icon- folder.gif   Conference Reports for NATAP  
 
  15th International HIV Drug Resistance Workshop
June 13- 17, 2006
Sitges, Spain
Back grey_arrow_rt.gif
 
 
 
Resistance to Enfuvirtide, Vicriviroc, and TMC125
 
 
  Resistance Workshop Review: Part 6
 
Mark Mascolini
 
The Resistance Workshop served up three intriguing reports of new or investigational antiretrovirals--the fusion inhibitor enfuvirtide, the CCR5 antagonist vicriviroc, and the nonnucleoside reverse transcriptase inhibitor (NNRTI) TMC125. But the conclusions proffered by study authors did not always win unanimous approbation.
 
Sustained CD4 gains with failing enfuvirtide?
Reports that CD4 gains made with enfuvirtide persist even after resistance to this fusion inhibitor emerges have been in the air--and in the literature [1]--for a few years. At the 2006 Conference on Retroviruses, Italian clinicians tied these abiding T-cell spurts to emergence of a V38A or V38E mutation in HIV's envelope gene [2]. People whose virus picked up an enfuvirtide resistance mutation at V38 maintained CD4 gains even as their viral load rebounded; people with different enfuvirtide mutations lost CD4 cells after rebound.
 
Sustained CD4 gains during virologically failing treatment are hardly a new phenomenon. But, if confirmed, that immunologic trick could have particular relevance for people taking enfuvirtide. Because enfuvirtide is the only antiretroviral that works by disrupting the envelope protein gp41, continuing the drug after virologic failure could seem a tempting strategy when keeping CD4 tallies high becomes the prime goal of therapy: In this scenario, further pileup of resistance mutations with an imperfect regimen would seem to matter less than keeping CD4 counts out of the danger zone.
 
To see whether mutations at V38 had the same CD4 effect in a bigger group, Tom Melby and Trimeris colleagues rounded up viral samples from the TORO trials of enfuvirtide-based salvage therapy [3,4]. The Trimeris team compared CD4 changes in three groups whose enfuvirtide regimen failed and who had CD4 data 8, 24, and 48 weeks after failure [5]:
 
- 58 people with a V38A/E mutation but without mutations at Q40 or N43
- 31 people with Q40H and/or N43D/K but without V38A/E
- 45 people with other or no mutations
 
The groups did not differ significantly in viral load or CD4 count when the TORO studies began. But a statistical analysis adjusted for viral load 8, 24, and 48 weeks after virologic failure determined that the V38 group actually gained CD4 cells after failure, while the other two groups lost CD4s (Table 1). Differences in CD4 swings between the V38 group and the other two groups were statistically significant 24 and 48 weeks after virologic failure.
 

Least-1.gif

Melby also found that CD4 counts waned faster in people with the Q40H mutation than in those with N43D or N43K.
 
Do these findings mean that--if your enfuvirtide is going to flop--it's better to end up with a V38A/E mutation than with others? No one at the workshop seemed ready to endorse that strategy. Melby himself pointed out that the CD4 comparisons could be thrown off by three factors: (1) "survivor bias," meaning people whose CD4 gains looked solid would tend to stay in the study while others would tend to drop out, (2) differing potency of the other drugs being taken with enfuvirtide, and (3) multiple virologic failure criteria, meaning virus was sampled for resistance testing at different points after viral rebound.
 
That last factor concerned some workshop attendees, who observed that other work shows V38A/E being replaced by N43D in the weeks after virologic failure of enfuvirtide. Because Melby searched for mutations only at virologic failure, N43D may have superseded V38A/E in some patients in this analysis during the time when their CD4 counts continued rising. If this pattern proved true in more than a few people, it could erase the mutation-based CD4 difference plotted by Melby.
 
Resistance to vicriviroc--or not?
CCR5 antagonists--only a year ago the brightest stars on the antiretroviral horizon--have endured a (sometimes fiery) fall from grace. Vicriviroc trial planners raised hackles, then groans, with an unusual study of this CCR5 plugger in people with a virginal antiretroviral record [6]. Treatment advocates worried about toxic risks of these novel drugs in people with no immediate need for such therapy. Others mulled the wisdom of a 2-week dose-ranging vicriviroc monotherapy run-in before the traditional randomized phase of the study.
 
When that trial ended early because vicriviroc plus Combivir did not match efavirenz/Combivir virologically, developer Schering-Plough found itself at pains to prove that resistance emerging during monotherapy did not account for vicriviroc's failure. And that was the point of Raphael Landovitz's report at the Resistance Workshop [7].
 
The study began when 92 people with no more than 2 weeks of antiretroviral experience took either placebo or 25, 50, or 75 mg of vicriviroc once daily. After 2 weeks people taking vicriviroc added Combivir (fixed-dose AZT/3TC) and people randomized to placebo started Combivir plus efavirenz. The trial ended when virologic breakthroughs in the vicriviroc group confirmed efavirenz as the superior first-line option [6]. In the control arm about 90% of enrollees had a viral load under 50 copies/mL 350 days after randomization, compared with about 80% in the 75-mg vicriviroc arm (not quite a significant difference at P = 0.13), about 50% in the 50-mg vicriviroc arm (P = 0.003), and about 25% in the 25-mg arm (P < 0.001).
 
