Tenofovir alafenamide (TAF) is a prodrug of tenofovir that provides higher intracellular tenofovir concentrations and greatly lower systemic drug exposure compared with tenofovir diproxil fumarate (TDF).1–3 A recent meta-analysis investigating the efficacy and safety of TAF vs. TDF in antiretroviral (ARV) regimens showed that TAF had significantly smaller reductions in estimated glomerular filtration rate, smaller changes in serum and urinary markers of renal dysfunction, and less reduction in spine and hip bone mineral density compared with TDF.4 The improvement in the renal and bone function was confirmed also in HIV-infected participants with moderate renal impairment who were switched from TDF-based ARV regimens to elvitegravir (ELV)/cobicistat (COBI)/emtricitabine/TAF coformulation.5
Coadministration of TAF with ritonavir or COBI resulted in increased tenofovir concentrations, given the inhibitory effect exerted by these boosting agents on the efflux transporter p-glycoprotein.6,7 For these reasons, the dose of TAF needs to be reduced from 25 to 10 mg daily when given concomitantly with ritonavir or COBI.6 Evidence is also available showing that administration of ritonavir-boosted HIV protease inhibitors (PIs) to patients with TDF-treated HIV and coadministration of COBI and TDF to healthy volunteers resulted in significantly increased tenofovir concentrations.8–12 However, at variance with TAF, no dose reduction is recommended for TDF when given with boosting agents.
Differences in the dosing strategies adopted between TAF and TDF when given with boosted ARVs may be relevant because of the growing and consistent evidence showing that patients exposed to high plasma tenofovir concentrations are at increased risk to experience drug-related complications.13–17 Accordingly, it could be speculated that the less nephro- and/or bone toxicity observed with TAF vs. TDF during registrative trials might have been driven, at least in part, by inappropriate dose selection for drugs comparison.
To address this issue, in this study, we aimed to (1) confirm in HIV-infected patients chronically treated with TDF the boosting effect of COBI on tenofovir concentrations previously documented in healthy volunteers,11 and (2) verify the potential contribution of boosting agents on the long-term durability of TDF-based ARV therapies in real life settings.
MATERIALS AND METHODS
Male and female HIV-infected patients on TDF-based antiretroviral therapy who underwent therapeutic drug monitoring (TDM) of tenofovir concentrations and referring to the Department of Infectious Diseases at Luigi Sacco University Hospital, Milan, Italy, were enrolled in this study. All patients had creatinine clearance >80 mL/min before starting TDF treatment and all were subsequently treated with TDF at 300 mg given orally every 24 hours. Pediatric subjects and patients with severe hepatic impairment (defined as Child–Pugh Class B or C) were excluded from this study.
Study Design and Statistical Analyses
This is a retrospective, cross-sectional study conducted in HIV-positive patients receiving TDF-containing antiretroviral therapies for at least 3 months and with at least 1 request of TDM of tenofovir plasma trough concentrations. The first TDM was considered in patients with more than 1 assessment of tenofovir plasma trough concentrations.
In the first part of the study, we sought to identify demographic, hematochemical, and clinical covariates potentially associated with tenofovir plasma trough concentrations. To address this issue, univariate and multivariate regression analyses were performed considering tenofovir concentration as the dependent variable and the clinical characteristics recorded at the time of TDM as independent covariates. Given the wide distribution of patients according to concomitant ARV agents, data were grouped in 4 main drug classes: PIs/ritonavir (PIs/r: atazanavir, darunavir, lopinavir, and fosamprenavir), non–nucleoside reverse transcriptase inhibitors (NNRTIs: efavirenz, rilpivirine, and nevirapine), integrase inhibitors (INSTIs: dolutegravir and raltegravir), and ELV/COBI coformulation (ELV/COBI at the time of the study conduction COBI was available on the market only coformulated with ELV). A general linear model to analyze the effect of independent clinical variables on tenofovir concentrations was applied. Independent variables significant at univariate analysis were introduced in the multivariate model. Tenofovir concentrations were log-transformed to comply with the model requirements.
In the second part of the study, we investigated the potential role of boosting agents (ritonavir or COBI) on TDF durability. Such outcome was defined as TDF discontinuation regardless of the cause. The time to the event in different groups was analyzed through a Kaplan–Meier survival analysis and log-rank test, whereas the effect of related factors was investigated with the Cox proportional hazard model. Besides the drug class coadministered with TDF, the following covariates were included also in the multivariable Cox model: serum creatinine, patients' weight, age, CD4+ T-cell count (all at the time of TDM), sex, days of TDF therapy, and plasma tenofovir concentrations.
