Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen.
Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)).
Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*).
Limited information is available on the pharmacokinetics of rifampicin (RIF) along with that of its active metabolite, 25-deacetylrifampicin (25-dRIF). This study aimed to analyse the pharmacokinetic data of RIF and 25-dRIF collected in adult patients treated for tuberculosis.
Methods
In adult patients receiving 10 mg/kg of RIF as part of a standard regimen for drug-susceptible pulmonary tuberculosis enrolled in the Opti-4TB study, plasma RIF and 25-dRIF concentrations were measured at various occasions. The RIF and 25-dRIF concentrations were modelled simultaneously by using a population approach. The area under the concentration–time curves of RIF and 25-dRIF were estimated on each occasion of therapeutic drug monitoring. Optimal RIF exposure, defined as an area under the concentration–time curve over 24 hours/minimum inhibitory concentration > 435, was assessed.
Results
Concentration data (247 and 243 concentrations of RIF and 25-dRIF, respectively) were obtained in 35 patients with tuberculosis (10 women, 25 men). Mycobacterium tuberculosis minimum inhibitory concentration ranged from 0.06 to 0.5 mg/L (median = 0.25 mg/L). The final model was a two-compartment model including RIF metabolism into 25-dRIF and auto-induction. Exposure to 25-dRIF was low, with a mean area under the concentration–time curve over 24 h ratio of 25-dRIF/RIF of 14 ± 6%. The area under the concentration–time curve over 24 h of RIF and 25-dRIF rapidly decreased during therapy, with an auto-induction half-life of 1.6 days. Optimal RIF exposure was achieved in only six (19.3%) out of 31 patients upon first therapeutic drug monitoring.
Conclusions
Exposure to both RIF and 25-dRIF rapidly decreased during tuberculosis therapy. The contribution of 25-dRIF to overall drug exposure was low. Attainment of the target area under the concentration–time curve over 24 hours/minimum inhibitory concentration for RIF was poor, supporting an increased RIF dosage as an option to compensate for auto-induction.
Lyon TB Study Group collaborators are listed in the Acknowledgements section.
Key Points
We report the first population pharmacokinetic analysis of rifampicin plus its active metabolite (25-deacetylrifampicin) in adult patients with tuberculosis.
Exposure to both rifampicin and 25-deacetylrifampicin rapidly decreases (about a 30–40% reduction in median exposure) over the first days of therapy (induction half-life of 1.6 days).
Most patients did not achieve the target area under the concentration–time curve/minimum inhibitory concentration for rifampicin.
1 Introduction
Rifampicin (RIF) stands as a cornerstone of the first-line regimen for drug-susceptible tuberculosis (TB) owing to its potent bactericidal activity against active and dormant Mycobacterium tuberculosis (Mtb) phenotypes [1]. The current standard dose of RIF (10 mg/kg) was historically selected based on pharmacokinetic (PK)/pharmacodynamic arguments, as well as toxicity concerns and cost considerations, all of which are now outdated [2]. Multiple studies have demonstrated that RIF exposure in patients often falls below the recommended efficacy-associated thresholds [3, 4]. This raises concerns for the management of TB, as subtherapeutic exposure to RIF has been associated with an increased risk of poor clinical outcomes and the emergence of drug resistance. [5, 6]
Anzeige
Therapeutic drug monitoring (TDM) is increasingly recognised as a valuable tool for the management of antimicrobial therapy. Clinical evidence supports RIF area under the concentration–time curve (AUC) and AUC over the minimum inhibitory concentration ratio (AUC/MIC) as reliable surrogate markers of therapeutic efficacy. Although reported in a few studies, the peak concentration appears less relevant [5, 7‐9].
The important role of TDM also extends to clinical trials by supporting the development and implementation of new treatment regimens into clinical practice. Notably, there is a growing interest in implementing high-dose RIF regimens that have been shown to achieve faster bacterial clearance without excessive toxicity [10, 11]. Such regimens are being evaluated in ongoing clinical trials [12, 13].
The pharmacokinetics of RIF is complex and considerable inter-individual and intra-individual variability has been reported [14]. Rifampicin is mainly metabolised into 25-desacetyl-rifampicin (25-dRIF), a metabolite described as active. Despite its potential clinical significance, only a few studies have evaluated the pharmacokinetics of both RIF and 25-dRIF concurrently, and limited information is available in patients with TB [15, 16]. The aim of this study was to evaluate the pharmacokinetics of RIF and 25-dRIF in patients treated for drug-susceptible TB.
2 Materials and Methods
2.1 Patient Enrollment
Patients were recruited in the Opti-4TB study whose main objective was to evaluate new immunological diagnosis and response biomarkers of active TB in adult patients. Written informed consent was obtained from all participants. This study was approved by our institutional ethics committee (no. 69HCL18_0757) and has been registered in the ClinicalTrials.gov website (NCT04271397). The study aims and design have been reported elsewhere [17]. This report focuses on the analysis of RIF and 25-dRIF exposure, a secondary objective of this study. Briefly, adult patients treated for drug-susceptible pulmonary TB at the Infectious Disease Department of Lyon University Hospitals, France were enrolled upon a TB diagnosis and received the four-drug standard regimen including 10 mg/kg of RIF. All patients were positive for either smear or Mtb complex nucleic acid detection in respiratory samples upon diagnosis, subsequently confirmed by a Mtb culture. Clinical Mtb isolates were stored for further MIC determination and genotypic characterisation.
