Head of HIV Infection
Infectious Diseases Division
Hospital Universitario Vall d’Hebr�n
Twenty-five years have passed since the onset of the AIDS epidemic, and considerable advances have been made in HIV treatment. The prognosis of HIV infection changed radically after 1996 with the development of protease inhibitors (PIs). These drugs, when used in combination with two nucleoside analogue reverse transcriptase inhibitors (NRTIs), are referred to as highly effective antiretroviral treatment (HAART). The introduction of HAART resulted in a considerable reduction in mortality and incidence of opportunistic diseases, and converted HIV infection into a chronic disease instead of a necessarily progressive, irreversible and fatal disease. However, there were several problems associated to chronic therapy, such as frequent adverse events and complexity of therapeutic regimens, which required administration of a large number of pills several times a day, and often with dietary restrictions.(1,2) Thus, adherence to treatment was difficult, drug efficacy was far from optimum and the patient’s quality of life was notably compromised.(3,4)
Since then, physicians’ and patients’ concern has been the availability of well-tolerated treatment regimens with convenient dosing schedules that can be adapted to the patient’s lifestyle. Intensive research by the main pharmaceutical laboratories has led to the development of new drugs, such as non-nucleoside analogues (NNRTIs) and new NRTIs (such as tenofovir, abacavir and emtricitabine), as well as improvements in the galenic formulation of older drugs, which has resulted in the notable simplification and improved tolerability of treatment for HIV infection in recent years.(1,5)
The main cause of therapeutic failure is poor compliance with treatment. Even small deviations from treatment adherence (5�20% of dose loss) quickly increase the risk of failure.(3) It has also been observed that, among patients who follow treatment adequately at first, adherence decreases over time and this has a negative impact on the durability of therapy.(3) Current treatments are powerful, and efficacy is high when optimal adherence (>95%) is achieved. Available treatments now involve a much smaller number of pills, dietary restrictions have notably decreased and more drugs can be administered in a single daily dose. Taking a simple treatment once a day (QD) is, in itself, an added value that can have positive repercussions on adherence, since it is easily adapted to the lifestyle of the majority of patients and improves their quality of life.(6,7)
In the past five years, administration of �antiretroviral treatment (ART) in a QD regimen has generated extraordinary interest in the scientific community and among patients receiving this treatment.(6,8-15)
Antiretroviral drugs that can be administered QD
In recent years there has been a substantial increase in the number of drugs available for QD administration. There are currently 13 drugs or coformulated drug combinations approved for QD administration. Efavirenz (EFV) was the first agent approved for use on a QD basis since the time of its introduction. Initially, the 600mg dose required three 200mg capsules per day, but now 600mg tablets are available.(16,17) Didanosine (ddI), which was initially administered BID, has undergone notable galenic modifications and is now given as a single capsule daily with much better tolerance than with the previous buffered tablets.(18-20) Tenofovir (TDF), the only available nucleotide analogue, has been marketed for administration as a single-daily tablet since its development.(21,22) Lamivudine (3TC) was contemplated for administration at a dose of 150mg BID, but later studies proved the equivalence of the 300mg QD dose, which is now available in 300mg tablets.(23-25) Abacavir (ABC) was initially commercialised for administration at a dose of 300mg BID; subsequent studies on intracellular pharmacokinetics, efficacy and safety demonstrated the equivalence of a 600mg dose QD (two tablets).(26,27) Stavudine CLP (d4T(CLP)) has been approved for QD administration; nevertheless, because of technical difficulties related to the manufacturing of the capsules, it is not available and may never be marketed.(28) Emtricitabine (FTC) was designed and commercialised for QD
administration.(29-33) Recently, two NRTI coformulations have appeared on the market, TDF/FTC (Truvada(R))(34) and ABC/3TC (Kivexa(R)),(35) which contain the complete daily dose of both drugs in a single tablet.
