Amprenavir (APV)

This is a new protease inhibitor from Glaxo Wellcome (GW)which may receive accelerated approval from the FDA by the time you read this newsletter. In NATAP Reports' last issue, we reported the results ofa study (n=173) comparing APV+AZT/3TC to AZT/3TC in treatment-naïveindividuals. After 16 weeks, individuals in this study were permitted tochange therapy. At week 16, 88% had <400 copies/ml using an On-Treatmentanalysis (n=74), 66% had <400 copies/ml using an Intent-To-Treat analysis,and 59% had <50 copies/ml using a less stringent Intent-To-Treat analysiscalled Intent-To-Treat observed data only.

APV is a gel-capsule taken with or without food. Preliminaryresearch shows it penetrates the CSF in humans. It appears well tolerated.It is taken twice daily but the pill burden is 8 pills per dose. APV willbe available in pill form or in liquid form. The PK profile is forgivingin that there is flexibility in dosing. It has a reportedly long half-lifeof10 hours. Drug levels stay elevated for a prolonged period lending itselfto adherence flexibility. Although it is recommended to take APV every12 hours missing a dose by a few hours should be ok. Each amprenavir pillcontains 109 International Units of Vitamin E but GW says only 15% is absorbeddue to the form of the Vitamin E used. Some of the more common sides effectsassociated with APV are GI related: oral parasthesia (numbness, tingling),nausea, flatulence, vomiting, diarrhea, headache; also, rash can occur ata rate of 11% or better.

It was reported at Retrovirus that of 606 individuals exposedto amprenavir for >12 weeks, and 41 individuals >48 weeks symptomsrelated to fat redistribution were reported in only 4 individuals, 3 ofwhom had previously used protease inhibitors. GW is suggesting that lipodystrophymay not occur, but it is pre-mature to presume lipodystrophy will not occuror even that it might occur with reduced incidence. In general, lipodystrophymay not emerge for 1 year or more while on a PI. In the APV expanded accessprogram, there is a group of participants with lipodystrophy who switchedto APV. Hopefully, data will eventually be available to address this question.

Resistance. The resistanceprofile of APV appears different than other protease inhibitors, suggestingif used as a first-line therapy it might be easier to salvage than indinaviror ritonavir, as is nelfinavir. As well, APV appears to have utility forindividuals with prior PI use and resistance. But, these points remain tobe proven. Regarding the first point of using APV for initial therapy, KeithHenry conducted a small study of individuals who had never taken any treatment,were PI-naïve, and who had failed nelfinavir in a previous study.These were nelfinavir failures who had remained on nelfinavir+AZT/3TC for 48 weeks with detectable viral load. In Henry's study these individualsreceived RTV+SQV+d4T/3TC as a salvage regimen; 71% (17/24) had <500 copies/mlat 24 weeks. Study participants adding and subtracting other drugs havecomplicated follow-up.

In ACTG 373, protease inhibitor naïve individualswho failed APV monotherapy (n=36) or APV+AZT/3TC (n=19) had their viralload monitored closely and were switched quickly to a salvage regimen ofindinavir+nevirapine+d4T/3TC upon detecting viral load rebound. After 76weeks, 80% of those who initially received APV monotherapy had <500 copies/ml,and about 70% who had initially received the triple regimen had <500copies/ml. It is not possible to discern whether the salvage regimen wassuccessful because of APV's lack of cross-resistance or because viral loadfailure was caught quickly and the quick switch to the new regimen. Successcould be due to both factors. M Tisdale and others reported at Retrovirusan analysis of 3 studies in which PI-naïve individuals received APV.They reported that when >4-fold APV resistance developed, phenotypiccross-resistance to other PIs did not appear. Except, there was about 3.5to 4-fold ritonavir resistance. When there was <4-fold APV resistancethere was no ritonavir cross-resistance. The I50V mutation appears to bethe one most associated with APV failure. But the I54V/L mutation also appearedas well as changes at position 46. This is in vitro data and additionalstudies in humans are required to prove that APV is more easily salvagablewhen used as a first-line PI.

APV may have utility in certain salvage therapy situationswhen individuals have resistance to other protease inhibitors. In vitro, the double mutation of M46I/L with I50V can reduce sensitivity to APV 7fold, while a virus with I47V added to those 2 mutations resulted in 14-20fold reduced sensitivity (IC50). A virus containing L10F with a transientI84V reduced in vitro sensitivity 3-fold. When >4-fold APV resistancehas been detected this L10F+ I84V mutant is more likely to be present. M. Tisdale and others reported on this at Retrovirus. In a study (CNA2007)of individuals with extensive prior therapy and with 72-84% >4-fold reducedsensitivity (phenotypic resistance) to the 4 approved proteases inhibitors,55% of the study individuals had some sensitivity (<4-fold phenotypicresistance) to APV prior to receiving APV in the study. Individuals withabout 20-fold or less resistance to the 4 approved PIs had <4-fold APVphenotypic resistance prior to receiving APV in this study. This informationis partially based on in vitro resistance data and preliminary. For thoseof you familiar with this study, interpretation of the overall resultsare complicated because participants entered with a good deal of broad NRTIresistance and NNRTI resistance. They received efavirenz +abacavir+APV inthis study. Obviously, individuals who were NNRTI-naïve, with low viralload at baseline, and with the least NRTI resistance were more likely toachieve undetectable viral load. To me, this data suggests a utility butdoes not characterize how much utility and in which circumstances APV wouldbe useful. Additional studies will have to be conducted.

