Once-Daily Saquinavir-sgc Plus Low-Dose Ritonavir
Once-Daily Saquinavir-sgc Plus Low-Dose Ritonavir
To the Editor: The use of nucleoside reverse-transcriptase inhibitor (NRTI)-sparing HAART is becoming increasingly common due to severe adverse events or extensive resistance to analogs. However, scanty information about combinations of protease inhibitors (PIs) and nonnucleoside reverse transcriptase inhibitors (NNRTIs) is currently available, and some of them present the possibility of significant pharmacokinetic interactions. One of these nonrecommended combinations is saquinavir (SQV) plus efavirenz (EFV), because EFV induces cytochrome P450 isoenzymes, decreasing SQV plasma levels. However, our hypothesis was that the concomitant administration of ritonavir (RTV) might reverse the effect EFV has on SQV metabolism induction. In this study, we evaluated the pharmacokinetics, safety, and efficacy of a once-daily regimen of SQV-sgc plus low-dose RTV (1200/100 mg) in combination with EFV (600 mg) in HIV-infected patients with prior antiretroviral therapy.
From February to December 2000, we recruited for this open-label prospective study every adult HIV-1-infected outpatient seen at the Infectious Diseases Service of our Hospital who had switched from a regimen of 2 NRTIs plus PIs or NNRTIs to the above-mentioned combination because of adverse effects attributable to NRTIs (42 patients; 32 male, mean age 41 years). There were no entry restrictions on plasma HIV-1 RNA viral load (VL), CD4 cell count, or previous antiretroviral treatments, so that some patients had an undetectable VL whereas others were in virologic failure. The median CD4 cell count and VL were 578/µL (range, 121-1133/µL), and <50 copies/mL (range, <50-696,000 copies/mL; <50 copies/mL: 22 patients), respectively. The median duration of prior antiretroviral therapy was 54 months. Every patient was PI-experienced (median: 30 months), and 22 of them had had at least one virologic failure while taking ≥1 PI. Furthermore, 28 patients (66%) were NNRTI-experienced (nevirapine: 8, EFV: 20), and 9 patients had a detectable VL (range, 146-764 copies/mL; 6 on EFV, and 3 on nevirapine) when treatment was switched.
Blood samples for plasma SQV and EFV concentrations were obtained from 35 unselected patients, including those 7 patients who experienced virologic failure during the follow-up. In 19 of them a 24-hour pharmacokinetic profile was obtained, and a single trough level in 16. SQV and EVF were quantitated by validated HPLC procedures. The lower limits of quantification were 0.04 for SQV and 0.1 µg/mL for EFV. The mean intra- and inter-assay coefficients of variation were 3.1% and 5% for SQV, and 3.2% and 4.7% for EFV, respectively. The median plasma concentration and interquartile ranges (IQRs) of SQV are shown in Figure 1. Pharmacokinetic parameters of SQV and EFV calculated from the concentration versus time curves by using noncompartmental methods are shown in Table 1 .
(Enlarge Image)
Mean and median (± interquartile range) saquinavir plasma concentration versus time curves after administration of 1200 mg saquinavir-sgc in combination with 100 mg ritonavir and 600 mg efavirenz once daily.
Efficacy data were analyzed by intent-to-treat, defining treatment failure as a virologic failure (a confirmed VL >50 copies/mL) or treatment discontinuation for whatever reason. Patients were assessed after 4 weeks of treatment, and every 3 months thereafter. The median follow-up was 15 months (range, 1 to 21 months). The number of CD4 cells was measured by flow cytometry, and plasma VL was measured by polymerase chain reaction method (lower limit of detection, 50 copies/mL (Amplicor HIV Monitor, Roche Diagnostic Systems) at each visit. The relationships between virologic failure and different variables were assessed by the
test for qualitative variables and by the Spearman rank correlation coefficients for quantitative variables.
