Dosing of Dolutegravir in TB/HIV Coinfected Patients on Rifampicin: Twice Is (Always) Better Than Once

To the Editors: We read with great interest the article by Modongo et al,1 which reported favorable tuberculosis (TB) and HIV outcomes in a large cohort of TB/HIV coinfected patients from Botswana treated with dolutegravir-based antiretroviral regimens. Interestingly, only 53% of the 739 TB/HIV coinfected patients who received rifampicin- and dolutegravir-based regimens received the twice daily dolutegravir dosing suggested by international guidelines (and foreseen in the study protocol), whereas the remaining were given the once daily dosing recommended for HIV patients not on rifampicin. Despite this deviation from the protocol, similar rates of viral suppression were found between those who were on once daily versus twice daily dolutegravir regimens while on rifampicin-based TB treatment (adjusted P value = 0.309). Although acknowledged by the authors as not conclusive, these findings might potentially provide a significant degree of confidence to clinicians and public health officials worldwide over the safety of once daily dolutegravir regimens in TB patients on rifampicin. However, we suggest caution in the interpretation of this study for a number of reasons that will be detailed below. In the first study published in 2013, Dooley et al2 showed that in healthy volunteers, concomitant administration of rifampicin 600 mg once daily resulted in a 33% reduction in the area under the curve (AUC) of dolutegravir given at 50 mg once daily; such inductive effect was significantly mitigated when the dose of dolutegravir was increased to 50 mg twice daily. These results were reinforced by Wang et al3; using a similar study design, the authors reported that coadministration of rifampicin and dolutegravir 50 mg once daily resulted in a 56% and 85% reduction, respectively, in the AUC and trough concentrations compared to values measured in absence of rifampicin. The doubling of dolutegravir dosage to 50 mg twice daily partially compensated the reduction in dolutegravir AUC (−26%) but not in the drug trough concentrations (−76%). As one main methodological limitation, both these studies were conducted in healthy volunteers.2,3 It is, indeed, well known that the pharmacokinetics of antiretrovirals, as well as their pattern of drug–drug interactions, may change significantly moving from healthy volunteers to HIV-infected patients.4–7 This gap has been recently bridged by Barcelo et al.8 By performing population pharmacokinetic modeling of data from more than 600 HIV-infected patients (including 10 concomitantly treated with rifampicin), the authors confirmed that the main driver of dolutegravir clearance was rifampicin coadministration. In particular, some simulations suggested that the average dolutegravir trough concentrations were, respectively, 63% or 92% lower after the administration of dolutegravir at 50 mg given twice daily or at 100 mg once daily compared with standard dosages of 50 mg once daily without rifampicin. Based on their findings, the authors proposed that the dose of dolutegravir should be increased up to 100 mg twice daily in selected patients concomitantly treated with rifampicin, ideally combined with routine therapeutic drug monitoring (TDM) of dolutegravir concentrations.8 To provide an assessment of this drug–drug interaction from real-life settings, we retrospectively looked at the database of our Infectious Diseases clinic searching for HIV-infected patients concomitantly treated with dolutegravir and rifampicin for any reasons and with at least one TDM available of dolutegravir trough concentrations. Such concentrations were then compared with values measured routinely in HIV-infected patients given dolutegravir without rifampicin.9,10 Ten patients fulfilling these criteria were identified. They were mostly males (n = 8, mean age 42 ± 11 years), Caucasians (n = 8) with normal renal (serum creatinine 0.9 ± 0.2 mg/dL) and liver function (aspartate aminotransferase 49 ± 25 IU/L), given rifampicin for lung TB (n = 4), bone TB (n = 2), atypical mycobacteriosis (n = 2), disseminated TB (n = 1), TB prophylaxis (n = 2), or prosthetic infection (n = 1). All the patients were treated with rifampicin at 600 mg once daily, with mean maximum drug concentrations of 6.6 ± 3.4 mg/L (normal values 8–24 mg/L). Seven out of the 