Atazanavir (ATV) is the only PI exclusively approved for QD administration at a dose of 400mg, which is given in two capsules per day.(36-41) In Europe, ATV use is only authorised in association with ritonavir (RTV) at doses of 300+100mg, a combination that presents pharmacokinetic advantages relative to 400mg/day of unboosted ATV. This benefit is particularly important when the drug is prescribed for patients with decreased susceptibility to PIs or when the drug is associated with other agents such as TDF or NNRTI, which can decrease ATV plasma concentrations due to pharmacokinetic interactions. In cases of substantial drug interactions, a dose of 400+100mg can provide a better pharmacokinetic profile. Amprenavir (APV) boosted with RTV was the first PI approved for QD and BID administration, although a large number of pills must be taken (10 capsules per day), which diminishes one of the potential advantages of QD treatment.(42-44) Fosamprenavir (FPV) boosted with RTV has been authorised by the FDA for QD administration in therapy-naive patients and requires four capsules per day (two 700mg tablets of FPV+ two 100mg capsules of RTV); QD use of this drug has not been approved by the EMEA.(45,46) Lopinavir/ritonavir (LPV/RTV) was recently authorised at a dose of 800/200mg (six capsules)(47-50) QD. In naive patients the efficacy of this agent in QD or BID regimens is similar, but there is a higher incidence of diarrhoea, lower LPV trough concentrations and greater variation in LPV minimum concentrations with QD use, factors that could be associated with a higher risk of failure in the case of ART-experienced patients. A new LPV/RTV formulation, which will soon be available in tablet form, improves the characteristics of the drug regarding simplicity (four tablets), tolerability and variability of plasma concentrations and increases the possibility of QD administration.
Nevirapine (NVP) is not as yet approved for QD administration, but pharmacokinetic studies and randomised and nonrandomised clinical studies have demonstrated similar efficacy with QD or BID use, thereby endorsing this possibility.(51-54)
The new 500mg formulation of saquinavir (SQV), recently placed on the market, improves the possibility of administration on a QD basis. Although SQV/RTV QD is still not authorised, various pharmacokinetic and clinical studies support its use.(55-62) The SQV/RTV dose most often applied in QD regimens is 1,600/100mg, but promising results have been obtained even at a dose of 1,200/100mg. This allows us to predict that, with the new formulation, a dose of 1,500/100mg (three + one tablets) may be adequate in QD regimens for patients who do not have decreased susceptibility to the drug. The dose of 2,000/100mg provides higher plasma SQV concentrations than those obtained with 1,600/100mg; thus this dose may be recommended for many patients under consideration to receive the SQV/RTV QD option.(63,64) Prospective studies are needed to determine what dose of the new SQV formulation is most suitable for QD administration.
Classically, the use of PIs in QD regimens has had the drawback of requiring an elevated number of tablets. This has meant that many patients prefer to divide it into two administrations, although the once-daily regimen can be useful for patients whose personal or social characteristics make QD treatment highly desirable. With the new PIs and the new formulations of older drugs, the viability of QD administration increases considerably, and it is possible to administer a total daily dose with only two (unboosted ATV), three (ATV/RTV) or four (SQV/RTV, FPV/RTV, LPV/RTV) tablets.
ART administered QD in various clinical scenarios
Patients starting ART (naive)
The choice of treatment for a naive patient is one of the most important interventions for the management of HIV infection. This is the time when therapy is most effective; hence, the decision can compromise subsequent treatments. The greatest accumulated experience with QD regimens is found in studies performed in patients who have not received prior therapy.
All the therapeutic guidelines recommend starting ART with two NRTIs plus one NNRTI or one PI. Nonetheless, there are some differences between guidelines with respect to the drugs considered to be of choice or alternatives for this purpose. The Spanish guidelines (GESIDA/PNS) consider the treatment of choice to be a combination of three drugs, one from each of the following groups:
* An NRTI, which could be zidovudine (AZT), ddI, ABC or TDF.
* An NRTI, which could be 3TC or FTC.
