CC BY-NC-ND
0Accepted Manuscript
Title: Urine Colorimetry for Therapeutic Drug Monitoring of Pyrazinamide during Tuberculosis Treatment
Authors: Isaac Zentner, Chawangwa Modongo, Nicola M. Zetola, Jotam G. Pasipanodya, Shashikant Srivastava, Scott K. Heysell, Stellah Mpagama, Hans P. Schlect, Tawanda Gumbo, Gregory P. Bisson, Christopher Vinnard
PII: DOI:
Reference:
S1201-9712(17)30329-6 https://doi.org/10.1016/j.ijid.2017.12.017 IJID 3125
To appear in:
International Journal of Infectious Diseases
Received date: Revised date: Accepted date:
6-9-2017
6-12-2017
9-12-2017
Please cite this article as: Zentner Isaac, Modongo Chawangwa, Zetola Nicola M, Pasipanodya Jotam G, Srivastava Shashikant, Heysell Scott K, Mpagama Stellah, Schlect Hans P, Gumbo Tawanda, Bisson Gregory P, Vinnard Christopher.Urine Colorimetry for Therapeutic Drug Monitoring of Pyrazinamide during Tuberculosis Treatment.International Journal of Infectious Diseases https://doi.org/10.1016/j.ijid.2017.12.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Urine Colorimetry for Therapeutic Drug Monitoring of Pyrazinamide during Tuberculosis Treatment
Authors: Isaac Zentner1, Chawangwa Modongo2, Nicola M. Zetola2, Jotam G. Pasipanodya3, Shashikant Srivastava3, Scott K. Heysell4, Stellah Mpagama5, Hans P. Schlect6, Tawanda Gumbo3, Gregory P. Bisson7, Christopher Vinnard1*
1Public Health Research Institute, New Jersey Medical School, Newark, New Jersey, United States of America 2Botswana-Upenn Partnership, Gaborone, Botswana
3Center for Infectious Diseases Research and Experimental Therapeutics, Baylor Research Institute, Baylor University Medical Center, Dallas, Texas, United States of America
4Division of Infectious Diseases and International Health, Department of Medicine, University of Virginia,
Charlottesville, VA, United States of America
5Kibong'oto National Tuberculosis Hospital, Sanya Juu, Tanzania
6Drexel University College of Medicine, Philadelphia, Pennsylvania, United States of America
7University of Pennsylvania, Perelman School of Medicine, Philadelphia, Pennsylvania, United States of America
Word count: 2,818 words
Corresponding author:
Dr. Christopher Vinnard, MD MPH MSCE
Public Health Research Institute
Rutgers, The State University of New Jersey
225 Warren Street
Newark, NJ 07103
Phone: (973) 854-3100
Fax: (973) 854-3101
christopher.vinnard@njms.rutgers.edu
Highlights
• Our objective was to determine whether a colorimetric test of urine can identify tuberculosis patients with adequate pyrazinamide exposures, as defined by the serum Cmax above a target threshold.
• The urine colorimetric assay was 97% sensitive and 50% specific to identify HIV/tuberculosis patients with pharmacokinetic target attainment, with an area under the ROC curve of 0.81 (95% confidence interval 0.60, 0.97).
• Future work will focus on refinement of the approach, with the goal of developing a simple, point-of-care, test that could be available for therapeutic drug monitoring during anti-tuberculosis therapy in high-burden settings.
ABSTRACT Objectives
Pyrazinamide is a key drug in the first-line tuberculosis treatment regimen, with a potent sterilizing effect. Although low pyrazinamide peak serum concentrations (Cmax) are associated with poor treatment outcomes, many resource-constrained settings do not have sufficient laboratory capacity to support therapeutic drug monitoring. Our objective was to determine whether a colorimetric test of urine can identify tuberculosis patients with adequate pyrazinamide exposures, as defined by the serum Cmax above a target threshold.
Methods
In the derivation study of healthy volunteers, we evaluated 3 dose sizes of pyrazinamide and performed intensive pharmacokinetic blood sampling over an 8-hour interval, with a timed urine void at 4 hours post-dosing. We performed an isolation of pyrazinamide in urine by spin column centrifugation with an exchange resin followed by colorimetric analysis, and measured the absorbance peak at 495 nm. The urine assay was then evaluated in a study of 39 HIV/tuberculosis patients in Botswana enrolled in an intensive pharmacokinetic study. Receiver-operating-characteristic (ROC) curves were used to measure diagnostic accuracy. We evaluated the guideline-recommended pyrazinamide serum Cmax target of 35 mg/L in the primary analysis, and in a secondary analysis evaluated the higher serum Cmax target of 58 mg/L, which was predictive of favorable outcomes in a clinical study.
