SHORT - FORM 1 Multipurpose prevention approaches with antiretroviral - based formulations 2 3

24 We compared the preclinical safety and efficacy of Tenofovir (TFV) 1% gel and 25 MZC gel containing 50 μM MIV-150 (M), 14 mM Zn(O2CCH3)2(H2O)2 (Z) and 3% 26 carrageenan (C) through a series of in vitro, ex vivo and in vivo assays. Both gels 27 showed good antiviral therapeutic indexes (>25-800). MZC showed greater anti28 SHIV-RT activity than TFV 1% gel in rhesus macaque vaginal explants. MZC 29 protected mice from vaginal HSV-2 challenge (p<0.0001), but the TFV 1% gel did 30 not. 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 on Jauary 8, 2018 by gest httpaac.asm .rg/ D ow nladed fom MAIN TEXT 47 The most recent lead microbicide formulations, Tenofovir (TFV) 1% gel and 48 dapivirine intravaginal ring (IVR), have been investigated as a means to prevent 49 primarily human immunodeficiency virus (HIV) acquisition. Recent clinical trials 50 like CAPRISA 004 (1, 2), conducted by the Centre for the AIDS Programme of 51 Research in South Africa, and the VOICE trial (Vaginal and Oral Interventions to 52 Control the Epidemic) (3) have investigated the effectiveness of TFV 1% gel to 53 prevent HIV acquisition. CAPRISA 004 was a study comparing TFV 1% vaginal 54 gel with a placebo when used before and after sex, while the VOICE trial looked 55 at daily use of an oral tablet (TFV or Truvada) or a vaginal gel (TFV 1% gel). 56 CAPRISA 004 and the VOICE trial showed a reduction in herpes simplex virus-2 57 (HSV-2) acquisition that correlated with TFV vaginal gel use. This was surprising 58 since TFV has modest anti-HSV activity in cell-based assays (4), ex vivo explants 59 (4) and murine models (4, 5). Herein we explored the preclinical safety and 60 efficacy of a dual compartment multipurpose prevention technology (MPT) 61 microbicide gel (MZC: 50 μM MIV-150 (M), 14 mM zinc acetate dihydrate (Z) and 62 3% carrageenan (C)) targeting HIV, HSV and human papillomavirus (HPV) (6), in 63 a side-by-side comparison with the TFV 1% gel. A recently completed Phase 3 64 trial (FACTS 001) of TFV 1% gel was designed following the CAPRISA 004 65 dosing strategy but with a larger number of participants. The gel was not proven 66 to be effective and poor user adherence to the dosing regimen likely contributed 67 to the outcome of the trial. A reduced glycerin version of this gel is currently in a 68 Phase 2 trial as a potential rectal microbicide (7). 69 on Jauary 8, 2018 by gest httpaac.asm .rg/ D ow nladed fom 70 MZC, TFV 1% and 3% carrageenan (CG) gels were formulated as previously 71 described (1, 6). MZC, like TFV 1% gel, is a clear and semisolid formulation. 72 However, based on physicochemical properties, the formulations differ mainly in 73 pH (TFV 1%= 5.0; MZC= 6.9) and osmolality (TFV 1% gel= 3358 mOsmol/kg; 74 MZC gel= 447 mOsmol/kg). MZC gel contains only 0.002% of the non-nucleoside 75 reverse transcriptase inhibitor (NNRTI) MIV-150 versus 1% of the NRTI TFV in 76 the 1% gel. 77 Antiviral activity against HIV-1 was tested using the standardized TZM-bl-based 78 assay (6) or the PBMC-based assay (6). Briefly, TZM-bl cells (1.5x10/ml) or 79 activated PBMCs (2x10/ml) were treated for 1h with dilutions of gels (triplicates) 80 before adding 100 focus forming units (FFU) or 100 TCID50 of virus, respectively. 