Oral administration of maraviroc, in infant rhesus macaques, fails to prevent SIVmac oral transmission Egidio Brocca-Cofano, PhD Center for Vaccine Research – University of Pittsburgh Good morning everybody and thank you to organizers to give me this opportunity to present and discuss my data.

Background The World Health Organization (WHO) estimates that approximately 240,000 children became HIV-infected annually. Most of these children live in underdeveloped countries, and the main route of HIV transmission is through breastfeeding (WHO 2014; Global update on the health sector response to HIV, 2014) Overall risk of breastfeeding transmission (BFT) is estimated at 5-20% (WHO 2013; Global update on HIV treatment 2013: Results, Impact and Opportunities) The exact correlates of BFT are unknown (i.e., lack of transmission from mothers with high VLs and/or cases of transmission from mothers with low levels of viral replication) The World Health organization estimates that annually around 2 hundred and forty thousand children became HIV-infect. Most of these children live in underdeveloped counties, and more than half of HIV transmission occurs during breastfeeding The risk of breastfeeding transmission is between 5 and 20 percent. The exact correlates of BFT are unknown (i.e., lack of transmission from mothers with high VLs and/or cases of transmission from mothers with low levels of viral replication)

Based on the following published data: Hypothesis Based on the following published data: PLoS Pathog. 2014 Mar; 10(3): e1003958. So, based on the following published data that show: The Paucity of CD4+ CCR5+ T cells in young natural host represent a factor that protects these animals from breast-feeding transmission of SIV The target cell available at the site of infection are important to reduce the risk of SIV transmission Combined with other data such as Maraviroc was the first US Food and Drug Administration-approved drug from a new class of antiretroviral agents that targets the coreceptor CCR5 This interaction prevent HIV entry in the cell Our hypothesis is that CCR5 blockade with Maraviroc will significantly impact the efficacy of oral SIV transmission to infant rhesus macaques (RMs) Our hypothesis is that CCR5 blockade with maraviroc (MVC) will significantly impact the efficacy of oral SIV transmission to infant rhesus macaques (RMs)

= CCR5 blockade with MVC study design 9 infant rhesus macaques (6 months old) 5 MVC-treated 4 Untreated Daily, 300mg/Kg of MVC, bid, orally At day of Challege 2 Wks 0 hour 4 hours Total of 6 times = Oral administration of MVC Bleeding for CCR5 coreceptor occupancy test Every two weeks SIVmac766XII oral Challenge To test our hypothesis in this study we enclose 9 rhesus macaques, six months old, and we split in 5 animals MVC-treated and in 4 animals untreated as control group. 300 mg/kg of MVR was daily administrated orally by food We started to administer the drug two weeks before of the challenge and at day of the challenge at 0 hour the drug was administrated to treated animals and after 4 hours we collected the blood from all animals for receptor occupancy assay and both groups were orally challenge with 10 thousand tissue culture infection dose of SIVmac766XII. This treatment was repeated every two weeks for a total of six times, until all animals of the control group were infected. Receptor occupancy by analysis of MVC binding to CCR5 (MIP-1β internalization assay) 10,000 TCID50 Oral infection with SIVmac766XII

SIVmac766XII a transmitted founder molecularly tagged infectious molecular clone used for oral challenge   R E G D Q L W K P A C T WT ― B F H I Comparisons of the infectivity of each molecularly tagged SIVmac239 clone with SIVmac239 WT LTR gag pol   env Comparisons of the molecularly tagged SIVmac239 clones with SIVmac239 WT For the challenge we used SIVmac766XII This one is a cocktail of different clones of SIVmac239 including the WT clone plus 9 clones carrying 2 or 3 mutations in the integrase gene These mutations not affect the infectivity of each clones. Also, each clone is present in similar proportion in the SIVmac766XII mixture Proportion of each molecularly tagged variant within SIVmac766XII Modified from Del Prete et al. J Virol. 2014

Maraviroc was effective in blocking CCR5 coreceptor on CD4+ T cells This graph summarizes all six coreceptor occupancy assays at day of the challenge. How we can see from in this graph Maraviroc was efficient to blocks CCR5 coreceptor on CD4 T cells in MVC treated group Oral administration of MVC efficiently blocks CCR5 coreceptor on CD4+ T cells.

SIVmac766XII transmission after oral inoculation in MVC treated infant RMs and untreated controls MVC-Untreated RM32 RM33 RM28 RM31 RM34 RM27 RM29 RM30 RM26 I no infected II III IV V VI      (3/5) 60%  (4/4) 100%  In this table we summarized the exposure to SIVmac766XII to infect infant RMs. After six inoculation all infant RMs in the control group and 60% of MVR-treated became infected with SIVmac766XII without statistical differences. There were no differences in the number of exposures needed to infect infant RMs in the two groups (1, 3, 5 and 6 inoculations for controls versus 2, 3 and 4 inoculation for the RMs receiving MVC).  None of the treated or control infant RMs were infected with more than one viral variant No statistical differences between the experimental groups and no differences in the number of exposures needed to infect infant RMs in the two groups observed None of the treated or control infant RMs were infected with more than one viral variant

