icon-    folder.gif   Conference Reports for NATAP  
  17th CROI
Conference on Retroviruses
and Opportunistic Infections
San Francisco CA
February 16-19, 2010
Back grey_arrow_rt.gif
Antiretrovirals For Prevention: tenofovir, FTC, maraviroc; rectal, vaginal, semen.
  Reported by Jules Levin
CROI 2010 Feb 16-19 SF
Efficacy of Intermittent Prophylaxis with Tenofovir and Emtricitabine against Rectal SHIV Transmission in Macaques and Relationship to Systemic and Mucosal Drug Levels
Gerardo Garcia-Lerma*1, M-E Cong1, Q Zheng1, A Holder1, A Martin1, C-C Lin1, R Otten1, W Lee2, C-P Pau1, and W Heneine1 1CDC, Atlanta, GA, US and 2Gilead Sci, Foster City, CA, US
Background: Intermittent pre-exposure prophylaxis (iPrEP) with long-acting drugs is a promising HIV prevention strategy. We evaluated in macaques the efficacy of iPrEP dosing regimens with Truvada (FTC and TDF) or GS7340, a next-generation tenofovir (TFV) pro-drug that at low doses results in high TFV-diphosphate (TFV-DP) levels in peripheral blood mononuclear cells (PBMC). We also assessed drug pharmacokinetics and their relationship to efficacy.
Methods: Efficacy was evaluated by a repeat-challenge macaque model consisting of up to 14 weekly rectal SHIV exposures. Truvada was given orally to groups of 6 macaques according to the following schedule (times are relative to virus exposure): +2h/+26h, -2h/+24h, -22h/+2h, -3d/+2h, and -7d/+2h. Two additional groups of 6 animals received one dose of Truvada or GS7340 3 days prior to exposure without a post-exposure dose. Drug levels were evaluated at first dose in plasma, PBMC, rectal secretions, and tissues. The Cox proportional hazards model was used to estimate risks of infection in treated animals relative to 32 untreated controls.
Results: iPrEP modalities with Truvada containing both a pre- and a post-exposure dose or 2 post-exposure doses significantly reduced the risk of infection by 4.1-fold (+2h/+24h, P =0.03), 4-fold (-2h/+24h, P =0.02), 16.7-fold (-22h/+2h, P =0.006), 15.4-fold (-3d/+2h, P =0.008) and 9.3-fold (-7d/+2h, P =0.003). The half-life of TFV-DP and FTC-TP in PBMC was 115 h and 24 h, respectively. FTC, but not TFV, was consistently detected in rectal secretions 2 to 5 h after dosing; concentrations of both FTC and TFV in secretions were highest at 24 hr (median = 743 and 1186 ng/mL, respectively). TFV-DP levels in rectal tissues were also high at 24 hr and remained high for 2 days to 7 days. Despite the long persistence of TFV-DP in PBMC and high TFV-DP levels in rectal tissues, a single - 3 day dose of Truvada was not effective (HR reduction = 0.5, P =0.16). A single -3d GS7340 dose was also ineffective (HR reduction = 0.5, P =0.24) despite the high TFV-DP levels seen at exposure (median = 773 fmol/106 PBMC). Median TFV-DP levels in PBMC after GS7340 dosing were 100-fold higher than after TDF dosing.
Conclusions: A single pre-exposure dose with Truvada or GS7340 was not sufficient to protect macaques despite the long persistence of TFV-DP. Combined with a post-exposure dose, iPrEP with Truvada was highly protective. The rapid FTC penetration in rectal tissues suggests that FTC plays an important role in the protection contributed by the post-exposure dose.

Protection of Rhesus Macaques from Vaginal Infection by Maraviroc, an Inhibitor of HIV-1 Entry via the CCR5 Co-receptor: "these findings validate maraviroc development as a vaginal microbicide for women"
R Veazey1, T Ketas2, J Dufour1, P Klasse2, and John Moore*2 1Tulane Natl Primate Res Ctr, Covington, LA, US and 2Weill Cornell Med Coll, New York, NY, US
Background: An effective vaginal microbicide could reduce human immunodeficiency virus type 1 (HIV-1) transmission to women. The detergent- or polyanion-based microbicide candidates have all failed to demonstrate efficacy in clinical trials, so the emphasis has finally switched to anti-retroviral drugs. Among microbicide candidates now in clinical development is Maraviroc (MVC), an FDA-approved small molecule drug that binds the CCR5 co-receptor and impedes HIV-1 entry into cells.
