Enfuvirtide (T-20), the twice-daily injectable antiretroviral that stops HIV from fusing to CD4 cells, poses difficult treatment decisions for people with highly drug-resistant virus. Because it blocks HIV replication by a mechanism entirely different from those exploited by reverse transcriptase inhibitors and protease inhibitors (PIs), enfuvirtide can control virus resistant to nucleosides, nonnucleosides, and PIs.

But HIV will evolve into strains resistant to any antiretroviral if a regimen does not shut down viral replication. So using enfuvirtide in a combination that lets HIV multiply long enough guarantees that the virus will become resistant to this drug. Yet the temptation is great to try enfuvirtide in people with the most resistant virus–the very people who can put together few, if any, drugs that will stop HIV.

The gp41 sugar protein that pokes from HIV’s coat helps the virus latch onto CD4 cells. Enfuvirtide mimics gp41’s HR2 region. Binding to HIV there, it fouls up the normal folding of HR2 onto the HR1 region and so frustrates viral fusion to CD4 cells. But HIV found ways to overcome this interference by hatching mutations in HR1 [1,2]. The only question was how fast these mutations appear if an enfuvirtide regimen start failing.

Several studies presented in the last few months suggest an answer–very fast indeed. As Stanford University’s Nancy Shulman explains in reviewing three reports from the 2005 Conference on Retroviruses [3], resistance to enfuvirtide emerges rapidly when the drug gets mixed into a suboptimal regimen. This speedy resistance, she observes, mirrors that seen with nonnucleosides and lamivudine (3TC) when regimens including those drugs falter.

Research reported at the Third European HIV Drug Resistance Workshop confirmed the evolution of resistance within weeks of starting enfuvirtide in a wobbly regimen and explored other clinical aspects of resistance to this still-unique agent.

Studying 11 heavily pretreated people who started a new combination including enfuvirtide, Cecilia Cabrera (irsiCaixi Foundation, Barcelona) recorded resistance mutations after as little as 1 week [4].This work has one advantage over earlier research: It tracked viral evolution over 96 weeks as people continued taking a poorly suppressive enfuvirtide regimen.

Viral load in this group dropped sharply 2 weeks after they started enfuvirtide, from an average 5.16 logs to 3.58 logs (about 145,000 to 3800 copies/mL). But no one pushed their load into undetectable territory, and everyone’s load rebounded from the lowest point reached.

Several familiar HR1 mutations at amino acid positions 36 to 45 arose rapidly– V38A/M/E in six people, N43D/S in six, G36V in two, N42T in two, and Q40H, L44M, and L45M in one each. Cloning virus from collected samples, Cabrera showed that mutations at positions 36 and 38 evolved between 1 and 4 weeks after therapy began.

Position 36 and 38 mutations disappeared as the virus continued to evolve, a finding suggesting that these changes may hurt HIV’s ability to replicate. But new HR1 mutations (N43D or Q40H plus L45M) appeared to replace the early changes. And complex genotypic patterns also evolved in the HR2 region of gp41. The HR2 changes, Cabrera speculated, may further resistance to enfuvirtide and crank up mutant HIV’s replicating power.

A study of six people who started a suboptimal enfuvirtide regimen showed that resistance to the fusion inhibitor evolved regardless of pretreatment susceptibility to the drug [5]. The viral load never fell in one person, while in the other five it dropped then rebounded. Fabrizio Mammano (Bichat-Claude Bernard Hospital, Paris) chalked up a wide range of pretreatment susceptibilities to enfuvirtide, yet mutations arose with no apparent regard to baseline susceptibility:

• Baseline susceptibility 56 ng/mL: G36V, V38A
• Baseline susceptibility 95 ng/mL: G36D, N43D
• Baseline susceptibility 135 ng/mL: G36D
• Baseline susceptibility 142 ng/mL: G36D, N43D
• Baseline susceptibility 489 ng/mL: G36D, N43H/E
• Baseline susceptibility 756 ng/mL: V38A

Mammano also tracked the entire HIV env gene, which carries the code for the gp41-studded viral envelope. The familiar HR1 mutations arose from an env population different from the one recorded before enfuvirtide therapy. Eventually, Mammano proposed, a single viral variant that is both highly resistant and relatively fit will emerge.

