclusions: Key populations were at risk for having a higher viraemic-time on second-line ART. Viraemic-time predicts clinical outcomes. Differentiated services should target subgroups at risk for a higher viraemic-time to control both HIV transmission and mortality. [ABSTRACT FROM AUTHOR]

Introduction: The number of individuals initiating antiretroviral pre‐exposure prophylaxis (PrEP) is increasing, but we do not fully understand who is coming forward for PrEP, how they use it and how they are followed‐up. The objective of this study was to examine PrEP user profiles, dynamics in PrEP use and follow‐up over time. Methods: We conducted a cohort analysis of longitudinally collected clinical record and questionnaire data among PrEP users at an HIV centre in Antwerp, Belgium, between June 2017 and March 2020. PrEP follow‐up and user profiles were examined using descriptive analyses and bivariate logistic regression. We compared early adopting PrEP users (started before June 2018) with late users. We also calculated the probabilities of switching between daily and on‐demand PrEP, and interruption, using a naïve estimator. Results and discussion: We included 1347 PrEP users in the analysis. After 12 months, retention in care was 72.3%. Median time between PrEP visits was 98 days (IQR 85–119 days). At screening visit, early adopting PrEP users (starting June 2017–May 2018) were significantly more likely to report one or more sexually transmitted infection in the prior 12 months, having used drugs during sex, a higher number of sexual partners and a history of paid sex and PrEP use prior to initiation, compared with PrEP users who initiated later (starting June 2018–February 2020). When taking PrEP daily, the probability of staying on daily PrEP at the next visit was 76%, while this was 73% when taking PrEP on‐demand. Those using on‐demand PrEP had a higher probability (13%) of interrupting PrEP care than daily PrEP users (7%), whereas those returning to PrEP care would mostly re‐start with on‐demand (35% vs. 13% for daily). Conclusions: The majority of PrEP users in this sample remained in care after 12 months. The probability of remaining on the same PrEP regimen at the subsequent visit was high. Though, we observed a diversity of transitions between regimens and interruptions in between visits. Our findings reaffirm the need to provide tailored PrEP services, counselling PrEP users across their life course. [ABSTRACT FROM AUTHOR]

In most low-resource settings, microscopy still is the standard method for diagnosis of cutaneous leishmaniasis, despite its limited sensitivity. In Ethiopia, the more sensitive molecular methods are not yet routinely used. This study compared five PCR methods with microscopy on two sample types collected from patients with a suspected lesion to advise on optimal diagnosis of Leishmania aethiopica. Between May and July 2018, skin scrapings (SS) and blood exudate from the lesion spotted on filter paper (dry blood spot, DBS) were collected for PCR from 111 patients of four zones in Southern Ethiopia. DNA and RNA were simultaneously extracted from both sample types. DNA was evaluated by a conventional PCR targeting ITS-1 and three probe-based real-time PCRs: one targeting the SSU 18S rRNA and two targeting the kDNA minicircle sequence (the 'Mary kDNA PCR' and a newly designed 'LC kDNA PCR' for improved L. aethiopica detection). RNAs were tested with a SYBR Green-based RT-PCR targeting spliced leader (SL) RNA. Giemsa-stained SS smears were examined by microscopy. Of the 111 SS, 100 were positive with at least two methods. Sensitivity of microscopy, ITS PCR, SSU PCR, Mary kDNA PCR, LC kDNA PCR and SL RNA PCR were respectively 52%, 22%, 64%, 99%, 100% and 94%. Microscopy-based parasite load correlated well with real-time PCR Ct-values. Despite suboptimal sample storage for RNA detection, the SL RNA PCR resulted in congruent results with low Ct-values. DBS collected from the same lesion showed lower PCR positivity rates compared to SS. The kDNA PCRs showed excellent performance for diagnosis of L. aethiopica on SS. Lower-cost SL RNA detection can be a complementary high-throughput tool. DBS can be used for PCR in case microscopy is negative, the SS sample can be sent to the referral health facility where kDNA PCR method is available. Author summary: Cutaneous leishmaniasis is a neglected tropical disease and causing a public health problem in Ethiopia. Microscopy is still the standard method for detection of the parasite in Ethiopia, and also in many other low resource settings. A more sensitive method is needed for timely diagnosis and treatment. In this study, we compared five molecular methods on samples collected from patients with a skin lesion suspected of cutaneous leishmaniasis to advice on optimal diagnosis of L. aethiopica. We collected two sample types from the same lesion (skin scrapings and lesion fluid on filter paper) and isolated both DNA and RNA of them. Majority (90.1%) of the samples from skin scrapings were positive in two or more methods and the molecular methods had a higher sensitivity than the conventional methods. Interestingly, we evaluated for the first time a new molecular method designed to improve L. aethiopica detection. Also, we showed that RNA detection performed well for samples that were collected under difficult field conditions. Samples collected on filter paper showed less positive results than skin scraped samples, but could still be the method of choice for easy sampling and transport in resource-limited settings as it performed better than microscopy. [ABSTRACT FROM AUTHOR]