* An NNRTI or PI, which could be EFV, NVP or LPV/RTV.(1)
Although in theory it would be possible to combine any of the NRTIs from the first group with any one from the second, some combinations are much more logical than others because of their pharmaceutical formulations: AZT would preferentially combine with 3TC because a coformulation exists (Combivir(R)) that allows their administration in one tablet twice a day; there are no coformulations with ddI, hence it can be administered with 3TC or FTC (two tablets a day); ABC would be preferentially administered with 3TC because a coformulation (Kivexa(R)) is available that allows administration of the two drugs in one tablet a day; and TDF would preferentially be combined with FTC in the available coformulation (Truvada(R)) that allows administration in one tablet a day. The third drug of choice is usually considered to be EFV, particularly when focusing on QD treatments, although the Spanish guidelines also offer LPV/RTV and NVP within the preferred regimens, with some limitations.(1)
Thus, one BID regimen and three QD regimens are included within the combinations of two NRTIs of choice for starting treatment. Efficacy and safety studies have been performed that directly compare the BID regimen with each of the three QD regimens, in all cases associated with EFV. The randomised clinical trial, GESIDA3903, compares ddI, 3TC and EFV with AZT, 3TC and EFV. The enrolment period was completed with 392 patients, and efficacy and safety results should soon be available. The randomised, double-blind CNA30024 trial compared ABC, 3TC and EFV (n=324) with AZT, 3TC and EFV (n=325).(65) The study included patients with a baseline CD(4) count greater than 50 cells/mm(3), and the 48-week results show that the percentage of patients with a viral load <50 copies/ml by intent-to-treat (ITT) was comparable in the two groups (70% versus 69%; 95% CI of the difference �6.3�7.9%), demonstrating that ABC is not inferior to AZT. Moreover, the increase in CD(4) lymphocytes was significantly higher in the ABC group (209 versus 155 cells/mm(3), p=0.005). The results of the randomised GS-934 study comparing TDF, FTC and EFV (n=258) with AZT, 3TC and EFV (n=259) were recently published.(66) The 48-week outcome demonstrated that the percentage of patients with a viral load <50 copies/ml by ITT was significantly higher in the TDF+FTC group than in the AZT+3TC group (77% versus 68%, p=0.03); in addition, the increase in CD4 lymphocytes was also significantly higher with TDF+FTC (190 versus 158 cells/mm(3), p=0.002). The main reason for these differences was a higher frequency of treatment interruptions due to toxicity, essentially anaemia, in the AZT+3TC group.
There are no direct comparisons among any of the three QD combinations of choice in patients starting treatment. Eight studies performed with the ddI+3TC or FTC combination, associated with EFV in seven cases and with NVP in one, have been analysed.(67-74) The combination of ddI, 3TC and EFV was used in five studies involving a total of 377 treatment-naive HIV-infected patients. At 48 weeks of treatment, 71% (63-80%) of patients presented a viral load <50 copies/ml (ITT noncompleter equals to failure [NC=F]).(67-70,74) The combination of ddI, FTC and EFV was used in two studies including a total of 326 naive patients and at 48 weeks of treatment, 79% (78-83%) of patients had a viral load of <50 copies/ml (ITT NC=F).(71,72) In one noncomparative study, the efficacy of a QD regimen with ddI, 3TC and NVP was investigated in 70 naive patients; at 48 weeks, 69% of the patients presented a viral load <50 copies/ml (ITT NC=F). Among the 56 patients with a baseline CD(4) lymphocyte count >100 cells/mm(3), efficacy was 77%.(73)
The results of seven randomised studies performed with the combination ABC+3TC associated with EFV or a PI have been analysed.(65,75-80) These studies included a total of 1,457 patients treated with the combination ABC, 3TC and EFV; at 48 weeks of treatment, 69% (64-76%) presented a viral load <50 copies/ml (ITT NC=F).(65,75,76,79,80) The various studies also included a total of 989 patients who received ABC+3TC associated with a PI (nelfinavir, APV or FPV); at 48 weeks of treatment, 54% (41-62%) of patients presented a viral load <50 copies/ml (ITT NC=F).(77,78)
