icon-folder.gif   Conference Reports for NATAP  
  5th IAS Conference on HIV Pathogenesis, Treatment and Prevention
July 19th-22nd 2009
Capetown, South Africa
Back grey_arrow_rt.gif
Management of Treatment-Experienced Patients
  Conference Reports for NATAP
5th IAS Conference on HIV Pathogenesis, Treatment and Prevention Cape Town, South Africa, 19-22 July 2009
Rafael E. Campo, MD
Professor of Clinical Medicine
Division of Infectious Diseases
University of Miami Leonard M. Miller School of Medicine
Miami, Florida
Abbreviations used
CI - confidence interval
c/ml - copies/milliliter
HAART - highly active antiretroviral therapy
ITT - intent to treat
NRTI - nucleoside reverse transcriptase inhibitor
NNRTI - non-nucleoside reverse transcriptase inhibitor
PI - protease inhibitor
PP - per protocol
TLOVR - time to loss of virologic response
The recently concluded 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention had a fair number of interesting studies dealing with the management of treatment-experienced patients. Some of the general themes presented included treatment modifications to simplify the long-term management of these individuals, greater tolerance to diminished adherence in patients with good viral suppression for prolonged periods of time, extended observations of previously reported novel salvage strategies, factors that influence the degree of CD4+ cell reconstitution among experienced patients, and a new set of observations regarding the use of IL-2 for the management of experienced patients. Please note - when discussing clinical trials in this summary, I have tried to focus mostly on the larger (≥25-50 patients) and longer-running (≥48 weeks) studies that had control arms. There were many interesting smaller studies of shorter duration and in some cases a single arm of therapy that are fine as pilot studies but probably not yet ready for prime time. Also, I have mostly stayed away from analyses of pooled data when individual sets of data have already been presented at other conferences with one notable exception.
We recognize that over a lifetime of antiretroviral therapy any decrease in the number of agents offers the theoretical advantage of decreasing pill-taking fatigue, medication-related adverse events, and costs. Thus, many studies over the years have looked at treatment de-intensification as a strategy for simplifying the management of experienced patients. We learned from work done a few years back that under certain conditions ritonavir-boosted PIs may function quite effectively as monotherapy for HIV infection. Because of its great potency and demonstrated efficacy among both treatment-na´ve and experienced patients, ritonavir (r)-boosted darunavir (DRV) is a protease inhibitor that may prove effective in this role. Two studies were presented that explored the use of DRV/r monotherapy in patients who had already achieved good viral control on conventional triple therapy. Both these studies have already been summarized and discussed by Jules Levin and Mark Mascolini on the NATAP website, so I will only briefly describe the studies and then focus on what I think are the take-home messages relevant to our clinical practice.
Dr. Jose Arribas from Madrid presented the results of the MONET study.1 A total of 256 patients with HIV RNA <50 c/ml for at least 24 weeks on standard HAART (43% NNRTI-based and 57% PI-based) were randomized to either DRV/r 800/100 mg daily (n=127) or DRV/r 800/100 mg daily plus 2 NRTIs (n=129). The primary outcome measure was a non-inferiority TLOVR analysis: 2 consecutive viral loads >50 c/ml by week 48 or switches off study treatment. By ITT analysis, HIV RNA suppression with DRV/r monotherapy (84.3%) was non-inferior to DRV/r + 2 NRTIs (85.3%) (difference -1%, lower limit of 95% CI -9.9%). The PP analysis showed similar results: HIV RNA suppression with DRV/r monotherapy 86.2% and with DRV/R + 2 NRTIs 87.8% (difference -1.6%, lower limit of 95% CI -10.1%). Seven DRV/r and 11 DRV/r + 2 NRTI patients had 2 HIV RNA >50 c/ml although in only 2 patients per arm were these sustained elevations (>400 c/ml). Nine patients in the monotherapy arm (4 for adverse events) and 9 patients in the triple therapy arm (0 for adverse events) had treatment modifications or study withdrawals.
