To the Editor:
It is well known that the nonnucleoside reverse transcriptase inhibitors nevirapine (NVP) and efavirenz are able to decrease the plasma levels of HIV-1 protease inhibitors (PIs) when the latter drugs are given as single PIs (i.e., not boosted by a low dose of ritonavir). Decreases in plasma trough levels and area under the plasma concentration vs. time curve in the range of 25–50% have been observed. (See Piscitelli and Gallicano 1 for a summary of these data).
Data on the effect of NVP and efavirenz on the plasma levels of ritonavir-boosted PIs are scarce. While the combination of efavirenz and indinavir/ritonavir, 800/100 mg BID, has been tested in a clinical drug-drug interaction study, 2 no such formal interaction study has been conducted to evaluate the effect of NVP on this indinavir/ritonavir combination. Recently, it was reported that addition of NVP does not significantly change indinavir and ritonavir trough levels. 3 These data were not confirmed by observations in our therapeutic drug monitoring (TDM) program of HIV-infected patients in Dutch hospitals. We set out to conduct a study on the effect of NVP on the pharmacokinetics of indinavir/ritonavir, 800/100 mg BID.
From the database of samples measured in the ongoing TDM program in the Netherlands, all records were extracted of patients who were using indinavir/ritonavir in a dose of 800/100 mg BID. Samples were regarded as real trough samples if the time between the last intake of medication and blood sampling was between 10–14 hours. If ≥1 trough sample was collected from a patient, only the first sample was included in this analysis. Based on the co-medication that was listed on the TDM application forms, 2 groups were constructed: the NVP group, including all patients who were concomitantly using NVP (200 mg BID), and the control group, including all patients who were not using NVP or any other interfering co-medication. Geometric means (GMs) and 95% CIs of indinavir and ritonavir troughs were calculated and tested for difference by an unpaired t test.
Twenty-one patients were identified in the NVP group and 139 in the control group. The GM (+ 95% CI) of the indinavir trough level in the NVP group was 0.49 mg/L (0.35–0.68 mg/L) vs. 1.14 mg/L (1.00–1.29 mg/L) in the control group (57% difference, P < 0.001). A similar effect of NVP on ritonavir troughs was observed: the GM in the NVP group was 0.22 mg/L (0.15–0.31 mg/L) vs. 0.52 mg/L in the control group (0.46–0.60 mg/L) (58% difference, P < 0.001). No patient in the NVP group or the control group had an indinavir trough level <0.10 mg/L, the proposed minimum effective level for indinavir in treatment-naive patients. 4
Thus, our data are in conflict with those of Moreno et al., 3 who could not find a significant effect of the addition of NVP (or efavirenz) on indinavir and ritonavir troughs. Possible explanations are the small number of NVP users and the different doses of indinavir and ritonavir that were used in their study. Furthermore, the effect of efavirenz and NVP was combined in their analyses, making it difficult to estimate the exact effect of NVP separately from efavirenz. The samples in the study of Moreno et al. were regarded as troughs, but actual time between blood sampling and drug intake was not recorded. Finally, our data are in line with published data of the effect of NVP and efavirenz on ritonavir-boosted regimens containing either indinavir, 2 lopinavir, 5 and amprenavir. 6
Evidence exists now that, in contrast to the effect of NVP and efavirenz on the pharmacokinetics of high-dose (i.e., 400–600 mg BID) ritonavir, these agents are able to decrease the plasma levels of low-dose ritonavir (100 mg BID). Therefore, the drug interaction between NVP and efavirenz vs. boosted PIs may be caused either by an effect directly on the boosted PI or indirectly through the effect on the pharmacokinetic enhancer ritonavir.
The clinical relevance of this interaction remains to be determined. In our analyses (and also in the efavirenz/indinavir/ritonavir interaction study 2) all subjects had indinavir troughs >0.10 mg/L, the proposed minimum effect trough level of indinavir. 4 Therefore, in indinavir-naive patients treated with a combination of NVP (or efavirenz) and indinavir/ritonavir 800/100 mg BID, there appears to be no need to modify the dose. In treatment-experienced patients, when virus with decreased susceptibility to indinavir may be present, it may be wise to increase either the indinavir dose (to 1000 mg) or the ritonavir dose (to 200 mg) to fully benefit from the boosting effect of ritonavir.
David M. Burger, PharmD, PhD*
Jan M. Prins, MD, PhD†
Marchina E. van der Ende, MD, PhD‡
Rob E. Aarnoutse, PharmD, PhD*
1. Piscitelli SC, Gallicano KD. Interactions among drugs for HIV and opportunistic infections. N Engl J Med. 2001; 344:984–996.
2. Aarnoutse RE, Grintjes KJ, Telgt DS, et al. The influence of efavirenz on the pharmacokinetics of a twice-daily combination of indinavir and low-dose ritonavir in healthy volunteers. Clin Pharmacol Ther. 2002; 71:57–67.
3. Moreno A, Casado JL, Marti-Belda P, et al. Concomitant use of non nucleoside reverse transcriptase inhibitors (NNRTI) does not decrease the inhibitory quotient of dual ritonavir/indinavir-based therapy. Paper presented at: 41th Interscience Conference on Antimicrobial Agents and Chemotherapy; December 14–18, 2001; Chicago. Abstract 1728.
4. Acosta EP, Kakuda TN, Brundage RC, et al. Pharmacodynamics of human immunodeficiency virus type 1 protease inhibitors. Clin Infect Dis. 2000; 30(suppl 2):S151–S159.
5. Product information Kaletra. Chicago: Abbott; 2000.
6. Duval X, Le MV, Longuet C, et al. Efavirenz-induced decrease in plasma amprenavir levels in human immunodeficiency virus-infected patients and correction by ritonavir. Antimicrob Agents Chemother. 2000; 44:2593.