GW has not yet conducted a PK study of combining APV withritonavir but it appears as if this might be a promising salvage combination.It needs to be tested and dosing needs to be identified before experimenting.It's not advisable to combine full doses of the two drugs and it's unknownwhat doses would be preferable, but I heard of two different dose combinationsbeing used: 900mg bid APV with 200mg bid RTV or 600mg bid APV with 400mgbid RTV, combined with other drugs. This is not a recommendation.

Preliminary APV Interactions With SQV, IDV, NFV andEFV. GW has conducted several preliminary interactionstudies looking at the effect on the blood levels of each drug when combiningtwo. It's important to bear in mind that the results of these types of interactionsstudies, which all drug companies conduct and report when a drug is approved,are not always confirmed in future studies. It can sometimes be risky tomake treatment decisions based on this data. For example, Agouron reportedpreliminary interaction data on NLF+IDV, but when Merck conducted the sameinteraction study in preparation for a double protease study of the 2 drugsthey saw different results. However, I've selected a few of the interactionsfrom the GW study to report to you which I think are worth taking noteof. The clinical significance of some of these interactions are uncertain.In GW's preliminary PK studies saquinavir reduced APV AUC by 36%. AUC isthe area under the curve and represents the amount of drug in the bloodover a dosing period which in the case of APV is 12 hours. However, APVCmin was not changed by saquinavir. Cmin is the lowest level of a drug ina person's blood which occurs at the end of the dosing period just beforeits time to take the next dose. This raises some questions regarding combiningAPV with saquinavir. APV reduced SQV AUC by 18%. Cmax is the highest levelof a drug in the blood and is usually reached a few hours after taking adrug.

APV blood levels were reduced (39% Cmax, 36% AUC, 43% Cmin) when combined with efavirenz. GW said there was a lot of inter-person variability.Individuals with the highest APV blood concentrations had the greatest reductionsin APV concentrations while those with lower APV blood concentrations hadless reductions when combining APV with EFV. They suggest it is pre-matureto draw conclusions about the clinical effect of these drug interactions. For example, it's been suggested that if using APV and EFV in a 4 drugregimen the effect of EFV on APV may not have clinical effect, althoughI'm not sure I agree with that. They are conducting a study now to betterunderstand the clinical effects of these interactions. In the meantime,I suggest caution if considering combining these two drugs. APV increasedEFV AUC by 15%. Dosing combinations preliminarily being considered fora study are 1200mg TID APV or 1800mg bid APV with the standard EFV dose(600mg once-daily).

Indinavir raises the blood levels of APV (6% Cmax, 26%AUC, 18% Cmin), and APV lowers IDV blood levels (22% Cmax, 38% AUC, 27%Cmin). Therefore one study I know of is combining 900 mg bid amprenavirwith indinavir in a 4-drug regimen. The actual effects of this interactiondata (the clinical effects) on a person taking this combination are againuncertain, but a study is being conducted that may allow us to understandhow these interactions will effect individuals. Again, I suggest cautionif considering combining these two drugs, although it's possible that theeffects of APV on IDV and the effects of IDV on APV may counter balanceeach other because they are about equal. Interactions studies with RTV,NVP, and DLV have not been conducted yet. I think a RTV study will beginsoon because there is interest in using RTV+APV as a salvage regimen. DLVwould also be expected to raise APV blood levels and I think should bestudied. Of note, NFV increased APV Cmin 167% ( Cmax 21%, no change inAUC). APV had little effect on NFV: APV NFV Cmax 12%, AUC 15%, Cmin 14%.

APV raised the blood levels of AZT (40% Cmax, 31% AUC,Cmin not yet known). If these interactions are confirmed they could accentuateAZT side effects. Rifampin reduced APV Cmax 70%, AUC 82%, and Cmin 92%.APV had no effect on Rifampin. Rifabutin reduced APV Cmax by 7%, AUC 15%,and Cmin 15%. APV increased Rifabutin Cmax by 127%, AUC 204%, and Cmin 349%.When APV actually receives accelerated approval the APV package insert,which is available in any pharmacy selling the drug, should outline recommendeddose modifications. Please consult with your doctor about these matters.