All statistical analyses were performed using SAS 9.4 statistical software (SAS Institute, Inc., Cary, NC). All P-values presented are 2 sided and a P-value of <0.05 indicated statistical significance.
This retrospective research was conducted on data collected for clinical purposes. All data used in the study were previously anonymized, according to the requirements set by Italian Data Protection Code (leg. decree 196/2003) and by the General authorizations issued by the Data Protection Authority. Approval by Ethics Committee was deemed unnecessary because, under Italian law, such an approval is required only in the hypothesis of prospective clinical trials on medical products for clinical use (art. 6 and art. 9, leg. decree 211/2003). Written informed consent for medical procedures/interventions performed for routine treatment purposes was collected for each patient.
Blood trough samples drawn into ethylenediaminetetraacetic acid–containing vacutainers were collected from all patients immediately before the next TDF intake (a time window of ±30 minutes was directly verified by the nurse staff and considered as acceptable). All samples were centrifuged at 3000g then plasma was separated, inactivated, and stored at −20°C until analysis. After purification of plasma samples through solid-phase extraction, tenofovir concentrations were determined by a liquid chromatography tandem mass spectrometry method, validated according to international guidelines.15,18 Chromatographic separation was achieved with a gradient (methanol and water with formic acid 1%) on a reversed phase analytical column (Atlantis HILIC Silica 2.1 × 150 mm, Waters, Milan, Italy). For quantification, isotope dilution analysis was performed by monitoring the transition m/z 288.3 >176.1 for tenofovir and the transition 294 >182 for the internal standard (2H6 tenofovir).
Five hundred and ten HIV-infected adult patients fulfilling the inclusion/exclusion criteria were included in this study. These patients were given TDF in combination with PIs/r (n = 212, 41.6%), NNRTIs (n = 176, 34.5%), INSTIs (dolutegravir or raltegravir, n = 46, 9.0%), or with ELV/COBI (n = 76, 14.9%). Demographic and clinical data of these patients (stratified according to the companion ARV regimens) at the time of TDM are listed in Table 1. Patients on ELV/COBI were at a significantly fewer days on TDF therapy compared with other regimens (P <0.01). Other significant differences at baseline were detected among groups for sex, presence of coinfections, viral load (>37 copies/mL), and CD4 (<250 cells/mL).
Distribution of Tenofovir Plasma Trough Concentrations
Overall, a wide distribution in the measured tenofovir plasma trough concentrations was observed with values ranging from 21 to 795 ng/mL resulting in a mean interpatient variability of 78.7%. As shown in Figure 1, the highest tenofovir concentrations were measured in patients given ELV/COBI (161 ± 113 ng/mL) being significantly higher than values measured in patients given PIs/r (147 ± 125 ng/mL), NNRTIs (109 ± 62 ng/mL), or INSTIs (113 ± 74 ng/mL).
By multivariate analysis we found that, among the clinical covariates studied, patients' age, sex, body weight, and renal function (expressed as serum creatinine) were significantly and independently associated with tenofovir plasma trough concentrations (Table 2). Univariate and multivariate analysis also evidenced a major contribution of ARV therapy on tenofovir exposure. Indeed, in the multivariate analysis, statistically significant differences were found between PIs/r and INSTIs (P = 0.035) and between PIs/r and ELV/COBI (P < 0.01) pointing the latter concomitant ARV regimen as the most significant predictor of high tenofovir concentrations (Table 2).
As exploratory analysis, we also looked at the distribution of tenofovir plasma trough concentration according to each single companion ARV. As shown in Figure 2, patients concomitantly treated with atazanavir/r or lopinavir/r had significantly higher tenofovir concentrations compared with amprenavir/r or darunavir/r (163 ± 145 or 164 ± 120 vs. 112 ± 96 or 107 ± 68 ng/mL, respectively; P < 0.05 for all paired comparisons).
