icon-    folder.gif   Conference Reports for NATAP  
  19th Conference on Retroviruses and
Opportunistic Infections
Seattle, WA March 5 - 8, 2012
Back grey_arrow_rt.gif
Antiretroviral Therapy at CROI - PHASE III, PHASE II, PHASE I
  Joseph Eron MD
Professor of Medicine
University of North Carolina at Chapel Hill

Elvitegravir/cobisistat/TDF/FTC fixed dose combination (FDC):

At this year's CROI we finally saw the detailed Phase III data on the fixed dose combination (FDC) of elvitegravir, cobisistat, tenofovir and emtricitabine; the pill that most people call the "QUAD". There were two large phase III studies presented and while we had seen the top line results in press releases last year most of us were hungry for more information. Would cobisistat just be ritonavir "not so light" or would there be apparent advantages to this booster that would add to the obvious convenience factor of this single pill regimen? What would the virologic failures look like? Would the frequency of resistance be similar to raltegravir and would the mutations that emerged be similar to raltegravir and result in viruses that remained sensitive to dolutegravir? The presentations addressed these questions and in my opinion made it clear that we will have an additional comparable first line choice when this quadruple single tablet regimen is approved by the FDA this summer (the 'drop-dead date' for the FDA is August 27, 2012 - though the agency could give their thumbs up earlier).

FDC ELV/Cobi/TDF/FTC vs. FDC efavirenz/TDF/FTC

In this presentation by Paul Sax approximately 700 patients were randomized to one of the two single tablet regimens (FDC ELV/Cobi/TDF/FTC vs. FDC efavirenz/TDF/FTC).[1] The study was placebo controlled and patients were instructed to take the FDC with efavirenz or placebo in the evening on an empty stomach. Therefore all the patients on the study were advised to take medications twice per day; ELV/Cobi/TDF/FTC with food and efavirenz/TDF/FTC on an empty stomach before bed. On the other hand this is the first large randomized comparison of two single tablet regimens. As would be expected the groups were well balanced. Mean baseline CD4 cell counts were > 380 cells/mm3. This number has been rising in our na夫e studies over the last several years - hopefully reflecting early entry of HIV-infected individuals into care. One third of subjects had HIV RNA values > 100,000 c/mL. Only approximately 10% of enrollees were women, which is clearly a disappointing number and will likely require a substantial commitment from the sponsor to study the quadruple tablet in women after FDA approval.

The study was well managed with less than 5% of patients lost-to-follow-up. The overall results in both arms were outstanding with 88% of patient on the elvitegravir/cobi-based regimen and 84% on the EFV-based regimen with HIV RNA < 50 c/mL in the intent-to-treat "snap-shot" analysis. This result clearly demonstrated non-inferiority of the quadruple regimen. Most subset analyses favored the ELV-cobi regimen and while statistics were not presented, patients over the age of 40 and those with higher CD4 cell counts seemed to accrue a substantial benefit when they received the ELV-cobi based regimen. The results in the high viral load group were similar between group 84% with the ELV-cobi regimen and 82% with the EFV regimen. In patients with CD4 < 350 the percentages were 83% and 84% for the quadruple and triple regimens respectively. As we have seen with other comparisons to EFV/TDF/TFC CD4 cell count increases were superior with the ELV-cobi regimen (mean change from baseline 239 vs. 206 cells/mm3; p =0.009). I have not heard a clear explanation for why the specific combination of EFV plus TDF/FTC has a somewhat less robust CD4 response than other first line regimens; an observation seen in multiple studies that appears to be specific to EFV with these specific nucleosides. However the difference seems unlikely to be clinically important.