The results of three studies carried out with the combination TDF+3TC or FTC associated with EFV or LPV/RTV have been analysed.(66,81,82) In two randomised studies, a total of 554 patients received the combination TDF, 3TC and EFV; 79% (77% and 82%) of patients presented a viral load <50 copies/ml (ITT) at 48 weeks of treatment; however, in one of the studies, substitution of nevirapine for efavirenz was not considered to be a failure (6% of patients).(66,81) In one randomised study, 190 patients received the combination TDF, FTC and LPV/RTV; at 48 weeks of treatment, 68% presented a viral load <50 copies/ml (ITT NC=F).(82)
Two studies have been conducted with the combination d4T(CLP) and 3TC.(83,84) The first one analyses this combination associated with EFV in 70 naive patients with a baseline CD(4) lymphocyte count of >100 cells/mm(3);at 48 weeks of treatment, 71% of the patients had a viral load <50 copies/ml (ITT).(83) The second trial was the randomised AI424-089 study, recently presented at a congress,(84) comparing the efficacy of ATV (400mg, n=105) and ATV/RTV (300/100mg, n=95) associated with d4T(CLP) and 3TC. At 48 weeks of treatment, no significant differences were seen between the two regimens (75% and 70%, respectively, of patients with viral load <50 copies/ml by ITT). Unfortunately, commercialisation of the new d4T formulation allowing QD use has been discontinued, apparently because of technical problems in the production of d4T(CLP) capsules.
Clinical experience related to the administration of PIs in full QD regimens has centred solely on LPV/RTV (six capsules + NRTI) and ATV alone (two capsules + NRTI) or boosted with RTV (three capsules + NRTI). A study conducted in Thailand (STACCATO), involving SQV/RTV administered QD in 200 treatment-naive patients at a dose of 1,600/100mg (nine capsules + NRTI), reported elevated efficacy at 24 months of treatment (viral load <50 copies/ml by ITT in 93% of patients).(55) The initial combination of NRTI in that study was d4T+ddI, which was later changed to TDF+3TC. The new formulations of SQV 500 and LPV/RTV tablets (soon available) will improve the possibility of QD administration of these PIs.
Some QD NRTI combinations that could be considered a priori very useful because of their dosage and intrinsic therapeutic potency, such as TDF+ddI or TDF+ABC, associated with a third drug, as well as the QD combinations of three NAs, are contra indicated in naive patients. Several studies have shown an elevated frequency of virological failure and the development of resistance mutations in naive patients initiating treatment with TDF, ddI and EFV or NVP, particularly patients presenting an elevated viral load and considerable baseline immunosuppression.(85-88) The combination TDF, ABC and EFV or LPV/RTV has also shown an elevated rate of therapeutic failures in highly immunodepressed patients (baseline CD(4) <100 cells/mm(3)) (ABATE and ADVANZ2 studies; personal communication by Dr Miro). A very high frequency of virological failures (around 50% after a few weeks of treatment) has been observed in naive patients with the combinations TDF+ddI+3TC and TDF+ABC+3TC, as well as the development of resistance mutations, which have often included the K65R mutation.(80,89-91)
Simplified ART regimens
Simplification of ART refers to switching from one regimen that has achieved an adequate virological and immunological response to another that maintains efficacy while reducing the complexity of the treatment, thereby providing a benefit in the patient’s quality of life that favours compliance.(5) This potential benefit depends on an increased ease of administration that can be better adapted to the patient’s lifestyle, as well as the possibility to reduce or reverse some of the toxic reactions caused by the previous treatment. The indication for simplified treatment is always relative and should first take into account the patient’s preferences. Moreover, it is important to determine whether the possible benefits of simplification are greater than the risk of presenting adverse effects associated with the new treatment. Simplification of a treatment regimen should be done on an individual basis according to the patient’s prior therapeutic experience, virological failures that may have selected resistance mutations and any adverse events that occurred. In general, if the simplified regimen is well selected, therapeutic efficacy will be higher than that observed in naive patients receiving the same ART regimen as initial therapy. This is a selected population that is able to follow periodic controls and has demonstrated sufficient adherence to achieve an undetectable viral load with the former regimen.