Methods: Delivered systemically, MVC reduces viral load in HIV-1-infected people, but its ability to prevent transmission is untested. We have now evaluated MVC as a vaginal microbicide, using a stringent model involving challenge of rhesus macaques with a high-dose of a CCR5-using virus, SHIV-162P3.
Results: Gel-formulated MVC derived from prescription-grade tablets provided dose-dependent protection, half-maximally at 0.5 mM (0.25 mg/mL) and fully at 6 mM (3 mg/mL). The duration of protection was transient; the longer the delay between MVC application and virus challenge, the less protection (T1/2 ∼ 4 h). As expected, MVC did not protect against vaginal challenge with a CXCR4-using virus, SHIV-KU1, but its presence also did not exacerbate post-infection viremia in animals infected with this virus.
Conclusions: These findings validate MVC development as a vaginal microbicide for women, and should guide clinical programs. Of note is that a single 300 mg tablet of prescription-grade MVC, which retails for ∼$15 in the USA, contains enough active drug to fully protect about 25 macaques, and by implication a similar number of women, if applied vaginally in a gel.

Antiretrovirals for Prevention: Maraviroc Exposure in the Semen and Rectal Tissue of Healthy Male Volunteers after Single and Multiple Dosing
Kevin Brown*, K Patterson, S Malone, N Shaheen, H Prince, J Dumond, M Spacek, P Heidt, M Cohen, and A Kashuba Univ of North Carolina at Chapel Hill, US
Background: MVC is a CCR5 antagonist approved for use in patients with R5-tropic HIV infection. The pharmacology of ARV in multiple compartments provides insight into the development of ARV resistance and prevention of sexual transmission of HIV. This is the first study to investigate MVC exposures in semen (SE) and rectal tissues (RT).
Methods: An open-label pharmacokinetic (PK) study was performed in 12 HIV negative men receiving MVC 300mg BID for 8 days. Seven blood plasma (BP) samples were collected over a 12 hr interval on Day 1 (PK1) and Day 7/8 (PK2); 1 RT sample from each subject was collected during PK1 and PK2. For PK1, 2 semen samples paired with 2 blood plasma samples were collected from each subject. For PK2, 6 semen samples were collected over 2 days paired with 6 blood plasma samples. MVC concentrations were measured by validated LC/MS assays with a 1 ng/mL LLQ. Data were analyzed by noncompartmental methods (WinNonlin 6). A RT density of 1.05 g/mL was assumed for comparisons. Demographics and PK parameters are reported as median (range), and PK ratios are reported as geometric means (GMR) (90% CI).
Results: Subject age was 22 (20 to 42) yrs and BMI was 25.1 (20.2 to 29.1) kg/m2. Eight were Caucasian. PK results are reported in the following table:

The MVC accumulation ratio (PK2 AUC12h: PK1 AUC12h) was 1.29 for BP, 1.34 for semen, and 4.10 for rectal tissue. After single and multiple dosing, MVC AUCs in semen were 56% and 62% of BP, respectively. After a single and multiple dosing, MVC AUC in RT was 9-fold and 28-fold that observed in blood plasma.
Conclusions: Contrary to our previous observation of 3-fold higher MVC exposures in cervicovaginal fluid compared to blood plasma, MVC exposures in semen were 38% lower than blood plasma after multiple doses. However, if MVC protein binding in semen is low, mucosal exposure of MVC may still exceed the protein-free HIV IC90 (0.5 ng/mL). Rectal tissue exposure after multiple dosing was 10-fold higher than what we previously measured for vaginal tissue (2-fold higher than blood plasma). This finding is likely due to fecal elimination and mucous trapping of MVC. These data suggest that MVC may prevent or reduce propagation of HIV in the gastrointestinal tract. Virologic and immunologic investigations are ongoing.

