icon-folder.gif   Conference Reports for NATAP  
 
  International AIDS Conference
 
July 13-16, 2003, Paris
Back grey_arrow_rt.gif
 
 
 
Antiretrovirals in First Line regimens: Trizivir, Sustiva, quad vs Triple therapy, Induction-Maintenance
 
Reported by Dr Graeme Moyle, Chelsea & Westminster Hospital, London.
 
  TOPICS
 
--ACTG 5095: Trizivir vs AZT/3TC/efavirenz
--abacavir/tenofovior/3TC: high failure rates, regimen to avoid
--Quad vs triple therapy: induction-maintenance
 
The antiretroviral therapy sessions at the IAS conference did not contain a large basket of revelations. However, the sessions provided an opportunity to scrutinize the results of ACTG 5095, the outcome of which had only previously be available from a press release. The study, and a second cohort study using the combination of tenofovir, 3TC and abacavir contributed to a general gloom about the prospects for triple NRTI therapy as initial antiretroviral treatment. Other data, however, showed at least one way in which compact triple NRTI pills may be used in the future. Other data on approved agents in first line regimens largely ‘recycled’ presentations from previous meetings.
 
ACTG 5095
 
The ACTG A5095 study (abstract 41) is an on-going randomized, double blind, placebo controlled study of 3 protease inhibitor (PI)-sparing regimens: AZT/3TC/ABC fixed dose tablets (Trizivir™) 1 BID, AZT/3TC (Combivir™) 1 BID + efavirenz 600mg QD or Trizivir1 BID+ efavirenz 600mg QD in initial therapy. The study defined viral failure as 2 consecutive viral load values >200cps/ml after week 16. The study was closely monitored by a data and safety monitoring board (DSMB) with pre-defined stopping criteria. The DSMB (February 6, 2003) recommended discontinuation of the Triziviralone arm but blinded continuation of the other 2 arms based on demonstrated inferior efficacy in the Triziviralone arm relative to the other 2 arms after a median of just 32 weeks follow-up. No safety concerns were raised.
 
The study randomized 1147 treatment naive individuals (81% men, 40% white) with viral loads >400cps/ml (mean baseline viral load of 4.85log (71,434) cps/ml) and any CD4 count (mean at baseline 238 cells/mm3) 1:1:1 to these 3 arms. In total, 382 patients were randomized to Triziviralone and 765 to the two efavirenz-based regimens. Forty two percent of subjects entered with a baseline viral load >100,000 cps/ml.
 
Over follow up, 83% of individuals remained on their initial randomised therapy. Protocol permitted drug substitutions included 8% of individuals replacing AZT with d4T, 5% substituting ddI for abacavir and 6% replacing efavirenz with nevirapine. 167 patients in the study reached the protocol defined virological failure; 21% in the Trizivirarm and 11% in the other two arms combined (p<0.001 by log rank test). Time to virological failure was shorter in the Trizivirarm than the two efavirenz-containing arms combined (p<0.001). Importantly, this was true not only in those individuals who were randomized to Trizivirwith a viral load >100,000 cps/ml at baseline (p<0.001) but also those with <100,000 cps/ml at baseline (p=0.001). For subjects through week 48 the proportions with a viral load <200 cps/ml by intent to treat analysis was 74% (95%CI 65-83%) in the Trizivirgroup (n=127) and 89% (95% CI 85-93%) for the efavirenz arms (n=256). CD4 cell count changes did not differ between groups.
 
Resistance at viral failure in the Trizivir arm included 22% wild type, 34% M184V alosne, 11% with M184V + other RTI resistance associated mutations, 2% with RTI-resistance associated mutations alone and 31% where sequence could not be obtained (mainly due t low viral load <500cps/ml). Data from virological failures in the efavirenz arms were not presented.
 
The data are consistent with other studies of Trizivir in persons with high viral load, where responses inferior to two-class regimens have been observed but raise new concerns about the use of Trizivir in persons with low viral load.
 
Abacavir/3TC/Tenofovir: a triple NRTI regimen to avoid
 
A pilot study (abstract 42) investigating the utility of a once-daily triple NRTI regimen containing abacavir, lamivudine, and tenofovir in 20 treatment naive individuals was also stopped prematurely. The investigators defined non-responders as individuals who failed to achieve a reduction in viral load of > 2-log by eight weeks of therapy or a rebound in viral load after initial suppression. Patients entering the study had a mean baseline viral load of 147,164 cps/ml and a mean CD4 cell count of 777 cells/mm3. Of 19 individuals with available follow-up data, only 27 percent were considered responders and specifically 58 percent of individuals were virological failures. Individuals who responded to this once-daily triple NRTI regimen had a significantly lower baseline viral load (mean 4.173 log) compared with the virological failures (mean baseline viral load 5.098 log). Genotypic analysis indicated 45 percent of failures had the M184V mutations alone and 36 percent had M184V plus K65R. Comments by this investigator and a medical adviser from GlaxoSmithKline during the session indicated that a second larger study using this combination in treatment naive individuals has also been stopped prematurely due to similar poor efficacy. Pharmacokinetic or intracellular drug interactions studies have not been performed with abacavir plus tenofovir but the present time the most likely explanation is that this regimen has only a two mutation genetic barrier.
 
