Translational pharmacology benchmarking of a novel drug candidate for an autoimmune disease
Usman Arshad (1), Silvia Maria Lavezzi (2), Gerben Bouma (1), Tracey Wright (1), Nicola Robertson (1), Martin Kolev (1), Laura Rapley (1), Peter Morley (1), Marina Zvartau-Hind (1), Chao Chen (1)
(1) GlaxoSmithKline (2) Parexel International
Objectives: When assessing the probability of success of drugs with new mechanisms, it is important to understand the pharmacological requirement for clinically meaningful efficacy and associated uncertainties. Translational modelling of drug exposure, target engagement (TE) and pathway pharmacology (PP) were conducted for a drug candidate for treating an autoimmune disease. The drug – a monoclonal antibody – binds to a soluble cytokine to prevent the cytokine’s interaction with its receptors on a certain population of pathogenic T-cells (i.e., TE), leading to reduced expression of an intracellular protein (i.e., PP), and consequently decreasing the count of those specific T-cells. The aim was to evaluate the therapeutic potential of this drug in terms of its Probability of Pharmacological Success (PoPS) [1, 2, 3], in this case defined as “the probability that 50% of the subjects can achieve required protein reduction in the target tissue, while the median systemic exposure of the drug in the population is kept within the safety limit identified in the relevant non-human species”.
Methods: Pharmacokinetics, TE and PP were assessed in the monkey. The TE was described by a target-mediated disposition model; and the PP was analysed with a turnover model. The TE and PP were translated from blood to the relevant tissue with the adjustment of the baseline levels of the cytokine and the protein, respectively, and with the consideration of the drug’s tissue distribution coefficient. Both TE and PP were then extrapolated from monkey to human by allometric scaling of the PK. Computer simulations of the PP in humans, including between-subject variability and translational uncertainties on relevant PK/PD parameters, were conducted for a wide range of doses administered every four weeks (Q4W). The uncertainty in potency was quantified in terms of the blood-tissue distribution coefficient of the compound. The uncertainty range of the required PP was defined as 70 to 95% of the maximal inhibition of the protein, benchmarked by clinical evidence from competitor drugs of relevant mechanisms and comparable indications. The safety limit was derived from systemic exposure at the no-observed-adverse-effect level in the monkey. All animal studies were reviewed and carried out in accordance with regional regulations and the GSK Policy on the Care, Welfare and Treatment of Animals.
Results: Under the most relaxed conditions (low requirement for pharmacology and high drug potency in tissue), high PoPS of >95% was estimated from 30 to 5000 mg Q4W. Under the most stringent conditions (high requirement for pharmacology and low drug potency in tissue), a moderate value of ~75% PoPS was estimated at 5000 mg Q4W. Using prior distributions for a range of pharmacology requirement and drug potency, the overall PoPS achieved a high value of >95% at doses of 2000 - 5000 mg Q4W. Beyond 5000 mg Q4W, a sharp decline in PoPS was evidenced under all conditions and for the overall PoPS, due to serum exposure exceeding the safety limit. These findings demonstrated a high confidence in the drug’s ability to deliver required pharmacology within the desired safety limit.
Conclusions: While translational pharmacology has long been used for clinical dose prediction from non-clinical data, understanding the pharmacology response required for clinical efficacy is a major challenge for predicting a drug candidate’s likelihood of success that informs the investment decisions. This understanding is particularly difficult for drugs with a new mechanism. This work shows the required pharmacology sometimes can be creatively benchmarked using clinical data of a pathway pharmacology endpoint shared with drugs with a similar mechanism and/or a similar indication. It also illustrates the importance of identifying and quantifying the uncertainties in the required pharmacological response and in the PK/PD translation between the site of observation and the site of action, to reliably predict a drug candidate’s therapeutic potential.
References:
[1] Zhou X, Graff O, Chen C. Quantifying the probability of pharmacological success to inform compound progression decisions. PLoS ONE. 2020; 15:10: e0240234.
[2] Lavezzi SM, Chen C, Upson J, Toomey J, French K, Chen L, Lindsay A. Translational modelling and estimation of probability of molecule success on balance of pharmacology benefit and safety risk. PAGE 29 (2021) Abstr 9702 [www.page-meeting.org/?abstract=9702].
[3] Lavezzi SM, Iavarone L, Rycker MD, Zhou X, Yang T, Zhang J, Miles T, Chen C. Evaluation of efficacy predictors and probability of pharmacological success for a novel compound to treat parasitic disease proposed for first-time-into-human. PAGE 29 (2021) Abstr 9597 [www.page-meeting.org/?abstract=9597].