Resistance emergence was uncommon and similar in both arms. Fourteen subjects in the ELV-cobi arm and 17 subjects in the EFV-arm meet criteria of the resistance analysis (essentially VL > 400 confirmed) and 8 (2%) in each arm had resistance mutations emerge. The were no surprises in the emergence of primary integrase resistance (7 patients) and primary NNRTI resistance mutations (8 patients) and all patients with integrase resistance emergence will likely have retain susceptibility to dolutegravir in my opinion (though this was not tested of course). All 8 patients with resistance in the ELV-cobi arm had a M184V/I mutation emerge and 3 had K65R. Only 2 patients on the EFV-arm had M184V/I emerge both had K65R. Whether this differential emergence of NRTI resistance is clinically important is not known. In clinical practice patients who fail EFV/TDF/FTC with resistance frequently have both K103N and M184V and if the M184V is not detected we frequently assume it is there. On the other hand it would be accurate to say that two class resistance appeared more common in the ELV-cobi treated patients who had emergence of resistance. What is clear is that "boosting" of elvitegravir with cobisistat does not result in a high barrier to resistance as is typically seen with boosted PI (see below).

One of the areas we were most curious about was how would therapy containing cobisistat be tolerated? The number of discontinuations due to adverse events was low in each arm: 4% vs. 5% in the ELV-cobi arm vs. EFV arm. These low rates are typical in clinical trial patients with relatively well-tolerated treatments and blinded therapy. Four patients on the ELV-cobi arm discontinued due to "renal abnormalities". The audience was told that 3 of these were typical TDF related renal abnormalities and I suspect we will hear more detail about these cases either at a future meeting or one the study is published. Nausea (all grades) occurred more commonly in the quadruple arm (21% vs. 14%) and in questioning we learned that this was mostly grade 1 nausea. Typical "efavirenz side effects" were more common in the EFV/TDF/FTC arm. These included abnormal dream, insomnia and dizziness. Rash was also significantly more common with efavirenz. The breakdown of the grades of these side effects was not presented.

Grade 3 and 4 lab abnormalities were relatively uncommon with no clear pattern favoring one treatment arm over the other. Creatinine does rise modestly on cobisistat (0.14 mg/dL on average in this study), which is related to the effect of cobisistat on renal tubular secretion of creatinine and not an effect on glomerular filtration rate. This effect occurred very early on therapy (as would be expected) and then plateaued over the next 46 weeks of the study. One of the more surprising results of the study was the modest effect the cobisistat-containing regimen had on lipids. I had assumed that the impact of cobisistat would be similar to that of ritonavir, but there was no difference between arms on triglyceride levels (median increase 7 mg/dL in both arms) and all cholesterol levels (total, LDL and HDL) increased less substantially on ELV-cobisistat/TDF/FTC than with EFV/TDF/FTC (all differences were statistically significant). Again in the question and answer period we learned that there was no difference in the total cholesterol to HDL ratio between the two arms.

CROI: The Efficacy and Safety of Elvitegravir/Cobicistat/Emtricitabine/Tenofovir DF ("Quad") Compared to Efavirenz/Emtricitabine/Tenofovir DF in Treatment Na夫e HIV-1 Infected Subjects: Primary Results of Study GS-US-236-0102 - (03/8/11)

FDC ELV/Cobi/TDF/FTC vs. atazanavir/r plus TDF/FTC

Edwin DeJesus presented the partner study to the study presented by Sax et al. This study compared FDC ELV/Cobi/TDF/FTC vs. atazanavir/r plus TDF/FTC and was presented in a poster session. [2]. The design was similar to the study by Sax et al. i.e., randomized and double-blinded with placebos. Therefore study patients had to take 4 pills in each arm though I assume that they could take them all once a day. Once again just over 350 subjects were randomized to each treatment arm and the mean CD4 cell count was again over 350 cells/mm3. In this study 40% of subjects have baseline HIV RNA > 100,000 c/mL and again only 10% of the study subjects were women. This study was also extremely well conducted with only 2% of subjects lost to follow-up.