Pretreatment 50% inhibitory concentration (IC50), fold change in IC50, percent maximal suppression, and relative percent maximal suppression before treatment began did not predict virologic breakthrough. So Landovitz tweezed out env (envelope gene) sequences from viral samples of 4 people in whom vicriviroc failed at baseline, on day 14, at week 24, and around the end of the study.
 
Although these 4 failures all took different trajectories, they shared certain features:
 
- Mutations in env cropped up in everyone.
- The M184V mutation, conferring resistance to 3TC, arose in everyone after they started Combivir (and the AZT-related M41L mutation reared up in 1 of the 4).
- Relative percent maximal suppression dropped in everyone.
 
Landovitz noted that the vicriviroc-related mutations in these people showed no consistent pattern and that they differed from those reported earlier for maraviroc and two investigational Schering CCR5 antagonists. Furthermore, evolution of these mutations did not appear to touch off drops in viral susceptibility to vicriviroc. (For details on the 4 cases of vicriviroc failure, see note 8.)
 
Workshop attendees, however, did read these data as evidence that HIV became resistant to vicriviroc during the trial, though resistance may not have arisen during vicriviroc monotherapy. Although HIV scouted out a different genotypic path to evade vicriviroc in each of these patients--paths different from those paved by maraviroc or earlier Schering CCR5 candidates--attendees surmised that may just mean predicting genotypic resistance to vicriviroc will be difficult.
 
The disconnect between emergence of env mutations and a shift in IC50, some attendees proposed, reflects the still-rudimentary phenotypic tool used to measure loss of susceptibility to CCR5 antagonists. The IC50 curves plotted by the Monogram susceptibility assay, they said, simply do not behave the way such curves do when resistance emerges to other antiretroviral classes.
 
Regardless of whether these critics are correct, much clearly remains to be learned about env mutations, loss of susceptibility, and virologic failure of CCR5 antagonists. (And no one was happy that M184V appeared in all 4 failures that Landovitz described.)
 
Which nonnuke failures can TMC125 rescue?
Evidence from a phase 2 study of TMC125, a Tibotec NNRTI, at a dose of 800 mg twice daily plus a background regimen, suggests that virus with the class-killing K103N mutation will respond to TMC125 as well as virus without that mutation (9). But it seems that virus laden with another common NNRTI resistance mutation, Y181C, will not cave in to TMC125 as readily as virus without that mutation. Resistance experts at the workshop cautioned, though, that findings based on a phase 2 study must be called preliminary, even though Johan Vingerhoets and Tibotec colleagues confirmed their results in a much larger batch of viral isolates on tap in Virco's vats.
 
The trial randomized 199 people with NNRTI resistance to an optimized regimen without TMC125 or to an optimized combo plus 400 or 800 mg of TMC125 twice daily. (A new formulation of TMC125 lowers the dose to 200 mg twice daily.) After 48 weeks a noncompleter-equals-failure analysis charted a 0.14-log copies/mL average drop in the control group versus 0.88 log in the 400-mg group (P = 0.018 versus control) and 1.01 log in the 800-mg contingent (P = 0.002 versus control).
 
Looking at prestudy resistance to TMC125 only in people taking 800 mg of the new nonnuke twice daily, Vingerhoets recorded an overall 1.7-fold change in 50% effective concentration (EC50) of TMC125 (compared with nonmutant virus) in all 197 viral isolates versus a virtually indistinguishable 1.6-fold change in EC50 for the 73 isolates marked by K103N. In contrast baseline fold change measured 3.1 in 75 isolates with Y181 C. All viruses bearing K103N or Y181C also carried other NNRTI-related mutations.
 
Measuring virologic response at week 24, Vingerhoets recorded similar viral load drops with and without K103N but a weaker response with Y181C in an analysis adjusted for baseline viral load and CD4 count, use of enfuvirtide, and total number of NNRTI mutations (Table 2).
 

TableAve-2.gif

Next Tibotec gauged susceptibility to TMC125 in 7582 viral samples sent to Virco for resistance testing, including 1895 with K103N with or without other NNRTI mutations and 1277 with Y181C with or without other NNRTI mutations. The average fold change in susceptibility to TMC125 measured 0.9 for 1407 viral samples with only one NNRTI mutation, 0.8 for 492 samples with only K103N, and 3.4 for 154 samples with only Y181C.
 
These analyses all indicate that the otherwise baleful K103N mutant will yield to TMC125 but Y181C will not respond as well as nonmutant virus. Whether this means TMC125 has a higher barrier to resistance than efavirenz or nevirapine--or just copes well with K103N--remains to be seen.
 
