François Riglet (1), Christelle Rodrigues (1), Myra Skerlep (1), Mathilde Romagnoli (1), Malek Okour (1), Agnès Hemon (1), Alberto Broniscer (2), Kristin Horton (2), Adrien Tessier (1)
(1) Translational Medicine, Servier, Gif sur Yvette, France (2) Therapeutic Area Oncology, Servier Pharmaceuticals, Boston, USA
Objectives:
S65487 is an intravenous potent and selective inhibitor of B-cell leukemia/lymphoma-2 (BCL-2), member of the BCL-2 family proteins which regulate apoptosis. The anti-apoptotic activity of BCL-2 is due to sequestration of pro-apoptotic BH3-only proteins, including BIM[1]. BCL2/BIM complex is disrupted by competitive binding with S65487, which triggers apoptosis through elevation in the levels of free BIM. Thereby, BCL2/BIM complex disruption is a reliable target engagement (TE) biomarker to evaluate drug activity, directly related to S65487 mechanism of action[2]. Based on Phase 1 BCL-2/BIM clinical data, a PK/PD model was built to describe TE dynamics in S65487-treated patients.
Thereby, the objective of this analysis consisted of identifying the TE saturating dose, or Optimal Biological Dose (OBD), using a PK/PD modeling approach, fitting biomarker data of these patients, to support selection of the Recommended Phase II Dose (RP2D).
Methods:
S65487 is undergoing evaluation in two Phase I studies: as a single agent in patients with relapsed or refractory (R/R) acute myeloid leukemia (AML), non-Hodgkin lymphoma, multiple myeloma or chronic lymphocytic leukemia [NCT03755154] and in combination with azacitidine (Aza) in newly diagnosed unfit patients with AML [NCT04742101].
For the monotherapy study, a weekly and an alternative schedule (with three administrations on the first week and then weekly administrations for the remaining cycle) were evaluated, with administered doses from 25 to 1200 mg. For the combination study, a weekly (schedule 1) or more intensified regimen (schedules 2 and 3) of S65487 with four and seven administrations concurrently with Aza were tested, with administered doses from 50 to 533 mg.
BCL-2/BIM complex disruption was measured in patients treated with S65487 in monotherapy or in combination with Aza. A meso scale discovery (MSD)-based ELISA assay was developed to follow the disruption of the complex in patients’ peripheral-blood mononuclear cells (PBMCs) collected before and after the S65487 infusion. Levels of BCL2-BIM complex were normalized to the protein concentration measured in the same PBMC sample and expressed as % disruption from pre-dose. A simultaneous PK/PD modeling approach was performed to describe complex disruption after treatment with S65487. Subsequently, simulations were performed from this model to identify OBD, assessing magnitude and duration of effect. Based on various preclinical studies, a threshold of 70% disruption from baseline, lasting for a minimum of 8 hours, was defined as the target to achieve in patients.
Results:
PK data were evaluable for 87 patients (mono: 55, combo: 32). BCL-2/BIM complex disruption was measured in 73 patients (mono: 47, combo: 26) treated with S65487. The complex was found disrupted to some extent in nearly all patients treated with 75 mg to 1200 mg of S65487 in monotherapy, with a significant relationship between the magnitude of BCL-2/BIM complex disruption and the plasma exposure of S65487. The duration of the complex disruption was consistent with the short half-life of S65487 and found more sustained with increasing doses of S65487. Drug kinetics was described using a 3-compartment model with linear clearance. The best way to describe TE was a turnover inhibition model, highlighting S65487 mechanism of action preventing BCL-2/BIM complex formation. A study effect on IC50 was estimated with a four-fold lower IC50 in combination compared to monotherapy study (48.2 vs 178.8 ng.mL-1). With the simulation results, both in terms of magnitude and duration, a plateau effect was identified at 533 mg in schedule 2 and at 300 mg in schedule 3, with no significant benefit to increase the dose in combination with Aza.
Conclusions:
A turnover inhibition PK/PD model was able to properly describe TE data after S65487 administration. According to simulation results, the target threshold of 70% of disruption could be achieved at doses used clinically, with a plateau reached at 533 and 300 mg at schedule 2 and 3, respectively, as part of Aza combo study. However, due to the lack of documented Aza direct effect on BCL-2/BIM complex disruption, specific effects in the combination study may be attributed to the different patient population between these studies. Nevertheless, a PK/PD approach of TE data allowed to identify the OBD in S65487-Aza combo study.
References:
[1] Parrondo RD, Paulus A, Ailawadhi S. Updates in the Use of BCL-2-Family Small Molecule Inhibitors for the Treatment of Relapsed/Refractory Multiple Myeloma. Cancers (Basel). 2022 Jul 8;14(14):3330. doi: 10.3390/cancers14143330. PMID: 35884390; PMCID: PMC9317574.
[2] Garciaz S, Miller T, Collette Y, Vey N. Targeting regulated cell death pathways in acute myeloid leukemia. Cancer Drug Resist. 2023 Mar 15;6(1):151-168. doi: 10.20517/cdr.2022.108. PMID: 37065864; PMCID: PMC10099605.
Reference: PAGE 32 (2024) Abstr 11163 [www.page-meeting.org/?abstract=11163]
Poster: Drug/Disease Modelling - Oncology