Raltegravir (MK-0518) phase 3 trials show successful viral suppression in multi-drug resistant patients

Simon Collins, HIV i-Base

Probably the most important results at the 2007 conference were the late breaker abstracts from the newly-named integrase inhibitor raltegravir (formerly MK-0518). [1, 2]

Based on early potency in experienced and naive patients in the phase 2 dose-finding studies seen last year [3], MK-0518 rolled out an early access programme in order to also be able to provide additional safety data as part of the regulatory submission. This fast-track application and early access, prior to results from phase 3 studies, indicated the potential and excitement generated by this new compound.

At CROI, David Cooper and Roy Steigbigel presented 16-week results from the phase 3 BENCHMRK-1 (Europe, Asia/Pacific, Peru) and BENCHMRK-2 (US, Canada and South America) studies. Both studies had the same design and similar patient characteristics and results, which were largely combined for these presentations.

Approximately 350 three-class resistant patients on failing therapy enrolled in each study. Patients optimised their background regimen (OBT) based on treatment history and resistance tests, and were randomised 2:1 to add raltegravir 400mg twice daily or placebo. Primary endpoints included viral load, CD4 and tolerability at week 16, after which patients could receive open-label raltegravir.

Mean baseline CD4 and viral loads were approximately 150 cells/mm3 and 4.6 log copies/mL respectively. Patients were highly treatment experienced with over 90% having had a previous AIDS diagnosis, and a median of ten years previous ARV therapy with a median of 12 antiretroviral drugs. Approximately 30% had a genotypic sensitivity score (GSS) of zero for the drugs in the OBT and 40% had a GSS of 1. When used for the first time, daruanavir and T-20 each counted as 1 for patients naive to these drugs. Approximately 25% of patients used T-20 across both studies. Darunavir use was higher in BENCHMRK-2 (approximately 50% vs 25% in BENCHMRK-1) due to the earlier access programmes for darunavir in the US. Mean age was 45, 85-90% of participants were male, and 75-80% and 55-65% were Caucasian in BENCHMRK-1 and -2 respectively.

At 16 weeks, using Non-Completer=Failure analysis, approximately 77% of patients achieved viral suppression to <400 copies/mL in the raltegravir arm compared to just over 40% in the placebo group. Results down to <50 copies/mL were just over 60% vs 33-35% and are detailed in Table 1. In both studies CD4 increases were approximately +80 cells/mm3 vs + 30 cells/mm3 in favour of raltegravir, and average viral load changes were -2 logs vs -1 log reductions, with these being sustained out to 24 weeks in patients for whom extended follow-up data was available (approximately half the study group). All differences between raltegravir and placebo were statistically significant (P<0.001).

Table 1: 16-week virologic response in raltegravir EAP Phase 3 studies

0518 + OBT placebo + OBT 0518 + OBT placebo + OBT
n 232 118 230 119
% <400 copies/mL 77% 41% 77% 43%
% <50 copies/mL 61% 33% 62% 36%
Combined results 0518 +OBT placebo + OBT
CD4 change +80 +30
Viral load drop -2 log -1 log
Virological failure 16% (n=76) 51% (n=121)

As with results from any new drug, an analysis of the choice of drugs in the optimised background regimen is essential in order to understand the individual impact of each component. Results from these sub-analyses are almost as important as the overall results. Results stratified by OBT and by baseline values are detailed below in Tables 2, 3 and 4.

Table 2 provides particularly important early results. 98% (44/45) patients using raltegravir with T-20 and darunavir for the first time achieved viral reductions to <400 copies/mL – compared to an impressive 87% in the placebo + OBT group. Using either one of these drugs in the OBT dropped response rates to 90% in the raltegravir arm (vs 55-63% in the placebo groups). Using neither drug still produced a 74% response in the raltegravir group (vs 29% with placebo).

While this may make understanding the contribution of each drug a challenge, the availability of other sensitive drug to support the investigational agent is both essential for sustained benefit and to minimise the short term wasting of new agents for patients who take part in registrational trials.

Table 2: % pts suppressed to <400 copies/mL by use of darunavir and T-20 as new drugs in OBT

T-20 DRV 0518 + OBT placebo + OBT
+ + 98% (n=44) 87% (n=23)
+ 90% (n=42) 63% (n=24)
+ 90% (n=80) 55% (n=47)
74% (n=191) 29% (n=90)
Overall 79% (n=447) 43% (n=230)

Table 3: % pts suppressed to <400 copies/mL by baseline genotypic (GSS) and phenotypic (PSS) sensitivity scores

PSS 0518 + OBT placebo + OBT
0 61% (n=62) 5% (n=44)
1 76% (n=141) 41% (n=68)
2 87% (n=242) 57% (n=110)
GSS 0518 + OBT placebo + OBT
0 57% (n=111) 10% (n=63)
1 85% (n=170) 43% (n=93)
2 89% (n=158) 71% (n=70)

Baseline viral load had an impact on response rate (88%/64% for raltegravir, when above/below 100,000 copies/mL versus 55%/19% in the placebo arm). Patients with baseline CD4 cells count <50 cells/mm3 were around 20% less likely to achieve a virological response in both arms, compared to patients with higher baseline counts.

