Correspondence

Tenofovir DF and Emtricitabine vs. Zidovudine and Lamivudine

N Engl J Med 2006; 354:2506-2508June 8, 2006

Article

To the Editor:

Gallant et al. (Jan. 19 issue)1 report the superiority of a regimen of tenofovir disoproxil fumarate (DF), emtricitabine, and efavirenz to a control regimen of zidovudine, lamivudine, and efavirenz in terms of suppression of replication of the human immunodeficiency virus (HIV). Two sources of bias compromise this conclusion: first, an open-label study design with a disproportionate number of discontinuations for reasons other than virologic failure in the control group and second, the issuance of a press release2 while the study was going on announcing the superiority of the tenofovir DF–emtricitabine regimen after an interim analysis at 24 weeks. In the recent review of this study by the Food and Drug Administration (FDA),3 10 sensitivity analyses were performed according to different methods to reclassify subjects who prematurely discontinued the control therapy for reasons other than virologic failure at any time or after the press release was issued. As shown in Table 1Table 1Subjects with Sustained Levels of HIV RNA below 400 Copies per Milliliter at Week 48 or Later., in two representative analyses the magnitude of the treatment difference between the two study groups was reduced, although these analyses generally supported the primary efficacy analysis (also see the Supplementary Appendix, available with the full text of this letter at www.nejm.org). Sensitivity analyses are an important means in examining the robustness of a treatment effect; however, sources of potential bias should be critically discussed in reporting clinical trials, especially when the treatment is not blinded, because these may substantially affect the interpretation of the results.

Katherine A. Laessig, M.D.
Linda L. Lewis, M.D.
Thomas S. Hammerstrom, Ph.D.
Food and Drug Administration, Silver Spring, MD 20993
katherine.laessig@fda.hhs.gov

3 References

1. 1

   Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs. zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med 2006;354:251-260
   Full Text | Web of Science | Medline

2. 2

   Gilead announces preliminary 24-week data from study 934 comparing Viread and Emtriva to Combivir both in combination with efavirenz in patients with HIV. Press release of Gilead Sciences, Foster City, Calif., August 26, 2004. (Accessed May 18, 2006, at http://www.gilead.com/wt/sec/pr_607254.)
   

3. 3

   Viread and Truvada scientific reviews, product labels, and approval letters. Rockville, Md.: Food and Drug Administration Office of Management Programs, Division of Freedom of Information, 2006.
   

To the Editor:

Gallant et al. explicitly excluded from their genotypic analysis the 22 enrolled patients subsequently shown to harbor baseline resistance to nonnucleoside reverse-transcriptase inhibitors (NNRTIs). Implicit in this exclusion was their assumption that baseline resistance would preclude virologic suppression. The authors did not, however, exclude patients with baseline resistance to the nucleoside reverse-transcriptase–inhibitor (NRTI) companion drugs that were being compared in the study. How such patients fared — both those with baseline NRTI resistance mutations and those without baseline NRTI resistance mutations — would shed light on the effect of baseline resistance on virologic outcome, potentially challenging the view that resistance is the cause, rather than the consequence, of residual viremia. Furthermore, although the authors showed the superiority of tenofovir to zidovudine in terms of the preservation of limb fat as measured by dual-energy x-ray absorptiometry, they omitted a comparison of bone mineral density (BMD) in the two study groups. Previously, a large efficacy trial comparing tenofovir with stavudine, with both drugs combined with efavirenz and lamivudine, demonstrated greater loss of BMD out to 144 weeks with tenofovir.1

Elizabeth R. Jenny-Avital, M.D.
Jacobi Medical Center, Bronx, NY 10461
jennyavita@earthlink.net

Dr. Jenny-Avital reports having received consulting fees and speaking fees from GlaxoSmithKline.

1 References

1. 1

   Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA 2004;292:191-201
   CrossRef | Web of Science | Medline

Author/Editor Response

Dr. Laessig and colleagues state that the conclusions of the study were compromised by an interim press release and the open-label design. They performed sensitivity analyses taking into account that 51 patients switched therapy and were classified as having treatment failure, despite having HIV RNA levels of less than 400 copies per milliliter after the issuance of the press release on August 26, 2004. In fact, we found that 45 of these 51 patients discontinued the study drugs before that date (median, 128 days before). Only six patients (three from each of the groups) switched therapy after that date.

