icon-folder.gif   Conference Reports for NATAP  
 
  6th Intl Congress on Drug Therapy in HIV Infection
 
Glasgow, Scotland Nov 17-20, 2002
Back grey_arrow_rt.gif
 
 
 
6th Intl Congress on Drug Therapy in HIV
 
Written by Graeme Moyle, MD, Chelsea & Westminster Hospital, London, UK
 
  Under the cold grey skies of Glasgow around 2500 HIV researchers meant to discuss the hot topics in HIV therapeutics over 17 - 21 November 2002. There was only modest amount of new data presented much of the discussion being developed by plenary speakers.
 
TOPICS:
--Gilead 903 study: tenofovir resistance outcomes
--Enfuvirtide (T-20): Predicting success and modelling survival benefits
--Induction-Maintenance with Trizivir
--Treatment interruption in the setting of resistance
 
Gilead 903 study: tenofovir resistance outcomes
 
The Gilead 903 study is a 600 patients randomised placebo-controlled study comparing tenofovir 245 mg once (300 mg pill) daily with d4T 40 mg twice daily (weight adjusted, immediate release formulation) combined with efavirenz and lamivudine, which has thus far completed 48 weeks of follow-up (in treatment-naive patients). By intention to treat missing equals failure analysis 82 percent of tenofovir recipients and 81 percent of d4T recipients had viral load is less than 50 copies/ml at week 48. CD4 benefits have also been similar in the two groups. Individuals meeting definition for virological failure were described. In the tenofovir group 29 individuals (9.7% of those randomised) and in the d4T group 25 individuals (8.3%) were defined as virological failure. Mutations associated with efavirenz resistance were found in 16 tenofovir failures and 12 d4T failures. The 3 TC signature mutations 184 V was found in 12 tenofovir failures and eight d4T failures. Wild type virus was detected in 10 tenofovir failures and 12 d4T failures. In vitro tenofovir is known to select for the K65R mutation, a mutation which may be associated with reduced viral fitness but when combined with other mutations may affect the activity of 3 TC, abacavir and ddI. No patient in either group had the K65R mutation. However, K65 was found in association with an efavirenz resistance mutation alone in 2 tenofovir failures and one d4T failure, and in combination with both an efavirenz resistance mutation and 184V in five tenofovir failures and one d4T failure. Overall, this meant that seven individuals in the tenofovir group of 29 with virological failure had the K65R mutation compared with two of the 25 virological failures in the d4T group having K65R. Assessment of the seven isolates from the tenofovir group using both the Virco and ViroLogic phenotype assays was performed. Using the Virco assay, three of the isolates with K65R were resistant to ddI, three resistant to abacavir and four resistant to tenofovir. A number of the isolates showed modest increases in susceptibility to AZT via both phenotypic assay. Follow-up information on all seven individuals with K65R associated failure in the tenofovir group were reported. All patients changed therapy, in general to regimens that contained a thymidine analog and protease inhibitor or ddI and a protease inhibitor. Five of these individuals had achieved viral load is less than 50 copies/ml at last follow-up the other two individuals had detectable virus below 2000 copies, with one individual lost to follow-up and a second individual known to be incompletely adherence to medication. These data would suggest that re-establishment of an optimal virological response is achievable in individuals who experience viral rebound associated with resistance mutations including the K65R mutation.
 
Enfuvirtide (T-20): Predicting success and modelling survival benefits
 
Enfuvirtide (formerly T-20) has been evaluated into randomised, comparative studies in which individuals who have been experienced with all three approved drug classes were given a new antiretroviral regimen optimised by the use of resistance testing with or without enfuvirtide in a 2:1 ratio. Both studies, known as TORO 1&2, demonstrated substantial virological and immunological advantage to including enfuvirtide in the new treatment regimen. Characteristics associated with treatment response in the Toro 1 study were recently reported at the IC AAC conference. A similar analysis of the Toro 2 study was presented at the Glasgow conference. In Toro 2, 335 individuals received enfuvirtide with optimised background therapy and 169 individuals received optimised background therapy alone. The changing baseline viral load over 24 weeks was -1.43 log in the enfuvirtide recipients and -.65 in those who received optimised background alone. This advantage of 0.78 log was highly statistically significant ( p<0.0001). Enfuvirtide recipients experienced a rise of 65 CD4 cells compared with just 38 in the control group. The number of individuals who reached a viral load less than 400 copies was 28 percent in the end for the enfurtide group and 14 percent in the control group, with 12 percent and 5% respectively achieving less than 50 copies/ml.
 
