Promising compounds for new treatments that were discussed at CROI 2013 included the integrase inhibitor dolutegravir, a new version of tenofovir, a new NNRTI, a new CCR5 inhibitor and several long-acting formulations.
With two phase 3 studies completed, dolutegravir has already been submitted for regulatory assessment in both Europe and the U.S. As a once-daily drug with a low milligram dose and no requirement for boosting, dolutegravir may have advantages over other integrase inhibitors (INSTIs) including raltegravir and the recently approved elvitegravir (as a component of Quad/Stribild).
Interim 24 weeks results were presented from the ongoing international phase 3 SAILING study in 715 treatment-experienced (integrase-naive) patients randomised to either 50 mg dolutegravir once-daily or 400 mg raltegravir twice-daily, each plus matching placebo.1 Patients could use an additional two investigator-selected ARVs, at least one of which had to be fully sensitive. The background combinations were generally robust (PI/r plus tenofovir 40%, lopinavir/r only 10%, darunavir/r plus etravrine 10%).
At baseline, median CD4 count and viral load were approximately 200 cells/mm3 and 15,000 copies/mL, respectively, with approximately half of participants having resistance to three or more classes and a median six years prior ART. Approximately 30% were women, 50% white and 40% African American and 15% had HIV/HCV coinfection.
At week 24, the dolutegravir arm had greater viral suppression compared to raltegravir (79% vs 70% with VL <50 copies/mL; difference 9.7% [95%CI: +3.4, +15.9], p=0.003). However, this was in an analysis that adjusted for baseline viral load, phenotype sensitivity and use of darunavir without PI mutations. The differences were based on fewer discontinuations in the dolutegravir arm (14% vs 17%) and lower rates of virological failure (4% vs 7%). Side effects were broadly similar with few treatment discontinuations in each arm.
In patients with hepatitis B or C coinfection, IRIS-related liver complications were reported more frequently in the patients using dolutegravir (6 vs 3 patients).
The primary endpoint for the study will be results at week 48.
A second late-breaker poster reported that CSF levels of dolutegravir were similar to the unbound fraction in plasma and that this was above the IC50 for wild-type virus (0.2 ng/mL) indicating likely therapeutic levels. This was an open label, single arm intensive PK study in 13 men receiving doultegravir with abacavir/3TC.2
Baseline viral load in CSF and plasma were 3.64 and 4.73 log copies/mL, with 12/13 men achieving undetectable levels at week 16 (using <2 c/mL and <50 c/mL cut-off tests for CSF and plasma respectively). Levels the patients with detectable levels were 5 c/mL and 77 c/mL respectively.
A lack of interaction between dolutegravir and either methadone or combined oral contraceptives (ethinyl estradiol 0.035 mg and norgestimate 0.25 mg) was also reported in a poster showing two drug interaction studies in HIV negative volunteers.3
Several studies were presented on a new prodrug of tenofovir that was previously in development as GS-7340 and now has the generic compound name tenofovir alafenamide (abbreviated to TAF). At a 25 mg dose, TAF results in 7-fold greater intracellular levels of tenofovir with 90% lower plasma levels, compared to 300 mg formulation of tenofovir disoproxil fumerate (TDF).
Early clinical data on TAF compared to TDF in treatment-naive patients was shown in a late breaker oral presentation. Both formulations were included in single tablet, four-drug combinations with elvitegravir, cobicistat and FTC.4
This is an ongoing, double-blind phase 2 study in 170 patients randomised 2:1 to TAF or TDF formulations respectively. The 4-drug combination uses a 10 mg TAF dose as cobicistat boosts TAF by 2.4-fold.
This was a largely male (97%), white (67%) group in early infection. Baseline CD4 and viral load were approximately 400 cells/mm3 (15% were <200) and 40,000 copies/mL (17-28% were >100,000 copies/mL), respectively. Entry criteria included eGFR >70 mL/min, with median baseline levels at 115 mL/min, as with previous studies using cobicistat and tenofovir.