10 patients were given dolutegravir at 50 mg twice daily, with mean trough dolutegravir concentrations of 369 ± 186 ng/mL. These concentrations were significantly lower compared with values measured in HIV-infected patients given dolutegravir without rifampicin who performed TDM in the same period (n = 200, mean dolutegravir trough concentrations 1225 ± 772 ng/mL; P = 0.0013), but still above the minimum efficacy drug concentrations (set in our laboratory at 100 ng/mL). The remaining 3 patients were treated with dolutegravir at 50 mg once daily. In the first case, rifampicin was prescribed by another center for TB prophylaxis (they were probably not aware of the HIV status); the second patient did not understand the instruction of the attending physician of doubling dolutegravir daily dose and, for the third case, we hypothesized a medical error. The trough dolutegravir concentrations measured in these 3 patients were 22, <20, and 94 ng/mL, respectively (Fig. 1). Remarkably, all the 3 patients, who had undetectable HIV viral load (<37 copies/mL) before starting rifampicin, showed an increase in the HIV viral load during concomitant rifampicin administration (67, 69,000 and 17,610 copies/mL, respectively). Just to provide an example on how rifampicin may not linearly impact on dolutegravir disposition, the trough concentrations of dolutegravir increased from 22 to 434 ng/mL in one of these 3 patients when the dose of dolutegravir was increased from 50 mg once daily to 50 mg twice daily; a third TDM, performed 6 months after rifampicin discontinuation, resulted in a dolutegravir trough concentration of 2584 ng/mL (the patients maintained the dose of 50 mg twice daily also after rifampicin discontinuation for the virological failure to dolutegravir and for the few therapeutic options available).FIGURE 1.: Box-plot of dolutegravir trough concentrations measured in HIV-infected patients given the drug with (n = 10) or without rifampicin (n = 200). Full and empty circles represent, respectively, patients given dolutegravir at 50 mg once daily and 50 mg twice daily. Dashed line represents the minimum effective drug concentration set at 100 ng/mL.Taken together, the findings of Barcelo and our real-life experience, by consistently showing that coadministration of these 2 drugs resulted in a highly significant reduction in the disposition of dolutegravir that was partly mitigated by doubling dolutegravir dosing, seem to argue against the experience from Botswana by Modongo et al.1 These apparent discrepant results can be eventually reconciled by considering some key differences in the study populations. Indeed, the study by Modongo et al involved exclusively African patients, whereas the large majority of the patients from the Swiss HIV Cohort Study,8 as well as those from our database, were Caucasians. This is not a trivial difference considering the available extensive literature showing that race/ethnicity can greatly influence plasma exposure of HIV-infected patients to antiretrovirals, as documented for efavirenz and tenofovir.11,12 Therefore, it can be reasonably speculated that patients from Botswana treated with dolutegravir at 50 mg once daily might be exposed, for their ethnicity, to higher drug concentrations, a condition that might have counterbalanced the inductive effect of rifampicin, eventually resulting in therapeutic drug concentrations and optimal clinical outcome without the need of doubling dolutegravir dosing. Conversely, the once daily dosing of dolutegravir given to Caucasian patients concomitantly treated with rifampicin is likely to result in suboptimal drug exposure, increasing the risk of HIV virologic failure as found in our 3 patients. In conclusion, the selection of the optimal dose of dolutegravir in patients concomitantly treated with rifampicin is still a work in progress, with some trials recently published and others still ongoing.13,14 Pending their final results, we believe that, presently, the safest approach for these patients, regardless of their ethnicity, is to use the doubled dose of dolutegravir. In this way, we can guarantee optimal exposure to dolutegravir in all patients, maintaining, at the same time, optimal drug tolerability. In fact, no increased incidence of dolutegravir-related adverse events was reported in HIV-infected patients from Botswana treated with dolutegravir at 50 mg twice daily.