ACTG 5095: Efavirenz-Based Regimens Outperform Triple Nucleoside Combination

• ACTG 5095: A Comparative Study of Three Protease Inhibitor-Sparing Antiretroviral Regimens for the Initial Treatment of HIV Infection (Oral 41)
  Authored by R.M. Gulick, H.J. Ribaudo, C.M. Shikuma, S. Lustgarten, W.A. Meyer, K. Klingman, K.E. Squires, S. Snyder, D.R. Kuritzkes
  

The ACTG 5095 trial is an ongoing phase III, randomized, double-blind, placebo-controlled study that enrolled antiretroviral-naive patients to compare the safety and antiviral activity of three antiretroviral regimens, two of which contain efavirenz (EFV, Sustiva, Stocrin) and one which contains only three nucleoside analogues.

The three arms of this trial consist of co-formulated abacavir (ABC, Ziagen) + AZT (zidovudine, Retrovir) + 3TC (lamivudine, Epivir) tablet (Trizivir) versus efavirenz plus co-formulated AZT + 3TC tablet (Combivir) versus efavirenz plus Trizivir.

As many clinicians are aware, the NIAID Data and Safety Monitoring Board (DSMB) recommended, based on a planned interim review, termination of the Trizivir arm and that the efavirenz-containing arms continue in blinded treatment. This recommendation was the subject of a letter sent to U.S. healthcare providers, and of an ACTG press release several months ago. The data presented at the 2nd IAS Conference gives the details of the data collected in the study to date.

ACTG 5095 enrolled 1,147 patients. Nineteen percent of these patients are women, 60 percent are non-white and 11 percent are injection drug users. The mean baseline viral load and CD4 count of the enrolled patients was 4.9 log10 copies/mL (43 percent with more than 100,000 copies/mL) and 238 cells/mm³, respectively, and were comparable across study arms.

After a median of 32 weeks of follow-up, 93 percent of the patients continued on the study and 91 percent continued on the study drugs. Grade 3 and 4 signs/symptoms occurred in 12 percent and 2 percent of patients, respectively, with comparable proportions across study arms.

The trial protocol allowed for switches if a patient was intolerant of the study medication. Eight percent of patients changed from AZT to stavudine (d4T, Zerit), 5 percent from abacavir to didanosine (ddI, Videx) and 6 percent from efavirenz to nevirapine (NVP, Viramune). In the study, virologic failure was defined as a confirmed viral load of more than 200 copies/mL after more than 16 weeks on therapy. The data were presented using an intent-to-treat analysis.

One hundred and sixty-seven (167) patients reached protocol-defined virologic failure, which was defined as a confirmed viral load greater than 200 copies/mL after week 16. The important findings in the analysis were:

• Of the virologic failures, 21 percent occurred in the Trizivir arm and 10 percent in the pooled efavirenz arms of the trial. The time to virologic failure was shorter with Trizivir compared to the pooled efavirenz arms (P<0.001), and this was true whether the baseline viral load was less than or greater than 100,000 copies/mL at baseline (P<0.001 for each).

• In the Trizivir and the pooled efavirenz arms, at week 48 of the trial 74 percent and 89 percent, respectively, achieved a viral load less than 200 copies/mL; and 61 percent and 83 percent, respectively, achieved a viral load less than 50 copies/mL (both with significant P values).

• In a post-hoc analysis of patients who achieved at least one viral load of less than 200 copies/mL, time to virologic failure was significantly shorter in the Trizivir patients (P<0.001). Also, in those who achieved at least one viral load less than 50 copies/mL, there was a strong trend toward increased failures in the Trizivir patients (P=0.083).

• The CD4 cell count increases were similar in the Trizivir and efavirenz-treated patients -- approximately 170 cells/mm³.

• Median self-reported adherence rates were 100 percent in all arms.

For the patients failing Trizivir, the resistance patterns were: 22 percent wild type, 34 percent 184V, 11 percent 184V with other nucleoside analogue mutations, 2 percent other nucleoside analogue mutations and 31 percent either not attempted or sequenced.

The authors conclude from these data that, in treatment-naive patients, Trizivir is inferior to efavirenz-containing treatment in terms of rates and time to virologic failure. So the question becomes: When should one use Trizivir or another triple-nucleoside-analogue regimen? Based upon ACTG 5095, and several other studies presented at this conference, the emerging consensus seems to be that Trizivir and other triple-nucleoside-analogue regimens should not be routinely used by themselves in clinical practice, regardless of the baseline viral load of the patient. There may remain a small subset of patients for whom these regimens are appropriate, but given the high rates of failure relative to the efavirenz-containing arms of ACTG 5095, clinicians should think long and hard before using Trizivir alone.