Minimum inhibitory concentrations of RIF were determined for Mtb clinical isolates from each patient by a standard microdilution method as previously described [18]. Briefly, serial two-fold dilutions of RIF were placed in the wells of a 96-well microtitre plate, with a concentration range from 16 to 0.0015 mg/L. Each well was inoculated with a final inoculum of approximately 2.105 CFU/mL. The plates were incubated at 37 °C for 12–18 days, and the wells were assessed for visible turbidity. The lowest concentration at which there was no visible turbidity was defined as the MIC.
2.3 Pharmacokinetic Sampling and Drug Assay
Therapeutic drug monitoring with measurement of the anti-TB drug concentration was performed on several occasions during therapy. The planned schedule for blood sampling was as follows: days 1–2, day 15, month 2 and month 6, with one pre-dose (trough) and one or two post-dose samples on each occasion. However, this schedule was individually adjusted in patients depending on the hospital discharge and follow-up visits, so there was more variability in actual days of TDM (see results).
Concentrations of RIF and 25-dRIF were measured in plasma by liquid chromatography (Acquity™; Waters Corporation, Milford, MA, USA) coupled with a tandem mass spectrometer (XEVO TQ S micro; Waters Corporation). Chromatographic separation was realised using a BEH C18 column (1.7 µm, 2.1 × 50 mm; Waters Corporation) maintained at 40 °C. Mobile phases consisted of solvent (A), 10 mM of ammonium formate in ultra-pure water and solvent (B), 0.1% (v/v) formic acid in acetonitrile. The lower limit of quantification of RIF and 25-dRIF were determined at 0.05 mg/L. A sample preparation was realised using the OSTRO® precipitation and dephospholipidation plate (Waters Corporation) as follows: to 30 µL of plasma were added 100 µL of 0.1% formic acid in acetonitrile containing internal standards (RIF-d8 and 25-dRIF-d8). Quantification transitions (m/z) were set as follow: RIF 823.6 > 791.5, RIF-d8 831.6 > 799.6, 25-dRIF 749.5 > 95.1 and 25-dRIF-d8 757.5 > 95. Further details on the assay methods are provided in the Electronic Supplementary Material (ESM).
2.4 Population Pharmacokinetic Modelling
A population PK analysis was performed with the Monolix® software (version 2023R1; Lixoft, Antony, France) using the stochastic approximation expectation maximisation algorithm. Concentrations of RIF and 25-dRIF were analysed simultaneously, with a parent-metabolite model. Concentrations below the quantification limit were not discarded but coded as censored observations and considered in the estimation of PK parameters. Various structural, variability, residual error and covariate candidate models were evaluated. As RIF was administered by the oral route in all patients, we estimated apparent clearance and apparent volume of distribution. However, bioavailability could not be estimated. Assuming F = 1, parameters will be denoted without reference to F throughout the article, for ease of presentation.
The best model was selected based on classical criteria including goodness-of-fit criteria (objective function, predicted vs observed concentrations plots, residual plots), precision of parameter estimates and simulation-based diagnostics (visual predictive checks). A nonparametric bootstrap with 200 case resampling was carried out to derive confidence intervals of parameter estimates.
In covariate modelling, the following variables were examined: age, sex, total body weight, ideal body weight [19], lean body weight [20], body surface area [21], body mass index, plasma creatinine, glomerular filtration rate (CKD-EPI equation) [22], plasma albumin, plasma protein, C-reactive protein, aspartate aminotransferase, alanine transaminase, alkaline phosphatase, gamma-glutamyltransferase and bilirubin. There were missing values for some covariates: total body weight (2%), albumin (20.6%), plasma creatinine (0.9%), plasma protein (3.9%), C-reactive protein (2.9%), aspartate aminotransferase (2%), alanine transaminase (2%), alkaline phosphatase (2.9%), gamma-glutamyltransferase (2.9%) and bilirubin (4.9%). We inputted the missing covariate data with the predictive mean matching method implemented in the mice R package [23]. Imputation quality was checked by comparing the density of the imputed and observed data.
2.5 Individual Pharmacokinetic Data Analysis
The final model was used to compute individual Bayesian posterior PK parameter values and estimates of the AUC over 24 hours (AUC24) of RIF and 25-dRIF on each TDM occasion, based on clearance estimates and the dose received in each patient. Individual values of RIF AUC24/MIC ratios were calculated and compared to the target value of > 435, which has been associated with an improved response in patients with TB, as reported by Zheng et al. [7].
We also examined the correlation between concentrations and AUC24 of RIF and 25-dRIF and the correlation between AUC24 of RIF and other exposure metrics including the RIF concentration, 25-dRIF concentration and ratio of RIF/25-dRIF concentration. This was done to evaluate whether RIF AUC24 could be predicted from concentrations without modelling in routine TDM. This analysis was performed in a subset of concentration data. Trough concentrations were discarded, and we considered only the highest post-dose measured concentration of RIF, i.e. the measured peak concentration and associated concentration of 25-dRIF measured at the same time. GraphPad Prism (version 10; Dotmatics, Boston, MA, USA) was used for performing ordinary least-square regression and plotting the results.
