Abacavir vs. Nelfinavir & Preliminary Data on Abacavir in >100,000 Viral Loads

This is a preliminary 24-week report of a 48-week open-label study comparing abacavir to nelfinavir 750 mg three times per day. The study is to assess safety and tolerance (including metabolic complications). The primary endpoint of the study is the percentage of patients with HIV RNA <50 copies/ml at week 48. Individuals with viral load between 1000 - 500,000 copies/ml are eligible. There are 98 patients in the abacavir arm and 97 in the nelfinavir arm. The study had 3 analyses: (1) ITT Switch Included; switch of randomized medication not included as failure; (2) ITT Switch=Failure; failure is changed or permanently off randomized study medication, missing data, plasma HIV RNA >50 copies/ml, HIV RNA >400 copies/ml; (3) As Treated (AT): on treatment. Median baseline viral load was the same in both arms; 4.2 log (15,000 copies/ml) (range 1.3-5.3). Median baseline CD4s were 387 in abacavir arm and 449 in nelfinavir arm.

Patient Disposition. In the abacavir arm, 87 patients completed 24 weeks, 77 did not switch, 10 did switch, and 9 discontinued randomized treatment. In the nelfinavir arm, 83 patients completed 24 weeks, 74 did not switch therapy, 9 switched, and 9 discontinued randomized treatment. Of the 19 in the abacavir arm that made a treatment change, 13 were due to an adverse event, 5 withdrew consent, and 1 other. In the nelfinavir arm, there were 18 who changed their treatment: 11 due to adverse event, 2 withdrew consent and 5 other.

Viral Load Suppresion. At week 24, 77% of the patients in the abacavir arm and 72% in the nelfinavir arm had <50 copies/ml (ITT Switch=Included). Using the ITT Switch=Failure analysis, 67% in the abacavir arm, and 66% in the nelfinavir arm had <50 copies/ml. Using the As Treated analysis, 90% (n=72) in the abacavir arm and 86% (n=70) in the nelfinavir arm had <50 copies/ml. Both arms had a median viral load reduction of about 2.4 log at week 24. The median CD4 change was about +90 in the abacavir arm and +65 in the nelfinavir arm at week 24.

Safety and Tolerance. 62% of patients (110/188) had ≥1 adverse event. The most frequent (≥10%) drug related AEs were: nausea & vomiting 38% (n=36) in the abacavir arm, and 32% (n=29) in the nelfinavir arm; diarrhea; 7% (7) in the abacavir arm and 41% (38) in the nelfinavir arm. 5% (9/188) had ≥1serious adverse events; there were 4 abacavir hypersensitivity reactions reported (4%). Early discontinuation of at least one study drug: 13 (14%) in abacavir arm, 11 (12%) in the nelfinavir arm. There was one serious AE in the nelfinavir arm, and none in abacavir arm.

Comments: The baseline viral load was very low (4.2 log).

Abacavir versus Indinavir in Combination with Combivir (AZT/3TC) in Therapy Na‘ve (CNA3014): patient group with baseline viral load >100,000 copies/ml. This is an open-label, randomized, multi-center 48-week study and Pedro Cahn from Buenos Aires reported preliminary 24 week data. The viral load endpoint is 400 copies/ml. The study looks at adherence, safety, tolerability, and "satisfaction". Patients have an option of switching therapy after week 16, if 2 consecutive viral loads are >400 copies/ml. For study entry, patients must have viral >5000 copies/ml and are stratified to viral load >5000 to 100,000 or >100,000. 342 treatment-na‘ve individuals were randomized to receive abacavir/combivir (n=169) or indinavir/combivir (n=173). The two analyses used are Intent-To-Treat (Missing=Failure) ITT: all subjects taking at least 1 dose of study treatment; As Treated (AT): all subjects while on randomized treatment, with the exception of major protocol violators (pregnancy or non-adherence; cumulative total of >6 weeks or continuous of >2 weeks).

There were about 60% men in each arm. Baseline median viral load was 4.8 in each arm (63,000 copies/ml). 64% in the abacavir arm and 62% in the indinavir arm were stratified to the 5,000-100,000 copies/ml group (n=216). 36% in the abacavir arm and 38% in the indinavir arm were stratified to the >100,000 copies/ml arm (n=126). CD4s at baseline were 323 in abacavir arm and 300 in the indinavir arm. 5 in the abacavir arm and 7 in the IDV arm did not start therapy. There were 14 (8%) discontinuations in the abacavir arm and 21 (12%) in the indinavir arm. Total subjects who changed therapy prior to week 24: 29 (18%) in the ABC arm, and 45 (27%) in the IDV arm. Here is the breakdown: those who switched therapy due to viral load >400 copies/ml; 4 in ABC arm and 15 in the IDV arm. Adverse event or lab abnormality: 6 in ABC arm and 7 in IDV arm. Listed as other reasons for switch of therapy: 17 in ABC arm, 23 in IDV arm. Overall, using the As Treated analysis, 87% in the ABC arm had <400 copies/ml and 83% had <400 copies/ml in the IDV arm. Over 100 were in each of these analyses. (See Tables 8, 9 & 10)


CD4s increased about the same in both groups; 110.

Adherence at Week 24:

Summary of Adverse Events:

Comments: It appears as though adherence and adverse events play a crucial role in the study. While indinavir is probably more potent than abacavir, abacavir is easier to adhere to and is probably more tolerable. This is attested to by the data in this study showing less adherence and more adverse events in the indinavir arm.