The fitness of enfuvirtide-resistant virus remains a puzzle. Work in the Harvard lab of Daniel Kuritzkes rated resistant virus highly unfit in the absence of enfuvirtide [6]. During enfuvirtide therapy these resistant mutants proved far more fit than nonmutant virus. Yet withdrawing enfuvirtide from a failing regimen–and so allowing the fitter resistant virus to revert to less fit nonmutant virus–had little virologic impact in a small study [7]. That finding suggests little advantage to continuing enfuvirtide as part of a failing regimen (especially given its high cost and taxing administration).

The clinical impact of reduced viral fitness (if any) is tough to reckon because most work measures fitness in the context of a single viral gene–in this case env. But the overall fitness of HIV depends on the total viral genome and perhaps on immunologic factors as well.

A detailed case report by Laura Maroldo (West Midtown Medical Group, New York) and ViroLogic colleagues documented the jackrabbit emergence of an unusual enfuvirtide mutation, V38E, in a person starting a salvage regimen containing enfuvirtide, 3TC, tenofovir disoproxil fumarate (TDF), and saquinavir/ritonavir [8]. He started the new combination with a viral load close to 150,000 copies/mL, but that load never dropped and his CD4 count never rose.

Genotyping and phenotyping showed no enfuvirtide-related mutations and full susceptibility to the fusion inhibitor on the first day of treatment. But by day 10 the viral population contained nonmutant-mutant mixtures at HR1 positions 36 (G36G/D) and 38 (V38V/E/A) and 7.6-fold resistance to enfuvirtide. By day 30 Maroldo recorded high-level (401-fold) resistance to enfuvirtide and a dominant V38E mutation, which persisted through 547 days of follow-up. The earlier work by Kuritzkes tied V38E to high-level resistance and reduced viral fitness [6].

HIV’s envelope gene evolves at a ferocious clip–a trait that frustrates neutralizing antibodies and vaccine researchers who would wield them. Long before a person may start enfuvirtide, workshop studies showed, env in general–and the part that encodes gp41 in particular–gyrates regularly.

Ali Si-Mohamed (Hôpital Européen Georges Pompidou, Paris) proved this point in a study comparing four groups: 50 people with mutations conferring resistance to nucleosides, nonnucleosides and PIs, 24 with mutations to nucleosides and/or nonnucleosides, 34 antiretroviral-naive people with no mutations, and 12 taking enfuvirtide in a salvage regimen [9].

The prevalence of gp41 polymorphisms–gene changes that may arise naturally without drug pressure–proved twice as high in the two groups with reverse transcriptase or protease mutations (41%) than in antiretroviral-naive people without mutations (20%) (P < 0.05). As would be expected, the rate of gp41 polymorphisms and enfuvirtide-linked mutations also proved significantly higher in enfuvirtide-treated people (75%) than in antiretroviral-naive individuals (17%, P < 0.004) or people with reverse transcriptase or protease mutations (35%, P = 0.05).

Genotyping virus from 53 HIV-infected, enfuvirtide-naive Russians spotted resistance mutations in four of them [10]. Elena Vazquez de Parga (Carlos III Hospital, Madrid) reported that two people infected with HIV-1 subtype A had virus bearing the Q46M and N42T mutations, and two people with subtype F had the R46M and V69I mutations. Forty-one of the 53 people tested (77%) had natural polymorphisms in HR1 positions.

One can probably safely assume that these people did not pick up HIV from someone already taking enfuvirtide. Nor would one expect much enfuvirtide experience in Galicia, the Spanish province north of Portugal. Yet three of 162 newly diagnosed, untreated people there (2%) had virus that carried the V69I mutation in gp41, reported Mercedes Muñoz-Nieto (Carlos III Hospital, Madrid) [11]. Ten of these 162 (6%) had virus laden with mutations in reverse transcriptase, and three (2%) had protease mutations.

Despite the just-reviewed findings, few clinicians are likely to test for resistance to enfuvirtide before starting the drug–especially since there is no commercial assay for gp41 resistance mutations. Indeed, resistance maven Daniel Kuritzkes argues there is little point in searching for gp41 mutations in people taking an ineffective enfuvirtide regimen [12]. The goal of genotyping is to figure out what drug to switch to, he observes, not whether current drugs are failing. And since there are no other fusion inhibitors that control enfuvirtide-resistant virus, nailing down what gp41 mutations emerged has little point.