Anzeige
3 Results
3.1 Population Characteristics
Patients’ characteristics are summarised in Table 1. A total of 247 concentrations of RIF and 243 concentrations of 25-dRIF measured in 35 patients with TB (10 women, 25 men) were available for the analysis. Several trough concentrations were below the lower limit of quantification: 50 (20%) for RIF and 91 (37%) for 25-dRIF. However, 52% of all samples were taken between 0 and 6 hours post-dose.
Data are given as median (minimum–maximum) or number
25-dRIF 25-deacetylrifampicin, RIF rifampicin
The median initial oral dose of RIF was 600 mg (minimum, 480 mg; maximum, 900 mg). Mycobacterium tuberculosis RIF MIC was obtained in 31 patients and ranged from 0.06 to 0.5 mg/L (median = 0.25 mg/L).
3.2 Population Pharmacokinetic Model
The final model was a two-compartment model including RIF metabolism into 25-dRIF. Log-normal distribution was retained for all random parameters. A mixed-error model with additional and proportional components best fitted the data. The final model adequately described the PK variability of the two drugs, as shown in Figs. 1 and 2. Parameter values are shown in Table 2 and the model structure is depicted in Fig. 2 in the ESM.
Fig. 1
Observed rifampicin (RIF) and 25-deacetylrifampicin (25-dRIF) concentrations versus individual model predictions. The coloured lines show the linear regression (solid lines) and confidence intervals (dotted lines). The solid black line is the line of identity
Prediction-corrected visual predictive checks of the final pharmacokinetic model. Top, visual predictive check for rifampicin (RIF); bottom, visual predictive check for 25-deacetylrifampicin (25-dRIF). Blue solid blue lines and shaded areas represent the 5th, 50th and 95th percentiles of the observed data and the 90% confidence intervals of model-based simulated observations (n = 1000), respectively. Blue dots are observed concentrations. h hours
Final estimates of population pharmacokinetic parameters
Parameter
Value
RSE
Bootstrap estimates
P2.5
Median
P97.5
Fixed effects
Ka (h-1)
0.95
1.0
0.56
0.97
1.51
V (L)
35.38
5.2
24.96
35.53
43.49
beta_V_IBW
0.90
37.7
0.2
0.81
1.51
CLRIF (L/h)
4.66
6.0
3.43
4.55
6.21
beta_CLRIF_BILIc
− 0.29
22.1
− 0.45
− 0.31
− 0.18
beta_CLRIF_IBWc
1.33
27.4
0.57
1.21
2.17
CL25-dRIF (L/h)
26.51
10.9
12.01
22.05
43.07
beta_ CL25-dRIF _BILIc
− 0.80
14.4
− 1.03
− 0.77
− 0.52
beta_ CL25-dRIF _IBWc
2.00
24.9
0.88
2.02
3.22
Fm
0.86 (fixed)
NA
NA
NA
NA
DAY50 (days)
1.64
7.4
0.36
1.72
3.87
MAXRIF
1.60
5.1
0.84
1.63
2.78
MAX25-dRIF
2.11
13.5
0.83
2.8
6.14
Random effectsa
omega_V
0.12 (11.6%)
34.9
0.031
0.13
0.32
omega_CLRIF
0.20 (20.5%)
23.1
0.028
0.18
0.26
omega_CL25-dRIF
0.28 (28.7%)
27.8
0.035
0.25
0.39
gamma_CLRIF
0.16 (16.5%)
23.4
0.07
0.16
0.24
gamma_CL25-dRIF
0.28 (28.5%)
24.8
0.11
0.24
0.41
Shrinkage
V
− 5.9%
NA
NA
NA
NA
CLRIF
− 7.6%
NA
NA
NA
NA
CL25-dRIF
− 1.0%
NA
NA
NA
NA
Error modelb
aRIF
0.06
14.5
0.0023
0.061
0.075
bRIF
0.42
7.3
0.34
0.41
0.5
a25-dRIF
0.02
15.3
0.001
0.024
0.031
b25-dRIF
0.51
9.5
0.39
0.51
0.64
25-dRIF 25-deacetylrifampicin, CLRIF apparent total body clearance of rifampicin, CL25-dRIF apparent total body clearance of rifampicin metabolite, CV% coefficient of variation, DAY50 time to half-maximum increase in clearance, Fm fraction of RIF dose metabolised into 25-dRIF, Ka absorption rate constant, MAXRIF maximal increase in RIF clearance from baseline, MAX25-dRIF maximal increase in 25-dRIF clearance from baseline, NA not applicable, P2.5 2.5th percentile, P97.5 97.5th percentile, RIF rifampicin, RSE percent relative standard error, V apparent volume of distribution of RIF and 25-dRIF
aOmega values are standard deviations of random effects (CV% of parameter) for inter-individual variability; gamma values are standard deviations of random effects (CV% of parameter) for inter-occasion variability
ba and b parameters are the additional and proportional error coefficients, respectively
cThe relationships between covariates and typical value of pharmacokinetic parameters were modelled as follows:
where _COV indicates the parameter value adjusted for the covariate, BILI is plasma bilirubin (µmol/L) and IBW is the ideal body weight (kg)
Inter-individual variability was estimated for the volume of distribution (V, which was assumed to be the same for both agents) as well as clearance of RIF and 25-dRIF. Population parameters were estimated with acceptable precision (relative standard errors < 50%). Shrinkage values were slightly negative, indicating that individual clearance and volume parameters were well estimated as well. Inter-occasion variability was also estimated for the two clearance parameters. It was quite similar to inter-individual variability for both agents, with inter-individual/inter-occasion coefficients of variation of 20.5%/16.5% and 28.7%/28.5% for RIF and 25-dRIF, respectively. We could not estimate inter-individual variability with acceptable precision for some parameters, including absorption rate constant, and thus only a population typical value was estimated. For the RIF fraction metabolised into 25-dRIF, it was fixed to a value estimated in preliminary runs (0.862) because of model stability issues. Inclusion of induction of both RIF and 25-dRIF improved the model fit. Induction was included into the model by considering a time-increasing clearance with a ceiling effect. The typical value of induction half-life was 1.6 days. The maximal increase in clearance from baseline was 160% and 211% for RIF and 25-dRIF, respectively. Regarding covariate modelling, V, clearance of RIF and 25dRIF increased with ideal body weight, while clearance of both agents decreased with the increasing bilirubin plasma level.