Table 4: % pts suppressed to <400 copies/mL by baseline CD4 and viral load

0518 = OBT placebo + OBT
n 447 230
Baseline VL <100,000 copies/mL 88% (n=228) 55% (n=155)
Baseline VL >100,000 copies/mL 64% (n=159) 19% (n=75)
CD4 <50 copies/mm3 63% (n=140) 24% (n=75)
CD4 >50-<200 copies/mm3 86% (n=168) 46% (n=82)
CD4 >200 copies/mm3 88% (n=138) 59% (n=73)

A summary of clinical and serious laboratory adverse events in this advanced population showed no significant differences between treatment arms (approximately 80% patients had at least one side effect, only 10% of which were classed as serious and <3% classed as a drug-related serious event, with more drug-related discontinuations in the placebo group). There were no specific side effects relating to raltegravir over placebo that showed a consistent concern in the two studies.

Virological failure was reported in 16% (n=76) of patients receiving raltregavir and 51% on placebo (n=121).

Genotypic analysis available from 41 patients showed that 75% (n=32) developed mutation in integrase and that this followed one of two main pathways: either via N155 (with additional mutations at E92Q, V151I, T97A, G163R, L74M) or Q148K/R/H (withG140S/A, andE138K). These were similar to mutations seen from earlier in vitro passaging studies.


Using RGV with T-20 and DRV reduced viral load to <400 copies/mL in 98% of these highly treatment experienced patients. These results are startlingly high – even given that this is only 16 week data – because we are accustomed to sub-optimal response to treatment in both first line and MDR therapy. But these are the results that we should be looking for from treatment, given that 3 sensitive drugs including two entirely new classes were being used, and that adherence is likely to be very high in such advanced patients in a clinical trial.

For patients with three-class MDR-HIV, using three sensitive drugs should now be standard of care. Since this programme was first available, maraviroc and etravirine have also become widely available in EAP programmes. Gambling with only two sensitive drugs will cause some patients to loose this drug and new class.

Raltegravir’s resistance barrier may be slightly higher than NNRTIs but it is not as resilient as protease inhibitors and mutations rapidly accumulated over 16 weeks in nearly three quarters of those people who failed to achieve viral suppression.

The conference also included in vitro data on candidate second-generation compounds (including MK-2048) which are planned to have once-daily PK and activity against raltegravir-resistant virus. [4]

Although raltegravir has the potential to be used strategically at many timepoints in a patients’ treatment history, pricing will determine whether this includes widespread use in first line therapy and as a treatment to switch for greater tolerability.

The expanded access programme for raltegravir will continue to run until the drug is approved. Merck have issued a statement saying that they will reimburse the difference between the current high EAP price (set close to that of T-20) and the final marketing price after raltegravir receives approval. They also say that this high EAP price was set in France when raltegravir was initially only being developed and a drug for use in last-line therapy.

For more details in the UK contact the medical information department at MSD on 01992-467272.

Patients can be registered for the expanded access programme at:


  1. Cooper D, Gatell J, Rockstroh J et al. Results of BENCHMRK-1, a Phase III Study Evaluating the Efficacy and Safety of MK-0518, a Novel HIV-1 Integrase Inhibitor, in Patients with Triple-class Resistant Virus. Oral abstract 105LBa.
  2. Steigbigel R, Kumar P, Eron J et al. Results of BENCHMRK-2, a Phase III Study Evaluating the Efficacy and Safety of MK-0518, a Novel HIV-1 Integrase Inhibitor, in Patients with Triple-class Resistant Virus. Oral abstract 105LBa.
  3. Results from the Phase 2 studies in naive patients (at CROI 2006) and experienced patients (at IAS 2006) were reported in the March and September 2006 issues of HTB.
  4. Wai J et al. Next Generation of Inhibitors of HIV-1 Integrase Strand Transfer Inhibitor: Structural Diversity and Resistance Profiles . Oral abstract 87. The oral presentations can be viewed online from the CROI website (see Tuesday, 6.30-7.30pm, New Antiretrovirals Late Breaker session; and Tuesday, 10.00-12.00am New ARV Agents, Resistance Mechanisms, and Clinical Resistance).

Links to other websites are current at date of posting but not maintained.