Response rates were very high with 90% of subjects on ELV-cobi regimen and 87% of those on ATV/r suppressed to < 50c/mL in the ITT "snap shot" analysis and very similar to what was seen in the BID arm of the QD MRK study in which raltegravir was given with TDF/FTC (89% < 50 c/mL; non-completer = failure analysis). [3] The ELV-cobi regimen was clearly non-inferior to the ATV/r regimen and while none of the sub-group analyses reached statistical significance all of them, except the analysis of women, favored ELV-cobi. The point estimate for the difference in effect in women favored ATV/r slightly and the confidence interval was obviously very broad given the very low number of women. There were no substantial differences between treatment regimens in the >100,000 and < 100,000 c/mL strata though numerically both regimens had a lower proportion < 50 c/mL in the higher VL stratum (8% lower in each). The 96 week data will be interesting as we will see whether a proportion of these patients were still declining and will eventually reach < 50 c/mL. CD4 counts rose by over 200 cells/mm3 in each arm and the difference at 48 weeks between arms was not significant. Resistance results were consistent with scores of ritonavir boosted PI studies. Only 8 patients met criteria for virologic failure and resistance analysis on the ATV/r arm and none had PI or NRTI resistance mutations. Twelve patients (3%) met criteria for virologic failure and resistance analysis on the ELV-cobi arm and 5 had resistance mutations documented. Four of the 5 had integrase resistance mutations and 4 had 184V/I.

The clinical and laboratory adverse event profiles were remarkably similar between the two treatment arms with the exception of scleral icterus and hyperbilirubinemia, which was higher in the ATV/r arm. The proportion with nausea was virtually the same (19% vs. 20%) and very few patients discontinued due to nausea. Only one patient in each arm discontinued due to renal abnormalities. Both arms had a small, rapid increase in estimated creatinine clearance that plateaued, which was slightly but significantly higher in the ELV-cobi arm (0.12 vs. 0.08; p < 0.001). Declines in bone mineral density in the hip and spine were similar between arms and appeared consistent with previous studies of TDF when given with a boosted PI. Cholesterol increases were small and very similar between the two arms. Triglyceride levels, however, rose significantly more with ATV/r (23 vs. 8 mg/dL; p = 0.006) suggesting that cobisistat has less of an impact on this parameter than ritonavir.

CROI: Efficacy and Safety of Elvitegravir/Cobicistat/Emtricitabine/Tenofovir DF "Quad" Compared to Ritonavir-boosted Atazanavir plus Emtricitabine/Tenofovir DF in Treatment Naive HIV-1 Infected Subjects - (03/9/11)

QUAD Summary

What is my overall assessment of the initial Phase III data on the single tablet regimen of elvitegravir/cobisistat/TDF/FTC? First of all the obvious should be pointed out - these are 48 week data and long term data will be needed to provide a more complete assessment of this combination. However, to date, 48 week treatment response data have been very predictive of 96 week virologic response data in virtually all phase III studies that I can recall. This combination is a highly effective, well-tolerated single tablet regimen and I would be very surprised if it did not eventually make it to the preferred initial treatment list of most guidelines. More data in women are clearly needed and the tension between a company's need for rapid accrual and more accurate representation of the HIV epidemic needs to be addressed.

How and in whom will the "Quad" be used? A crystal ball would be useful here. As a single tablet regimen it does not have the early CNS issues of the efavirenz-based single tablet and it does not have the high viral load concern of the rilpivirine-based tablet. The pregnancy category of this quadruple regimen will be better than any EFV-based tablet but the lack of data in women is a potential weakness. Clinicians will have to balance the long term data and clear comfort with the EFV-based single tablet regimen compared to the improved early side effect profile, which may improve medication adherence and the persistence of the initial regimen (i.e. less switching) in the clinic. How clinicians will weigh the ELV-cobisistat single tablet regimen in patients in whom they are using twice daily raltegravir will likely depend on how "well-tolerated" the quadruple therapy is in the clinic. Raltegravir based therapy is frequently prescribed when clinician want a regimen that has very few side effects and whether the low grade nausea will influence clinicians will only come with clinical experience. Clinician may get a sense of this if patients suppressed on a raltegravir/TDF/FTC regimen come in after hearing about a one pill daily alternative. We are in a "data free zone" here and more cautious clinicians are likely to wait for clinical trial or descriptive data. However, patients who are switched will likely report quickly if the side effects of the single pill are tolerable or not compared to what they were used to. Cobisistat-based regimens will not be free of most ritonavir-related drug-drug interactions and this fact will need to stay in the forefront of our minds and we do not have data as far as I know on drug interaction with the HCV protease inhibitors. What about patients who we now start on initial therapy with a boosted PI such as ATV/r or DRV/r? I use boosted PI as initial therapy predominantly in two groups of patients. One, are patients who have transmitted-drug resistance and the second are patients who are at risk for treatment interruption and resistance emergence. Given the data from study of DeJesus et al confirming once again the barrier of the boosted protease inhibitors I don't think practitioners with a similar mind set as mine will change much. Data with the single tablet regimen in patients with transmitted NNRTI resistance would be useful.

