(T-054) Retrospective analysis of systemic and dermal PK of several anti-inflammatory topicals using SimCYP MechDermA model explains the observed clinical efficacy of successful and failed drugs

Shibin Mathew – PDM – Pfizer Inc.; Jaymin Shah – Pharmaceutical Sciences – Pfizer Inc.

Objectives: Predicting dermal and systemic PK of topical drugs has been challenging due to the difficulty in accurately capturing the processes of drug permeation within different skin layers. Therefore, most preclinical drug development workflows rely on simple mass balance equations to predict average systemic exposure to ensure safe margins and do efficacy testing in the clinic. Skin PK is also ignored due to lack of proper methods to capture time course of skin PK causing a lack of Pillar 1 (exposure at site of action). In this work, we provide a strategy to predict dermal and systemic PK for topical drugs utilizing the latest developments in the field, namely the in silico SimCYP PBPK model (MechDermA) [1] to predict systemic and dermal PK and the dermal open flow microperfusion (dOFM) experimental technique [2] to validate the model. We performed a retrospective analysis of successful and failed clinical candidates from our portfolio and the new workflow is capable of making sense of the efficacy results using in silico PK and pharmacology data alone.

Methods: We started with a list of drugs with known dermal and systemic PK measurements, previously published by our experimental group [2]. This dataset includes drugs like brepocitinib, tofacitinib, diclofenac with known clinical efficacy as well as discovery compounds that failed clinically and were discontinued. MechDermA model was parametrized for each compound, with diffusivity, partitioning and binding parameters determined by QSAR methods [1]. The MechDermA systemic predictions were compared to clinical values. Since dOFM studies were done in pig, we converted the MechDermA dermal PK to pig PK by reparametrizing the MechDermA parameters to pig skin physiology using a published database [3]. In addition to this data, each compound had extensive preclinical flux data (IVPT) generated in ex vivo human skin.

Results: Firstly, simulation of human systemic PK using default QSAR models in MechDermA significantly overpredicted (100-fold) the PK from clinical observations. In addition, the predicted dermal PK was about 100-fold higher than observed from dOFM. This certainly did not explain the efficacy of failed drugs. Alternately, we ran the MechDermA model in IVPT mode, and reestimated the most sensitive parameters using the steady state flux and lag time method instead of the QSAR methods [4]. Evidently, the new workflow gave better predictions of systemic and dermal PK of these compounds (within 2 to 5-fold). Importantly, the dermal PK profile of brepocitinib, tofacitinib and diclofenac were sufficiently higher than their potency values while those of the failed compounds were significantly lower than the potency values, validating our method. Therefore, calculated target modulation (Pillar 3) from MechDermA Pillar 1 aligned with efficacy data.

Conclusions: This retrospective analysis proves the utility of MechDermA model, with skin IVPT data and potency information to predict clinical efficacy. This provides a simple workflow for dermal drug development especially for predicting target modulation at the site of action.

Citations: [1] Patel N, Clarke JF, Salem F, et al. Multi-phase multi-layer mechanistic dermal absorption (MPML MechDermA) model to predict local and systemic exposure of drug products applied on skin. CPT Pharmacometrics Syst Pharmacol. 2022;11(8):1060-1084. doi:10.1002/psp4.12814
[2] Bodenlenz M, Yeoh T, Berstein G, et al. Comparative Study of Dermal Pharmacokinetics Between Topical Drugs Using Open Flow Microperfusion in a Pig Model Pharm Res. 2024;41(2):223-234. doi:10.1007/s11095-023-03645-3
[3] Krumpholz L, Clarke JF, Polak S, Wiśniowska B. An open-access data set of pig skin anatomy and physiology for modelling purposes. Database (Oxford). 2022;2022:baac091. doi:10.1093/database/baac091
[4] Shah et al., Analysis of percutaneous permeation data: II. Evaluation of the lag time method, International Journal of Pharmaceutics, Volume 109, Issue 3, 1994, Pages 283-290, https://doi.org/10.1016/0378-5173(94)90390-5.

Keywords: topical drug discovery, Skin PBPK, SimCYP MechDermA, topical efficacy