Results
At the optimal cut-off identified in the derivation sample, the urine colorimetric assay was 97% sensitive and 50% specific to identify 35 of 39 HIV/tuberculosis patients with pharmacokinetic target attainment, with an area under the ROC curve of 0.81 (95% confidence interval 0.60, 0.97). Diagnostic accuracy was lower at the 58 mg/L serum Cmax target, with an area under the ROC curve of 0.68 (95%
confidence interval 0.48 to 0.84). Men were less likely than women to attain either serum pharmacokinetic target.
Conclusions
The urine colorimetric assay was sensitive but not specific for the detection of adequate pyrazinamide pharmacokinetic exposures among HIV/tuberculosis patients in a high-burden setting.
Keywords: Tuberculosis; Pyrazinamide; Pharmacokinetics; Human immunodeficiency virus; Point-of-care testing
BACKGROUND
Standardized drug regimens using weight-based dosing algorithms are the cornerstone of the global public health response to the tuberculosis epidemic1. There is wide variability in the absorption and metabolism of the anti-tuberculosis drugs, and low drug concentrations in blood are associated with inferior tuberculosis treatment outcomes, including treatment failure and relapse2. Pyrazinamide is the key sterilizing-effect drug in the first-line tuberculosis treatment regimen3,4. Prospective clinical studies have demonstrated that pyrazinamide efficacy is concentration driven, with the peak plasma or serum concentration (Cmax) predictive of a successful clinical outcome5,6.
Tuberculosis patients co-infected with HIV are more likely to experience inferior treatment outcomes, and these outcomes include delayed sterilization of sputum, relapse following treatment completion, and death7-10. According to the most recent CDC/ATS/IDSA guidelines for tuberculosis treatment, therapeutic drug monitoring should be routinely considered among tuberculosis patients in high-risk groups, which includes individuals co-infected with HIV11. However, therapeutic drug monitoring
requires sophisticated laboratory techniques unavailable for millions of patients receiving tuberculosis therapy in resource-limited settings where the vast majority of tuberculosis cases are treated12.
Urine colorimetry was first evaluated as a method to assess the bioequivalence of different fixed-dose combinations of anti-tuberculosis drugs. In these descriptions, the approach was based on timed urine sampling after the oral administration of drug formulations to healthy volunteers, with the performance colorimetric assays (measuring sample absorbance at specific wavelengths) to quantify drug excretion in urine13,14. More recently, a colorimetric approach to detect isoniazid in urine has been commercialized as a tool to monitor isoniazid adherence15. As a tool to perform therapeutic drug monitoring, urine colorimetry could be performed on urine collected at one or more timed intervals following dose administration. Potential advantages of urine colorimetric methods include a noninvasive sampling approach, improved patient acceptability, and the low cost and stability of chemical reagents.
We have recently demonstrated proof-of-concept for urine colorimetry to identify patients with low serum rifampin exposures during tuberculosis treatment16. Our objective with the current study was to determine whether urine colorimetry accurately predicts adequate pyrazinamide serum pharmacokinetic exposures during anti-tuberculosis therapy.
METHODS
Derivation study design
We first performed a non-randomized, open-label, cross-over study of the first-line anti-tuberculosis drugs (rifampin, isoniazid, ethambutol, pyrazinamide) in 6 healthy volunteers. We sequentially evaluated 3 anti-tuberculosis drug dose sizes in separate study visits, with pyrazinamide dosed at 500 mg, 1000 mg, and 2000 mg. Each study visit was separated by a wash-out period of at least 1 week thereby supporting a single-dose design at each study visit. Subjects presented for the study visit after
an overnight fast. Blood samples were collected prior to oral administration of the study drugs, and then at 1, 2, 4, 6, and 8 hours following oral administration of the study drugs. Breakfast was provided 30 minutes following dosing, and lunch was provided 4 hours following dosing. Frozen serum samples were shipped to the Infectious Disease Pharmacokinetics Laboratory at the University of Florida for measurement of drug concentrations by liquid chromatography-mass spectrometry. The serum Cmax obtained from the pyrazinamide concentration-versus-time profile was defined to be the "gold standard" pharmacokinetic assessment in the statistical analysis plan. All urine was collected during the study visit, and the volume and time of collection were noted. Timed voids were obtained at 4 hours and 8 hours post-dosing. Urine was aliquoted into single-use 3 mL conical vials and stored at -80C until ready for analysis.