81 TZM-bl cells were incubated for 72h before staining with X-Gal to count FFU. The 82 supernatant was replaced in PBMCs with fresh stimulation media on d 1 and 4 83 post infection. The p24 level in the supernatant was determined on d 7 after 84 infection by p24 ELISA (Zeptometrix, Buffalo, NY). Cytotoxicity (CC50 values) for 85 each gel formulation was estimated using XTT and CyQuant by running the 86 antiviral assay in the absence of virus (6). The EC50 values were calculated 87 based on gel dilution factor in order to compare efficacy of the two gels, each 88 containing a different active pharmaceutical ingredient (API). By comparing EC50 89 based on gel dilution we observed how MZC with only 0.002% of MIV-150 can 90 achieve better or similar antiviral activity than TFV gel containing 1% of the API. 91 MZC gel was generally more potent than TFV 1% gel in blocking HIV-1 infection 92 on Jauary 8, 2018 by gest httpaac.asm .rg/ D ow nladed fom in TZM-bl or PBMC with the clear exception of one multidrug resistant strain 93 (MDR; OL-1/4(II)d4) containing 2 mutations (K101E and Y181I) associated with 94 decreasing susceptibility of viruses to NNRTI (Table 1). Similarly, TFV 1% gel 95 showed an increase in EC50 values for two strains (71361-1 and 56252-1) 96 containing the 65R amino acid change associated with HIV resistance to TFV. 97 Although NNRTIs are known to select resistant viruses rapidly, MIV-150 seems 98 to select resistance at a slower pace compared to other NNRTIs and requires 99 two or more mutations in a single genome to decrease HIV susceptibility (8). 100 Additionally, it is important to mention that resistance development in topically 101 applied antiretrovirals is not fully understood yet. 102 103 Cell-based assay are excellent tools for screening potential microbicides, testing 104 antiviral properties against a variety of isolates/ MDR strains and monitoring the 105 stability of formulations. However, the testing of a lead formulation in the explant 106 model allows for assessment of preclinical safety and efficacy in a more relevant 107 HIV target cell and architectural context. We tested MZC and TFV 1% gel in our 108 ex vivo rhesus macaque (RM) vaginal explant model using not only cell-free virus 109 inoculum but also cell-associated virus. 110 111 Macaque vaginal mucosa (biopsies or necropsy tissues) were collected and 112 transported overnight as previously described (9). Histological analysis was 113 performed on polarized macaque vaginal tissue explants after overnight 114 treatment (~18h) with neat MZC, TFV 1% or CG gels following the procedure 115 on Jauary 8, 2018 by gest httpaac.asm .rg/ D ow nladed fom described in Barnable et al (9). Neither gel induced vaginal epithelial damage 116 (Figure 1A) or decreased viability (MTT data not shown). The antiviral activity of 117 diluted MZC, CG (1:100) or TFV 1% (1: 30 or 1:100) gels against ex vivo cell-free 118 SHIV-RT challenge was performed as described in Barnable et al (9) in 119 immersion culture model. Briefly, explants were immersed overnight (~18h) in 120 1:30 and 1:100 diluted MZC, 1%TFV or CG (vs. untreated controls) in the 121 presence of PHA/IL2. The diluted gels did not decrease tissue viability (MTT data 122 not shown). After exposure to the gels, the explants were washed and then 123 challenged with SHIV-RT (10 TCID50/explant) 24h or 4d after gel exposure. 