There is now sufficient information to know that it is possible and safe to simplify some ART regimens. In most cases the physician analyses the efficacy and safety of changing a regimen with a PI, generally unboosted, to another with EFV, NVP or ABC, maintaining the initial NA combination. In some cases the change contemplated is the substitution of one PI for another that is more convenient and has some advantages in the toxicity profile, such as ATV, while also maintaining the regimen of NAs. There are few studies in which the entire simplification regimen adapts to QD dosing. We will now review the studies in which treatment is simplified to a QD regimen.
The randomised ALIZE(92) study analysed the efficacy and safety of simplifying treatment to ddI, FTC and EFV versus continuing with a treatment that included PIs, which had resulted in an undetectable viral load. Patients who were suspected to have resistance to any of the drugs in the new regimen, particularly 3TC or EFV, were excluded. At 48 weeks of treatment the percentage of patients with a viral load <50 copies/ml (ITT) was higher in the QD simplification arm than in those continuing treatment (95% versus 87%, p<0.01), with a better lipid profile in the simplification group. The VEST-QD(93) study was conducted in 300 patients with a viral load <50 copies/ml under stable treatment that included a PI. The PI was substituted for EFV in all cases, and patients were randomised to continue with their previous NA combination or switch to ddI and 3TC. At 24 weeks of treatment the percentage of patients with an undetectable viral load was high and similar in the two groups (87% and 85%, respectively). The majority of patients in both groups noted a better quality of life with the new treatment (89% and 91%, respectively). No study is currently available in which treatment was simplified to QD dosing with TDF, FTC and EFV or with ABC, 3TC and EFV. Nevertheless, on the basis of the efficacy and safety of these regimens when used in naive patients and when administered individually in different regimens for treatment changes, it is reasonable to assume that they will also be safe and effective for simplification regimens, as long as there is no previous resistance or intolerance to any of the components.
Patients presenting an undetectable viral load after having followed one or several treatments including 3TC or FTC with virological failure at any point are very likely to harbour virus with the M184V/I mutation, which confers resistance to 3TC and FTC. These patients will not be eligible to receive any of the three NRTI combinations of choice for naive patients or for simplification regimens. TDF+ddI is one of the NRTI combinations that could be useful in a QD regimen in this situation because of its resistance profile. Various studies have analysed the efficacy and safety of QD TDF+ddI associated with NVP or EFV for treatment simplification.(94-96) The percentage of patients under these regimens with a viral load <50 copies/ml at 48 weeks of treatment was high and similar to that observed in patients continuing with the same prior treatment. However, a marked decrease in CD(4) lymphocyte count was seen after six to nine months of treatment, despite a persistently undetectable viral load. When TDF and ddI were combined, an a priori unexpected pharmacokinetic interaction occurred between the two agents, with increases in plasma and intracellular ddl concentrations of 40-60%, without changes in TDF concentrations.(97-100) This interaction occurs at the level of the enzyme purine nucleoside phosphorylase, which intervenes in the catabolism of both drugs.(101) This interaction is clinically relevant, resulting in an increase in the toxicity of ddI,(102-105) as well as the aforementioned decrease in lymphocyte regeneration. To reach plasma concentrations similar to those achieved with the standard dose of both drugs administered individually, the initial dose of ddI has to be reduced to 250mg in individuals weighing 60kg or more and 200mg in patients weighing less than 60kg. With the reduction in ddI dose, patients who presented a decrease in CD(4) lymphocyte count with the TDF+ddI combination at standard doses showed an increase in CD(4) count, but baseline values were not recovered. Because of the problems found with the TDF+ddI combination (toxicity, CD(4) decrease and virological failure in naive patients), the EMEA and the pharmaceutical company have issued a warning about the use of this combination in any clinical situation. When administration of TDF+ddI is considered necessary, strict control should be applied to detect the possible appearance of the above-mentioned complications. When TDF is associated with ddI at the adjusted dose from the beginning, the decrease in CD(4) lymphocytes does not occur.(106,107)