Validation study design
The validation study was nested within a prospective cohort study of anti-tuberculosis drug pharmacokinetics in HIV/tuberculosis patients at 22 public clinics and Princess Marina Hospital in Gaborone, Botswana17. In brief, HIV-infected adults (21 years of age and older) were eligible for enrollment in the parent study if they were citizens of Botswana, naïve to antiretroviral therapy, and newly diagnosed with pulmonary tuberculosis. Patients must have been initiated on a standard first-line tuberculosis treatment regimen, following WHO guidelines for weight-based dosing bands. The diagnosis of pulmonary tuberculosis was established by either a positive sputum smear, a positive GeneXpert MTB/RIF assay (Cepheid, Sunnyvale, CA, USA), or the presence of WHO criteria for smear-negative pulmonary tuberculosis. Exclusion criteria included pregnancy, renal insufficiency (defined as a creatinine clearance less than 50 mL/min), and hepatic dysfunction (defined as either an alanine aminotransferase or aspartate aminotransferase greater than 3 times the upper limit of normal).
The pharmacokinetic study visit was scheduled prior to the completion of the intensive phase of antituberculosis therapy. After an overnight fast, oral doses of the anti-tuberculosis drugs were directly
administered to the participant on the morning of the study visit. A baseline blood sample was drawn prior to dosing, and then at 0.3, 0.9, 2.2, 4.5, and 8 hours post-dosing. These timepoints were selected based on optimal sampling theory of isoniazid pharmacokinetics, which was the focus of the parent study in Botswana17. At each time point, 10 mL of blood was drawn and transported to the Botswana Harvard Partnership Laboratory. Breakfast was provided 30 minutes following dosing, and lunch was provided 4 hours following dosing. After centrifugation, serum was stored at -80C, and drug concentrations were measured using liquid chromatography-tandem mass spectrometry methods in the Gumbo Laboratory at the Baylor Research Institute (Dallas, TX). For the performance of the urine colorimetric assay, a single urine sample was obtained 4 hours after dosing, based on the diagnostic accuracy of this time point observed in the derivation study. Urine samples were frozen and shipped to the Infectious Disease Clinical Research Laboratory at Drexel University College of Medicine (Philadelphia, PA).
Urine colorimetric assay
The steps of the urine colorimetric assay were identical for the derivation and validation samples, and followed the method published by Gurumurthy et al, with pyrazinamide isolation by spin column centrifugation with an exchange resin18. In brief, 0.13 g of Dowex 1- x8 (Cl-form, 200-400 mesh) was added to a 1 mL spin column, and washed with 500 ^L of water via centrifugation for 1 minute at 8,000 rpm. For each sample, 1.5 mL was added to the column in 500 ^L increments spinning the sample for 1 minute at 8,000 rpm. Elution was carried out by incubating resin with 500 ^L of water for 5 minutes and centrifugation for 1 minute at 8,000 rpm in a fresh collection tube. From each elution sample, 100 ^L was transferred to a 96-well plate and 83 ^L of a 0.2% sodium nitroprusside in water was added. Following the 0.2% sodium nitroprusside, the samples were treated with an equal volume (83 ^L) of 2M sodium hydroxide. The 96-well plate was left to incubate for 5 minutes at room temperature and the optical densities were measured at 495 nm in a Multiskan GO Microplate Spectrophotometer (Thermo Fisher Scientific).