124 Following 18h of viral challenge, the tissues were washed and cultured for 14d in 125 the presence of IL-2. SIV p27 release was measured at 0, 3, 7, 11, 14d of culture 126 by ELISA. Cell-associated SHIV-RT infection was performed following a different 127 model (10) where macaque vaginal explants were immersed overnight in media 128 containing diluted MZC, 1%TFV or CG gels (vs. untreated controls). The samples 129 were then washed and cultured for 24h or 4d before being challenged with 130 mitomycin-C-treated, SHIV-RT-infected PBMC (10 infected PBMC/27 TCID50 131 per explant; 2-4 replicates) for ~18h, washed and maintained in culture media in 132 the presence of IL-2 (100 U/ml) (vs. 10μM 3TC control). Mitomycin-C-treated, 133 SHIV-RT-infected PBMC cultured alone were included as controls. SHIV-RT p27 134 concentration was determined in supernatants during 14d culture. Soft endpoint 135 (SOFT) analysis demonstrated that MZC and TFV 1% gels significantly reduce 136 SHIV-RT infection from both cell free (Figure 1B) and cell-associated (Figure 1C) 137 challenges, with MZC being more effective against cell-free virus 24h after 138 on Jauary 8, 2018 by gest httpaac.asm .rg/ D ow nladed fom diluted gel (1:100) application (p<0.0001). Similar results were observed with 139 cumulative analysis (not shown). 140 141 We have previously shown that the combination of CG and zinc acetate (as in 142 the MZC formulation) results in antiviral synergy (in vitro and in vivo) against 143 HSV-2 (11). We explored the anti-HSV-2 activity of MZC and TFV 1% gels in a 144 murine model. Depo-Provera-treated Balb-C mice were dosed with 10 μl of test 145 gel intravaginally 1h prior to HSV-2 infection plus 1h after HSV-2 infection to 146 mimic the BAT24 dosing strategy used in CAPRISA 004 and FACTS 001 (12). 147 Mice were challenged with 10 μl of HSV-2 G (5x10 pfu/mouse) and examined 148 and scored daily for 21d as previously described (11). Despite being tested in a 149 less stringent murine model (lower virus inoculum compare to previous 150 evaluation in this model (6, 11)), the TFV 1% gel did not protect mice from HSV-2 151 infection while MZC gel protected 100% of the animals (Figure 2). 152 153 A possible explanation for these divergent results (when compared to CAPRISA 154 004 in humans) is that TFV phosphorylation and/or TFV uptake may be less 155 efficient in mice than in human cells. In fact, sub-therapeutic (below LLOQ < 100 156 ng/g) TFV diphosphate (TFV-DP) levels were found in murine cervicovaginal 157 tissue, and even TFV-only levels were low (median = 5400 ng/g) as determined 158 by LC-MS/MS (13). 159 160 on Jauary 8, 2018 by gest httpaac.asm .rg/ D ow nladed fom MZC’s potent in vitro and in vivo anti-HPV activity makes this formulation a very 161 appealing MPT candidate targeting three non-curable viral STIs (6). However, 162 poor adherence in clinical trials is an important issue that has overshadowed 163 success of microbicide gels in the HIV pre-exposure prophylaxis field. In light of 164 this, the MZC combination could be explored not only as a gel (with potential for 165 a rectal microbicide) but also an IVR that incorporates levonorgestrel (LNG) to 166 prevent unintended pregnancy (14) (a similar approach is being tested in Phase 167 1 trial with a TFV + LNG IVR [NCT02235662] (15)). Importantly, the results 168 shown in this paper inform about API levels that need to be released from 169 alternatives delivery systems (e.g. IVR) in order to be safe and achieve 170 protection against HIV infection. Adding a contraceptive, targeting more than one 171 STI and providing different choices for drug delivery may increase 172 demand/uptake as well as efficiencies in delivery and access. The MZC173combination is a promising MPT that was successfully evaluated in a Phase 1174 trial (Population Council #558), and the results shown herein support moving175forward with its clinical evaluation.176177ACKNOWLEDGMENTS178This research