After 52 weeks, 30 patients (71%) were still on treatment with a VL <50 copies/mL, and an additional 5% (2 patients) with VL of 71 and 460 copies/mL, respectively. Five patients dropped out during follow-up even though they were maintaining an undetectable VL. One of the five, who was affected by chronic hepatitis B and C viruses, dropped out because of a flare of hepatitis at the 6th month, although he restarted his treatment with EFV with no further adverse effects after recovering from the acute episode. Another of the five dropped out because of tuberculosis during the 9th month of treatment. Three of the five stopped treatment on their own, although they continued on follow-up. Seven patients, all of them entering the study with a high VL, were considered virologic failures after 6 months on treatment (3 patients mantaining VL <500 copies/mL, 3 patients with VL >5000 copies/mL, and 1 patient with a viral rebound).
There was no correlation between SQV or EFV plasma levels and virologic failures. A relationship was observed between virologic failure and the following variables: previous virologic failure on ≥2 PIs (p < .000; RR: 64. CI95: 5.6-723), previous virologic failure on NNRTIs (p = 0.005; RR: 15. CI95: 2.2-99), and entering the study with a detectable VL (p = 0.003; RR: 2.7. CI95: 1.7-4.1). In fact, only 1 virologic failure (a patient with previous failures on SQV-hgc and on nevirapine) was observed among 33 patients who had previously failed with ≤1 PI, whereas 6 of 9 patients with previous virologic failure with ≥2 PIs failed on this regimen. The median increase in CD4 cells was 39 cells/mm at week 4; 44 cells/mm after 3 months; 97 cells/mm after 6 months; 147 cells/mm after 9 months, and 215 cells/mm after 1 year (IQR, 119-276 cells/mm).
The combination of SQV-sgc plus low dose RTV and EFV qd was well tolerated. The most commonly reported adverse effects included transient dizziness, lightheadedness, and drowsiness related to EFV administration, although no patient required treatment interruption because of them. No clinically relevant alterations in laboratory safety parameters were noted during the study, except for the patient with hepatitis mentioned above. The median values of fasting triglycerides and total cholesterol concentrations at inclusion and after one year were 169 mg/dL to 255 mg/dL, and 208 mg/dL to 243 mg/dL, respectively.
SQV-sgc is indicated for use in combination with other antiretroviral agents at a dose of 1200 mg three times daily. However, when it is given with low-dose RTV, higher SQV plasma levels are reached, and the trough plasma levels obtained would support a once-daily regimen. SQV-sgc doses of 1200 to 1800 together with 100 or 200 mg of RTV daily resulted in very similar trough plasma concentrations for all combinations. That is why we chose the 1200/100 mg daily dose, with the additional advantage of fewer pills a day.
As in other pharmacokinetic studies involving SQV and EFV, we have observed a large interpatient variability. The SQV levels were lower than those previously described with higher doses of SQV qd. A lower SQV-sgc dose and the presence of EFV could explain these differences. On the other hand, EFV plasma levels were higher than those previously reported in HAART regimens without RTV (data on file; DuPont Pharmaceuticals, USA); however, we have observed an EFV t½
similar to that seen in patients without PIs (7), suggesting that the effect RTV has on EFV is limited to its absorption and first-pass metabolism.
The data of efficacy of this combination are very encouraging, in spite of the fact that many patients had had virologic failures on previous PIs and NNRTIs, suggesting that the plasma drug levels of both SQV and EFV obtained are adequate. The relation between virologic failure and entering the study with a detectable VL may reveal only that those patients had previously undergone more virologic failures on different PIs and NNRTIs rather than an actual correlation. In fact, it had been desirable to have a resistance test before initiating this combination in those patients who entered the study with a high VL. Unfortunately, baseline genotypes were only retrospectively available in two cases, and multiple mutations associated to PI resistance were present in both patients, including the L90M and G48V mutations. In summary, the pharmacokinetic, efficacy, and safety data of once-daily SQV-sgc/RTV (1200/100 mg) plus EFV 600 mg support the administration of such regimen in those patients without resistance to SQV or EFV in whom an NRTI-sparing regimen is considered. Futhermore, the once-daily administration would reduce dosing frequency and increase the long-term adherence.