Statistical analysis
The goal of the statistical analysis plan for the derivation study was to define the accuracy of the urine colorimetric assay to identify volunteers with pyrazinamide serum Cmax target attainment across a range of possible cutoff values. The receiver-operating-characteristic curve is a plot of the assay sensitivity versus 1-specificity across the range of possible test cut-off values. Sensitivity and specificity were defined by the following equations:
True positive
Sensitivity =---
True positive + False negative
True negative
Specificity = ---—---——
True negative + False positive
The area under the receiver-operating-characteristic (ROC) curve provides a summary measure of the diagnostic test accuracy. An area under the ROC curve equal to 1 demonstrates perfect discrimination, whereas an area of 0.5 demonstrates that the diagnostic test performs no better than chance alone19. In this approach, a higher urine colorimetric assay result indicates a greater probability of achieving a serum Cmax target greater than a certain threshold. As the goal of the urine assay is to identify tuberculosis patients that would benefit from increased pyrazinamide dosing, we reasoned that a highly sensitive test would be required to minimize the number of false negative results, which could lead to inappropriate dose increases and potential toxicities.
Bootstrapping with 2000 replicates was performed to estimate the 95% confidence intervals for the area under the ROC curve20. In this approach, we sampled n=40 individuals from the study cohort (with replacement), repeated for 2000 replicate datasets, and determined the ROC curve for each replicate dataset. The distribution of the area under the ROC curve for these replicates provided an estimate of the 95% confidence interval (i.e. corresponding to the 2.5th and 97.5th percentiles in the distribution).
For the primary analysis, we evaluated the target pyrazinamide serum Cmax of 35 mg/L, which was predictive of treatment success among adult tuberculosis patients in Botswana, most with HIV co-infection5. In secondary analysis, we evaluated the serum Cmax target of 58 mg/L, identified in a cohort of South African tuberculosis patients to be the strongest predictor of sputum conversion after 2 months of treatment21. Given a priori knowledge that sex is a predictor of pyrazinamide pharmacokinetics, we evaluated the relationship of patient sex with serum Cmax target attainment. Predictive models of pharmacokinetic target attainment were constructed with multivariate logistic regression analysis. The predictive accuracy was evaluated with Brier scores, and goodness-of-fit was assessed by the Hosmer-Lemeshow test. Statistical analysis was performed in Stata 13 (StataCorp, College Station, TX), and plots were generated using GraphPad Prism v 7.00 (GraphPad Software, La Jolla, CA).
RESULTS
Development of the pyrazinamide colorimetric assay
The correlation of the colorimetric assay with the working standards of pyrazinamide concentrations is shown in Figure 1. We observed assay linearity across the range of pyrazinamide concentrations from 7.9 to 1,000 mg/L (R2 0.99).
Derivation study
Each of the 6 healthy volunteers completed 3 study visits, for a total of 18 pharmacokinetic visits with simultaneous sampling of blood and urine compartments. The pyrazinamide serum Cmax was above the target of 35 mg/L in 5 of 18 instances (Figure 2a). A 4-hour urine concentration threshold of 48.7 mg/L, as measured by the colorimetric assay, was 100% sensitive and 85% specific to identify healthy volunteers with a pyrazinamide serum Cmax exceeding the 35 mg/L target (Figure 2b), with an area under the ROC curve of 0.94 (95% confidence interval 0.78 to 1.0).
Validation study
Thirty-nine HIV/tuberculosis patients completed the intensive pharmacokinetic study and provided a 4hour urine sample for analysis, and all of these patients were included in the validation sample. Baseline demographic and clinical characteristics for these patients have been previously described17. The pyrazinamide serum Cmax exceeded the 35 mg/L target in 35 of 39 instances (Figure 3a). At the cut-off identified in the derivation study, a 4-hour urine pyrazinamide concentration of 48.7 mg/L was 97% sensitive but only 50% specific to identify 35 of 39 HIV/tuberculosis patients with pyrazinamide Cmax greater than 35 mg/L (Figure 3b), with an area under the ROC curve of 0.81 (95% confidence interval 0.60 to 0.97). Secondary analysis
Urine colorimetric assay for the Cmax target of 58 mg/L
We evaluated a pyrazinamide serum Cmax target of 58 mg/L, identified to be the most significant pharmacokinetic predictor of favorable clinical outcome among South African tuberculosis patients treated with a combination drug regimen that included pyrazinamide21. Because the pyrazinamide serum Cmax was greater than 58 mg/L in only 1 of 18 instances in the derivation study, we used the HIV/tuberculosis patient cohort to both derive and validate the diagnostic accuracy for this higher target, and performed bootstrapping (2000 replicates) for internal validation. Ten of 39 (26%) HIV/tuberculosis patients attained a pyrazinamide serum Cmax greater than 58 mg/L. The distribution of the urine colorimetric assay according to target attainment among HIV/tuberculosis patients is shown in Figure 4a, and the corresponding ROC curve is shown in Figure 4b. The area under the ROC curve was 0.68 (95% confidence interval 0.48 to 0.84), which indicated that the predictive accuracy of the assay was not significantly different from chance alone.