was supported by United States Agency for International179Development (USAID), under the terms of Award No. GPO-A-00-04-00019.180181REFERENCES182onJauary8,2018bygesthttpaac.asm.rg/Downladedfom 1. Abdool Karim Q, Abdool Karim SS, Frohlich JA, Grobler AC, Baxter C,183Mansoor LE, Kharsany AB, Sibeko S, Mlisana KP, Omar Z, Gengiah184TN, Maarschalk S, Arulappan N, Mlotshwa M, Morris L, Taylor D. 2010.185Effectiveness and safety of tenofovir gel, an antiretroviral microbicide, for186the prevention of HIV infection in women. Science 329:1168-1174.1872. Abdool Karim SS, Abdool Karim Q, Kharsany AB, Baxter C, Grobler188AC, Werner L, Kashuba A, Mansoor LE, Samsunder N, Mindel A,189Gengiah TN, Group CT. 2015. Tenofovir Gel for the Prevention of Herpes190Simplex Virus Type 2 Infection. N Engl J Med 373:530-539.1913. Marrazzo JM, Ramjee G, Richardson BA, Gomez K, Mgodi N, Nair G,192Palanee T, Nakabiito C, van der Straten A, Noguchi L, Hendrix CW,193Dai JY, Ganesh S, Mkhize B, Taljaard M, Parikh UM, Piper J, Masse B,194Grossman C, Rooney J, Schwartz JL, Watts H, Marzinke MA, Hillier195SL, McGowan IM, Chirenje ZM. 2015. Tenofovir-based preexposure196 prophylaxis for HIV infection among African women. The New England197journal of medicine 372:509-518.1984. Andrei G, Lisco A, Vanpouille C, Introini A, Balestra E, van den Oord199J, Cihlar T, Perno CF, Snoeck R, Margolis L, Balzarini J. 2011. Topical200 tenofovir, a microbicide effective against HIV, inhibits herpes simplex201virus-2 replication. Cell host & microbe 10:379-389.2025. Nixon B, Jandl T, Teller RS, Taneva E, Wang Y, Nagaraja U, Kiser PF,203Herold BC. 2014. Vaginally delivered tenofovir disoproxil fumarate204provides greater protection than tenofovir against genital herpes in a205onJauary8,2018bygesthttpaac.asm.rg/Downladedfom murine model of efficacy and safety. Antimicrob Agents Chemother20658:1153-1160.2076. Kizima L, Rodriguez A, Kenney J, Derby N, Mizenina O, Menon R,208Seidor S, Zhang S, Levendosky K, Jean-Pierre N, Pugach P, Villegas209G, Ford BE, Gettie A, Blanchard J, Piatak M, Jr., Lifson JD, Paglini G,210Teleshova N, Zydowsky TM, Robbiani M, Fernandez-Romero JA. 2014.211A potent combination microbicide that targets SHIV-RT, HSV-2 and HPV.212PloS one 9:e94547.2137. Microbicides Trial Network. 2015. Accessed November 16, 2015.214 http://www.mtnstopshiv.org/news/studies/mtn017.2158. Hsu M, Keele BF, Aravantinou M, Krawczyk N, Seidor S, Abraham CJ,216Zhang S, Rodriguez A, Kizima L, Derby N, Jean-Pierre N, Mizenina O,217Gettie A, Grasperge B, Blanchard J, Piatak MJ, Jr., Lifson JD,218Fernandez-Romero JA, Zydowsky TM, Robbiani M. 2014. Exposure to219 MIV-150 from a high-dose intravaginal ring results in limited emergence of220drug resistance mutations in SHIV-RT infected rhesus macaques. PloS221one 9:e89300.2229. Barnable P, Calenda G, Ouattara L, Gettie A, Blanchard J, Jean-Pierre223N, Kizima L, Rodriguez A, Abraham C, Menon R, Seidor S, Cooney224ML, Roberts KD, Sperling R, Piatak M, Jr., Lifson JD, Fernandez-225Romero JA, Zydowsky TM, Robbiani M, Teleshova N. 2014. A MIV-226150/zinc acetate gel inhibits SHIV-RT infection in macaque vaginal227explants. PloS one 9:e108109.228onJauary8,2018bygesthttpaac.asm.rg/Downladedfom 10. Barnable P, Calenda G, Bonnaire T, Menon R, Levendosky K, Gettie229A, Blanchard J, Cooney ML, Fernandez-Romero JA, Zydowsky TM,230Teleshova N. 2015. MIV-150/zinc acetate gel inhibits cell-associated231simian-human immunodeficiency virus reverse transcriptase infection in a232macaque vaginal explant model. Antimicrob Agents Chemother 