Presented in part at the 9th Conference on Retroviruses and Opportunistic Infections, Seattle, Washington, U.S.A., February 2002.
To the Editor: The use of nucleoside reverse-transcriptase inhibitor (NRTI)-sparing HAART is becoming increasingly common due to severe adverse events or extensive resistance to analogs. However, scanty information about combinations of protease inhibitors (PIs) and nonnucleoside reverse transcriptase inhibitors (NNRTIs) is currently available, and some of them present the possibility of significant pharmacokinetic interactions. One of these nonrecommended combinations is saquinavir (SQV) plus efavirenz (EFV), because EFV induces cytochrome P450 isoenzymes, decreasing SQV plasma levels. However, our hypothesis was that the concomitant administration of ritonavir (RTV) might reverse the effect EFV has on SQV metabolism induction. In this study, we evaluated the pharmacokinetics, safety, and efficacy of a once-daily regimen of SQV-sgc plus low-dose RTV (1200/100 mg) in combination with EFV (600 mg) in HIV-infected patients with prior antiretroviral therapy.
From February to December 2000, we recruited for this open-label prospective study every adult HIV-1-infected outpatient seen at the Infectious Diseases Service of our Hospital who had switched from a regimen of 2 NRTIs plus PIs or NNRTIs to the above-mentioned combination because of adverse effects attributable to NRTIs (42 patients; 32 male, mean age 41 years). There were no entry restrictions on plasma HIV-1 RNA viral load (VL), CD4 cell count, or previous antiretroviral treatments, so that some patients had an undetectable VL whereas others were in virologic failure. The median CD4 cell count and VL were 578/µL (range, 121-1133/µL), and <50 copies/mL (range, <50-696,000 copies/mL; <50 copies/mL: 22 patients), respectively. The median duration of prior antiretroviral therapy was 54 months. Every patient was PI-experienced (median: 30 months), and 22 of them had had at least one virologic failure while taking ≥1 PI. Furthermore, 28 patients (66%) were NNRTI-experienced (nevirapine: 8, EFV: 20), and 9 patients had a detectable VL (range, 146-764 copies/mL; 6 on EFV, and 3 on nevirapine) when treatment was switched.
Blood samples for plasma SQV and EFV concentrations were obtained from 35 unselected patients, including those 7 patients who experienced virologic failure during the follow-up. In 19 of them a 24-hour pharmacokinetic profile was obtained, and a single trough level in 16. SQV and EVF were quantitated by validated HPLC procedures. The lower limits of quantification were 0.04 for SQV and 0.1 µg/mL for EFV. The mean intra- and inter-assay coefficients of variation were 3.1% and 5% for SQV, and 3.2% and 4.7% for EFV, respectively. The median plasma concentration and interquartile ranges (IQRs) of SQV are shown in Figure 1. Pharmacokinetic parameters of SQV and EFV calculated from the concentration versus time curves by using noncompartmental methods are shown in Table 1 .
(Enlarge Image)
Mean and median (± interquartile range) saquinavir plasma concentration versus time curves after administration of 1200 mg saquinavir-sgc in combination with 100 mg ritonavir and 600 mg efavirenz once daily.
Efficacy data were analyzed by intent-to-treat, defining treatment failure as a virologic failure (a confirmed VL >50 copies/mL) or treatment discontinuation for whatever reason. Patients were assessed after 4 weeks of treatment, and every 3 months thereafter. The median follow-up was 15 months (range, 1 to 21 months). The number of CD4 cells was measured by flow cytometry, and plasma VL was measured by polymerase chain reaction method (lower limit of detection, 50 copies/mL (Amplicor HIV Monitor, Roche Diagnostic Systems) at each visit. The relationships between virologic failure and different variables were assessed by the
test for qualitative variables and by the Spearman rank correlation coefficients for quantitative variables.