Sex differences in pyrazinamide pharmacokinetic target attainment
Pharmacokinetic target attainment differed by patient sex. The median pyrazinamide serum Cmax among men was 42.6 mg/L (interquartile range 38.8 to 51.0 mg/L), and among women was 58.1 mg/L (interquartile range 45.4 to 66.5 mg/L, p=0.002 by Kruskal-Wallis test). Eighteen of 18 women (100%)
achieved a pyrazinamide serum Cmax greater than 35 mg/L, while 17 of 21 men (81%) reached that target (p=0.051 by chi-squared test). Similarly, the Cmax target of 58 mg/L target attainment was achieved in 9 of 18 women (50%) but only 1 of 21 men (5%, p<0.01 by chi-squared test). The highly skewed distribution by sex precluded the formal evaluation of separate urine colorimetry cut-offs for men and women for each pharmacokinetic target in an adjusted ROC analysis. The association between patient sex and attainment of the 58 mg/L serum Cmax target remained significant after adjusting for weight, serum creatinine, and creatinine clearance (as estimated by Cockroft-Gault equation), suggesting that sex differences in pyrazinamide pharmacokinetics was not a reflection of confounding by these characteristics.
DISCUSSION
In this paired derivation and validation study, we evaluated urine colorimetry as a tool to assess pyrazinamide pharmacokinetic target attainment during tuberculosis treatment. We observed that the colorimetric assay was sensitive but not specific to identify HIV/tuberculosis patients with pyrazinamide peak serum concentrations above the target of 35 mg/L. In a secondary analysis of the higher pyrazinamide serum Cmax target of 58 mg/L, we observed a decrease in diagnostic accuracy of the urine assay, with a 95% confidence interval that included 0.5, indicating that the test was not better than chance alone. Men were less likely than women to attain the serum Cmax 58 mg/L target, independent of total body weight or renal function.
As outlined in the End TB goals of the World Health Organization, the task of the tuberculosis research community in the decade ahead is to optimize existing tools for the treatment and prevention of tuberculosis. Therapeutic drug monitoring is a tool to optimize anti-tuberculosis drug dosing based on pharmacokinetic target attainment, which has been linked to successful tuberculosis treatment outcomes2,5,6,21. However, the laboratory requirements to perform the standard drug concentration assays, such as high performance liquid chromatography or liquid chromatography-tandem mass
spectrometry, present a formidable challenge to the performance of therapeutic drug monitoring in many high-burden settings12. With a simple test performed at the point-of-care on a timed urine sample, delivering actionable data at the time of the patient encounter, the tuberculosis clinician could adjust drug doses to optimize the patient's treatment regimen. If the clinician were to act on this information by increasing the dose size for patients with a low urine colorimetric test result, low specificity would relate to "missed opportunities" for dose increases, while the high sensitivity would minimize the number of inappropriate dose increases, which could lead to exposure-related toxicities. Potentially, urine colorimetry could also be used to re-evaluate patients after drug dose adjustment has been made, to re-assess target attainment at the new dose level.
Interestingly, we observed that men were less likely than women to attain pyrazinamide pharmacokinetic targets, evaluating two targets that were related to treatment outcomes in clinical cohorts of tuberculosis patients. The pharmacokinetics of several anti-tuberculosis drugs are influenced by sex, including isoniazid, rifampin, and pyrazinamide22,23. Given that body weight is already incorporated into anti-tuberculosis drug dosing regimens, in the form of weight-based dosing bands, the observed relationship between patient sex and anti-tuberculosis drug pharmacokinetics suggests a mechanism other than differences in body weight, although there may have been differences in lean body mass between men and women with the same total body weight. Furthermore, we observed that the sex differences were independent of differences in serum creatinine or creatinine clearance. More work is needed to understand the mechanism of sex differences in anti-tuberculosis drug pharmacokinetics, and the potential use of this knowledge in dosing decisions.