Anzeige
3.3 Individual Drug Exposure
Exposure to 25-dRIF was low, with a mean individual AUC24 ratio of 25-dRIF/RIF of 0.14 ± 0.06 for all patient-occasion values (minimum, 0.04; maximum, 0.38). Table 3 shows exposure to RIF and AUC24/MIC ratio over time. Optimal RIF exposure (AUC/MIC > 435) was achieved in only six (19%) out of 31 patients upon the first TDM, which was performed on day 1 for 20 out of 35 patients (interquartile range, 1–5 days).
Table 3
Exposure to rifampicin and AUC/MIC over time
Time period
Dose (mg)
AUC24 (mg.h/L)
AUC24/MIC
Day 1
600 (480–900)
81 (32–157)
[n = 20]
251 (126–628)
[n = 18]
Days 4–6
600 (480–600)
73 (47–139)
[n = 8]
259 (93–1609)
[n = 7]
Days 8–14
600 (480–900)
57 (17–82)
[n = 17]
170 (67–549) [n = 16]
Days 15–30
600 (480–720)
59 (25–87)
[n = 13]
236 (50–1455)
[n = 11]
Day ≥30
600 (480–900)
55 (23–103)
[n = 44]
149 (64–822)
[n = 40]
Data are given as median (minimum–maximum)
AUC area under the concentration–time curve, AUC24 area under the concentration–time curve over 24 hours, MIC minimum inhibitory concentration
Area under the concentration–time curve over 24 hours of both RIF and 25-dRIF rapidly decreased over the first days of therapy, as shown in Fig. 3. The median (interquartile range [IQR]) of individual AUC24 of RIF decreased from 81 (65–97 mg·h/L) on day 1 (n = 20) to 73 (54–102 mg·h/L) on days 4–6 (n = 8) and 57 (45–69) on days 8–14 (n = 17) and was stable afterwards. For 25-dRIF, the median AUC24 was stable between day 1 (10; IQR, 8–16 mg·h/L) and days 4–6 (10; IQR, 6–23 mg·h/L) and then declined on days 8–14 (6, IQR, 4–8 mg·h/L) and was stable hereafter.
Fig. 3
Boxplot of individual area under the concentration–time curve (AUC) over 24 hours of rifampicin (RIF) and 25-deacetylrifampicin (25-dRIF) over time
A correlation analysis of RIF exposure is shown in Fig. 3 of the ESM. The correlation between RIF and 25-dRIF concentration was moderate (R2 = 0.35, p < 0.0001). A stronger correlation was observed between the AUC24 of the two agents (R2 = 0.58, p < 0.0001). The correlation between RIF AUC24 and RIF concentration was weak (R2 = 0.13, p = 0.0005), while there was no correlation between the RIF AUC24 and RIF/25-dRIF concentration ratio (R2 = 0.02).
4 Discussion
Although the pharmacokinetics of RIF in patients with TB has been described in several studies, limited information exists about that of its active metabolite, 25-dRIF. In their systematic review of population PK studies of RIF published in 2022, Muda et al. identified 16 population studies performed in adult patients, with only one reporting data on both RIF and 25-dRIF [24]. That study from Seng et al. was performed in 34 healthy volunteers, not in patients with TB [25]. An older study described the pharmacokinetics of RIF and that of two metabolites (25-dRIF and 3-formyl-RIF) in two groups of six patients who received multiple doses of oral and intravenous RIF, respectively, without modelling results [15, 26]. To our knowledge, this is the first population PK modelling analysis of RIF and 25-dRIF in patients treated for TB.
Anzeige
The inclusion of an increase in RIF clearance over time improved the model fit, which is consistent with previous reports and the known auto-induction of the drug. However, the speed and magnitude of RIF induction somewhat vary between studies. Smythe et al. estimated an induction half-life of 7.8 days and a 1.85-fold increase of oral clearance of RIF at steady state from baseline [27]. Svensson et al. reported a dose-dependent increase in apparent clearance over time (1.7 for a standard dose of 10 mg/kg) and an induction half-life of 4.8 days [28]. Chirehwa et al. reported very similar values: induction half-life of 4.5 days and a 1.9-fold maximal increase in CL from baseline [29]. We found a slightly higher maximal increase in RIF apparent oral clearance over time (+ 160%, i.e. 2.6-fold increase) and a shorter induction half-life of 1.6 days, with the upper bound of the bootstrap confidence interval of 3.9 days. The difference between studies may be explained in part by the sampling design. The longest induction half-life reported by Smythe et al. may be explained by the prolonged period between sampling days (day 1 and day 28). By contrast, patients were sampled on days 7 and 14, and days 1, 8, 15 and 29 in studies form Svensson et al. and Chirehwa et al., respectively. In our study, 20 patients were sampled on day 1, and eight were sampled between days 4 and 6. Those early data probably influenced the estimation of a shorter half-life.