The other DRV/r monotherapy study was presented by Christine Katlama from Paris and the results were not as encouraging as those of the MONET study.2 In the MONOI-ANRS 136 study, patients on HAART with HIV RNA <400 c/ml for at least 18 months and <50 c/ml at entry had DRV/r 600/100 mg bid added to their therapy for 8 weeks and were then randomized to either DRV/r monotherapy (n=113) or DRV/r plus 2 NRTIs (n=112). The primary endpoint was, once more, a non-inferiority analysis: 2 HIV RNA >400 c/ml within 2 weeks of each other by week 48 or ART modification or study withdrawal. However, this study was somewhat different to MONET in that the pre-specified lower limit of the 90% CI could not be greater than -10%. By ITT analysis, HIV RNA suppression with DRV/r monotherapy (87.5%) was not non-inferior to DRV/r + 2 NRTIs (92%) (difference -4.5%, lower limit of 90% CI -11%). The PP analysis, however, did show similar results for both treatment arms: HIV RNA suppression with DRV/r monotherapy 94.1% and with DRV/R + 2 NRTIs 99.0% (difference -4.8%, lower limit of 90% CI -9%). There were 3 virologic failures in the monotherapy and none in the triple therapy arm and 11 and 9 treatment modifications or study withdrawals, respectively, although it was not reported how many of these were due to adverse events.
Why the difference in outcomes between these 2 studies? Other than the subtle differences in the entry criteria, definitions of virologic failure, and allowed lower limit of the 95% vs. 90% CI, the one major difference in the studies was the use of DRV/r 800/100 mg daily vs. 600/100 mg twice daily. It is possible, especially if the discontinuations and treatment modifications in the MONOI study were due to adverse events, that the difference in outcomes may have been driven by greater drug intolerance secondary to higher total daily doses of DRV/r. Certainly, the rate of virologic failure was very low in both studies and the differences between the treatment arms were non-significant. Thus, this may be a case in which less is more. Certainly, as the ARIES study also seems to suggest (see below), in patients with sustained viral suppression lower PI exposure may not be a critical issue.
The ARIES study explored whether atazanavir (ATV) without ritonavir boosting may be non-inferior to ATV/r in maintaining viral suppression once viral suppression is achieved with ATV/r induction, all with 2 NRTIs.3 Individuals who had undergone a 36 week induction with ATV/r 300/100 mg daily plus abacavir/lamivudine (ABC/3TC) 600/300 mg daily were randomized to either continue the original regimen (n=209) or de-intensify to ATV 400 mg without ritonavir (n=210) plus ABC/3TC for 48 weeks (total follow-up of 84 weeks although the study will be extended to a total of 144 weeks). The primary study endpoint was the proportion of patients with HIV RNA <50 c/ml at week 84 by a TLOVR analysis. By ITT analysis, the ATV (86%) and the ATV/r (81%) arms were comparable in maintaining HIV RNA <50 c/ml (difference +5%, 95% CI -1.75, +12.48). Interestingly, with an HIV RNA <400 c/ml analysis, the ATV regimen (92%) was actually superior to the ATV/r regimen (86%) (difference +6%, 95% CI +0.44, +12.22). There were no differences in outcome when patients were stratified by viral loads less than or greater than 100,000 c/ml, and the difference in the rate of virologic failure was non-significant between the ATV (0.5%, n=1) and the ATV/r (3%, n=7) arms as was the rate of any grade 2-4 drug-related adverse events (ATV 30%, ATV/r 33%). Thus, this study suggests that after an adequate period of viral load suppression, a de-intensification of therapy such as removal of ritonavir boosting with the appropriate increase in the primary PI's dose may be successful in maintaining viral suppression. It should be noted, however, that of the commonly used PIs, ATV is the only one that can be given without ritonavir boosting. Also, in this study the NRTIs were continued. Finally and most importantly, this and the DRV/r monotherapy strategies should be considered experimental and not ready for prime time as it will be critically important to assess whether viral suppression is maintained for longer periods of time.