For the primary endpoint of virological suppression at 24 weeks, 87% vs 90% in the TAF vs TDF arms had viral load <50 copies/mL (weighted difference: -4.9%, 95%CI -15.7, +5.9, p=0.36). CD4 increases were similar (+163 vs +177 cells/mm3).
With efficacy expected to be high, the focus on side effects showed similar short-term results. The five side effects occurring in ≥10% of patients were: nausea (18% vs 12%), diarrhea (12% vs 12%), fatigue (12% vs 9%), headache (10% vs 10%), and upper respiratory tract infection (7% vs12%); any grade, TAF vs TDF.
Both arms had an increase in serum creatinine and reduction in eGFR related to use of cobicistat. These occurred by week 2 but then stablised to week 24, and were greater with TDF (-4.9 mL/min vs -11.8 mL/min, p = 0.032).
Mean (+/-SD) bone mineral density (BMD) was reduced less in the TAF arm for both spine [-0.8 (+/-3.4) vs -2.5 (2.5], p = 0.002 and hip [-0.3 (+/-1.8) vs -2.0 (+/-2.7)], p <0.001.
There were no cases of proximal renal tubulopathy or discontinuations for renal events.
First efficacy and safety data in HIV positive people were presented for a new NNRTI with the development name MK-1439 (from Merck) that has in vitro activity against common NNRTI drug resistant mutations (K103N, Y181C and G190A).5
This was a double-blind, placebo controlled, single-site, phase Ib study in 18 treatment-naive men randomised (1:1:1) to 25 mg (n=6), 200 mg (n=6) or placebo (n=3 for each placebo), taken once-daily for seven days as monotherapy. All participants started standard ART from day eight for 10 days to minimise risk of drug resistance during the washout phase.
Mean viral load reductions (90%CI) compared to placebo of -1.37 (-1.60, -1.14) and -1.26 (-1.51, -1.02) log copies/mL in the 25 and 200 mg arms, respectively, with non-significant differences between active doses at all time points.
A total of 21 non-serious side effects were reported in 13/18 participants, including headache (n=5), nausea (n=2), common cold (n=2) and sore throat (n=2). Night sweats, headache (at 200 mg) and loss of appetite (at 25 mg) were considered possibly related to MK-1439. The single serious event was a increase in LFT in one patient on day 7, judged related to acute HCV infection between screening and study entry.
Pharmacokinetic results were similar to those seen in HIV negative studies, with mean concentrations at 24 hours post dose that were 14-fold (25 mg dose) and 87-fold (200 mg dose) higher than the adjusted IC95 for wild-type virus (19 nM, in 50% serum).
Results from a phase 1a safety study in HIV negative people including multiple doses up to 750 mg for ten days were presented as a separate poster, reported a lack of significant interactions with or without food, and that at steady-state, a 12 mg dose produced 24-hour post dose drug levels that remained above the adjusted IC95 for wild-type virus.6 Other phase 1 studies in 140 HIV negative individuals included 14 young women and 12 elderly women, without reports of clinically relevant side effects, including rash or CNS events.
Phase 2b studies continue using 25, 50, 100 and 200 mg doses.
A late-breaker oral presentation of 24-week primary endpoint results from a randomised, double-blind, double placebo, phase 2b of the CCR5/CCR2 inhibitor cenicriviroc in 143 treatment-naive patients, was presented by Joseph Gathe.7
This compound has been in development in various formulations by Tobira for several years (previously as TBR-652). The current study used a 50 mg formulation and randomised patients 2:2:1 to either 100 mg or 200 mg cenicriviroc compared to efavirenz 600 mg, all with matching placebo and plus open label tenofovir/FTC. This was a twice-daily combination with a requirement for cenicriviroc/placebo to be taken as a morning dose following breakfast and efavirenz/placebo to be taken at night.