Pharmacokinetic analysis of plasma MVC concentration combined with viral load at time of the SIV infection MVC-Treated MVC-Treated infected MVC-Treated uninfected This graph shows the level of MVC in plasma in each animals-treated. The PK profile of MCV was evaluated by measuring the plasma levels of MVC at 4 hours after the oral administration of the drug, which corresponded to the time of viral challenge (when the plasma levels were expected to be maximal). Additional testing of the plasma levels of MVC was performed just before the drug administration (when plasma levels were expected to be minimal) at 2, 3 and 7 days after viral challenge. We observed that the levels of MVC in both evaluation were similar to therapeutic levels observed in HIV-infected patients MVC-treated and the level of the drug were never under the limit of quantification of the assay After maraviroc treatment two infant RMs were resistant to the infection and 3 were infected For this reason we generated two groups: MVC-treated uninfected group (n=2) and MVC-treated infected group (n=3) to better compare the effect of the drug. Levels of MVC were similar to therapeutic levels observed in HIV-infected patients (Brown, K. et al. 2011. The J. Infect. Dis.; Dumond, J. B. et al. 2009 J Acquir Immune Defic Syndr. )

Groups compared and contrasted in this study MVC-Treated infected MVC-Untreated We next compared and contrasted key parameters of SIVmac infection between MVC-treated and control RMs in order to assess whether or not MVC administration has any impact on the natural history of SIV infection in MVC-treated SIV-infected RMs

Viral Load * Peak of acute viremia and development of chronic viremia were similar in MVC-treated infected and untreated infected group All infected animal in MVC-treated and untreated show similar trend of acute and chronic viral load. Peak of acute viremia and development of chronic viremia were similar in MVC-treated infected and untreated infected group However we found that oral administration of MVC delay ramp-up and the peak of the acute viremia Oral administration of MVC delays ramp-up and the peak of the acute viremia

Gating strategy – PBMC - (RM 34, MVC-treated day 0) FSC-A FSC-H SSC-A CD3 CD4 CD8 Ki-67 CD4 CD8 Proliferation CCR5 expression CCR5 CD4 CD8 CD4 CD3 Absolute counts We analyzed by flow cytometry the impact of MRV on main T cell in the blood: In this slide there is showed the gating used in a representative uninfected animal. We selected the CD4 and CD8 T cells from CD3, lymphocytes and singlets And we analyzed: the dynamic of CD4 absolute count, CCR5 expression on CD4 and CD8, And CD4 and CD8 proliferation by expression of Ki-67

Longitudinal analysis of circulating CD4+ T cells in infant RMs Similar levels of CD4+ T cell depletion after oral infection with SIVmac in both MVC treated and control RMs In spite of a moderate decrease of CD4 absolute counts We didn’t observe any statistical difference between MVC-treated infected and Untreated infected

Longitudinal analysis of the expression of CCR5 on CD4+ and CD8+ T cell And we didn’t observe any statistical difference on the expression of CCR5 on CD4 and CD8. CCR5 coreceptor expression on CD4+ and CD8+ T cells is not affected by oral treatment of MVC

Longitudinal analysis of the percentage of proliferation on CD4+ and CD8+ T cell in infant RMs’ blood. The levels of proliferation increase after infection in both groups but no statistical difference was observed No significant difference in the levels of cell proliferation between MVC treated and control RMs

Conclusions MVC was safe and well tolerated in infant RMs. MVC was efficient in blocking CCR5 on CD4+ T cells Blocking CCR5 with MVC did not significantly impact SIV oral transmission These results might be due to the fact that SIVmac is more promiscuous than HIV-1 with regard to coreceptor usage (i.e., being able to use alternative coreceptors, such as BOB/GPR15 (Pohlmann S, et al. J Infect Dis 1999) and Bonzo/STRL33 (Zhang YJ, et al. AIDS Res Hum Retroviruses 2001). However, one should consider that previous studies using SHIV did not report any impact of MVC administration on the efficacy of intra-rectal transmission. CCR5 blockade in humans might be more effective in preventing MTIT alone or in combination with other antiretroviral drugs. MVC administration had no significant impact on the natural history of SIVmac infection in infant RMs. However, the fact that the ramp-up viremia is delayed in the MVC-treated infants, may provide a window of opportunity for other interventions aimed at preventing HIV BFT. In the light of these results: MVC was safe and well tolerated in infant RMs. MVC was efficient in blocking CCR5 on CD4+ T cells Blocking CCR5 with MVC did not significantly impact SIV oral transmission These results might be due to the fact that SIVmac is more promiscuous than HIV-1 with regard to coreceptor usage, because SIV is able to use BOB/GPR15 and Bonzo/STRL33 as alternative coreceptors. However, one should consider that previous studies using SHIV did not report any impact of MVC administration on the efficacy of intra-rectal transmission. CCR5 blockade in humans might be more effective in preventing MTIT alone or in combination with other antiretroviral drugs. MVC administration had no significant impact on the natural history of SIVmac infection in infant RMs. However, the fact that the ramp-up viremia is delayed in the MVC-treated infants, may provide a window of opportunity for other interventions aimed at preventing HIV BFT.

Acknowledgements Center for Vaccine Research, University of Pittsburgh Drs. Apetrei and Pandrea’s Labs Dr. Cuiling Xu Dr. Dongzhu Ma Benjamin Bruno Policicchio Kevin David Raehtz Tammy Dunsmore George Sebastian Haret-Richter Dr. Brandon F. Keele AIDS and Cancer Virus Program Eshelman School of Pharmacy, Chapel Hill, NC Dr. Angela D. Kashuba Mackenzie L. Cottrell I would say thanks to Professors Apetrei and Pandrea and my coworker To Dr. Keele from AIDS and Cancer Virus Program from NCI at Frederick And to Dr. Angela kashuba from Eshelman School of Pharmacy at Chapel Hill.