PT-004 - INTEGRATING BIOMARKERS AND PBPK MODELING TO UNCOVER THE MECHANISM OF A MULTI-TRANSPORTER DDI.

A. Riselli¹, X. Liang², R. Liu², Y. Lai², J. Chun¹, K. Giacomini¹; ¹UCSF, San Francisco, CA, USA, ²Gilead Sciences, Foster City, CA, USA.

Background: Eltrombopag is a thrombopoietin receptor agonist that significantly increases rosuvastatin levels, prompting a product label warning against co-administration with substrates of BCRP and OATP1B1. However, whether eltrombopag inhibits both transporters clinically remains uncertain. After confirming this clinical DDI, we leveraged transporter biomarker and PBPK modeling approaches to investigate the roles of intestinal BCRP and hepatic OATP1B1 in the observed interaction.

Methods: A crossover study (NCT04542382) was conducted in 11 healthy volunteers receiving 10 mg rosuvastatin with and without 75 mg eltrombopag. Plasma levels of rosuvastatin, eltrombopag, CP-I, and CP-III were measured using LC-MS. PK data were analyzed using non-compartmental methods, with geometric means (95% CI) and geometric mean ratios (95% CI) reported.

A PBPK model for eltrombopag with mechanistic absorption was developed using Simcyp (v22), incorporating in vitro transporter inhibition data from uptake studies in HEK293 cells and additional parameters from the literature. Simulations were conducted to mirror the clinical study, with 10 trials of 10 healthy volunteers. The OATP1B1 Ki parameter was excluded in simulations isolating BCRP inhibition.

Results: Eltrombopag significantly increased rosuvastatin levels, consistent with inhibition of BCRP, OATP1B1, or both. However, the Cmax ratios for CP-I (0.95, CI: 0.86-1.05) and CP-III (0.83, CI: 0.68-1.02) remained below the 1.25-fold threshold, suggesting that eltrombopag did not inhibit OATP1B1 (Fig 1A).

The PBPK model for eltrombopag accurately predicted clinical data (Fig 1B). DDI simulations closely aligned with clinical findings (Fig 1C), showing that inhibition of BCRP alone (Ki = 0.21 µM) increased rosuvastatin AUC by 1.42-fold, with only a modest additional increase (12%) when OATP1B1 inhibition (Ki = 0.37 µM) was included (Fig 1C).

Conclusion: This study confirmed the DDI between eltrombopag and rosuvastatin, quantified plasma CP-I and CP-III levels, and developed a validated PBPK model to elucidate the mechanisms responsible for the interaction. Together, the findings strongly suggest that BCRP inhibition by eltrombopag is the primary driver of the DDI, whereas clinical inhibition of OATP1B1 is unlikely.