Simplification of ART to a regimen of three NRTIs administered QD is contraindicated because of the elevated rate of virological failure seen in several studies.(108,109) No pharmacokinetic interactions or antagonism have been found among TDF, ABC and 3TC; it is likely that the inefficacy of this regimen is due to the low genetic barrier of the combination as a whole.(110,111)
The current status quo for ART has changed considerably compared with some years ago. The first HAART regimens involved administration of a very large number of tablets several times a day. The difference between having to take 15 or 20 tablets a day divided into two or three administrations and taking two to five tablets once a day represents a substantial simplification, but with the currently available PI formulations it is not likely that treatment simplification can be achieved by changing the PI. In any case, the main reason for substituting a PI or another drug in patients with virological suppression is to elicit an improvement in the toxic effects. To this end, some studies have substituted, for example, a PI for ATV to achieve a better lipid profile or improve digestive intolerance, in addition to simplifying the treatment.(112)
Situations of therapeutic failure (salvage therapy)
In situations of virological failure it is particularly important to individualise ART. The possible cause of failure should be taken into account (eg, poor adherence due to the patient’s characteristics or the complexity of the treatment) and the accumulated viral mutations. The priority in patients with therapeutic failure is to design an ART regimen that is effective against viruses with resistance mutations. Classically, salvage treatments were very complex, an aspect that could make adherence even more difficult. Although efficacy takes precedence over convenience when planning a salvage therapy, it is also important to prescribe the most amenable treatments possible. In the past, it was almost impossible to construct a treatment that could be administered QD for salvage therapy. Now some QD regimens are possible when viral resistance is moderate. We will probably never be able to use 3TC or FTC, but it may be possible to combine two of the other three NAs that are administered QD – TDF, ABC and ddI – taking into account the limitations indicated in the sections above. In a large number of patients that present virological failure with any of the three QD regimens of choice in naive patients, the profile of the resistances that develop advise administration of a thymidine analogue and TDF. Since a QD formulation for d4T (CLP) is not available, these regimens (TDF+d4T or AZT) must be administered BID.
In many cases, NNRTIs cannot be used as the third drug because of viral resistance, but it may be possible to use a boosted PI (r) once daily. It should be highlighted that, except in the case of atazanavir, when a PI (lopinavir/r, saquinavir/r) is administered once a day, the trough concentration is lower than that reached when the same dose is administered twice daily. Decreases in the inhibitory ratio of up to 50% have been observed when a boosted PI is used in a QD regimen as compared with the same dose given BID.(49) This fact is irrelevant when there are no PI-resistant mutations, but when there has been a prior virological failure with PIs and mutations have appeared it is advisable to prescribe the PI in two administrations per day, to provide more elevated minimum concentrations.
Impact of regimens administered once a day on adherence to therapy and quality of life
The experience gained with drug therapy in other areas of medicine assessed in two meta-analyses shows an inverse relationship between the number of administrations prescribed per day and treatment adherence.(113,114) Information on adherence in the area of QD ART is still sparse. In an analysis of clinical trials involving treatment in naive patients, an inverse relationship was found between the number of tablets prescribed per day and the efficacy of the regimen.(115) In a large, prospective, observational study (the CUVA study) conducted in Spain, adherence was significantly higher in patients receiving QD therapy than in those treated with BID regimens.(116) An opinion survey, also carried out in Spain and including both physicians and patients, indicated that reductions in the number of administrations as well as the number of tablets were the main interventions that could improve ART adherence.(117,118)
In patients with HIV infection, the durability of ART can be compromised by a decrease in adherence over time. In a study on treatment simplification conducted in patients with elevated adherence at the baseline visit, the decrease in compliance was lower in patients who were given a simplified QD regimen than in those continuing with BID administration.(119)
The patient’s perception of quality of life is a subjective parameter that is quantified by considering the impact of the disease and its treatment on several personal factors. Patients are unarguably very interested in receiving a treatment administered QD.