Subgroup analyses indicated that the benefits of the enfuvirtide were consistently observed in both male and female, white and non-white and all patients that were older and younger than 40 years of age. Of note, individuals who entered the study with an age less than 40 years experienced a more modest and not statistically significant advantage for enfuvirtide relative to background therapy alone, whereas individuals who entered the trial with an age greater than 40 years, a greater and statistically significant advantage was observed. When groups were stratified by baseline viral load less than or greater than 40,000 copies/ml and by baseline CD4 less than or greater than 100 cells/mm3, significant and consistent advantages to enfuvirtide over optimised background therapy alone were observed.
 
Using resistance information derived from baseline samples patients were given genotypic and phenotypic sensitivity scores whereby a score of one was given for every drug included in the optimised background that was likely to be active in the regimen. Individuals with a genotypic sensitivity score of two or more achieved the greatest reductions in viral load from baseline, however, individuals randomised to enfuvirtide had consistently larger reductions in viral load from baseline relative to those who only received optimised background regardless of the genotypic sensitivity score. Similar benefits were seen when evaluating the phenotypic sensitivity scores, although individuals with phenotypic sensitivity scores of three or more in the optimised background therapy group in this analysis achieved similar responses to the enfuvirtide group, whereas the enfuvirtide group had significant advantage in those individuals entering the study with a phenotypic sensitivity score of two or less.
 
As reported in the analysis of the Toro 1 study, individuals naive to Kaletra at baseline achieved larger reductions in viral load than those with prior Kaletra experience. When stratifying individuals by prior Kaletra experience, individuals randomised to enfuvirtide had significantly greater reductions in viral load from baseline both in individuals naive to Kaletra and experienced with Kaletra at entry to the study.
 
In a multiple regression analysis investigating factors associated with the change in viral load from baseline to week 24, inclusion of enfuvirtide in the regimen was associated with an estimated 0.8 log reduction in viral load, factors including lower baseline viral load, higher baseline CD 4 count, phenotypic or genotypic sensitivity score indicating greater number of active drugs in the optimised background regimen, and better adherence were associated with 0.1 to 0.3 log reductions in viral load. Individuals with prior Kaletra experience were estimated to have a 0.86 log smaller reduction in viral load than those naive to Kaletra at study entry.
 
These subgroup analyses of the TORO 1 and 2 studies provide useful practical guidance as to how we should use enfuvirtide in clinical practice. The studies indicate that the best response is to a regimen containing enfuvirtide are seen in individuals with at least two and preferably 3 active agents to include in the accompanying regimen, and that treatment response may also be better if commenced with a relatively lower viral load and a higher CD 4 count. Support for adherence appears critical, with a 10 percent increase in adherence in the Toro 2 study being associated with an estimated 0.11 log greater reduction in viral load. There are also issues about the use of Kaletra raised by the studies. If being naive to Kaletra at the time of initiation of enfuvirtide is truly critical to achieving the greatest virological response to an enfuvirtide containing regimen, then physicians may need to consider "saving" Kaletra for the time in the treatment sequence that they may plan to use enfuvirtide.
 
Results from the Toro 1 and 2 studies were placed in a complex mathematical model to estimate the advantage of enfuvirtide relative to optimised background therapy alone in delaying the time to development of an AIDS defining event or death. The model suggested that inclusion of enfuvirtide in the regimen would delay the development of a new AIDS defining event by approximately 1.5 years (from 3.3 years in the optimised background regimen to 4.8 years in the enfuvirtide containing regimen) and increase overall survival by 1.6 years (from 4.8 years in the optimised background alone regimen to 6.2 years in the enfuvirtide containing regimen). These information suggest that enfuvirtide is likely to be a highly cost-effective therapy and that it is likely to Œbuyı not only years of good quality (due to the prevention of new AIDS defining illnesses) but also time for new treatment options to become available.
 