CD4 cell count does not improve," Kuritzkes says, "or if they had a viral load reduction and then experienced a rebound back to baseline, this should be sufficient evidence to suggest that the drug has failed and is probably not worth continuing" [12].

At the Resistance Workshop, Bonaventura Clotet (University Hospital Germans Trias i Pujol, Badalona) suggested one circumstance when enfuvirtide resistance testing may make sense: If enfuvirtide is part of a failing regimen, determining whether gp41 resistance mutations have emerged will let a clinician know whether to stop enfuvirtide fast and try it again if it can be added to potent drugs that become available later. But the perky emergence of resistance to enfuvirtide–reviewed above–strongly suggests the mutations will be there.

Mark Mascolini writes about HIV infection (markmascolini@earthlink.net).

References
(To view slides and posters from the Third European HIV Drug Resistance Workshop, go to http://www.hivpresentation.com.)

1. Wei X, Decker JM, Liu H, et al. Emergence of resistant human immunodeficiency virus type 1 in patients receiving fusion inhibitor (T-20) monotherapy. Antimicrob Agents Chemother 2002;46:1896-1905.

2. Sista PR, Melby T, Davison D, et al. Characterization of determinants of genotypic and phenotypic resistance to enfuvirtide in baseline and on-treatment HIV-1 isolates. AIDS 2004;18:1787-1794.

3. Shulman N. Enfuvirtide (ENF, T-20) resistance, fast and pretty complete, studies from 12th CROI. NATAP. 2005. http://www.natap.org/2005/CROI/croi_49.htm.

4. Cabrera C, García E, Marfil S, et al. Genotypic evolution of T-20 resistance-associated mutations in heavily treated HIV-infected patients on long-term treatment with T-20. Third European HIV Drug Resistance Workshop. March 30-April 1, 2005. Athens. Abstract 61. Poster 9.6.

5. Labrosse B, Morand-Joubert L, Goubard A, et al. Continued selection of different viral quasi-species carrying similar HR1 mutations, in enfuvirtide-treated patients. Third European HIV Drug Resistance Workshop. March 30-April 1, 2005. Athens. Abstract 37. Poster 7.1.

6. Lu J, Sista P, Giguel F, Greenberg M, Kuritzkes DR. Relative replicative fitness of human immunodeficiency virus type 1 mutants resistant to enfuvirtide (T-20). J Virol 2004;78:4628-4637.

7. Deeks S, Lu J, Hoh R, et al. Interruption of enfuvirtide in patients with enfuvirtide resistance. 12th Conference on Retroviruses and Opportunistic Infections. February 22-25, 2005. Boston. Abstract 680.

8. Maroldo L, Coakley, Chappey C, et al. Rapid selection of high-level resistance to enfuvirtide. Third European HIV Drug Resistance Workshop. March 30-April 1, 2005. Athens. Abstract 95. Poster 10.5.

9. Si-Mohamed A, Piketty C, Tisserand P, et al. Increased polymorphism in the HR-1 gp41 env gene encoding the enfuvirtide target in HIV-1 variants harboring multiple antiretroviral drug resistance mutations in the pol gene. Third European HIV Drug Resistance Workshop. March 30-April 1, 2005. Athens. Abstract 99. Poster 10.9.

10. Vazquez de Parga E, Rakhmanova A, Pérez-Álvarez L, et al. Natural resistance-associated mutations to T-20 and polymorphisms in HIV-1 gp41 sequences of different genetic forms from Russia. Third European HIV Drug Resistance Workshop. March 30-April 1, 2005. Athens. Abstract 8. Poster 1.8.

11. Muñoz-Nieto M, Perez-Alvarez L, Carmona R, et al. Persistence and prevalence of drug-resistance strains to T-20, protease and retrotranscriptase inhibitors, in newly diagnosed HIV-1 patients from Galicia, Spain. Third European HIV Drug Resistance Workshop. March 30-April 1, 2005. Athens. Abstract 19. Poster 1.19.

12. Kuritzkes DR, Squires KE, Zolopa AR. Resistance and management of experienced patients. Clinical Care Options/HIV. 2005.