One concern that has clearly been addressed by the phase III studies is the impact of the ELV-cobisistat single tablet regimen on lipids. There appears to be little impact of this combination on triglycerides and the overall effect on cholesterol and its subsets (LDL and HDL) is modest. Management of changes in estimated GFR should be pretty straightforward in most patients who have a normal eGFR above 70 or 80 mL/min. Patients could come in at their first VL check at 2-4 weeks post initiation and also get a creatinine and that level could then be used as a new baseline. Management should be the same but might be trickier in individuals who have eGFR in the 50-70 range. They may be fine with the therapy, but a 0.12 to 0.14 increase in eGFR may push them into a range where one would normally dose adjust TDF. Their true GFR would likely be unchanged but it might be harder to maintain the fixed dose regimen without definitive knowledge of true GFR. This fact did not escape Gilead as the inclusion criterion for both phase III studies discussed above was > 70 mL/min as opposed to the > 60 mL/min criterion used in previous studies of TDF. The FDA indication for use of the quadruple single tablet regimen will likely follow the study inclusion criterion.


Dolutegravir, a potent, once daily, integrase inhibitor that does not require boosting, is likely to be the next one of the next drugs approved by the FDA potentially some time in 2013 as multiple phase III studies are ongoing in treatment na夫e and treatment experienced patients. In part because of all these ongoing trials there has not been a lot of new data on dolutegravir over the last year. At CROI we saw the 96 week data from the phase II dose ranging study of DTG in treatment na夫e patients. [4] We have seen this study before. It is a randomized study partially blinded study comparing 3 doses of dolutegravir or EFV each combined with 2NRTI. The three doses (10, 25 and 50 mg once daily) were blinded, the EFV was open label. There are approximately 50 subjects per arm (total 155 subjects received DTG) and baseline characteristics were relatively well balanced with a median CD4 cell count of 324 cells/mm3. Approximately 2/3 of subjects were given TDF/FTC and 1/3 received ABC/3TC. At 96 weeks antiretroviral response outcomes were similar with 88%, 78% and 79% on 50 mg, 25 mg and 10 mg of DTG < 50 c/mL compared to 72% on EFV-containing regimen. Virologic failure (rebound by the TLOVR criteria) was relatively uncommon occurring in 2, 4 and 7 subjects in the 50mg, 25 mg and 10 mg DTG-arms and 4 subjects on the EFV arm. Notably 1, 2 and 3 subjects on the DTG arms re-suppressed without changing therapy, as did 3 of the 4 on the EFV arm. No subject on the DTG 50 mg arm rebounded above 400 c/mL and only 3 subjects on DTG at the lower doses and 1 subject on EFV met this definition of virologic failure. No subjects on DTG on this study have had emergence of a virus with an integrase inhibitor resistance mutation. One subject of the 155 who received DTG developed virus with an M184V mutation. CD4 cell counts rose as expected in all 4 arms.