Risk of TDF Discontinuation
The last available follow-up of the enrolled patients was recorded at a mean of 1149 ± 3537 days after starting TDF therapy and 350 ± 534 days after the TDM assessment with great differences on the time to TDF therapy between ELV/COBI and other ARV groups (Fig. 3). Overall, a total of 149 cases of TDF discontinuations were recorded during this period. Such events were recorded in 75/212 (35%), 41/176 (23%), 13/46 (28%), and 20/76 (26%) patients given TDF in combination with PIs/r, NNRTIs, INSTIs, and ELV/COBI, respectively. As shown in Figure 3, the Kaplan–Meier survival analysis revealed a significant difference between ELV/COBI and other ARV regimens (log-rank P = 0.0002), which resulted particularly evident in the first 1 year after starting treatment with a number of events of TDF discontinuation resulting 3-fold higher compared with PIs/r, NNRTIs, and INSTIs. In particular, the probability to develop the event was 0.156, 0.131, 0.106, and 0.436 for PIs/r, NNRTIs, INSTIs, and ELV/COBI, respectively.
Results of a multivariable Cox regression analysis assessing the time fixed factors at baseline associated with the risk to experience TDF discontinuation are given in Table 3. In this model, ELV/COBI concomitant therapy and tenofovir plasma trough concentrations were both associated with a significantly higher risk to develop TDF discontinuation (ELV/COBI: hazard ratio = 2.284; tenofovir concentrations: hazard ratio = 1.021 per 10 ng/mL increment).
In the first part of the study, we confirmed the association of demographic, laboratory, and clinical covariates with increased tenofovir concentrations. Indeed, in agreement with previous findings, we demonstrated that female sex, aging, and low patients' body weight were all significantly and independently associated with high tenofovir plasma trough concentrations.13,15,19 Disappointingly, despite these evidences, HIV-infected adult patients are still treated with the same dose of 300 mg once daily, irrespectively of their sex, age, and body weight. The clinical implications of this “1 size (dose) fits all” approach have been clearly demonstrated in 2 large cohort studies involving Japanese patients. Thus, in the first study, it was shown that the risk to experience TDF-associated renal toxicity increased by nearly 20% per each 5 kg decrements in the patients' body weight,20 whereas in the second one, it was documented that the incidence of TDF-related renal dysfunction was twice as high as in patients with low body weight compared those treated with abacavir.21 On the same line, we have provided evidence that HIV-infected women with low body weight (≤50 kg) treated with TDF at 300 mg daily have a 1.8-fold hazard ratio to experience drug complications compared with those with body weight >50 kg.15
Also the concomitant ARVs can greatly impact on tenofovir exposure. A significant effect of PIs on tenofovir plasma concentrations was described previously.8 This trend was confirmed also in our cohort study: tenofovir plasma trough concentrations measured in patients given PIs/r were significantly and numerically higher compared with those given INSTIs or NNRTIs, respectively. Moreover, we extended previous finding by documenting for the first time in HIV-infected patients that coadministration with COBI—given coformulated with ELV (at the time of the study conduction COBI was available on the market only coformulated with ELV)—resulted in significantly higher tenofovir plasma trough concentrations compared with all other antiretroviral therapy regimens including also ritonavir-boosted PIs. These results cannot be ascribed to different effects of ARVs on tenofovir metabolism because this drug is not metabolized to a great extent by phase I or phase II metabolic enzymes.22 Rather, it is likely that the boosting effect of COBI (and, to a lesser extent of ritonavir) on tenofovir exposure may be driven by an inhibition on p-glycoprotein, an efflux pump that normally limits the intestinal absorption of tenofovir.6,23 Accordingly, booster-induced inhibition of p-glycoprotein results in increased tenofovir absorption and systemic exposure, as confirmed by our results.
In the second part of the study, we investigated the potential implications of a prolonged overexposure to tenofovir in patients concomitantly treated with COBI or ritonavir as compared with other ARV regimens. To address this issue, we looked at the number of events of TDF discontinuations happened in the different ARV regimens up to the last available visit. Despite the patients given COBI had a lower follow-up, because of the more recent introduction on the market of the ELV/COBI/TDF/FTC coformulation, they experienced overall a similar number of events compared with PIs, NNRTIs, and INSTIs. As consequence, patients concomitantly treated with ELV/COBI had a nearly 3-fold higher rate to experience TDF discontinuation in the first year of therapy compared with other ARV regimens, whereas no differences were detected when comparing PIs with NNRTIs or INSTIs.