In addition, our analysis provided evidence of induction of 25-dRIF as well, which was even more marked, with a maximal increase of metabolite clearance of 211% from baseline. We could not estimate a separate half-life for induction of RIF and 25-dRIF, which is likely to have influenced the estimation as well. Induction of 25-dRIF was previously reported by Loos et al. who found a 50% decrease in relative AUC24 of 25-dRIF on days 8–9 compared with days 1–2 [15].
Of note, the magnitude of observed changes in individual AUC24 values (Fig. 3) was lower than the maximal increase in clearance estimated by the model. This may be due to the use of variable doses of RIF (in milligrams) between patients and the number of individual AUC24 estimates, which was not consistent over TDM days. The large inter-individual and intra-individual variability of drug clearance may also explain this difference. Interestingly, in the study from Loos et al., the decrease in RIF exposure over time was greater when RIF was administered by oral route compared with the intravenous route, which suggests that both pre-systemic (gut and liver first-pass effect) and systemic clearances may undergo induction [26].
We identified ideal body weight and bilirubin as covariates influencing RIF and 25d-RIF pharmacokinetics. The influence of body size on RIF clearance and V has been previously reported in some studies [30‐32], but inconsistently as shown in the review from Muda et al. [24]. Although this influence appears as a rational basis for weight-based dosing of RIF, body size explains a limited fraction of PK inter-individual variability, and the clinical relevance of weight-based dosing in adults remains controversial [33]. The decrease of RIF and 25-dRIF clearance with increasing plasma bilirubin is consistent with previous reports and may be explained by a competition between RIF and metabolites with bilirubin for conjugation and biliary excretion [34, 35].
Our results indicate that exposure to 25-dRIF was low, representing about 14% of RIF exposure on average, although this ratio varied largely between patients and occasions. This average AUC ratio was consistent with the results from Loos et al., who reported an AUC ratio of about 10% [15]. We observed no trend in this ratio over time (data not shown) and the AUC24 of 25-dRIF correlated well with that of RIF (Fig. 3 of the ESM). This means that the decrease in RIF exposure over time was not associated with an increased exposure in 25-dRIF. In other words, the metabolite cannot compensate for reduced RIF exposure and the whole active fraction decreases over time.
Although 25-dRIF has been described as active, limited information exists about its antimicrobial activity on Mtb. Only one study explored the activity of rifapentine and its metabolite 25-O-desacetylrifapentine on various Mtb complex isolates, showing weaker antimicrobial activity of the metabolite with approximately four-fold higher MIC [36]. Recently, our group has performed an in vitro study of the activity of rifamycin (RIF, rifabutin and rifapentine) metabolites against various lineages and strains of MTB and observed an activity of 25-deacetyl metabolites similar to that of the parent drugs, and additivity when combined with them (manuscript under review). Further research is necessary to clarify whether exposure to 25-dRIF has clinical relevance for TB therapy.
The AUC24 of RIF estimated on day 1 (85 ± 29 mg·h/L) and on day ≥30 (56 ± 18 mg·h/L) in our study are quite consistent with results of the meta-analysis from Stott et al. who reported AUC24 pooled estimates of 73 and 39 mg·h/L after a single dose and at steady state, respectively, with large inter-study variability [37]. Considering an AUC24/MIC ratio of 435 as a reference, optimal exposure to RIF was achieved in only six (19%) out of 31 patients upon the first TDM. This AUC24/MIC cut-off was associated with treatment outcome at 6–8 months in the study from Zheng et al. [7]. It is noteworthy that AUC estimates in that study were based on TDM performed after 2 weeks. As most patients in our study had the first TDM before, their concentrations after 2 weeks were probably lower because of RIF induction. Many studies reported low concentrations of anti-TB drugs in adult patients who received standard dosages, especially for RIF [38]. Although the clinical relevance of accepted concentration targets can be questioned, the 600-mg/kg or 10-mg/kg dose of RIF has been highlighted as too low for a long time [2] and higher dosages of RIF have shown promising results [11, 13, 39].
Therapeutic drug monitoring of anti-TB drugs may be relevant to identify patients with low exposure to RIF and implement higher dosages through a more individualised approach [40, 41]. We believe that TDM should not be performed once and too early during therapy but repeated, as RIF exposure is not stable and decreases with time, unless a model incorporating auto-induction is used to analyse the results and can predict the decrease in clearance. The complexity of RIF pharmacokinetics makes the interpretation of TDM results and dosing decision difficult. The clearance of RIF is both time and dose dependent [28], with a more than proportional increase in AUC when the dose increases [10]. Our results suggest that a single RIF concentration and a single RIF/metabolite ratio are not predictive of the AUC of RIF. Multiple linear regression based on limited sampling strategies or Bayesian modelling approaches are required to accurately estimate AUC [42, 43]. Of note, sampling and measuring trough concentrations of RIF should be discouraged in routine TDM, as the result is quite often below the quantification limit, providing no relevant information on drug exposure for an individual patient.