Baseline characteristics included approximate baseline CD4 and viral load of 400 cells/mm3 (range 77-1090) and 25-40,000 copies/mL (14-25% >100,000), respectively. The study was 94% male; 62% Caucasian, 32% African American; 24% Hispanic. Mean age was 36 (range 19-63)
At week 24, viral suppression to <50 copies/mL (ITT analysis, snapshot algorithm) was achieved by 76% and 73% vs 71% of patients in the 100 mg and 200 mg vs efavrienz arms respectively. Virologic non-response was higher in the cenicrivoc arms (12% and 14% vs 4% efavirenz). Cenicriviroc arms appeared less effective compared to efavirenz in the small percentage of patients with baseline viral load >100,000 copies/mL (50% and 60% vs 75%) although discontinuations due to non-response were similar (20% and 29% vs 25%). In the results stratified by baseline viral load, a range of non-responders due to lack of virologic data at week 24 related to early discontinuation (from 0% with efavirenz at >100,000 to 29% with efavirenz at <100,000), complicated the interpretations of these results.
Better viral responses were reported with highest quartile (141-400 ng/mL) of modeled Cmin trough concentrations of cenicriviroc (100%) with 12%, 9%, and 17% non responders in Q3 (70-141 ng/mL), Q2 40-71 ng/mL) and Q1 13-40 ng/mL), respectively, showing a wide range of interpatient variability.
CD4 changes from baseline were similar (+147 and +170 vs +135 cells/mm3).
Discontinuation related to side effects was significantly more frequent with efavirenz (0% and 2% vs 18%) as were grade 3 events (2% and 4% vs 11%). There were no grade 4 events, serious events or deaths in the study.
Laboratory abnormalities were higher in the 200 mg arm, principally increased creatinine phosphokinase, but these generally resolved without treatment discontinuation.
Resistance mutations in patients with viral load rebounding to >400 copies/mL were predominantly M814V/I in 5 patients taking cenicriviroc (vs 0% in the efavirenz arm).
The impact of CCR2 blocking on the monocyte activation pathways was seen by dose related increases in the CCR2 ligand MCP-1 of approximately 450 ng/L in the 100 mg arm and 750 ng/L in the 200 mg arm. Both cenicriviroc arms also reported a reduction in levels of the monocyte activation marker of soluble CD14 of -0.2 vs +1.3 x 10(6) pg/mL in the efavirenz group. Soluble CD14 has been associated with an increased risk of all-cause mortality independent of CD4 and viral load and this potential was highlighted in the conclusion as a property of cenicriviroc that warranted additional research.
A new formulation of cenicriviroc will be used for phase 3 studies although the dose for fuuture research has still to be decided. The company intends to coformulate cenicriviroc with other ARVs although this is currently only at a preliminary planning stage.
Studies on long acting parental and nanoformulations are reported below in a separate article,8 but an oral presentation included in vitro data on maturation inhibitor showing that this potential drug class remains a focus for ongoing research.9
Maturation inhibitors target the final stage of HIV gag processing that inhibit release of fully formed capsid, with early studies focused on the compound beviramat (as PA-457). This was of most interest for drug-resistant HIV, with viral load reductions of approximately -1.2 log in treatment responders, but that common polymorphisms at baseline, principally V370A, present in 50% of patients, correlated with non-response. Although early phase 1/2 studies raised no safety concerns, the development of beviramat was discontinued in June 2010 (by Myriad who had bought the compound from Panacos).
The study at CROI provided in vitro data on second-generation maturation inhibitor molecules developed to overcome V370A. Results were presented by Carl Wild from DFH Pharma (who was previously involved with the Panacos team) and the group collaborated with researchers at the U.S. National Cancer Institute.
The IC50 for DFH-055 had similar activity at (0.032 uM) to wild type and V370A (compared to <0.08 and >32.0 uM for wild-type and V370 respectively for beviramat). Current best compounds (DFH-068 and DFH-070) further improved on activity against V370A with 5-fold greater sensitivity compared to DFH-055 at 30.1 and 38.3 nM, respectively. Although these results are encouraging the presenter was cautious in not announcing whether either molecule had been selected as a lead compound for further development.
Unless stated otherwise, references are to the Programme and Abstracts of the 20th Conference on Retroviruses and Opportunistic Infections (CROI), 3-6 March 2013, Atlanta.
Links to other websites are current at date of posting but not maintained.