Beside COBI, the only factor resulting significantly associated with TDF discontinuation was tenofovir plasma trough concentrations. This was not an unexpected finding considering the available literature reporting associations between tenofovir overexposure and the risk to experience TDF toxicity. Therefore, if (1) concomitant administration of COBI (and to a lesser extent of ritonavir) greatly increases tenofovir plasma trough concentrations and (2) patients experiencing TDF-related toxicity have high plasma tenofovir exposure, it can be reasonably anticipated that patients concomitantly treated with COBI might have the highest risk to experience TDF toxicity leading to drug discontinuation. This is indirectly supported also by recent findings from Costarelli et al.24 Thus, by analyzing data from the ICONA Foundation Study Cohort, they found that coadministration of TDF with ritonavir-boosted PIs was associated with an increased risk of TDF discontinuation compared with NNRTI. Although they do not measure tenofovir concentrations and do not include in their analysis data on COBI-treated patients, they equally gave the proof that coadministration of the standard dose of TDF (300 mg once daily) with boosting agents greatly impact on drug tolerability and durability.
The study, given its retrospective, observational design is not without limitations. The most important pitfall of our research is represented by the lack of detailed information on the reasons leading the attending physicians to discontinue TDF in some of their patients. The fact that TDF discontinuations in the ELV/COBI group were found early after initiation of therapy (within the first year) seems to speak against chronic TDF toxicity. Moreover, significant differences in the incidence of TDF discontinuations were observed in patients treated with ELV/COBI vs. those given boosted PIs, despite the same range of tenofovir concentrations. Taken together, these evidences suggest that episodes of TDF discontinuations in the COBI arms might have been overestimated by other reasons beside frank drug toxicity, such as COBI-related increase in serum creatinine concentrations, high tenofovir trough concentrations, or by the availability of less expensive novel single tablet regimens (such as dolutegravir-based formulations). However, as potential strength, this study depicted what is likely to have happened in real life settings.
Therefore, we are confident that, taken together, our findings add further evidence that the dose of TDF should be reduced when combined with COBI-based regimens.6,12 This concept is important not only when considering the tolerability of TDF per se, but also when comparing it with that of TAF. Actually, the dose of TAF, but not of TDF, is reduced from 25 to 10 mg daily when given with ritonavir or COBI, to compensate for their boosting effects.6 Accordingly, it cannot be excluded that the lack of proper dose adjustment for TDF when given with COBI could have biased the safety results of TAF and TDF during registrative trials.
As final remark, we believe that our results may be relevant also from an economical viewpoint, mainly in countries with economic constraints, considering the next arrival on the market of generic TDF formulations. Indeed, as previously performed in the ENCORE1 trial showing the noninferiority and less toxicity of a reduced 400 mg dose of efavirenz vs. the standard 600 mg dose,25 if the non inferiority of a reduced dose of generic TDF (ie, 200 mg) vs. standard dose of TDF (300 mg) or TAF (10 mg) all given with COBI will be demonstrated, we could reduce the cost of toxicity of TDF-based therapies while maintaining at the same time the optimal safety profile observed with TAF.
1. Ray AS, Fordyce MW, Hitchcock MJ. Tenofovir
alafenamide: a novel prodrug of tenofovir
for the treatment of human immunodeficiency virus. Antiviral Res. 2016;125:63–70.
2. Antela A, Aguiar C, Compston J, et al. The role of tenofovir
alafenamide in future HIV management. HIV Med. 2016;17(suppl 2):4–16.
3. De Clercq E. Tenofovir
alafenamide (TAF) as the successor of tenofovir
disoproxil fumarate (TDF). Biochem Pharmacol. 2016;119:1–7.
4. Wang H, Lu X, Yang X, et al. The efficacy and safety of tenofovir
alafenamide versus tenofovir
disoproxil fumarate in antiretroviral regimens for HIV-1 therapy: meta-analysis. Medicine (Baltimore). 2016;95:e5146.
5. Post FA, Tebas P, Clarke A, et al. Switching to tenofovir
alafenamide, coformulated with elvitegravir, cobicistat, and emtricitabine, in HIV-infected adults with renal impairment: 96-Week results from a single-arm, multicenter open-label phase 3 study. J Acquir Immune Defic Syndr. 2016;74:180–184.
6. Vitoria M, Hill AM, Ford NP, et al. Choice of antiretroviral drugs for continued treatment scale-up in a public health approach: what more do we need to know? J Int AIDS Soc. 2016;19:20504.
7. Zack J, Doyle E, Graham H, et al. Bioequivalence of a fixed dose formulation of rilpivirine/emtricitabine/tenofovir
alafenamide and rilpivirine. Paper presented at: 15th European Aids Conference; October 2015; Barcelona, Spain. Abstract PE10/9.