This study has several limitations. The sample size was limited and PK data, obtained in routine clinical conditions, were sparse. A study with more samples for each individual study participant would be necessary to confirm our findings. We did not control some factors that may influence RIF PK and intra-individual variability such as food intake or albumin levels. Pharmacokinetic sampling, although repeated on several occasions, was sparse. However, each sample provided two forms of concentration data (RIF + metabolite) and the concentrations of the metabolite increased the overall PK information and contributed to the estimation of RIF PK profile and clearance. Indeed, shrinkage values were low, indicating that individual data were informative enough to estimate individual clearance and volume parameters. We performed an internal but not external validation of the population PK model because of the small study size. In our PK analysis, we could not identify complex absorption mechanisms, nor saturable elimination and dose-dependent clearance of RIF that have been described elsewhere [28, 44]. For this reason, we did not perform dosing simulations based on the model, as the AUC prediction for higher dosages would probably be inaccurate. We did not examine relationships between drug exposure and outcomes.
5 Conclusions
This study showed that exposure to both RIF and 25-dRIF rapidly decreases during the first days of RIF therapy in adult patients treated for TB, probably because of auto-induction. The contribution of 25-dRIF to the overall exposure to rifamycin active fraction is low. Attainment of target AUC24/MIC for RIF was poor, suggesting an increased RIF dosage as an option to compensate for auto-induction during TB therapy.
Acknowledgements
This study was supported by a grant from the French National Research Agency (Agence Nationale pour la Recherche, ANR) (project PRIM-TB, ANR-18-CE17-0020). Lyon TB Study Group collaborators: F. Ader, O. Bahuaud, R. Bayaa, A. Becker, E. Braun, P. Chabert, P. Chauvelot, C. Chedid, A. Conrad, O. Dumitrescu, T. Ferry, C. Genestet, S. Goutelle, E. Hodille, J. Hoffmann, C. Javaux, G. Lina, C. Pouderoux, T. Perpoint, S. Roux, M. Simon, F. Valour.
Declarations
Funding
Open access funding provided by Hospices Civils de Lyon. This study was supported by a grant from the French National Research Agency (Agence Nationale pour la Recherche, ANR) [project PRIM-TB, ANR-18-CE17-0020].
Conflict of interest
Sylvain Goutelle, Olivier Bahuaud, Charlotte Genestet, Aurélien Millet, François Parant, Oana Dumitrescu and Florence Ader have no conflicts of interest that are directly relevant to the content of this article.
Ethics approval
This study was approved by our institutional ethics committee (n°69HCL18_0757) and has been registered in the ClinicalTrials.gov website (NCT04271397).
Consent to participate
Written informed consent was obtained from all participants.
Consent for publication
Not applicable.
Availability of data and material
The study data are available from the corresponding authors upon reasonable request.
Code availability
Not applicable.
Author contributions
OD and FA initiated the project and designed the Opti-4TB trial. OB, OD and FA participated in the clinical data acquisition. CG, AM and FP participated in the collection of biological data. SG performed the PK data analysis and drafted the manuscript. All authors performed a critical review of the manuscript draft and approved the final version of the manuscript.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc/4.0/.
Peloquin CA, Davies GR. The treatment of tuberculosis. Clin Pharmacol Ther. 2021;110(6):1455–66.PubMedCrossRef
2.
van Ingen J, Aarnoutse RE, Donald PR, Diacon AH, Dawson R, van Balen GP, et al. Why do we use 600 mg of rifampicin in tuberculosis treatment? Clin Infect Dis. 2011;52(9):e194–9.PubMedCrossRef
3.
Akkerman OW, Dijkwel RDC, Kerstjens HAM, van der Werf TS, Srivastava S, Sturkenboom MGG, et al. Isoniazid and rifampicin exposure during treatment in drug-susceptible TB. Int J Tuberc Lung Dis. 2023;27(10):772–7.PubMedPubMedCentralCrossRef
4.
Trentalange A, Borgogno E, Motta I, Antonucci M, Pirriatore V, Costa C, et al. Rifampicin and isoniazid maximal concentrations are below efficacy-associated thresholds in the majority of patients: time to increase the doses? Int J Antimicrob Agents. 2021;57(3): 106297.PubMedCrossRef
5.
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(9):1464–73.PubMedPubMedCentralCrossRef
6.
Nahid P, Mase SR, Migliori GB, Sotgiu G, Bothamley GH, Brozek JL, et al. Treatment of drug-resistant tuberculosis: an official ATS/CDC/ERS/IDSA clinical practice guideline. Am J Respir Crit Care Med. 2019;200(10):e93-142.PubMedPubMedCentralCrossRef
7.
Zheng X, Bao Z, Forsman LD, Hu Y, Ren W, Gao Y, et al. Drug exposure and minimum inhibitory concentration predict pulmonary tuberculosis treatment response. Clin Infect Dis. 2021;73(9):e3520–8.PubMedCrossRef
8.