We have felt for many years that in order for HAART to be successful the degree of adherence to therapy must be as close as possible to perfect. Certainly that is the message we convey and should continue to convey to patients even though there are some indications that modern HAART has become so potent that close-to-perfect adherence may not be indispensable. A study by Lima et al explored the level of adherence and duration of viral suppression needed to prevent viral rebound in a British Columbia cohort of patients on HAART.4 A primary outcome of viral rebound (the first of ≥2 consecutive HIV RNA >400 c/ml after achieving viral suppression) among 1305 patients occurred in 274 (21%) individuals with a median time of suppression before rebound of 2 years. They found that the likelihood of viral rebound was associated with certain factors that have been previously described such as female gender, active drug use, and earlier treatment periods (e.g. 2000-1 vs. 2005-6). They also found, however, that for any one of the 4 levels of adherence they studied (<40%, 40-<80%, 80-<95%, and ≥95%), the likelihood of a viral rebound decreases the longer a patient has been suppressed. For instance, with adherence of <40%, the likelihood of a viral rebound is close to 80% with viral load suppression of only 12 months but drops to 10% after suppression of 72 months. On the other hand, with adherence of ≥95%, the likelihood of a viral rebound is 10% with viral load suppression of 12 months but drops to close to 0% after suppression of 72 months. An observational study such as this one provides real-world proof that prospective studies such as the following one do have a basis for working.
The FOTO study enrolled 60 patients with viral load suppression (HIV RNA <50 c/ml for ≥90 days) in order to explore whether 5 consecutive days of efavirenz (EFV), tenofovir, and emtricitabine therapy per week is as effective in maintaining viral suppression as 7 days per week. In 2008, the study had previously reported no difference between the treatment arms in virologic failure (confirmed HIV RNA >400 c/ml) out to 24 weeks and now reported 48 week outcomes.5 At week 24, the patients on 7 days of therapy/week crossed over to 5 days/week and of the 50 individuals who made it to week 48, 45 (90%) had HIV RNA <50 c/ml. Five patients had blips <200 c/ml at week 48 but were all <50 c/ml at the final study visit. Five days of therapy a week was vastly preferred to 7 days by patients. What are the clinical implications of this observation? In my opinion, the data reassure us that missing up to 2 consecutive days of EFV-based therapy every so often will not be associated with viral failure invariably. My preference would be, however, to strive for as close to perfect adherence as possible as we always have. Among the patient population I work with, I would be concerned about the impact that forgetfulness due to cognitive impairment or just plain "patient-determined extension" of the drug holiday might do to long-term efficacy, although I recognize that in some highly involved and dependable populations the FOTO strategy seems to work quite well.
At CROI earlier this year, the French Agency for AIDS Research (ANRS) had presented 24-week outcome data for its ANRS 138 EASIER study in which 170 patients with triple-class-resistant HIV on an enfuvirtide (ENF)-containing regimen for at least 3 months and with HIV RNA <400 c/ml were randomized to either stay on ENF (n=85) or substitute the ENF with raltegravir (RAL) (n=84). Final 48 week data were now presented at the IAS meeting.6 Baseline patient characteristics were similar for both groups; it should be noted that this was a heavily pre-treated patient population with a median prior HAART and ENF exposure of 13.6 and 2.3 years, respectively. Through week 24 by an ITT analysis, only 1 failure (HIV RNA >500 c/ml) occurred in each of the treatment arms and 89% of patients in both arms had HIV RNA <50 c/ml demonstrating non-inferiority of the RAL arm. After the week 24 time point, patients on ENF were all switched to RAL. By week 48 there were no subsequent viral failures in either arm, 90% of patients in both arms had HIV RNA <50 c/ml, and there were no differences in the incidence of grade 3-4 adverse events or laboratory abnormalities. Thus, substitution of ENF with RAL seems to be a safe and effective one-for-one switch.