Only 4 subjects (3%) withdrew due to adverse events on all the DTG arms combined compared to 5 subjects (10%) on the EFV arms. Grade 2 or greater adverse events, in general, were uncommon on all the arms with no particular pattern of AE seen with DTG. Grade 2 or greater laboratory AE were also quite uncommon, again with no particular pattern seen in the DTG arms. Like with cobisistat, there is an 0.1-0.15 mg/dL increase in creatinine with dolutegravir that is also due to a decrease in creatinine secretion in the proximal tubule of the kidney and not due to an effect on glomerular filtration rate.

These results on a relatively large number of subjects for phase II trial (155 total treated with one of three doses of DTG) heighten the anticipation as we await the results of the phase III studies. We know from the phase II VIKING studies that DTG has antiretroviral activity in patients who harbor viruses with integrase inhibitor resistance mutations selected for by raltegravir (and elvitegravir) and that the antiretroviral effect can be prolonged especially if DTG is paired with at least one fully active agent. Importantly, DTG has several attractive characteristics as a first line agent. DTG at 50 mg once daily (the dose used in the phase III na夫e studies) is potent and appears very well tolerated and does not require boosting so it is likely to have a drug-drug interaction profile similar to raltegravir and may also have a limited impact on lipids. The results from the phase II study presented at CROI suggest that DTG may also have a higher barrier to resistance than RAL or ELV-cobi. To date, no na夫e patient has developed integrase inhibitor resistant virus and in vitro, DTG resistance seems harder to select than resistance to RAL or ELV. In addition, the relatively small mg dose of DTG should allow co-formulation into a single tablet regimen. Spring II - which is a head-to-head comparison of DTG vs. RAL in treatment na夫e patients, both with 2 NRTI, is likely to be the first phase III study that will have 48 week results. These results should be available this summer or fall. It will be interesting to see if the promise of phase II is borne out in phase III.

CROI: Dolutegravir ViiV Integrase Inhibitor phase 2b SPRING Study 96-Week Results in Treatment-Naives - (03/8/11)


Gilead Sciences (GS) - 7340 is in interesting molecule that requires some explanation for at least some of our readers. At first glance the company appears to be taking a nucleotide analog that has very potent antiretroviral activity in short term studies and deescalating the doses. To some this may seem counterintuitive. As some readers may know when we talk about studies of "tenofovir" in combination with other oral antiretrovirals we are really talking about studies of "tenofovir disoproxil fumarate" or TDF, which is a pro-drug of a pro-drug. TDF is the pro-drug for tenofovir and tenofovir is the pro-drug of tenofovir-DP (TFV-DP). TFV-DP is actually the active drug intracellularly. TDF is converted pretty rapidly to tenofovir in plasma so plasma levels of tenofovir are relatively high. The renal toxicity (and potentially the bone toxicity) of TDF (stay with me here) is correlated with plasma tenofovir levels. Here is where GS-7340 comes in. GS-7340 (like TDF) is a pro-drug of a pro-drug. It must first be converted to tenofovir and then to TFV-DP to exert its activity. Unlike TDF however very little GS-7340 is converted into tenofovir in plasma and very high concentrations of TFV-DP (the active metabolite) are reached inside lymphocytes. GS-7340 is about 50 times more active than TDF in vitro in lymphocyte cultures and about 4x more active in macrophage. In vivo, 150 mg of GS-7340 has greater short-term antiretroviral activity than 300 mg of TDF and plasma concentrations of tenofovir are lower [5]. Since our major concern with TDF is not activity but very long term safety testing doses of GS-7340 that produce similar or greater activity than TDF but result in very low plasma concentrations of tenofovir might result in an agent that has similar or better activity but substantially less long term toxicity. An added bonus would be that with a small milligram dose co-formulation may be easier and results in smaller pills. A co-formulation with a boosted PI into a single tablet regimen may be possible.