Sekaggya-Wiltshire C, von Braun A, Lamorde M, Ledergerber B, Buzibye A, Henning L, et al. Delayed sputum culture conversion in tuberculosis-human immunodeficiency virus-coinfected patients with low isoniazid and rifampicin concentrations. Clin Infect Dis. 2018;67(5):708–16.PubMedPubMedCentralCrossRef
9.
Te Brake L, Dian S, Ganiem AR, Ruesen C, Burger D, Donders R, et al. Pharmacokinetic/pharmacodynamic analysis of an intensified regimen containing rifampicin and moxifloxacin for tuberculous meningitis. Int J Antimicrob Agents. 2015;45(5):496–503.CrossRef
10.
Boeree MJ, Diacon AH, Dawson R, Narunsky K, du Bois J, Venter A, et al. A dose-ranging trial to optimize the dose of rifampin in the treatment of tuberculosis. Am J Respir Crit Care Med. 2015;191(9):1058–65.PubMedCrossRef
11.
Onorato L, Gentile V, Russo A, Di Caprio G, Alessio L, Chiodini P, et al. Standard versus high dose of rifampicin in the treatment of pulmonary tuberculosis: a systematic review and meta-analysis. Clin Microbiol Infect. 2021;27(6):830–7.PubMedCrossRef
12.
Marais S, Cresswell FV, Hamers RL, Te Brake LHM, Ganiem AR, Imran D, et al. High dose oral rifampicin to improve survival from adult tuberculous meningitis: a randomised placebo-controlled double-blinded phase III trial (the HARVEST study). Wellcome Open Res. 2019;4:190.PubMedCrossRef
13.
Perumal Kannabiran B, Palaniappan NA, Manoharan T, Paramasivam PK, Saini JK, Ansari MS, et al. Safety and efficacy of 25 mg/kg and 35 mg/kg vs 10 mg/kg rifampicin in pulmonary TB: a phase IIb randomized controlled trial. Open Forum Infect Dis. 2024;11(3):ofae034.PubMedPubMedCentralCrossRef
14.
Abulfathi AA, Decloedt EH, Svensson EM, Diacon AH, Donald P, Reuter H. Clinical pharmacokinetics and pharmacodynamics of rifampicin in human tuberculosis. Clin Pharmacokinet. 2019;58(9):1103–29.PubMedCrossRef
15.
Loos U, Musch E, Jensen JC, Mikus G, Schwabe HK, Eichelbaum M. Pharmacokinetics of oral and intravenous rifampicin during chronic administration. Klin Wochenschr. 1985;63(23):1205–11.PubMedCrossRef
16.
Vu DH, Koster RA, Bolhuis MS, Greijdanus B, Altena RV, Nguyen DH, et al. Simultaneous determination of rifampicin, clarithromycin and their metabolites in dried blood spots using LC-MS/MS. Talanta. 2014;121:9–17.PubMedCrossRef
17.
Bahuaud O, Genestet C, Hoffmann J, Dumitrescu O, Ader F. Opti-4TB: a protocol for a prospective cohort study evaluating the performance of new biomarkers for active tuberculosis outcome prediction. Front Med (Lausanne). 2022;9: 998972.PubMedPubMedCentralCrossRef
18.
Genestet C, Ader F, Pichat C, Lina G, Dumitrescu O, Goutelle S. Assessing the combined antibacterial effect of isoniazid and rifampin on four Mycobacterium tuberculosis strains using in vitro experiments and response-surface modeling. Antimicrob Agents Chemother. 2018;62(1):e01413-e1417.PubMedCrossRef
19.
Devine BJ. Gentamicin therapy. Drug Intell Clin Pharm. 1974;8:650–5.
20.
James WPT, Waterlow JC. Research on obesity: a report of the DHSS/MRC Group. HM Stationery Office; 1976.
21.
Gehan EA, Georges SL. Estimation of human body surface area from height and weight. Cancer Chemother Rep. 1970;54(4):225.PubMed
22.
Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–12.PubMedPubMedCentralCrossRef
23.
van Buuren S, Groothuis-Oudshoorn K. Multivariate imputation by chained equations in R. J Stat Softw. 2011;45(3):1–67.CrossRef
24.
Muda MR, Harun SN, Syed Sulaiman SA, Sheikh Ghadzi SM. Population pharmacokinetics analyses of rifampicin in adult and children populations: a systematic review. Br J Clin Pharmacol. 2022;88(7):3132–52.PubMedCrossRef
25.
Seng KY, Hee KH, Soon GH, Chew N, Khoo SH, Lee LS. Population pharmacokinetics of rifampicin and 25-deacetyl-rifampicin in healthy Asian adults. J Antimicrob Chemother. 2015;70(12):3298–306.PubMedCrossRef
26.
Loos U, Musch E, Jensen JC, Schwabe HK, Eichelbaum M. Influence of the enzyme induction by rifampicin on its presystemic metabolism. Pharmacol Ther. 1987;33(1):201–4.PubMedCrossRef
27.
Smythe W, Khandelwal A, Merle C, Rustomjee R, Gninafon M, Bocar Lo M, et al. A semimechanistic pharmacokinetic-enzyme turnover model for rifampin autoinduction in adult tuberculosis patients. Antimicrob Agents Chemother. 2012;56(4):2091–8.PubMedPubMedCentralCrossRef
28.