Validating Measures of Tenofovir Drug Exposure in a US Pre-exposure Prophylaxis Trial
Albert Liu*1,2, E Vittinghoff2, M Gandhi2, Y Huang2, K Chillag3, R Wiegand3, P Anderson4, R Grant2, R Greenblatt2, and S Buchbinder1,2 1San Francisco Dept of Publ Hlth, CA, US; 2Univ of California, San Francisco, US; 3CDC, Atlanta, GA, US; and 4Univ of Colorado Denver, US
Background: Several pre-exposure prophylaxis (PrEP) trials are evaluating the safety and efficacy of regimens including tenofovir (TFV) for HIV prevention. Accurate measures of drug exposure that capture differences in adherence as well as pharmacokinetics are critical for interpreting study outcomes.
Methods: Single samples of plasma, peripheral blood mononuclear cells, and hair were collected at the same or adjacent visit from 89 HIV-uninfected men in a placebo-controlled PrEP trial and tested for TFV in a blinded fashion. Adherence measures including pill counts, electronic monitoring (MEMS), and self-report were collected at the corresponding visit. TFV in plasma and hair and intracellular tenofovir diphosphate (TFV-DP) in peripheral blood mononuclear cells (PBMC) were quantified using LC/MS/MS. Spearman correlation coefficients between drug levels in different specimens and with adherence measures were calculated.
Results: Median TFV levels were 83 (range <10 to 367) ng/mL in plasma, 0.05 (<0.01 to 0.21) ng/mg in hair, and 40 (<5 to 102) fmol/million cells in PBMC for the active arm. The sensitivity and specificity of TFV detection was 94% (44/47 active arm samples detectable) and 98% (1/41 placebo samples detectable) for plasma; 94% (44/47) and 92% (1/12) for PBMC; and 98% (46/47) and 98% (1/42) for hair. One of 42 placebo subjects had detectable TFV levels in plasma, PBMC, and hair, suggesting TFV exposure occurred. Five of 47 active-arm subjects had undetectable drug in at least one sample, most commonly plasma and PMBC (both 3/5) followed by hair (1/5). The Spearman correlation was 0.62 for plasma and PBMC levels, 0.41 for PBMC and hair, and 0.21 for plasma and hair. Low levels of correlation were observed between drug exposure and adherence measures: correlation with pill counts, MEMS, and self-report were 0.36, 0.27, and 0.06 for plasma; 0.17, 0.03, and 0.12 for PBMC, and 0.36, 0.22, and 0.14 for hair. Moderate correlations were seen between adherence measures: pill counts/self-report (0.45); pill counts/MEMS (0.51); and MEMS/self-report (0.30).
Conclusions: Analyses of plasma, PBMC, and hair samples were highly sensitive and specific for TFV exposure in this PrEP trial. The modest correlation between drug exposure and adherence measures may reflect reporting bias, or variations in drug absorption, clearance, or transport. Indices of long-term drug exposure, afforded with hair and PBMC analysis, are candidate surrogate markers for PrEP activity, which require further validation in efficacy trials.

PRO2000 Vaginal Gel is Ineffective in Preventing HIV Infection: Results of the MDP301 Phase III Microbicide Trial
M Chisembele1, A Crook2, M Gafos3, R Hayes4,5, U Jentsch6, A Kamali7, C Lacey8, Sheena McCormack*9, G Ramjee10, H Rees11, and the MDP Team 1Univ Teaching Hosp, Lusaka, Zambia; 2Med Res Council Clin Trials Unit, London, UK; 3Africa Ctr for Hlth and Population Studies, Mtubatuba, South Africa; 4London Sch of Hygiene and Tropical Med, UK; 5NIMR/AMREF/LSHTM Collaborative Res Projects, Mwanza, Tanzania; 6Contract Lab Svcs, Johannesburg, South Africa; 7Med Res Council/Uganda Virus Res Inst, Entebbe; 8Univ of York, UK; 9Imperial Coll London, UK; 10Med Res Council, Durban, South Africa; and 11Univ of the Witwatersrand, Johannesburg, South Africa
Background: PRO2000 is a synthetic naphthalene sulphonated polymer. A concurrent Phase IIb trial (HPTN035) reported a non-significant 30% reduction in HIV incidence in women using 0.5% PRO2000 vaginal gel compared to placebo. MDP301 assessed 0.5% and 2% PRO2000 with 80% power to detect a 35% reduction.