59:3829-2333837.23411. Fernandez-Romero JA, Abraham CJ, Rodriguez A, Kizima L, Jean-235Pierre N, Menon R, Begay O, Seidor S, Ford BE, Gil PI, Peters J, Katz236D, Robbiani M, Zydowsky TM. 2012. Zinc acetate/carrageenan gels237 exhibit potent activity in vivo against high-dose herpes simplex virus 2238 vaginal and rectal challenge. Antimicrobial agents and chemotherapy23956:358-368.24012. Rees H, Delany-Moretlwe SA, Lombard C, Baron D, Panchia R, Myer241L, Schwartz JL, Doncel GF, Gray G. 2015. FACTS 001 Phase III Trial of242 Pericoital Tenofovir 1% Gel for HIV Prevention in Women, abstr CROI2432015, Seattle, Washington,24413. Clark MR, Friend DR. 2012. Pharmacokinetics and topical vaginal effects245of two tenofovir gels in rabbits. AIDS research and human retroviruses24628:1458-1466.24714. Ugaonkar SR, Wesenberg A, Wilk J, Seidor S, Mizenina O, Kizima L,248Rodriguez A, Zhang S, Levendosky K, Kenney J, Aravantinou M,249Derby N, Grasperge B, Gettie A, Blanchard J, Kumar N, Roberts K,250Robbiani M, Fernandez-Romero JA, Zydowsky TM. 2015. A novel251onJauary8,2018bygesthttpaac.asm.rg/Downladedfom intravaginal ring to prevent HIV-1, HSV-2, HPV, and unintended252pregnancy. J Control Release 213:57-68.25315. Clark JT, Clark MR, Shelke NB, Johnson TJ, Smith EM, Andreasen254AK, Nebeker JS, Fabian J, Friend DR, Kiser PF. 2014. Engineering a255segmented dual-reservoir polyurethane intravaginal ring for simultaneous256prevention of HIV transmission and unwanted pregnancy. PloS one2579:e88509.258259Figure 1. MZC and TFV gels have no apparent effect on tissue architecture260and reduce cell-free and cell-associated SHIV-RT infection of macaque261vaginal explants. A) MZC and 1% TFV gels do not induce histopathological262changes in macaque vaginal tissues using the previously described polarized263 macaque vaginal explant model (9). Results representative of 3 experiments are264shown at 10X magnification. B) Tissue challenge with cell-free SHIV-RT infection265was performed as in Barnable et al. (9). Briefly, explants were immersed in266diluted gels (1:30 or 1:100) (vs. untreated controls) for 18h in the presence of267PHA/IL-2. Then tissues were washed and challenged with SHIV-RT (10 TCID50)26824h or 4d after exposure to the gels. The tissues were washed again 18h after269 virus challenge and cultured for 14 d in the presence of IL-2. SIV p27 levels270measured over the culture period. C) Tissue challenge with cell-associated SHIV-271RT was performed as previously described [9]. Briefly, explants were immersed272in media containing diluted gels (vs. untreated controls) for 18h in the presence273of PHA/IL-2. Then tissues were washed and challenged (18h) with mitomycin-C-274onJauary8,2018bygesthttpaac.asm.rg/Downladedfom treated, SHIV-RT-infected PBMCs (10 infected PBMCs/27 TCID50 per explant;2752-4 replicates) 24h or 4d after exposure to the gels. Then tissues were washed276and cultured for 14d in the presence of IL-2 (vs. 10μM 3TC control) and analyzed277as in B). No released p27 was detected in cultures of control mitomycin-C-treated,278SHIV-RT-infected PBMCs cultured alone (not shown). The analysis was done279using a log-normal generalized linear mixed model with the individual replicate280data. (B, C) Shown are SOFT analyses (Mean ± SEM) of n=5-9 (24h) and n=3-9281(4d) experiments. SHIV-RT p27 concentrations of individual replicate values ≥282lower limit of quantification (LLOQ) were assumed log-normal. Type 3 F tests283 were used to determine the overall effect of treatment. Tukey’s adjusted t tests284 were used for pairwise comparisons