Svensson RJ, Aarnoutse RE, Diacon AH, Dawson R, Gillespie SH, Boeree MJ, et al. A population pharmacokinetic model incorporating saturable pharmacokinetics and autoinduction for high rifampicin doses. Clin Pharmacol Ther. 2018;103(4):674–83.PubMedCrossRef
29.
Chirehwa MT, Rustomjee R, Mthiyane T, Onyebujoh P, Smith P, McIlleron H, et al. Model-based evaluation of higher doses of rifampin using a semimechanistic model incorporating autoinduction and saturation of hepatic extraction. Antimicrob Agents Chemother. 2016;60(1):487–94.PubMedCrossRef
30.
Chang MJ, Chae JW, Yun HY, Lee JI, Choi HD, Kim J, et al. Effects of type 2 diabetes mellitus on the population pharmacokinetics of rifampin in tuberculosis patients. Tuberculosis (Edinb). 2015;95(1):54–9.PubMedCrossRef
31.
Kim ES, Kwon BS, Park JS, Chung JY, Seo SH, Park KU, et al. Relationship among genetic polymorphism of SLCO1B1, rifampicin exposure and clinical outcomes in patients with active pulmonary tuberculosis. Br J Clin Pharmacol. 2021;87(9):3492–500.PubMedCrossRef
32.
Gao Y, Davies Forsman L, Ren W, Zheng X, Bao Z, Hu Y, et al. Drug exposure of first-line anti-tuberculosis drugs in China: a prospective pharmacological cohort study. Br J Clin Pharmacol. 2021;87(3):1347–58.PubMedCrossRef
33.
Susanto BO, Svensson RJ, Svensson EM, Aarnoutse R, Boeree MJ, Simonsson USH. Rifampicin can be given as flat-dosing instead of weight-band dosing. Clin Infect Dis. 2020;71(12):3055–60.PubMedCrossRef
34.
McIlleron H, Wash P, Burger A, Norman J, Folb PI, Smith P. Determinants of rifampin, isoniazid, pyrazinamide, and ethambutol pharmacokinetics in a cohort of tuberculosis patients. Antimicrob Agents Chemother. 2006;50(4):1170–7.PubMedPubMedCentralCrossRef
35.
Acocella G, Segre G, Conti R, Pagani V, Pallanza R, Perna G, et al. Pharmacokinetic study on intravenous rifampicin in man. Pharmacol Res Commun. 1984;16(7):723–36.PubMedCrossRef
36.
Rastogi N, Goh KS, Berchel M, Bryskier A. Activity of rifapentine and its metabolite 25-O-desacetylrifapentine compared with rifampicin and rifabutin against Mycobacterium tuberculosis, Mycobacterium africanum, Mycobacterium bovis and M. bovis BCG. J Antimicrob Chemother. 2000;46(4):565–70.PubMedCrossRef
37.
Stott KE, Pertinez H, Sturkenboom MGG, Boeree MJ, Aarnoutse R, Ramachandran G, et al. Pharmacokinetics of rifampicin in adult TB patients and healthy volunteers: a systematic review and meta-analysis. J Antimicrob Chemother. 2018;73(9):2305–13.PubMedPubMedCentralCrossRef
38.
Mota L, Al-Efraij K, Campbell JR, Cook VJ, Marra F, Johnston J. Therapeutic drug monitoring in anti-tuberculosis treatment: a systematic review and meta-analysis. Int J Tuberc Lung Dis. 2016;20(6):819–26.PubMedCrossRef
39.
Boeree MJ, Heinrich N, Aarnoutse R, Diacon AH, Dawson R, Rehal S, et al. High-dose rifampicin, moxifloxacin, and SQ109 for treating tuberculosis: a multi-arm, multi-stage randomised controlled trial. Lancet Infect Dis. 2017;17(1):39–49.PubMedPubMedCentralCrossRef
40.
Seijger C, Hoefsloot W, Bergsma-de Guchteneire I, Te Brake L, van Ingen J, Kuipers S, et al. High-dose rifampicin in tuberculosis: experiences from a Dutch tuberculosis centre. PLoS ONE. 2019;14(3): e0213718.PubMedPubMedCentralCrossRef
41.
Pasipanodya JG, Gumbo T. Individualizing tuberculosis (TB) treatment: are TB programs in high burden settings ready for prime time therapeutic drug monitoring? Clin Infect Dis. 2018;67(5):717–8.PubMedCrossRef
42.
Saktiawati AMI, Harkema M, Setyawan A, Subronto YW, Sumardi, Stienstra Y, et al. Optimal sampling strategies for therapeutic drug monitoring of first-line tuberculosis drugs in patients with tuberculosis. Clin Pharmacokinet. 2019;58(11):1445–54.
43.
Sturkenboom MGG, Martson AG, Svensson EM, Sloan DJ, Dooley KE, van den Elsen SHJ, et al. Population pharmacokinetics and Bayesian dose adjustment to advance TDM of anti-TB drugs. Clin Pharmacokinet. 2021;60(6):685–710.PubMedPubMedCentralCrossRef
44.
Wilkins JJ, Savic RM, Karlsson MO, Langdon G, McIlleron H, Pillai G, et al. Population pharmacokinetics of rifampin in pulmonary tuberculosis patients, including a semimechanistic model to describe variable absorption. Antimicrob Agents Chemother. 2008;52(6):2138–48.PubMedPubMedCentralCrossRef