Data Tibotec unveiled at the European Resistance Workshop earlier this year suggest that novel NNRTI resistance mutations may conspire to defeat this new nonnucleoside (10). In this phase 2 trial analysis the Tibotec team found that seven mutations--always combined with up to four other mutations--jacked up resistance to TMC125 at least 10-fold: K101P, V179E, V179F, Y181I, Y181V, G190S, and M230L. Y181C always appeared as one of the additional mutations with V179E, V179F, G190S, and M230L. This analysis also showed a waning 24-week virologic response to TMC125 as mutations piled up, but even virus with three or more NNRTI mutations (in 29 people) averaged a 0.66-log drop in viral load.
 
Mark Mascolini writes about HIV infection
 
References and Notes
 
1. Poveda E, Rodes B, Labernardiere JL, et al. Evolution of genotypic and phenotypic resistance to enfuvirtide in HIV-infected patients experiencing prolonged virologic failure. J Med Virol 2004;74:21-28.
2. Aquaro S, Svicher V, D'Arrigo R, et al. Characterization of gp41 evolution in a large cohort of HIV-1-infected patients receiving long-term T-20 treatment as a single active drug. 13th Conference on Retroviruses and Opportunistic Infections. February 5-8, 2006. Denver. Abstract 596.
3. Lalezari JP, Henry K, O'Hearn M, et al. Enfuvirtide, an HIV-1 fusion inhibitor, for drug-resistant HIV infection in North and South America. N Engl J Med 2003;348:2175-2185.
4. Lazzarin A, Clotet B, Cooper D, et al. Efficacy of enfuvirtide in patients infected with drug-resistant HIV-1 in Europe and Australia. N Engl J Med 2003;348:2186-2195.
5. Melby T, DeSpirito M, DeMasi R, et al. Impact of enfuvirtide resistance genotype on CD4 increases in patients with ongoing viral replication while receiving enfuvirtide. XV International HIV Drug Resistance Workshop. June 13-17, 2006. Sitges, Abstract 35.
6. Greaves W, Landovitz R, Fatkenheuer G, et al. Late virologic breakthrough in treatment-naive patients on a regimen of Combivir + vicriviroc. 13th Conference on Retroviruses and Opportunistic Infections. February 5-8, 2006. Denver. Abstract 161LB.
7. Landovitz R, Faetkenhauer G, Hoffmann C, et al. Characterization of susceptibility profiles for the CCR5 antagonist vicriviroc in treatment-naive HIV-infected subjects. XV International HIV Drug Resistance Workshop. June 13-17, 2006. Sitges, Abstract 18.
8. First vicriviroc failure: A person who started treatment with 50 mg of vicriviroc once daily had a 1-log drop in viral load but then a rebound at day 14 of monotherapy. Adding Combivir pushed the viral load down to 58 copies/mL at day 113, but the effect didn't last. M184V and an increased relative percent maximal suppression were recorded as the rebound continued. V3 envelope sequencing revealed the K10R mutation, which appeared to confer a significant drop in relative percent maximal suppression. Second vicriviroc failure: This person had a good virologic response to 50 mg of vicriviroc; adding Combivir drove the load below 50 copies/mL, but only by day 80. M184V emerged upon viral rebound at day 150, and relative percent maximal suppression increased. V3 sequencing revealed the mutations Q18E and D25G, which moderately lowered relative percent maximal suppression. Addition of G17A to the first two mutations conferred high-level resistance to vicriviroc. Third vicriviroc failure: Viral load fell 1 log copies/mL with 75 mg of vicriviroc monotherapy, but adding Combivir did not push the load down farther. As in the first two cases, M184V arose on viral rebound and relative percent maximal suppression increased. V3 sequencing turned up the N29D and Y34H mutations, which correlated with a drop in relative percent maximal suppression. The later appearance of P21I and T23P further decreased relative percent maximal suppression. Fourth vicriviroc failure: The 75-mg dose of vicriviroc lowered this person's viral log 1 log, but adding Combivir failed to push viremia below 50 copies/mL. Viral load lingered around 1000 copies/mL through 230 days of follow-up. The M184V and M41L reverse transcriptase mutations evolved upon continued treatment with this faulty regimen. The vicriviroc-related mutations H14L, A20F, D24N, R25GT, E/Q32K, and G17W decreased relative percent maximal suppression.
9. Vingerhoets J, Janssen K, Welkenhuysen-Gybels J, et al. Impact of baseline K103N or Y181C on the virological response to the NNRTI TMC125: analysis of study TMC125-C223. XV International HIV Drug Resistance Workshop. June 13-17, 2006. Sitges, Abstract 17.
10. Vingerhoets J, Peeters M, Corbett C, et al. Impact of baseline resistance on the virologic response to a novel NNRTI, TMC125, in patients with extensive NNRTI and PI resistance: analysis of study TMC125-C223. 4th European HIV Drug Resistance Workshop. March 29-31, 2006. Monte Carlo. Abstract 52.