icon-folder.gif   Conference Reports for NATAP  
  50th ICAAC
Boston, MA
September 12-15, 2010
Back grey_arrow_rt.gif
ICAAC 2010 Report by David Margolis MD University of North Carolina
  50th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), September 12-15, 2010, Boston
ICAAC, the very large, very broad infectious disease meeting is held annually in an enormous conference center somewhere. This year's meeting was held in the Boston Convention Center, located in an isolated, cavernous monstrosity on cheap land in the harbor hinterlands well outside the city's charming center. I will highlight some of my impressions of new data presented here, where I found a surprising amount of HIV therapy-related material. Although there was nothing that I found truly earth-shattering, there were several useful nuggets of information, and several reports on the steady development of antiretroviral agents that will broaden therapeutic choice, though not likely to radically change treatment paradigms.
Coreceptor tropism and CCR5 antagonist antivirals
The use of the CCR5 antagonist maraviroc is coupled to testing to document sensitivity to the drug (CCR5-tropic virus), showing the absence of detectable levels of virus that can enter the cells without the use of the CCR5 receptor (dual tropic or X4 virus). Heera and colleagues presented a study (H-1165) examining data and specimens from the Motivate I and II, and A4001029 studies of maraviroc. Patients were screened using the older, less-sensitive, first-generation Trofile viral phenotype assay for entry into these studies. They compared co-receptor tropism as determined by the now-obsolete first-generation phenotypic assay testing and genotypic methods. Sequencing was done of the V3 loop of the HIV envelope glycoprotein by what is perhaps the gold standard "deep" sequencing technique, using a 454 mass sequencing machine capable for sequencing thousands of molecules of genetic information in microwells --- a tool not available in clinical laboratories in the near future, although someday it might be. This study would have been much more informative and useful if the old samples were re-run on the newer, more sensitive Trofile ES.
Overall, of the 889 patients evaluated, for most patients the results of the two tropism testing methods were the same: 84% had concordant tropism calls, with about 10% having non-R5 virus, and 74% having R5-using HIV as determined by either method. Of the patients with discordant results (the two tests gave opposite results), most were R5 by Trofile but 454 sequencing detected non-R5 HIV.
Such discordant results are of course troubling, as the use of maraviroc in the presence of X4-tropic or dual-tropic virus would be expected to lead to a poor therapeutic outcome. Confirming this concern, Heera et al. showed that the median decline of HIV RNA at week 8 of therapy in these studies was similar in the patients that had R5 by Trofile, but non-R5 by 454 sequencing to those that had non-R5 virus by both tests (1.2 vs. 1.3 log declines at week 8). Patients that had R5 virus by both tests had the best responses (2.5 log declines) and discordant R5 patients --- those that had R5 virus by the more sensitive test but appeared to have non-R5 virus by Trofile --- had a slightly reduced magnitude of response (2.0 log decline). It is also confusing as to why the more sensitive test (454) would miss non-R5 virus populations detected by Trofile, but this could be due to the differences in the way the assays work (behavior of a virus clone in the Trofile assay vs. sequence information interpretation after 454 sequencing). The difference in viral load decline is not statistically significant, as this patient group was small, and so perhaps it is most comforting to think of this as a chance observation.
Heera et al. tried to derive clinical predictors that would portend a poor performance of the Trofile assay. Demographics, viral clade, initial viral load, and number of antivirals did not predict discordant test results. Median baseline CD4+ cell counts trended higher in patients with concordant R5 calls vs. concordant non-R5 calls (179 vs. 58 cells/mm3). However, it is not clear that this is useful currently, as many patients with lower CD4 counts had concordant calls, and deep sequencing is not currently clinically available. Perhaps, however, this might induce some providers to favor the use of maraviroc when CD4 counts were higher. The authors concluded that deep sequencing appears optimized to select maraviroc responders. Of course, a comparison to Trofile ES would have been more clinically relevant.
Vicriviroc is another CCR5-antagonist, initially developed by Schering, now Merck. Its further development seems somewhat in doubt. Nevertheless, using data from prior studies, Mann and colleagues (poster H-924) presented sequencing and phylogenetic data to discover the origin of viruses that could use CXCR4 (X4-tropic virus), which emerged during vicriviroc therapy. They showed, as with maraviroc, that these viruses originated from small clonal subpopulations of X4 virus that were present prior to the initiation of vicriviroc, and that a phenotypic switch or evolution of the virus to use CXCR4 had not occurred.