8. Calcagno A, Gonzalez de Requena D, Simiele M, et al. Tenofovir
plasma concentrations according to companion drugs: a cross-sectional study of HIV-positive patients with normal renal function. Antimicrob Agents Chemother. 2013;57:1840–1843.
9. Kearney BP, Mathias A, Mittan A, et al. Pharmacokinetics and safety of tenofovir
disoproxil fumarate on coadministration with lopinavir/ritonavir. J Acquir Immune Defic Syndr. 2006;43:278–283.
10. Hoetelmans RM, Mariën K, De Pauw M, et al. Pharmacokinetic interaction between TMC114/ritonavir and tenofovir
disoproxil fumarate in healthy volunteers. Br J Clin Pharmacol. 2007;64:655–661.
11. Custodio J, Garner W, Jin F, et al. Evaluation of the drug interaction potential between the pharmacokinetic enhancer and tenofovir
disoproxil fumarate in healthy subjects. Paper presented at: 14th International Workshop on Clincal Pharmacology of HIV Therapy; April 22–24, 2013; Amsterdam, the Netherlands. Abstract O_07.
12. Hill A, Khoo S, Back D, et al. Should the dose of tenofovir
be reduced to 200–250 mg/day, when combined with protease inhibitors? J Int AIDS Soc. 2014;17(4 suppl 3):19583.
13. Rodríguez-Nóvoa S, Labarga P, D'avolio A, et al. Impairment in kidney tubular function in patients receiving tenofovir
is associated with higher tenofovir
plasma concentrations. AIDS. 2010;24:1064–1066.
14. Scherzer R, Estrella M, Li Y, et al. Association of tenofovir
exposure with kidney disease risk in HIV infection. AIDS. 2012;26:867–875.
15. Gervasoni C, Meraviglia P, Landonio S, et al. Low body weight in females is a risk factor for increased tenofovir
exposure and drug-related adverse events. PLoS One. 2013;8:e80242.
16. Baxi SM, Scherzer R, Greenblatt RM, et al. Women's interagency HIV study (WIHS). Higher tenofovir
exposure is associated with longitudinal declines in kidney function in women living with HIV. AIDS. 2016;30:609–618.
17. Kunimoto Y, Ikeda H, Fujii S, et al. Plasma tenofovir
trough concentrations are associated with renal dysfunction in Japanese patients with HIV infection: a retrospective cohort study. J Pharm Health Care Sci. 2016;2:22.
18. D'Avolio A, Sciandra M, Siccardi M, et al. A new assay based on solid-phase extraction procedure with LC-MS to measure plasmatic concentrations of tenofovir
and emtricitabine in HIV infected patients. J Chromatogr Sci. 2008;46:524–528.
19. Baxi SM, Greenblatt RM, Bacchetti P, et al. Common clinical conditions—age, low BMI, ritonavir use, mild renal impairment—affect tenofovir
pharmacokinetics in a large cohort of HIV-infected women. AIDS. 2014;28:59–66.
20. Nishijima T, Komatsu H, Gatanaga H, et al. Impact of small body weight on tenofovir
-associated renal dysfunction in HIV-infected patients: a retrospective cohort study of Japanese patients. PLoS One. 2011;6:e22651.
21. Nishijima T, Gatanaga H, Komatsu H, et al. Renal function declines more in tenofovir
- than abacavir-based antiretroviral therapy in low-body weight treatment-naive patients with HIV infection. PLoS One. 2012;7:e29977.
22. Kearney BP, Flaherty JF, Shah J. Tenofovir
disoproxil fumarate: clinical pharmacology and pharmacokinetics. Clin Pharmacokinet. 2004;43:595–612.
23. Lepist EI, Phan TK, Roy A, et al. Cobicistat boosts the intestinal absorption of transport substrates, including HIV protease inhibitors and GS-7340, in vitro. Antimicrob Agents Chemother. 2012;56:5409–5413.
24. Costarelli S, Cozzi-Lepri A, Lapadula G, et al. Long-term durability of tenofovir
-based antiretroviral therapy in relation to the co-administration of other drug classes in routine clinical practice. PLoS One. 2016;11:e0160761.
25. Puls R, Amin J, Losso M, et al. Efficacy of 400 mg efavirenz versus standard 600 mg dose in HIV-infected, antiretroviral-naive adults (ENCORE1): a randomised, double-blind, placebo-controlled, non-inferiority trial. Lancet. 2014;383:1474–1482.