Another interesting study that explored RAL switching was one by Garrido et al from Madrid.7 Patients who had switched their anchor drug (either a PI or an NNRTI) to RAL because of tolerability issues with undetectable viral loads (<50 c/ml) for at least 12 months prior to the switch and who were then followed for at least 6 months after the switch (n=29) were compared to a group of control patients on their original HAART regimens with undetectable viral loads for at least 18 months (n=22). All RAL switches maintained viral suppression. The CD4+ cell counts of both groups of patients had been stable over the 12 months preceding the switch in the study group or continuation of therapy in the control group. However, by month 6, a statistically significant rise from 364 to 483 cells/mm3 (p=0.018) had taken place among the RAL switches. There was no change in CD4+ counts of the patients who stayed on their original therapies (381 to 420 cells/mm3; p=0.685). The authors explored whether this increase may have been due to decreases in integrated proviral DNA but were not able to detect any changes. This is a finding similar to that reported by Gandhi et al in ACTG A5244 in which addition of RAL to an already suppressive HAART regimen as an intensification strategy was not associated with further decreases in viremia by ultrasensitive, single copy assays.8 There are anecdotal reports that RAL seems to be associated with greater CD4+ cell increases than other agents. The Garrido study is the first one, to my knowledge, to explore this with a control population albeit in a retrospective fashion. It is a step in the right direction and a prospective trial would now seem to be justified to corroborate this observation.
The other ANRS study for which 48-week data were presented was the ANRS 139 TRIO trial for heavily experienced patients treated with novel agents.9 A total of 103 patients with HIV RNA >1,000 c/ml with a history of NNRTI failures, ≥ 3 primary PI and NRTI mutations but ≤3 DRV and NNRTI mutations were enrolled into a single arm study consisting of DRV/r 600/100 mg, RAL 400 mg, and etravirine (ETV) 200 mg, all given twice daily, ▒ an optimized background regimen at the individual investigators' discretion. At week 24, as reported last fall, 93 (90%) patients had HIV RNA <50 c/ml. The definition of viral failure for the week 24-48 phase of the study (HIV RNA ≥50 c/ml on 2 consecutive measures for those with HIV RNA <50 c/ml at week 24) was met by only 4 patients and 100 of the original 103 patients remained on the study at week 48 for a success rate of 86% (n=89). The median CD4+ cell increase was 108 cells/mm3 and the therapy was well tolerated with only 15 patients having a grade 3-4 adverse event. In my opinion, what is most interesting about the study is the virologic success rate approaching 90% in a heavily experienced, hard-to-treat patient population. A success rate this high 10 years ago in na´ve patients would have been gratifying; to think this can now be achieved in experienced patients is frankly amazing.
Although most patients on effective HAART have a satisfactory increase in CD4+ cells within a reasonable time frame, there is a minority who experience poor immune restoration. One of the studies presented at the conference dealt with the speed at which 95 individuals with advanced AIDS and multidrug resistant virus reconstituted their CD4+ cells once they achieved viral suppression (HIV RNA <50 c/ml) for at least 12 weeks.10 Patients were followed longitudinally after the first time point of suppression. Their median nadir, baseline, and start-of-suppression CD4+ cell counts were 53, 145, and 248 cells/mm3, respectively. The investigators' analysis showed that with continuous suppression, the mean CD4+ counts increased 83 cells/ mm3 in the first year, 56 cells/ mm3 in the second year, and 32 cells/mm3 per year subsequently. As I would have expected, this rate of rise seems to be somewhat slow relative to what would be seen in a treatment-na´ve patient without resistance and with good viral suppression. Furthermore, this was a mean increase; there undoubtedly will be individuals with an even slower or perhaps insignificant degree of immune restoration.
And this brings us to the last paper I wanted to review. At the CROI meeting in February the results of the ESPRIT and SILCAAT IL-2 trials were presented. In short, close to 6,000 patients were enrolled into both studies with the goal of determining whether IL-2 added to HAART would be more effective than HAART alone in reducing the number of study-specified clinical endpoints (opportunistic diseases or deaths). This hypothesis seemed valid at the time because it is known that the numbers of CD4+ cells are raised effectively by IL-2; however, it has not been clear until now whether this rise is also associated with better clinical outcomes. Both ESPRIT (conducted in patients with CD4+ ≥300 cells/mm3) and SILCAAT (conducted in patients with CD4+ 50-299 cells/mm3) were prematurely terminated by their Data Safety Monitoring Boards. Interim analyses had shown either no difference in the rates of clinical endpoints between the treatment arms (HAART alone or HAART plus IL-2) (i.e. SILCAAT) or this same finding plus a greater frequency of serious adverse events in the arm with IL-2 (i.e. ESPRIT).