Now we get to the study presented by Ruane et al at CROI 2012. [6] This was a Phase Ib randomized, partially-blinded, placebo and active controlled 10-day mono-therapy study comparing three doses of GS-7340; 8, 25 and 40 mg with TDF 300 mg (standard dose) and placebo in patients who were na夫e or off therapy and had no history of tenofovir resistance. In the primary analysis, which was the median DAVG11 [log10 c/mL] (essentially average viral load to day 11), the 25mg and 40 mg GS-7340 had significantly greater activity than TDF 300 mg (-1.08, -0.94 and -0.48 respectively). The 8 mg dose had numerically greater activity (-0.76) than 300 mg of TDF but the difference was not statistically significant. In a number that might be more familiar, the median decrease in viral load at day 11 was 1.73 log10- with 40 mg of GS-7340, 1.46 log10- with 25 mg and 1.08 log10- with 8 mg compared to 0.97 log10- with 300 mg of TDF. The decreases with the two higher doses were significantly greater than the decline with TDF.

Plasma concentrations (area under the curve) of tenofovir were 79%, 86% and 96% lower with 40mg, 25mg and 8 mg of GS-7340 compared to with TDF 300 mg yet PBMC intracellular tenofovir-DP concentration were > 20x and approximately 7x for the 40 mg and 25 mg GS-7340 dose and about the same between the 8 mg dose and TDF 300 mg. Adverse events were not presented in detail but there were no discontinuations over 11 days, no serious adverse events and no "clinically significant" laboratory abnormalities.

The key question is whether substantially lower plasma TFV concentrations will result in substantially less impact on the kidney and on bone over long term treatment. The toxicity of TDF is modest and for the most part can be monitored but there is clearly some renal risk and measureable bone loss. A substantial improvement will be welcome especially as we may be treating patients for 4 or more decades (!) depending on when they begin therapy. The exact mechanism of tenofovir renal and bone toxicity is not known so the lower plasma levels of tenofovir seen with GS-7340 are not guaranteed to have less toxicity with bone perhaps being the bigger unknown. The dose chosen for further development is in the 20 - 25 mg range and a fixed dose tablet with elvitegravir, cobisistat and FTC is already in a phase II clinical trial and, according to Dr. Ruane, a phase II study with a fixed dose combination of darunavir, cobisistat, GS-7340 and FTC will start soon. The results should be quite interesting.

CROI: GS-7340 25 mg and 40 mg Demonstrate Greater Antiviral Activity Compared with TDF 300 mg in a 10-Day Monotherapy Study of HIV-1 Infected Patients - (03/9/11)


1. Sax P, et al. Elvitegravir/cobicistat/emtricitabine/tenofovir disoproxil fumarate has non-inferior efficacy and favorable safety compared to efavirenz/emtricitabine/tenofovir disoproxil fumarate in treatment-na夫e HIV-1+ subjects. 19th CROI 2012, Seattle, WA Abst. #101.

2. DeJesus E, et al. Week 48 results of an ongoing global phase 3 study comparing elvitegravir/cobicistat/emtricitabine/tenofovir disoproxil fumarate with atazanavir/ritonavir-boosted plus emtricitabine/tenofovir disoproxil fumarate in treatment-na夫e HIV-1+ subjects showing efficacy, safety, and pharmacokinetics. 19th CROI 2012, Seattle, WA Abst. #627.

3. Eron JJ et al. Raltegravir once daily or twice daily in previously untreated patients with HIV-1: a randomised, active-controlled, phase 3 non-inferiority trial. Lancet Infect Dis. 2011 Dec;11(12):907-15.

4. Stellbrink, H-J, et al. Dolutegravir in combination therapy exhibits rapid and sustained antiviral response in ARV-na夫e adults: 96-week results from SPRING-1 (ING112276). 19th CROI 2012, Seattle, WA Abst. #102LB.

5. Markowitz M et al. GS-7340 Demonstrates Greater Declines in HIV-1 RNA than TDF during 14 Days of Monotherapy in HIV-1-infected Subjects CROI 2011; Paper#152LB

6. Ruane P, et al. GS-7340 25 mg and 40 mg demonstrate superior efficacy to tenofovir disoproxil fumarate 300 mg in a 10-day monotherapy study of HIV-1+ patients. 19th CROI 2012, Seattle, WA Abst. #103.