Ivy Song (1), Kimberly Adkison (2), Mark Lovern (3), Joannellyn Chiu (3), Jenny Huang (4), Cindy Vavro (1), Mounir Ait-Khaled (5), Brian Wynne (6), Sherene Min (1)
(1) GlaxoSmithKline, RTP, NC, US; (2) PAREXEL, RTP, NC, US; (3) Quantitative Solutions, Raleigh, NC, US; (4) GlaxoSmithKline, Mississauga, Canada; (5) GlaxoSmithKline, Stockley Park, UK;(6) GlaxoSmithKline, Collegeville, PA, US
Objectives:To develop a population PKPD model describing HIV-1 virologic response during the functional monotherapy phase in Phase 2b/3 studies of DTG in combination with background therapy in integrase inhibitor resistant (INIr) patients and apply the model to determine the viability of various DTG dosing regimens in treating HIV infection in INIr patients.
Methods: Short-term HIV-1 RNA(viral load) data from four Phase 2a/3 studies were described using a circular/proliferative(viral dynamic) system similar to that described by Jacqmin et al [1]. The model described the dynamics of healthy (target), actively infected, and latently infected CD4+ cell populations. Complementary models were developed in which minimum(Cmin) and average(Cavg) plasma concentration respectively served as the measure of drug exposure. All covariates were tested on the IC50 parameter using a two-step (forward inclusion/backward elimination) procedure. Covariates evaluated included measures of baseline INI resistance, previous experience with elvitegravir or raltegravir, and viral susceptibility to background therapy. The impact of viral load observations below the quantitation limit(BLQ) was assessed through implementation of the M3 method [2]. The final model was evaluated using visual predictive checks. Simulations were performed to predict the response rates for various scenarios including 100mg BID vs the approved dose 50mg BID, and dosing with food (increasing DTG exposure), metal cation-containing dietary supplements (reducing DTG exposure), or enzyme-inducing drugs.
Results: A total of 751 viral load observations from 266 HIV-1 patients (247 INIr patients and 19 naïve to integrase inhibitors [INIn]), were included in the model building. The available data were well-described by the final population PKPD models, and the fits of Cmin and Cavg models were virtually identical. Of the covariates tested, only baseline INI resistance and Study Population (INIn vs INIr) were found to be significant. Simulations predicted that increasing the twice-daily DTG dose from 50mg to 100mg would yield small additional declines in Day 8 change from baseline in log10 viral load: -0.07 log, -0.05 log, -0.11 log, and -0.12 log for the overall population and the subpopulations with no Q148 mutation, with Q148+1 mutations, and with Q148+³2 mutations, respectively. Dosing with food, metal cation-containing supplements, or enzyme inducers marginally affected Day 8 response..
Conclusions: This analysis supports the current dosing recommendation of DTG 50mg BID in INIr patients.
References:
[1] Jacqmin P, McFayden L, Wade J. Basic PK/PD principles of drug effects in circular/proliferative systems for disease modeling. J Pharmacokinet Pharmacodyn (2010) 37: 157-177.
[2] Bergstrand M, Karlsson MO. Handling data below the limit of quantification in mixed effect models. J AAPS (2009) 11: 371-380.
Reference: PAGE 24 (2015) Abstr 3540 [www.page-meeting.org/?abstract=3540]
Poster: Drug/Disease modeling - Infection