In a European survey, 504 patients with HIV infection were asked to score from 1 to 10 their interest in taking all their medication QD. Among the total surveyed, 81% scored their interest above 7. The survey also showed that interest in taking a QD regimen was related to the number of pills in the patient’s treatment. When the number was high, the majority of patients preferred to take their medication in two administrations a day. When the number was seven, eight or more pills, only 38% and 31% of patients, respectively, preferred to take the treatment QD. With a regimen of six pills, 59% preferred QD administration and with three or four pills, 92% and 84%, respectively, preferred QD treatment. Most of the patients surveyed considered that QD administration was the regimen that would best adapt to their lifestyle. Among patients who were already taking ART, the rate of missed administrations was 36% lower in those taking their regimen QD than in those taking it BID.(120)
Opinion polls also show that patients prefer regimens taken entirely QD. In one survey, 505 patients considered that the greatest positive impact on adherence to treatment and quality of life would be obtained with regimens in which all the drugs were administered together once a day. In a survey conducted in 256 patients on the internet in the USA, most patients (73%) preferred to receive a treatment consisting of four pills QD than one requiring one pill in the morning and two at night.(6,121)
- Enferm Infecc Microbiol Clin 2004;22:564-642.
- Ann Intern Med 1999;131:81-7.
- Clin Infect Dis 2002;34:1115-21.
- AIDS Care 2005;17:10-22.
- Enferm Infecc Microbiol Clin 2002;20 Suppl 2:48-57.
- J Antimicrob Chemother 2005;56:808-18.
- Enferm Infecc Microbiol Clin 2004;22:106-12.
- Clin Infect Dis 2003;36:1186-90.
- J HIV Ther 2003;8:7-11.
- J Acquir Immune Defic Syndr 2002;31 Suppl 1:S10-5.
- HIV Clin Trials 2003;4:193-201.
- Int J STD AIDS 2003;14 Suppl 1:1-5.
- Clin Infect Dis 2002;34:686-92.
- Clin Pharmacokinet 2003;42:1179-91.
- Ann Pharmacother 2004;38:1924-34.
- Drugs 1998;56:1055-64.
- Int J STD AIDS 2003;14 Suppl 1:6-14.
- AIDS 2000;14:2485-94.
- J Acquir Immune Defic Syndr 2000;24:418-24.
- J Acquir Immune Defic Syndr 2001;28:150-3.
- Clin Infect Dis 2003;37:944-50.
- Drugs 2005;65:413-32.
- HIV Clin Trials 2002;3:361-70.
- Antimicrob Agents Chemother 2004;48:176-82.
- Clin Infect Dis 2004;39:411-8.
- Antivir Ther 2005;10:239-46.
- Ann Pharmacother 2005;39:1302-8.
- Drugs 2002;62:2667-74.
- Drugs 2003;63:2413-24.
- Ann Pharmacother 2004;38:1006-14.
- AIDS Res Hum Retroviruses 2004;20:1173-82.
- Drugs 2005;65:1427-48.
- Clin Infect Dis 2006;42:126-31.
- Drugs 2004;64:2075-82.
- Drugs 2005;65:285-302.
- Expert Rev Anti Infect Ther 2003;1:403-13.
- Pharmacotherapy 2004;24:1732-47.
- Drugs 2003;63:1679-93.
- Clin Pharmacokinet 2005;44:1035-50.
- Ann Pharmacother 2004;38:1664-74.
- Drugs 2005;65:2309-36.
- Ann Pharmacother 2002;36:102-18.
- Expert Opin Investig Drugs 2000;9:371-82.
- HIV Med 2004;5:364-70.
- J HIV Ther 2004;9:87-91.
- Drugs 2004;64:2101-24.
- AIDS 2005;19:1105-7.
- Expert Opin Pharmacother 2005;6:1573-85.
- J Infect Dis 2004;189:265-72.
- AIDS 2006;2