CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Conclusion: Our novel method indicates significantly higher penetration of efavirenz in the female genital tract than previously reported. This provided data for successful qualification of a PBPK model of efavirenz in pregnant women genitalia. 461 PK/PD STUDY OF RALTEGRAVIR ALONE OR COMBINED WITH LAMIVUDINE AS PrEP: AN RCT Julie Fox 1 , Carolina Herrera 2 , Laura Else 3 , Julianne Lwanga 4 , Ming Lee 4 , Alieu Amara 3 , Laura Dickinson 3 , Marta Boffito 5 , Robin J. Shattock 2 , Saye Khoo 3 1 King's College London, London, United Kingdom, 2 Imperial College London, London, United Kingdom, 3 University of Liverpool, Liverpool, United Kingdom, 4 Guy's Hospital, London, United Kingdom, 5 Chelsea andWestminster NHS Foundation Trust, London, United Kingdom Background: Neither the added value of lamivudine or the effect of raltegravir as PrEP are well known. We evaluated raltegravir +/- lamivudine (RGV/3TC) pharmacokinetics and pharmacodynamics (PK/PD) using a tissue explant model. Methods: Open label trial of 36 HIV- females and males (1:1) randomised to 7d raltegravir 400mg bd followed by 7d raltegravir 400mg/lamivudine 150mg bd (after washout), in 6 sampling blocks to capture different times post-dose. Blood, saliva, rectal fluid (RF)/tissue (RT), vaginal fluid (VF)/tissue (VT) sampled at baseline, on PrEP (day2, 4 or 6) and off PrEP (day8, 10 or 12) for PK (RGV, 3TC, 3TC-triphosphate) and antiviral activity (ex vivo challenge R5-tropic HIV-1BaL virus ;p24 levels at 15d). Protection was defined as >50% reduction in p24 compared to baseline. Results: RGV and 3TC were detectable in all tissue samples at day 2 PrEP. On day 6, GM RGV levels were 247.9 ng/g in VT and 589.2 ng/g in RT; GM tissue-to- plasma accumulation ratios 0.75 (VT) and 2.6 (RT). After PrEP cessation, 50/7% of VT and 86/58% of RT samples remained above RGV IC 95 (15 ng/mL) day 10. Extensive 3TC VT (1397 ng/g) and RT (2662 ng/g) accumulation: GM Tissue- to-plasma accumulation ratios were 7.3 (VT) and 17.1 (RT) day 6. Off PrEP, 3TC persisted in VT (102 ng/g) and RT (275 ng/g) until day12. G. Plasma explained a greater variability in VT level (r2 >0.759; P<0.001) compared with VF. Whereas RF explained more of the variability for 3TC RT levels (r2 =0.591; p<0.001), than plasma. Raltegravir provided maximum ex vivo protection at day 2-8 (83% of rectal; 100% of vaginal samples) Raltegravir/Lamivudine provided 100% protection in rectal tissue from day 2-10, and in vaginal tissue from day 8-12. Conclusion: Following discontinuation, high concentrations of RGV remained in RT (but rapid decline in plasma and VT concentrations) with persistent inhibitory activity in RT up to 4 days later. Addition of lamivudine increased inhibitory activity in RT and VT, with similar persistent inhibition associated with high 3TC RT concentrations 4 days after discontinuation. 462 MODELING-SUPPORTED ISLATRAVIR DOSE SELECTION FOR PHASE III Deanne J. Rudd 1 , Youfang Cao 1 , Pavan Vaddady 1 , Jay Grobler 1 , Ernest Asante- Appiah 1 , Tracy Diamond 1 , Stephanie O. Klopfer 2 , Anjana Grandhi 3 , Peter Sklar 2 , Carey Hwang 2 , Ryan Vargo 1 1 Merck & Co, Inc, West Point, PA, USA, 2 Merck & Co, Inc, Upper Gwynedd, PA, USA, 3 Merck Research Laboratories, Rahway, NJ, USA Background: Islatravir (ISL) is the first nucleoside reverse transcriptase translocation inhibitor (NRTTI) in development for the treatment and prevention of HIV-1 infection. Single doses of ISL as low as 0.5 mg showed robust efficacy in a proof-of-concept (POC) clinical trial and established an IQ (ratio of drug exposure to potency) of 5 for ISL for wild-type HIV-1. In a Ph2 clinical trial (NCT03272347), participants who initiated ISL+doravirine (DOR) in combination with 3TC and switched to ISL+DOR no earlier than Wk 24 had high efficacy at Wk 48 as measured by HIV-1 RNA <50 c/mL. Data through Wk 48 showed that exposure-response was flat, indicating an achievement of maximal efficacy at the ISL doses examined (0.25, 0.75, and 2.25 mg). Modeling and simulation, along with in vitro potency data, were used to select the dose for further clinical development of ISL most appropriate for HIV-1 treatment-naïve, virologically suppressed, and highly-treatment experienced (HTE) populations. Methods: A population pharmacokinetic model for ISL and its active moiety, ISL-triphosphate (ISL-TP), has been developed based on Ph1 and Ph2 data in healthy participants and people living with HIV-1 (PLWH) and used to examine the Ph2 exposure-response relationship. The population pharmacokinetic model was also used to predict the percentage of participants expected to have ISL-TP concentrations sufficient to have antiviral activity against common NRTI- resistant viruses (e.g., M184V, etc.).