We recognize the potential for bias due to differential rates of discontinuation in an open-label study. However, we think that this disadvantage is offset by the advantage in an open-label trial of being able to assess effects on efficacy of the differences in pill burden or dosing frequency, which would not be possible in a placebo-controlled trial. As discussed in our report, the most common adverse event leading to discontinuation was anemia, which was an objective measure. In the control zidovudine–lamivudine group, 9 percent of patients discontinued participation because of adverse events, as specified by the FDA's analysis of the time to loss of virologic response.1,2 This finding is consistent with what has been seen in other clinical trials in which the same control regimen was used. For example, in a large, double-blind, placebo-controlled study (CNA 30024) that used the same control regimen, 16 percent of patients discontinued participation because of adverse events.3,4 In addition, in our study (Study 934) and CNA 30024, a similar proportion of patients in the control groups (14 percent and 11 percent, respectively) discontinued treatment because of “other reasons” (e.g., pregnancy, withdrawal of consent, loss to follow-up, and nonadherence, among others).2-4 Also, a similar proportion of patients in the control groups (73 percent and 71 percent, respectively) in those trials achieved a virologic response (HIV RNA level, <400 copies per milliliter) at week 48.2,3 Thus, we remain confident of the integrity of the reported study results.

Dr. Jenny-Avital requested efficacy results that include patients with and without baseline NRTI resistance mutations. Such an analysis is less straightforward than the comparison of results for patients with and without NNRTI resistance, since the effect of NRTI mutations on NRTI susceptibility is far more heterogeneous. However, excluding patients with baseline NRTI resistance yields similar statistically significant differences favoring the tenofovir DF–emtricitabine regimen. The results including patients with baseline NRTI resistance mutations are further clarified in Table 1 in the Supplementary Appendix, available with the full text of this letter at www.nejm.org. Data on BMD were not collected in the study.

Joel E. Gallant, M.D., M.P.H.
Johns Hopkins University School of Medicine, Baltimore, MD 21287
jgallant@jhmi.edu

Edwin DeJesus, M.D.
Orlando Immunology Center, Orlando, FL 32803

Andrew K. Cheng, M.D., Ph.D.
Gilead Sciences, Foster City, CA 94404

4 References

1. 1

   Food and Drug Administration, Center for Drug Evaluation and Research. Guidance for industry: antiretroviral drugs using plasma HIV RNA measurements — clinical considerations for accelerated and traditional approval. October 2002. (Accessed May 18, 2006, at http://www.fda.gov/cder/guidance.)
   

2. 2

   Tenofovir. Foster City, Calif.: Gilead Sciences, 2006 (package insert). (Accessed May 18, 2006, at http://www.gilead.com.)
   

3. 3

   Abacavir. Philadephia: GlaxoSmithKline, 2006 (package insert). (Accessed May 18, 2006, at http://www.gsk.com.)
   

4. 4

   DeJesus E, Herrera G, Teofilo E, et al. Abacavir versus zidovudine combined with lamivudine and efavirenz, for the treatment of antiretroviral-naive HIV-infected adults. Clin Infect Dis 2004;39:1038-1046
   CrossRef | Web of Science | Medline

Citing Articles (6)

Citing Articles

1. 1

   Alicen Spaulding, George W Rutherford, Nandi Siegfried, Alicen Spaulding. 2010. Tenofovir or zidovudine in three-drug combination therapy with one nucleoside reverse transcriptase inhibitor and one non-nucleoside reverse transcriptase inhibitor for initial treatment of HIV infection in antiretroviral-naïve individuals. .
   CrossRef

2. 2

   Alicen Spaulding, George W Rutherford, Nandi Siegfried, Alicen Spaulding. 2010. Stavudine or zidovudine in three-drug combination therapy for initial treatment of HIV infection in antiretroviral-naïve individuals. .
   CrossRef

3. 3

   Kevin Carroll, Hilary Chaudri Ross, Dan Evans, Lesley France, Rob Hemmings, Sara Hughes, Frances Lynn, Paul Mahoney, Alan Phillips. (2008) Conditional approval: discussion points from the PSI conditional approval expert group. Pharmaceutical Statistics 7:4, 263-269
   CrossRef

4. 4

   Michael D Gitman, David Hirschwerk, Cindy H Baskin, Pravin C Singhal. (2007) Tenofovir-induced kidney injury. Expert Opinion on Drug Safety 6:2, 155-164
   CrossRef

5. 5

   (2007) Current awareness in geriatric psychiatry. International Journal of Geriatric Psychiatry 22:1, 91-98
   CrossRef

6. 6

   S. Esser. (2006) Antiretrovirale Therapieregime. Der Hautarzt 57:11, 961-968
   CrossRef