of treatments. The analysis was performed285with SAS V9.4, SAS/STAT V13.1 with α=0.05. P values <0.05 (*), <0.01 (**), and286 <0.001 (***) for relevant comparisons are indicated. MZC gel was tested only at2871:100 dilution in the cell-free model due to tissue availability.288289Figure 2. MZC but not 1% TFV gel protects mice from HSV-2 vaginal290challenge. Depo-Provera-treated Balb/C mice were treated intravaginally with29110μl of the indicated formulations 1h before and after HSV-2 challenge with2925x10 pfu (n=20/formulation). The percentages of uninfected animals are shown293for each treatment group. Fisher’s exact test was used for statistical comparison,294 P values <0.05 were taken as statistically significant.295296onJauary8,2018bygesthttpaac.asm.rg/Downladedfom Table 1. Antiviral Activity of MZC and 1% TFV gels against HIV-1. HIV-1EC50 based on gel dilution factor (95% confidence interval)TIMZC1% TFVMZC 1% TFVBaL3x10 (2x10 to 4x10)3.7x10-4 (3x10 to 5x10)>133 >10ADA-M2x10 (1x10 to 3x10)2.2x10 (1x10 to 3x10)>200 >18MN0.4x10 (0.3x10 to 0.6x10)3.2x10 (1x10 to 7x10)>1000 >12MG505.WOM.ENV.CZ 2x10 (1x10 to 4x10)6x10 (3x10 to 1x10)>200 >66NL4-32x10 (8x10 to 7x10)4x10 (2x10 to 7x10)>250 >12592UG0294x10 (3x10 to 5x10)3x10 (1x10 to 1x10)>125 >166.791US0564x10 (2x10 to 7x10)3x10 (1.5x10 to 7x10)>125 >166.792BR0143x10 (1x10 to 7x10)1x10 (5x10 to 3x10)>166.7 >50092HT5931x10 (5x10 to 2x10)1x10 (6x10 to 2x10)>500 >50097ZA0091x10 (9x10 to 2x10)5x10 (2x10 to 1x10)>500 >10097USNG306x10 (5x10 to 9x10)5x10 (3x10 to 7x10)>83.3 >10096USNG316x10 (2x10 to 3x10)4x10 (1x10 to 9x10)>83.3 >125CMU061x10 (1x10 to 2x10)1x10 (7x10 to 4x10)>500 >50092TH0206x10 (4x10 to 1x10)1x10 (7x10 to 3x10)>833.3 >50093TH0514x10 (2x10 to 1x10)3x10 (1x10 to 6x10)>125 >166.735764-28x10 (3x10 to 2x10)4x10 (2x10 to 1x10)>625 >1257295-1 g5x10 (4x10 to 7x10)6x10 (4x10 to 8x10)>100 >83.329129-2 g2x10 (9x10 to 3x10)7x10 (5x10 to 1.1x10)>250 >71.456252-1 g5x10 (2x10 to 1.4x10)~1x10>100 >54755-5 g2x10 (1x10 to 5x10)~3x10>250 >166.71617-1 g1x10 (6x10 to 4x10)2x10 (5x10 to 5x10)>500 >257324-4 g3x10 (2x10 to 5x10)4x10 (1x10 to 1.3x10)>166.7 >1257324-1 g1x10 (7x10 to 3x10)4x10 (2x10 to 4x10)>500 >166.78415-2 g4x10 (1x10 to 1.2x10)5x10 (2x10 to 1.1x10)>125 >1006463-13 g4x10 (2x10 to 1x10)3x10 (1x10 to 7x10)>1250 >166.77136-1 g3x10 (1x10 to 9x10)~1x10>166.7 >50V16770-2 g5x10 (4x10 to 7x10)5x10 (2x10 to 1.3x10)>100 >100V17763-5 g4x10 (1x10 to 1.6x10)2x10 (2x10 to 1.3x10)>125 >250W1023892-2 g2x10 (6x10 to 5x10)1x10 (3x10 to 2x10)>250 >500J18-1(2) g2x10 (1x10 to 7x10)3x10 (1x10 to 8x10)>250 >166.7OL-1/4(II)d4 g>2.5x103x10 (1x10 to 7x10)ND>166.7C18-15d7 g1.5x10 (1.1x10 to 1.9x10)1x10 (4x10 to 2x10)>33.3 >50a The HIV-1MN, HIV-1ADA-M and HIV-1BaL laboratory strains were provided by Dr. J.D. Lifson at the AIDS and Cancer Virus Program, LeidosBiomedical Research, Inc. HIV-1 transmitted/founder virus, clone MG505.WOM.ENV.CZ, was provided by Dr. James Arthos at NIAID, NIH,Bethesda, MD, USA. Additionally a panel of 28 viruses/clones (Kizima et al., 2014) representing different HIV-1 clades and multidrugresistant (MDR) strains were testedb EC50 values in TZM-bl and PBMC were calculated using a dose-response-inhibition analysis on GraphPad Prism v5.0c software(GraphPad Software, San Diego, CA).c Cytotoxicity and antiviral assays were performed in TZM-bl, all other viruses in PBMCsd Published data [4]e Therapeutic indexes (TI=CC50/EC50). CC50 > 5E-02 in PBMCs and > 4E-03 in TZM-blf clade A; g clade B; h clade C; i clade END= Not determinedonJauary8,2018bygesthttpaac.asm.rg/Downladedfom