dysbiosis on cervicovaginal fluid (CVF) TFV concentrations in US women taking oral TDF/FTC for PrEP. Methods: Southern Californian women on oral TDF/FTC in the Adherence Enhancement Guided by Individualized Texting and Drug Levels (AEGiS) study had CVF collected at week 24 to evaluate (i) sexually transmitted bacterial (gonorrhea, chlamydia, gardnerella and trichomonas), viral (HPV, CMV and HSV-1/2) and fungal (Candida) infections, (ii) microbiome composition by 16S sequencing (V3-V4 region) and (iii) cytokine profiles by ELISA (IL-8, MIP-1a, MIP-1b, and IP-10). Microbiome Community State Types (CSTs) were assigned based on described characteristics (i.e. CST I, II, III, V dominated by lactobacilli and CST IV containing mostly non-lactobacilli species). TFV in CVF and tenofovir diphosphate (TFV-DP) in dried blood spots (DBS) were also measured at week 24. CVF TFV of 100-1000ng/mL benchmarked typical genital concentrations; DBS TVF-DP ≥700fmol/punch suggested long-term adherence. Univariate statistical analysis was used to determine factors associated with low and high CVF TFV. Results: 34 women had CVF specimens collected at week 24. Median age was 43 (IQR 35-47) years. 15% (5/34) had discordant tenofovir concentrations (i.e. DBS TFV-DP≥700 and CVF TFV<100). No inflammatory process was associated with lower CVF TFV concentrations or tenofovir discordance. Notably, among the 26 participants assessed for vaginitis (Candida, Gardnerella or Trichomonas), women with possible vaginitis (n=13) were more likely to have high (>1000 ng/ mL) CVF TFV concentrations compared to those without vaginitis (77% versus 31%, p<0.05). No difference was seen in CVF TFV concentrations by vaginal microbiome type. 3 of the 5 women with discordance had non-lactobacillus dominant microbiomes; however, they were dominated by non-vaginitis organisms (2 E. cloacae , 1 E. coli ). Conclusion: Presence of genital viruses, cytokines or Gardnerella spp. were not associated with low CVF TFV levels in women taking oral PrEP. Women with vaginitis may actually have higher CVF TFV levels, perhaps due to inflammatory processes augmenting blood flow and TFV penetration into the vaginal compartment. 460 ARV PENETRATION INTO FEMALE GENITAL TRACT DURING PREGNANCY: EFAVIRENZ AS A CASE STUDY Oluwasegun I. Eniayewu 1 , Shakir A. Atoyebi 1 , Jacinta Nwogu 2 , Alieu Amara 3 , Oluseye Bolaji 1 , Damien Anweh 4 , Ebunoluwa Adejuyigbe 1 , Marco Siccardi 3 , Saye Khoo 3 , Andrew Owen 3 , Adeniyi Olagunju 1 1 Obafemi Awolowo University, Ile-Ife, Nigeria, 2 University of Ibadan, Ibadan, Nigeria, 3 University of Liverpool, Liverpool, UK, 4 Federal Medical Centre, Makurdi, Nigeria Background: Proper characterisation of antiretroviral pharmacokinetics in the female genital tract is crucial in developing effective pre-exposure prophylaxis and prevention of intrapartummother-to-child transmission interventions. Progress in this area has been limited by sampling constraints. Methods: A novel assay to quantify efavirenz in cervicovaginal fluid from flocked swabs using LC-MS/MS was developed and validated as per FDA guidance. Efavirenz was quantified from cervicovaginal swabs (CVS) collected from HIV infected pregnant women enrolled in the VADICT study (NCT03284645) receiving 600 mg daily. To further characterise efavirenz penetration into the female genital tract, we extended a previously described pregnancy PBPK model constructed and implemented in SimBiology® (MATLAB® version 2018b) to include a multi-compartmental cervicovaginal unit (vagina fluid, epithelium, stroma blood and tissues). Variables representing drug and system characteristics were obtained from the literature for model parametrization. Efavirenz movement within the cervicovaginal compartments was by passive diffusion. The model was qualified by comparing predictions with data from the VADICT study. Results: Mean CVS efavirenz concentration with the newmethod in this cohort (n = 39, mean gestational age 33.8 weeks) at 14.8 h post-dose was 1.237 µg/mL (95% CI: 0.138, 6.397), giving CVS:plasma concentration ratio of 0.64, more than previous reports. This was adequately predicted by the model, predicted CVF concentration being 1.190 µg/mL (0.542, 2.430) in a virtual cohort (n = 100, 29.5 weeks’ gestation). Trough (C trough ), maximum (C max ) efavirenz concentration and area under the concentration-time curve (AUC0-24h) were 0.62 µg/mL (0.29- 1.33), 1.67 µg/mL (1.05-2.67), and 28.4 (µg.hr/mL) (16.89-41.41) respectively. The corresponding parameters in vaginal epithelium, stroma blood and tissue were 7-22% higher. Importantly, both observed and predicted efavirenz C trough were above reported protein-binding adjusted IC 95 of 126 ng/mL for wild-type HIV-1 in all patients.

Poster Abstracts

CROI 2020 162