This proof-of-concept study had several important limitations. The derivation and validation samples were selected from different populations, and differences between these populations likely contributed to differences in the observed assay performance. Although all participants in both the derivation and validation cohorts received combination drug regimens that included co-administered tuberculosis
drugs (rifampin, isoniazid, and ethambutol), the HIV/tuberculosis patients may have been treated with additional therapies as part of their HIV care (such as prophylaxis against opportunistic infections), and the potential interfering effects of these additional drugs were not evaluated. Additionally, in the derivation study we evaluated the serum and urine pharmacokinetics of pyrazinamide in a single-dose design (with a wash-out period between dose sizes), rather than under steady-state conditions, which may have lowered the proportion of subjects attaining pharmacokinetic targets (with only 1 of 18 visits demonstrating a pyrazinamide Cmax exceeding 58 mg/L). Strengths of our approach included a priori decisions regarding the target serum Cmax thresholds, the enrollment of patients in the validation cohort from a relevant patient population (HIV/tuberculosis patients in sub-Saharan Africa), and the use of intensive pharmacokinetic sampling to define the gold standard of serum target attainment.
In summary, a urine colorimetric assay was sensitive but not specific in identifying HIV/tuberculosis patients with pyrazinamide serum pharmacokinetic target attainment. Future work will focus on refinement of the approach, with the goal of developing a simple, point-of-care, test of timed urine samples that could be available for therapeutic drug monitoring during anti-tuberculosis therapy in high-burden settings.
Conflict of Interest Statement: All authors report no conflicts of interest.
Funding Source: This work was supported by a Grand Challenges and Explorations grant from the Bill and Melinda Gates Foundation (to Dr. Vinnard) and NIAID (R21AI104441 to Dr. Bisson, K23AI102639 to Dr. Vinnard). Dr. Vinnard receives internal support from the Public Health Research Institute, Rutgers, The State University of New Jersey. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Ethical Approval: The derivation study was approved by the institutional review board at Drexel University College of Medicine. The institutional review boards of the University of Pennsylvania, Botswana Ministry of Health, and the Princess Marina Hospital approved the validation study. All participants provided written informed consent.
REFERENCES
1. Organization WH. Treatment of tuberculosis guidelines. 2010.
2. Pasipanodya JG, Srivastava S, Gumbo T. Meta-analysis of clinical studies supports the pharmacokinetic variability hypothesis for acquired drug resistance and failure of antituberculosis therapy. Clin Infect Dis 2012;55:169-77.
3. Chigutsa E, Pasipanodya JG, Visser ME, et al. Impact of nonlinear interactions of pharmacokinetics and MICs on sputum bacillary kill rates as a marker of sterilizing effect in tuberculosis. Antimicrob Agents Chemother 2015;59:38-45.
4. Gumbo T, Dona CS, Meek C, Leff R. Pharmacokinetics-pharmacodynamics of pyrazinamide in a novel in vitro model of tuberculosis for sterilizing effect: a paradigm for faster assessment of new antituberculosis drugs. Antimicrob Agents Chemother 2009;53:3197-204.
5. Chideya S, Winston CA, Peloquin CA, et al. Isoniazid, rifampin, ethambutol, and pyrazinamide pharmacokinetics and treatment outcomes among a predominantly HIV-infected cohort of adults with tuberculosis from Botswana. Clin Infect Dis 2009;48:1685-94.
6. Swaminathan S, Pasipanodya JG, Ramachandran G, et al. Drug Concentration Thresholds Predictive of Therapy Failure and Death in Children With Tuberculosis: Bread Crumb Trails in Random Forests. Clin Infect Dis 2016;63:S63-S74.
7. Khan FA, Minion J, Pai M, et al. Treatment of active tuberculosis in HIV-coinfected patients: a systematic review and meta-analysis. Clin Infect Dis 2010;50:1288-99.
8. El-Sadr WM, Perlman DC, Denning E, Matts JP, Cohn DL. A review of efficacy studies of 6-month short-course therapy for tuberculosis among patients infected with human immunodeficiency virus: differences in study outcomes. Clin Infect Dis 2001;32:623-32.
9. Korenromp EL, Scano F, Williams BG, Dye C, Nunn P. Effects of human immunodeficiency virus infection on recurrence of tuberculosis after rifampin-based treatment: an analytical review. Clin Infect Dis 2003;37:101-12.
10. Mallory KF, Churchyard GJ, Kleinschmidt I, De Cock KM, Corbett EL. The impact of HIV infection on recurrence of tuberculosis in South African gold miners. Int J Tuberc Lung Dis 2000;4:455-62.
