Rasmus Jansson-Löfmark (1), Markus Fridén (2), Owen Jones (3)
(1) Drug Metabolism and Pharmacokinetics; Cardiovascular, Renal and Metabolism, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden, (2) Drug Metabolism and Pharmacokinetics; Respiratory, Inflammation and Autoimmunity, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden, (3) Drug Metabolism and Pharmacokinetics; Oncology, IMED Biotech Unit, AstraZeneca, Cambridge, UK
Introduction/Objectives: Pre-clinical data and predictive translational modelling are extensively used in support of candidate drugs taken into the clinic. This includes the prediction of human dose commonly anchored upon an understanding of the level of target modulation required to drive efficacy in an animal model or using clinical competitor data. A fundamental part in the human dose prediction is to have a solid prediction of human exposure-response for target engagement. Given the big leap in extrapolation from non-clinical data to clinic, this step in translation can be associated with uncertainties.
There are publications focusing on best practice [1-4], efficacy and pharmacokinetic pharmacodynamic translation of individual compounds [5,6]. But based on our knowledge there are very few publications reporting an analysis that evaluates multiple compounds, targets and therapy areas. Despite the implementation in pharma of model based solutions to predict human dose and exposure-response for target engagement, it is perhaps surprising that there are relatively few publications reporting an analysis that evaluates the pre-clinical to clinical translation at a cross therapy level. This work presents such an analysis looking across the AstraZeneca portfolio and examines how pre-clinical data / models translates to clinical pharmacodynamic/target engagement response.
Methods: A retrospective analysis of internal data across oncology, respiratory and cardiovascular renal or metabolism therapy areas was done. Inclusion of a compound for evaluation were driven by (1) the availability of sufficient clinical data to enable a derivation of exposure-response relationship, (2) the same biomarker measured both in the animal model used and the clinic. The compiled dataset composed of 19 small molecule drugs with various drug targets, mode of action, receptor types and biomarkers. The majority of pre-clinical pharmacokinetic-pharmacodynamic datasets were either analyzed by direct exposure response models or by indirect turnover models. For the pre-clinical dataset, a naïve data pool data analysis was deemed sufficient in most cases to retrieve exposure-response relationships. For the translation, interspecies differences in plasma protein binding, affinity for the target receptor and time delays were also considered.
Results: For those projects that had clinical data available to enable back translation of a target engagement biomarker, 83% of the compounds showed an exposure response relationship that translated within 2-fold. The 50th (5th, 95th ) percentile of the ratio between predicted exposure response to measured was 1.1 (0.55; 3.1). For those compounds that deviated, there was a trend of a binary deviation suggesting that biological or experimental assumptions were flawed.
Conclusions:This exercise demonstrates that applying translational pharmacokinetic-pharmacodynamic modelling can predict human exposure-response in the majority of cases. It also provides the modeller with some quantitative understanding of the confidence to predict human exposure-response for taget engagement. Additionally, it suggests that miss-predictions in translations could be due to flawed biological or experimental assumptions rather than technical miss specifications in the PKPD modelling process. Attrition in the clinic continues to be dominated by lack of efficacy, and therefore, much work is still required to improve the translation and prediction of human drug efficacy, linking target modulation to effects on disease biology.
References:
[1]. Visser, Sandra A G et al. “Model-based drug discovery: Implementation and impact.” Drug Discovery Today 2013: 764-775.
[2]. Marshall, Sf et al. “Good Practices in Model-Informed Drug Discovery and Development: Practice, Application, and Documentation.” CPT: Pharmacometrics & Systems Pharmacology 5.3 (2016): 93-122.
[3]. Schuck, E et al. “Preclinical pharmacokinetic/pharmacodynamic modeling and simulation in the pharmaceutical industry: an IQ consortium survey examining the current landscape.” The AAPS Journal 17.2 (2015): 462-73.
[4]. Tuntland, T et al. “Implementation of pharmacokinetic and pharmacodynamic strategies in early research phases of drug discovery and development at Novartis institute of biomedical research.” Frontiers in Pharmacology 2014; 5:174
[5]. Wong, Harvey et al. “Learning and confirming with preclinical studies: Modeling and simulation in the discovery of GDC-0917, an inhibitor of apoptosis proteins antagonist.” Drug Metabolism and Disposition 41.12 (2013): 2104-2113
[6]. Maurer et al “Pharmacodynamic Model of Sodium–Glucose Transporter 2 (SGLT2) Inhibition: Implications for Quantitative Translational Pharmacology” AAPS J. 2011 Dec; 13(4): 576–584.
Reference: PAGE 28 (2019) Abstr 9036 [www.page-meeting.org/?abstract=9036]
Poster: Drug/Disease Modelling - Other Topics