Quad Therapy as initial therapy: induction-maintenance
 
Two clear roles for Trizivirremain. As a simple maintenance therapy after a sustained complete virological response to an initial regimen and as part of subsequent multidrug regimen where a resistance tests supports inclusion of the components of Trizivirin the regimen.
 
Additionally, some physicians consider 4 drugs (such as Trizivir+ efavirenz, perhaps Trizivir+ tenofovir) may be preferred for initial therapy in individuals presenting with high viral load. Further follow-up of ACTG A5095 is likely to shed light on this view. Another on-going study is investigating initiation with a quadruple regimen Trizivir+ efavirenz regimen but then deintensifying half of the participants to Trizivirafter 48 weeks of therapy. The remainder of the patients will stay on Quad therapy for 96 weeks. This study (ESS40013) will help further clarify the role of Triziviras simplified maintenance therapy in a de-intensification or ‘induction-maintenance’ approach. However, compared to well chosen triple regimens, quad regimens are more costly, may risk more adverse effects and may add complexity without benefit. Their case remains to be proven.
 
The preliminary 48 week results for the ‘induction’ phase of ESS40013 (abstract 42), where 448 individuals with a mean baseline viral load of 5.04log (56% >100,000) cps/ml and CD4 cell count of 245 cells/mm3 received Trizivir 1 BID + Efavirenz 600mg QD, indicated a high rate (37%) of study discontinuation, most commonly (12%) secondary to drug-related adverse events. A further 5% were protocol defined virological failures and 1% investigator defined. In 22 individuals in which genotype results were available, the median VL at viral failure was 20,000 c/ml (i.e these were generally not low level failures). Twenty three percent of samples were wild-type or ‘as at baseline’ the remaineder included 14% with M184V alone, 23% with EFV resistance alone, 32% with both M184V+EFV resistance plus a futher 9% who had M184V+EFV + thymidine selected mutations. Two samples contained the K65R mutation, in one case just with EFV resistance, the other case with both M184V and EFV resistance.
 
Regarding adverse events, abacavir HSR was diagnosed in 7% of participants, this higher rate now commonly reported in clinical trials arising following a change in reporting recommendations.
 
Due to the large number of drop outs the ITT M=F analysis reported a disappointing 61% of patients had <50 copies/ml at week 48 whereas by on treatment analysis the response rate was 90% <50 copies/ml. Response at week 48 did not differ by baseline viral load although a significant difference in responders was observed between those who started therapy with a CD4 count >350/mm3 (97% <50cps/ml at week 48 as treated) and those with a baseline CD4 <50/mm3 (87% <50cps/ml at week 48 as treated, p=0.012). Thus for patients presenting with very low CD4 cell counts quadruple drug therapy may not be the optimal approach.
 
Consistent with previous trial analyses the patients who entered the study with very high (>750, 000) viral loads took a prolonged period to achieve <50cps/ml, the majority of patients in this group taking longer than 24 weeks (median 35 weeks) to reach <50cps/ml. This has some implications for consideration of the results of ACTG 5095 and raises the possibility that the definition of response in that study (<200cps/ml by week 16) may have been a hard endpoint if the speed of response differed between regimens.
 
Quad vs. Triple therapy
 
A small randomized open label study (abstract 615) involving 53 treatment naēve patients compared treatment responses to a standard triple therapy regimens of AZT/3TC (as Combivir) + EFV (n=26) with a quadruple therapy of AZT/3TC/ABC (as Trizivir) + EFV (n=27). No significant baseline differences were reported although CD4 cell count tended to be higher in the triple therapy group (155/mm3) compared to quadruple therapy (82/mm3). Viral load declined somewhat more slowly in the first 10 days of therapy with triple therapy (-1.79) vs. quad therapy (-1.92). Over 48 weeks the response to <50cps/ml by ITT analysis was 76.9% for the 3-drug regimen compared with the quad response of 63% (p value not significant). The median time to a viral load <5cps/ml was similar at 336 days for triple and 349 days for quad with a similar proportion (26.1 and 26.3%, respectively) <5cps/ml at week 48. Only 1 abacavir HSR was diagnosed. The authors suggested a sample size of 52 patients/arm would be required to show a 10% difference with 80% certainty. Thus, ACTG 5095 with over 600 patients on-going should be powered adequately to address this question.
 
These preliminary data certainly suggest more comparative studies of triple vs. quadruple therapy are needed and potentially alternative quad regimens proposed (such as AZT/3TC/ABC/Tenofovir) to move this approach forward.
 
Summary
 
The results of several studies indicate that caution should be used when considering the use of triple nucleoside analogues alone in initial therapy. In particular, the combination of abacavir/3TC/tenofovir appears associated with a high rate of viral failure/non-response and the emergence of both M184V and K65R mutations. Induction maintenance approaches are under investigation but, thus far, quadruple therapy has not demonstrated advantage over triple therapy. Patients presenting with viral loads >750,000cps/ml require prolonged follow-up (>24 weeks minimum) before considering whether treatment response is adequate. Those presenting with CD4 cell counts <50/mm3 require close monitoring for treatment response and may require multi-class regimens.