Methods: Healthy women were randomized to receive 0.5%, 2% or placebo gel for 12 months (up to 24 months in Uganda) to be applied pre-sex at 6 centers in East and Southern Africa. All women received condoms and safe sex counselling. The 2% gel was discontinued in Feb 2008 on the recommendation of the independent Data Monitoring Committee. HIV status was determined using rapid tests and/or ELISA at screening and weeks 12, 24, 40, and 52 (up to 104 in Uganda). Seroconversions were confirmed in a central laboratory including review of status at enrolment and week 4. The primary efficacy analysis was modified intention to treat (MITT) up to week 52 and censored for pregnancy; secondary analyses included ITT using all available data. Women were categorized as consistent gel users if they reported using gel during the last sex act at >92% of their visits and had attended a minimum of 7/13 visits. Analyses comparing 2% to placebo were censored at 14 Feb 2008.
Results: In this study, 15818 women were screened and 9385 enrolled; 95% of women contributed to the efficacy analysis resulting in 84% of the maximum possible person-years (person-years) for the MITT analysis (88% for ITT). The overall HIV incidence was 4.6/100 person-years with 418 seroconversions; 253 were included in the MITT comparison of 0.5% and placebo, incidence 4.5 (130/2873) and 4.3 (123/2836)/100 person-years, respectively. The HR for 0.5%:placebo was 1.05 (95%CI 0.82 to 1.34); for ITT: HR=1.00 (95%CI 0.79 to 1.26). The mean percentage reported gel use during the last sex act was 89.3%. The HR for consistent users (58% of women) was 0.97 (95%CI 0.87 to 1.35) and for inconsistent users 1.17 (95%CI 0.42 to 1.72). In the comparison of 2% to placebo, the incidence was 4.7 (82/1741) and 3.9 (67/1717)/100 person-years respectively; HR = 1.21, 95%CI 0.88 to 1.68. There were no differences in primary safety, local or systemic toxicity endpoint rates between the 0.5%, 2%, or placebo arms.
Conclusions: Neither concentration of PRO2000 was protective against HIV infection in this large Phase III trial, which definitively ruled out substantial benefit. Retention and adherence were both high, suggesting that PRO2000 had insufficient in vivo potency.

Association of Expanded HAART Coverage with a Decrease in New HIV Diagnoses, Particularly among Injection Drug Users in British Columbia, Canada
Julio Montaner*, E Wood, T Kerr, B Yip, V Lima, K Shannon, R Harrigan, and R Hogg BC Ctr for Excellence in HIV/AIDS, Vancouver, Canada
Background: We recently described an association between expanded HAART coverage, decreased "community plasma HIV-1-viral load" and decreased HIV transmission within two long-standing research injection drug users (IDU) cohorts in the downtown eastside (DTES) of Vancouver. However, prospective populational evaluation of this association is lacking.
Methods: We used administrative records to evaluate the association between expansion of HAART coverage, plasma HIV-1-viral load and new HIV diagnoses in BC. HAART and medical services are available free of charge for all British Columbia residents. HIV testing, HAART distribution, and plasma HIV-1-viral load determinations are centralized in British Columbia. A targeted effort to expand HAART outreach among IDU was initiated in 2007. Data for second half of 2009 is preliminary due to delayed reporting, therefore only the first half of 2009 was used for statistical analyses.
Results: Following the initial HAART roll out in 1996 to 1999, the number of individuals on HAART remained remarkably stable in British Columbia until 2003 (Fig 1, blue line, left axis). The number of individuals on HAART increased from ∼2500 to ∼5000, from 2004 to 2009. As previously described, the initial HAART roll out was associated with a ∼50% decrease in new HIV diagnoses (first half of top red line, right axis, P =0.027). The HAART expansion in 2004 to 2009 was associated with a second decrease in new HIV diagnoses (second half of top red line, right axis slope calculated quarterly counts until June 2009, P =0.001). Of note, a ∼50% a decrease in new HIV diagnoses among IDU (bottom red line, right axis) was evident after 2007 and the proportion of HIV infected IDU in BC with plasma HIV-1-RNA levels above 1500 copies/mL, as a surrogate for high "community plasma HIV-1-viral load, decreased steadily from ∼50% between 2000 to 2004 to ∼20% in 2009 (Fig 2 P <0.001).
Conclusions: The results of this prospective population study demonstrate an association between expanded HAART coverage, decreased "community plasma HIV-1-viral load", and decreased new HIV diagnoses, temporally related to a HAART outreach effort targeting IDU.