Standard data analysis pipelines for digital PCR estimate the concentration of a target nucleic acid by digitizing the end-point fluorescence of the parallel micro-PCR reactions, using an automated hard threshold. While it is known that misclassification has a major impact on the concentration estimate and substantially reduces accuracy, the uncertainty of this classification is typically ignored. We introduce a model-based clustering method to estimate the probability that the target is present (absent) in a partition conditional on its observed fluorescence and the distributional shape in no-template control samples. This methodology acknowledges the inherent uncertainty of the classification and provides a natural measure of precision, both at individual partition level and at the level of the global concentration. We illustrate our method on genetically modified organism, inhibition, dynamic range, and mutation detection experiments. We show that our method provides concentration estimates of similar accuracy or better than the current standard, along with a more realistic measure of precision. The individual partition probabilities and diagnostic density plots further allow for some quality control. An R implementation of our method, called Umbrella, is available, providing a more objective and automated data analysis procedure for absolute dPCR quantification. [ABSTRACT FROM AUTHOR]

Background: Digital polymerase chain reaction (dPCR) is an increasingly popular technology for detecting and quantifying target nucleic acids. Its advertised strength is high precision absolute quantification without needing reference curves. The standard data analytic approach follows a seemingly straightforward theoretical framework but ignores sources of variation in the data generating process. These stem from both technical and biological factors, where we distinguish features that are 1) hard-wired in the equipment, 2) user-dependent and 3) provided by manufacturers but may be adapted by the user. The impact of the corresponding variance components on the accuracy and precision of target concentration estimators presented in the literature is studied through simulation. Results: We reveal how system-specific technical factors influence accuracy as well as precision of concentration estimates. We find that a well-chosen sample dilution level and modifiable settings such as the fluorescence cut-off for target copy detection have a substantial impact on reliability and can be adapted to the sample analysed in ways that matter. User-dependent technical variation, including pipette inaccuracy and specific sources of sample heterogeneity, leads to a steep increase in uncertainty of estimated concentrations. Users can discover this through replicate experiments and derived variance estimation. Finally, the detection performance can be improved by optimizing the fluorescence intensity cut point as suboptimal thresholds reduce the accuracy of concentration estimates considerably. Conclusions: Like any other technology, dPCR is subject to variation induced by natural perturbations, systematic settings as well as user-dependent protocols. Corresponding uncertainty may be controlled with an adapted experimental design. Our findings point to modifiable key sources of uncertainty that form an important starting point for the development of guidelines on dPCR design and data analysis with correct precision bounds. Besides clever choices of sample dilution levels, experiment-specific tuning of machine settings can greatly improve results. Well-chosen data-driven fluorescence intensity thresholds in particular result in major improvements in target presence detection. We call on manufacturers to provide sufficiently detailed output data that allows users to maximize the potential of the method in their setting and obtain high precision and accuracy for their experiments. [ABSTRACT FROM AUTHOR]