Portsmouth (poster H-923) also presented long-term follow-up on 15 patients from the original MERIT study in whom X4-tropic viruses emerged during maraviroc therapy. Such viruses were detected in blood samples taken an average of 85 days after therapy initiation. Therapy was switched in response to this finding an average of 98 days later. For these 15 patients, median CD4 cell count gain during the MERIT study was 218 cells/Ál. Long-term follow-up was reported for 8 patients, but 5 patients were lost to follow-up before 48 weeks and 2 between weeks 48 and 96. This was said to be due to the fact that follow-up therapy was not provided by the study, and patients returned to other providers. For the 8 patients with long-term follow-up, the median follow-up reported was 165 weeks after the emergence of X4 virus (more than 3 years later). For all of these patients, virologic and immunologic response to subsequent therapy was good, with sustained virologic suppression and CD4 cell gain. For two patients in whom recoverable viral RNA was available, viral tropism reverted from X4 to R5 after maraviroc was stopped.
Finally, Wilkin and colleagues (poster H-932a) presented a comparison of a commercial tropism assay developed by Quest Laboratories to that of the second-generation Monogram Trofile-ES assay. 326 patient samples were collected from patients entering studies of maraviroc that either enrolled patients who at screening were found to have R5 virus or were found to have X4 or dual tropic virus. In this way, the cohort represented a mix of viral tropism types. Samples were typed for viral tropism by both assays. There were multiple, somewhat complex analyses performed, but the overall answer was simple and clear. In every analysis the Quest assay performed more poorly than the Trofile ES. The authors, somewhat generously, concluded that the Quest test did not achieve non-inferiority to the old Trofile test, and appeared inferior to the Trofile ES.
ICAAC: Clinical, Virological and Immunological Characteristics of Patients With Discordant Phenotypic and Genotypic (Ultra-Deep Sequencing) Tropism Test Results: Analysis of Tropism Calls in MOTIVATE and 1029 - (09/19/10)
ICAAC: Comparison of Quest Diagnostics to Trofile and Trofile-ES Coreceptor Tropism Assays for Predicting Virologic Response to Maraviroc (MVC), a CCR5 Antagonist - (09/19/10)
ICAAC: Long-Term Outcome of Individuals Experiencing a Phenotypic Switch in HIV-1 Co-receptor Use in the MERIT Study - (09/19/10)
New drugs advancing through a trickling pipeline
Fortunately, currently available antiretroviral therapy for patients that are integrated into a healthcare system and able to access care is generally quite good. Of course, there is always room for improvement. Several presentations at ICAAC gave us an update in the forward motion in this area of the field.
Joe Eron from UNC reported the results of two studies of rilpivirine (TMC 278), a new NNRTI being prepared for licensure and marketing as a co-formulated single tablet with Truvada (tenofovir/FTC) in a total of nearly 1400 patients (abstract H-1810). A new drug application for the single pill has been filed with the FDA.
ECHO was a study of 25 mg of rilpivirine daily given with Truvada, while THRIVE studied rilpivirine given with the NRTIs chosen by the patient and provider. In THRIVE about 60% of the study subjects chose to take Truvada, while about 30% took Combivir (AZT/3TC) and 10% took Epzicom (abacavir/3TC). Entrants were screened by standard testing for NNRTI and NRT resistance at baseline, and excluded if significant resistance was discovered.
In the pooled TLOVR analysis of the two studies, at 48 weeks 84% of the patients in the rilpivirine arms were below 50 copies of HIV RNA, while 82% in the efavirenz arms had achieved and sustained suppression. The TLOVR analysis is an intent-to-treat analysis that examines end points by using specific criteria for determining treatment failure in patients who have achieved HIV RNA viral loads below the limit of quantification. Looking in a post-hoc analysis (which makes claims of significance more tenuous) at patients with baseline viral loads of > 100,000 copies, 77% vs. 81% of rilpivirine vs. efavirenz patients were suppressed, and in the <100,000 strata the ranking were reversed at 90% vs. 84%.
Overall, the somewhat puzzling take-home message appeared to be that the two regimens performed well and about as well as each other, but the efavirenz arms seemed to fail most often due to adverse events, with neurological events predictably in predominance, while the rilpivirine arms suffered virologic failure more often (10% vs. 6%), with the emergence of drug resistance mutations seen more frequently. Rilpivirine seemed to fail more often with higher baseline viral loads or perhaps poorer adherence. Rilpivirine failures displayed the NNRTI resistance mutation E138K mutation in reverse transcriptase most often, or (in decreasing frequency) K101E, H221Y, V189I, Y181C, or V90I. Also NRTI mutations, most often the alternative M184I mutation against FTC/3TC, rather than the well-known M184V mutation, were seen more often in rilpivirine failure. When this fixed-dose combination therapy is available, it will present patients and providers with another good therapy option, but the relative strengths and weaknesses will have to be judged on an individual basis.