Induction-Maintenance with Trizivir
 
Patients who commenced therapy with CD4 cell counts less than 100/mm3 or viral load > 100,000 copies/ml may not respond as completely or as durably to some treatment regimens. Unboosted protease inhibitor based triple therapy and triple nucleoside regimens may both be included in this category, whereas diminution of treatment effect in these patients with challenging baseline characteristics has not been observed in a number of studies with efavirenz or Kaletra based regimens. Some physicians, however, consider the individuals who present with poor baseline characteristics may be best initially treated with a quadruple therapy regimen. It is not currently known whether this provides greater efficacy relative to well chosen triple therapy regimens or if quadruple regimens simply add an additional burden of toxicity, complexity and cost to antiretroviral therapy. A cohort study which recruited individuals with CD 4 cell counts less than 200/mm3 was reported at conference. Participants initiated on Trizivir (ZDV/3TC/ABC) plus efavirenz (with the option to switch to nevirapine) and followed over 24 weeks. The study included 55 individuals, with a median baseline viral load of 554,548 copies/ml and CD 4 cell counts of 64 at baseline. By week 24, 53% individuals by ITT analysis had achieved a viral load less than 50 copies/ml, 72% by as treated analysis. The rise in CD4 count was 142 cells/mm3. Five percent of individuals discontinued the study due to suspect had abacavir hypersensitivity reactions.
 
Previous analyses of treatment response in individuals initiating with high baseline viral those indicate that many individuals may take > 24 weeks before a viral load less than 50 copies/ml is achieved. The results may be prematurely being presented to be considered useful at present.
 
A second study of quadruple therapy in 51 individuals treated over 48 weeks was reported. In the study individuals received Trizivir plus nelfinavir for 48 weeks and then discontinued the nelfinavir if their viral load was less than 50 copies/ml. Eighteen of the originally randomised individuals had discontinued nelfinavir and had been followed on Trizivir alone for a further 24 weeks. All 16 of the individuals who remained on study had sustained viral suppression below 50 copies/ml. Two patients discontinued the study for reasons other than virological rebound or adverse event (one for pregnancy, and one for protocol violation). While it is unclear why data from an ongoing study were presented in this way, rather than waiting until the study was complete, they provide support for the idea that individuals on first-line therapy who had achieved optimal treatment response on their initial regimen may safely switch or deintensify to a trizivir maintenance regimen.
 
Treatment interruption in the setting of resistance
 
Treatment interruptions are under investigation in a range of settings although in clinical practice, probably the most common use is in individuals who have virus resistant to the agents that currently receiving. After an interruption of treatment and the removal of drug selection pressure the undergrowth of a more sensitive and sometimes fully sensitive, more fit virus may be seen. Some of studies and clinical cohorts have suggested that outgrowth of a more sensitive virus during treatment interruptions may lead to a better treatment response when a new regimen is commenced relative to simply switching from the failing regimen to the new regimen. It is however well evidenced that the resistant virus remained archived within the individuals viral pool and may therefore readily re-emerge when the appropriate selection pressure is re-established. Research from Italy described the analysis of a clinical cohort in which some patients had an interruption of therapy for at least 60 days commencing when the viral load was > 400 copies/ml on therapy and then commenced a new regimen on which 48 weeks of follow-up was available. Amongst the 56 individuals described the median viral load at the time of treatment interruptions was 4.4 log copies/ml, the individuals had been exposed to a median of eight drugs and two or three approved drug classes. Patients interrupted for a median of 112 days during which time the viral load increased 55.18 log copies/ml. Perhaps more important with regards to risk were the changes in CD 4 cell count where the mean CD4 value declined from 344/mm3 to 266/mm3. The authors did not report whether clinical events were observed during this time, however, a previous analysis from the EuroSIDA database has indicated that individuals with a CD 4 count of 200/mm3 or less at the time of treatment interruption have a high risk of experiencing HIV disease related events even during short periods of interruption.
 
When a new treatment was commenced in these individuals 34 percent of individuals had achieved and sustained a viral load less than 400 copies/ml through to week 48. The average viral load at week 48 had declined to 3.8 log copies/ml. In a multivariate analysis, lower viral load before treatment interruption and longer duration of treatment interruption were both associated with a greater chance of achieving a viral load less than 400 copies/ml at week 48. There were four AIDS defining events reported during the 48 weeks of therapy. All occurred in those patients who had CD4 counts below 200/mm3 at the time of interruption.
 
The study provides some support to the idea that in individuals with CD 4 nadirs > 200/mm3 who have began to experience low-level virological rebound with resistant virus on a current treatment regimen, that a treatment interruption prior to the initiation of the next line of therapy may be a reasonable approach to consider. A large randomised study is currently ongoing to evaluate this approach.