11. Nahid P, Dorman SE, Alipanah N, et al. Official American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America Clinical Practice Guidelines: Treatment of Drug-Susceptible Tuberculosis. Clin Infect Dis 2016;63:e147-e95.
12. Nwobodo N. Therapeutic drug monitoring in a developing nation: a clinical guide. JRSM Open 2014;5:2054270414531121.
13. Ellard GA. The colorimetric analysis of anti-tuberculosis fixed-dose combination tablets and capsules. Int J Tuberc Lung Dis 1999;3:S343-6; discussion S51-2.
14. Pillai G, Ellard GA, Smith PJ, Fourie PB. The potential use of urinary excretion data for assessing the relative bioavailability of rifampicin in fixed dose combination anti-tuberculosis formulations. Int J Tuberc Lung Dis 2001;5:691-5.
15. Nicolau I, Tian L, Menzies D, Ostiguy G, Pai M. Point-of-care urine tests for smoking status and isoniazid treatment monitoring in adult patients. PLoS One 2012;7:e45913.
16. Zentner I, Schlecht HP, Khensouvann L, et al. Urine colorimetry to detect Low rifampin exposure during tuberculosis therapy: a proof-of-concept study. BMC Infect Dis 2016;16:242.
17. Vinnard C, Ravimohan S, Tamuhla N, et al. Isoniazid clearance is impaired among human immunodeficiency virus/tuberculosis patients with high levels of immune activation. Br J Clin Pharmacol 2017;83:801-11.
18. Gurumurthy P, Nair NG, Sarma GR. Methods for the estimation of pyrazinamide and pyrazinoic acid in body fluids. Indian J Med Res 1980;71:129-34.
19. MS P. The statistical analysis of medical tests for classification and prediction. New York, NY: Oxford University Press; 2003.
20. Carpenter J, Bithell J. Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med 2000;19:1141-64.
21. Pasipanodya JG, McIlleron H, Burger A, Wash PA, Smith P, Gumbo T. Serum drug concentrations predictive of pulmonary tuberculosis outcomes. J Infect Dis 2013;208:1464-73.
22. Hemanth Kumar AK, Kannan T, Chandrasekaran V, et al. Pharmacokinetics of thrice-weekly rifampicin, isoniazid and pyrazinamide in adult tuberculosis patients in India. Int J Tuberc Lung Dis 2016;20:1236-41.
23. McIlleron H, Rustomjee R, Vahedi M, et al. Reduced antituberculosis drug concentrations in HIV-infected patients who are men or have low weight: implications for international dosing guidelines. Antimicrob Agents Chemother 2012;56:3232-8.
FIGURE LEGENDS
Figure 1. Calibration assay for pyrazinamide colorimetric assay
Pyrazinamide concentration (mg/L)
Figure 2A. Individual pyrazinamide serum concentrations-versus-time among healthy volunteers administered varying doses. Legend: dot-dashed black line: 500 mg dose size; solid black line: 1000 mg dose size; solid grey line: 2000 mg dose size; dashed black line: 35 mg/L target threshold
Time (hours)
Figure 2B. Receiver-operating-characteristic curve of urine colorimetry to identify healthy volunteers with pyrazinamide serum Cmax greater than 35 mg/L
1 - Specificity (%)
Figure 3A. Individual pyrazinamide serum concentrations-versus-time among HIV/tuberculosis patients in Botswana. Legend: solid black line: pyrazinamide Cmax below 35 mg/L; solid grey line: pyrazinamide Cmax above 35 mg/L; dashed black line: 35 mg/L target threshold.
от E
Ф О С
о тз
§ 20 с
Time (hours)
Figure 3B. Receiver-operating-characteristic curve for pyrazinamide urine assay to identify HIV/tuberculosis patients with pyrazinamide serum Cmax greater than 35 mg/L
1 - Specificity (%)
Figure 4A. Box plot of pyrazinamide urine concentrations estimated by colorimetric assay, grouped by serum pharmacokinetic target attainment (serum Cmax greater than 58 mg/L)
0) o c o o
E rc c
FIGURE 4B. Receiver-operating-characteristic curve for pyrazinamide urine assay to identify HIV/tuberculosis patients with pyrazinamide serum Cmax greater than 58 mg/L
20 40 60
1 - Specificity (%)