Marty St Clair from GSK presented findings on early studies of GSK2248761, a new NNRTI initially developed by Idenix (abstract H-1814). In a short-course monotherapy phase II studies of a range of doses, viral load declines of one to almost 2 logs were seen, and the rapid emergence of NNRTI resistance was not measured. In the test tube, HIV requires three mutations, K103N, Y181C, and G190A, to possess high-level resistance to this drug. Further studies using 100 to 200 mg of the drug are planned.
Investigators working with Gilead presented 48 week data from two Phase II studies: one of the developing "Quad" tablet, a one-pill once-a-day combination of Truvada, the integrase inhibitor elvitegravir, and the pharmacological booster cobicistat, and the other of Truvada/atazanavir boosted by either ritonavir or cobicistat (poster H-938b). Virological efficacy rates were high and similar in both studies, and Quad therapy or atazanavir boosting with cobicistat were well tolerated. There were early mild increases in GFR in arms containing cobicistat, similar to that in the ritonavir-containing arm. These drugs seem to be advancing steadily towards licensure, and three large (700 patient) Phase II studies are underway. The determination of the place of Quad therapy or cobicistat boosting in the choice of and sequence of HIV therapy will rest on the data developed in further studies.
The integrase inhibitor under development by Shionogi/GlaxoSmithKline, S/GSK1349572, was discussed in several presentations. Deanda (abstract H-1166) presented laboratory data supporting the concept that advanced integrase inhibitors such as 572 remain bound to the HIV integrase enzyme target for longer than raltegravir, even in the presence of drug-resistance mutations. Further, polymorphisms in the HIV integrase gene, mutations that occur naturally in the gene in the absence of drug pressure, do not confer drug resistance to 572 (poster H-935). Further, examining clinical data from research studies of 572 this far, there was no evidence that the presence of these polymorphisms reduced the response to 572 in clinical studies. Finally, another study (poster H-934) showed that the pharmacokinetics of 572 are not changed when 572 is given with omeprazole to healthy adults.
Data on one truly new drug was presented. Jim Mullens presented a first-in-man study of the Koronis Pharmaceuticals drug candidate KP1461 (abstract H-1170a). This prodrug is metabolized after administration into the deoxycytidine analog KP1212. In tissue culture experiments, KP1212 increases the mutation rate of HIV to such an extent that after 13 passages in culture (about 6 weeks) viral replication collapses as the virus is too defective to replicate. 10 patients received KP1461 for up to 124 days and an extensive analysis of viral sequences was conducted from these patients in comparison to 10 control patients.
No significant diminution of viral load was observed in patients receiving KP1461. A significant increase in mutations was seen after 56 and 124 days, suggesting the potential of the drug to induce mutations.
Although a novel approach, it is not clear to this author that this drug will prove efficacious. The known toxicity of deoxy nucleoside analogs is of concern, and it is unclear how long the drug might need to be given to have a clinical effect. No affect on viral load was seen in vivo after more than 15 weeks of administration, while the drug acts after a much shorter time in culture. The size of the pool of cells able to produce wild-type (unmutated) virus may be large, and so unmutated virus may be able to emerge for long periods of time. It would be reassuring if modeling could be done to predict the duration of therapy required to induce collapse of the viral population. Published testing of this drug in a SIV model system in vivo could not be found.
ICAAC: Characterization of the resistance profile of TMC278: 48-week analysis of the Phase III studies ECHO and THRIVE - (09/15/10)
ICAAC: The Single-Tablet Regimen Elvitegravir/Cobicistat/Emtricitabine/Tenofovir Disoproxil Fumarate (EVG/COBI/FTC/TDF; "QUAD") Maintains a High Rate of Virologic Suppression, and Cobicistat (COBI) is an Effective Pharmacoenhancer Through 48 Weeks -(09/15/10)
ICAAC: GSK2248761, an NNRTI with Activity Against Common NNRTI Resistance Mutants, Did Not Select for NNRTI Resistance Mutations in Two 7-Day Monotherapy Studies - (09/15/10)
Expanding options for use of current antiretrovirals in special situations
Several presentations provided more information for providers to use "old" drugs. The Aires study (poster H-204) provided outcomes for patients suppressed on therapy with Atazanavir/ritonavir who "simplified" their therapy by discontinuing ritonavir; these patients had achieved suppression over 36 weeks after induction with ATV/r + ABC/3TC, and so unboosted ATZ was acceptable as the regimen did not contain tenofovir. Such a strategy might be useful for patients that could not tolerate continued tenofovir or ritonavir. In the study the patients were randomized to stay on ATV/r + ABC/3TC or switch to ATV + ABC/3TC after week 36. Results of a planned analysis at 120 weeks are presented, and study to week 144 is planned. Sustained virologic suppression was seen; 84 % of ATV- and 83 % of ATV/r-treated subjects maintained <50 c/mL. This strategy may be a viable option for some patients.