After 52 weeks, 30 patients (71%) were still on treatment with a VL <50 copies/mL, and an additional 5% (2 patients) with VL of 71 and 460 copies/mL, respectively. Five patients dropped out during follow-up even though they were maintaining an undetectable VL. One of the five, who was affected by chronic hepatitis B and C viruses, dropped out because of a flare of hepatitis at the 6th month, although he restarted his treatment with EFV with no further adverse effects after recovering from the acute episode. Another of the five dropped out because of tuberculosis during the 9th month of treatment. Three of the five stopped treatment on their own, although they continued on follow-up. Seven patients, all of them entering the study with a high VL, were considered virologic failures after 6 months on treatment (3 patients mantaining VL <500 copies/mL, 3 patients with VL >5000 copies/mL, and 1 patient with a viral rebound).
There was no correlation between SQV or EFV plasma levels and virologic failures. A relationship was observed between virologic failure and the following variables: previous virologic failure on ≥2 PIs (p < .000; RR: 64. CI95: 5.6-723), previous virologic failure on NNRTIs (p = 0.005; RR: 15. CI95: 2.2-99), and entering the study with a detectable VL (p = 0.003; RR: 2.7. CI95: 1.7-4.1). In fact, only 1 virologic failure (a patient with previous failures on SQV-hgc and on nevirapine) was observed among 33 patients who had previously failed with ≤1 PI, whereas 6 of 9 patients with previous virologic failure with ≥2 PIs failed on this regimen. The median increase in CD4 cells was 39 cells/mm at week 4; 44 cells/mm after 3 months; 97 cells/mm after 6 months; 147 cells/mm after 9 months, and 215 cells/mm after 1 year (IQR, 119-276 cells/mm).
The combination of SQV-sgc plus low dose RTV and EFV qd was well tolerated. The most commonly reported adverse effects included transient dizziness, lightheadedness, and drowsiness related to EFV administration, although no patient required treatment interruption because of them. No clinically relevant alterations in laboratory safety parameters were noted during the study, except for the patient with hepatitis mentioned above. The median values of fasting triglycerides and total cholesterol concentrations at inclusion and after one year were 169 mg/dL to 255 mg/dL, and 208 mg/dL to 243 mg/dL, respectively.
SQV-sgc is indicated for use in combination with other antiretroviral agents at a dose of 1200 mg three times daily. However, when it is given with low-dose RTV, higher SQV plasma levels are reached, and the trough plasma levels obtained would support a once-daily regimen. SQV-sgc doses of 1200 to 1800 together with 100 or 200 mg of RTV daily resulted in very similar trough plasma concentrations for all combinations. That is why we chose the 1200/100 mg daily dose, with the additional advantage of fewer pills a day.
As in other pharmacokinetic studies involving SQV and EFV, we have observed a large interpatient variability. The SQV levels were lower than those previously described with higher doses of SQV qd. A lower SQV-sgc dose and the presence of EFV could explain these differences. On the other hand, EFV plasma levels were higher than those previously reported in HAART regimens without RTV (data on file; DuPont Pharmaceuticals, USA); however, we have observed an EFV t½
similar to that seen in patients without PIs (7), suggesting that the effect RTV has on EFV is limited to its absorption and first-pass metabolism.
The data of efficacy of this combination are very encouraging, in spite of the fact that many patients had had virologic failures on previous PIs and NNRTIs, suggesting that the plasma drug levels of both SQV and EFV obtained are adequate. The relation between virologic failure and entering the study with a detectable VL may reveal only that those patients had previously undergone more virologic failures on different PIs and NNRTIs rather than an actual correlation. In fact, it had been desirable to have a resistance test before initiating this combination in those patients who entered the study with a high VL. Unfortunately, baseline genotypes were only retrospectively available in two cases, and multiple mutations associated to PI resistance were present in both patients, including the L90M and G48V mutations. In summary, the pharmacokinetic, efficacy, and safety data of once-daily SQV-sgc/RTV (1200/100 mg) plus EFV 600 mg support the administration of such regimen in those patients without resistance to SQV or EFV in whom an NRTI-sparing regimen is considered. Futhermore, the once-daily administration would reduce dosing frequency and increase the long-term adherence.
Presented in part at the 9th Conference on Retroviruses and Opportunistic Infections, Seattle, Washington, U.S.A., February 2002.