At this conference, Babiker et al presented a pooled analysis of the studies that quantified the difference between the decrease in clinical endpoints that would have been expected based on the rise of CD4+ cells and the observed increase, if any. The average increase in CD4+ cells during follow-up in the IL-2 vs. no IL-2 arms was 160/mm3 (+25%) in ESPRIT, 59/mm3 (+18%) in SILCAAT, and 23% in a pooled analysis. Based on what is known from historical data about the clinical impact of such rises in CD4+ cells, it was estimated that the hazard ratio (HR) for clinical endpoints should have decreased to 0.74, 0.79, and 0.75 in ESPRIT, SILCAAT, and the pooled analysis, respectively. However, the actual observed HR decrease was much less at 0.93, 0.92, and 0.92 in ESPRIT, SILCAAT, and the pooled analysis, respectively, demonstrating that IL-2 did not protect at all from clinical endpoints. In fact, further analysis showed that for the same latest CD4+ cell count in the 2 treatment arms of either study, the immediate risk of a clinical endpoint was actually higher for the IL-2 treated patients. The take home message here is that not all forms of CD4+ cell expansion are equal, and the clinical impact of any given level of CD4+ cell expansion brought about by HAART alone seems to be better than that brought about by HAART plus IL-2. What is not clear is what this will mean for other forms of CD4+ cell expansion. For instance, there is the previously described Garrido study with better CD4+ restoration among RAL recipients or the observation that in the MERIT study maraviroc-treated patients boost their CD4+ cells better even when virologic responses are not as good as in EFV-treated patients. Further work will be needed to determine the clinical impact of these differences in CD4+ expansion.
1. Arribas J, Horban A, Gertstoft J, et al. The MONET trial: darunavir/ritonavir monotherapy shows non-inferior efficacy to standard HAART, for patients with HIV RNA <50 copies/mL at baseline [Abstract TUAB106-LB]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
2. Katlama C, Valentin MA, Algarte-Genin M, et al. Efficacy of darunavir/ritonavir as single-drug maintenance therapy in patients with HIV-1 viral suppression: a randomized open-label non-inferiority trial, MONI-ANRS 136 [Abstract WELBB102]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
3. Squires K, Young B, DeJesus E, et al. Similar efficacy and tolerability of atazanavir (ATV) compared to ATV/ritonavir (RTV, r), each in combination with abacavir/lamivudine (ABC/3TC), after initial suppression with ABC/3TC + ATV/r in HIV-1 infected patients: 84 week results of the ARIES trial [Abstract WELBB103]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
4. Lima V, Harrigan R, Rosenblum M, et al. Duration of viral suppression strongly modifies the adherence-viral rebound relationship [Abstract WEPEB283]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
5. Cohen C, Colson A, Pierone G, et al. The FOTO study: The 48 week extension to assess durability of the strategy of taking efavirenz, tenofovir and emtricitabine Five days On, Two days Off (FOTO) each week in virologically suppressed patients [Abstract MOPEB063]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
6. De Castro N, Braun J, Charreau I, et al. Switch from enfuvirtide to raltegravir in virologically suppressed multidrug-resistant HIV-1 infected patients: final results of the randomized ANRS 138 trial (EASIER) [Abstract MOPEB066]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
7. Garrido C, Benito JM, Lozano S. CD4 recovery after switching to raltegravir in a simplified strategy in patients with undetectable viremia [MOPEB069]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
8. Gandhi R, Zheng S, Bosch R, et al. Raltegravir (RAL) intensification does not reduce low-level residual viremia in HIV-1-infected patients on antiretroviral therapy (ART): results from ACTG A5244 [Abstract WELBB104]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
9. Fagard C, Descamps D, Dubar V, et al. Efficacy and safety of raltegravir plus etravirine and darunavir/ritonavir in treatment-experienced patients with multidrug-resistant virus: 48-week results from the ANRS 139 TRIO trial [Abstract TUPDB204]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
10. Anis AH, Cameron DW, Sun H, et al. The effect of treatment on immune reconstitution of multi-drug resistant HIV/AIDS patients [Abstract MOPEB074]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.
11. Babiker AG. An analysis of pooled data from the ESPRIT AND SILCAAT studies; findings by latest CD4+ count [Abstract TUAB101]. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. Cape Town, South Africa, 19-22 July 2009.