Joe Gathe presented results of the VERxVE study of sustained-release nevirapine (poster H-207, abstract H-1808). XR NVP 400 mg qd was compared to standard immediate release (IR) NVP 200 mg BID, both co administered with FTC/TDF. The investigators studied 1011 patients with a screening viral load >1000 copies/mL and a CD4+ cell count <400 cells/mm3 (males) and <250 cells/mm3 (females). A double blind, double dummy non-inferiority study design was used. Randomization stratified by baseline VL (≤100,000 or >100,000 c/mL). The primary endpoint was again confirmed virologic response (<50 c/mL) at week 48 using TLOVR. NVP XR qd was non-inferior to NVP bid, as 75.9% (384/506) for bid NVP and 81.0% (409/505) for NVP XR qd had HIV RNA < 50 copies/ml at week 48. If it becomes available, this may offer yet another options to patients seeking daily therapy.
The Odin trial (poster H-1811) offered another option for once-daily protease inhibitor therapy. A Phase III, randomized, open-label study, Odin demonstrated noninferiority in virologic response of darunavir/r 800/100mg qd vs. 600/100mg bid (+ >2 NRTIs) in treatment-experienced HIV-1-infected adults with no darunavir resistance-associated mutations (RAMs) at screening. DRV has been previously approved for qd therapy in na´ve patients. This study suggested that qd DRV/r may be used in therapy-experienced patients with no DRV-associated mutations.
Another PI regimen that avoided ritonavir was presented (poster H-205). In a small, single center prospective 48-week single arm switch study, 30 patients switched to RAL/ATZ. with extension to 96 weeks. RAL 400 BID and ATV 400mg QD might provide an effective suppressive maintenance regimen for a few very selected, highly compliant patients. However, in this small study 4 of the 30 patients did not complete week 48 successfully, and 3 more before week 96: 4 had virological failure or frequent blips, 3 withdrew due to adverse effects, and 1 was lost. This regimen choice could not be considered preferred, given the sparse data.
ICAAC: 48 Week Efficacy, Pharmacokinetics, and Safety of Once a Day 400 mg Nevirapine (Viramune) Extended Release Formulation for Treatment of Antiretroviral Na´ve HIV-1 Infected Patients - (09/15/10)
ICAAC: Induction/Maintenance-Simplification: Sustained Virologic Efficacy of Atazanavir (ATV) Versus Atazanavir/Ritonavir (ATV/r), each in Combination with Abacavir/Lamivudine (ABC/3TC) over 120 Weeks: the ARIES Trial - (09/15/10)
ICAAC: 48-week resistance and efficacy subgroup analysis of once- versus twice-daily darunavir/ritonavir in ODIN - (09/19/10)
Useful Pharmacologic information on current ART:
There has long been concern about protease inhibitors and the induction of cardiac conduction disturbances. The SMART study team examined associations between protease inhibitors and QTc and PR interval durations in their cohort. (abstract H-218). In 3767 participants in the SMART Trial, of whom 1902 were randomized to receive antiretroviral therapy intermittently, and 1865 received ART continuously, (VS group), heart rate-corrected QT (QTc) was calculated using Bazett's (QTcB) and Frederica's (QTcF) formulae. QTc was variably affected by PI therapies, while all PI-based regimens were associated with prolongation of PR interval. Drug discontinuation did not immediately reverse QTc and PR interval durations. The authors conservatively concluded that as increased QTc is associated with excess sudden cardiac death in the general population, the clinical impact of QT and PR interval changes in HIV-positive individuals should be investigated. However, most changes observed were minor. Perhaps more importantly, attention should be paid to other drugs that alter conduction (eg. calcium-channel blockers) and in such patients who also receive PIs attention should be heightened. This message was amplified by Miller and colleagues (poster H-210) who found that of 229 people taking their first antiretroviral combination about a third of patients had clinically significant interactions involving antiretrovirals. Also, new data was presented involving raltegravir (RAL) and pregnancy (abstract H-1668a). As this is a drug that is safe in pregnancy, the data is important. Consistent with previous reports, RAL PK showed extensive variability. RAL exposure was not consistently altered during 3rd trimester compared to PP and historical data, and the standard dose appears appropriate during pregnancy. Finally, RAL readily crossed the placenta.
ICAAC: Raltegravir Pharmacokinetics during Pregnancy (09/16/10)