Apolline Lefèvre (1), Zinnia P. Parra-Guillen (1), Iñaki F. Trocóniz (1,3,4), Christophe Boetsch (2), Nicolas Frances (2)
(1) Pharmacometrics & Systems Pharmacology,Department of Pharmaceutical Science. School of Pharmacy and Nutrition,University of Navarra, Pamplona, Spain; (2) Roche Pharma Research and Early Development pRED, Pharmaceutical Sciences PS, Roche Innovation Center Basel, Switzerland; (3) IdiSNA, Navarra Institute for Health Research, Spain; (4) Institute of Data Science and Artificial Intelligence, DATAI, University of Navarra, Pamplona, Spain
Objectives:
T-cell engagers (TCE) are anti-cancer therapeutic agents that reactivate the patient’s immune system. They are promising therapies with remaining challenges associated with their clinical safety profiles. In particular, Cytokine Release Syndrome (CRS) is one of the most common adverse events following TCE dosing and is expected to be associated with the levels of cytokine released [1]. It was recently observed in vitro that when two subsequent TCE stimulation were performed seven days apart, low amounts of cytokine were released at the second stimulation while tumor cell killing was maintained [2]. Those findings support the use of new dosing strategies like step-up dosing to minimize the risk of CRS in the clinical setting. A previously developed PKPD model [3] has been updated to include tumor cell dynamics and to uncouple tumor cell killing from cytokine release after two consecutive TCE administrations. This work has the objective of opening new doors towards the use of PKPD models for the selection of the best dosing regimen to maximize efficacy while minimizing the occurrence of CRS.
Methods:
The previously published PKPD model has been expanded to describe tumor cell killing following TCE administration. The dynamics of the different entities (T-cells, tumor cells and TCB) were modeled using ordinary differential equations (ODE) describing the mass balance between the synthesis, transition and degradation or apoptosis processes. Tumor growth was modeled as an exponential growth and killing was dampened using transit compartments as in [4] Parameter values characterizing the above mentioned processes for the mechanistic PKPD model were taken from literature or fine tuned to describe the available data of in vivo IL6 following two TCE doses [2]. A parameter scan analysis was performed to evaluate the effects of varying each parameter individually (+/- 10%) on maximum cytokine release at second TCE administration and identify most influential processes. All simulations and analysis were performed on Rstudio V.4.1.3 using publicly available libraries.
Results:
In the presented model, different T-cells states were modeled: upon TCE exposure, naive T-cells transit to activated T-cell state (via a 5-transit compartment), which then transit to a desensitized T-cell state for a longer period of time. The latter state corresponds to a T-cell phenotype that allows them to exert their cytotoxic effects while not releasing cytokines. Thus, only the activated T-cells can release cytokines while both activated and desensitized T-cells can eradicate tumor cells. Upon first TCE stimulation, the pool of baseline T-cells gets depleted as the cells transition to the activated then desensitized states. In the case of a second stimulation, close to the first one, no naive T-cells are available for activation and cytokine release. On the other hand, desensitized T-cells are still available for killing, allowing for cytotoxic effects while no cytokines are released.
Furthermore, the parameter scan analysis highlighted the importance of the rate constants for baseline naive T-cells (Kin,naive and Kapo,naive), as well as the parameters involved in the cytokine release function versus plasma concentration (alpha, EmaxCyt and EC50Cyt) with respect to the quantities of cytokine released.
Conclusions:
The updated model was successfully able to describe the cytokine release following subsequent TCE administrations at different dosing intervals. In addition, the model was also able to describe the uncoupling of tumor cell killing and cytokine release. In contrast to the previously published model [5] – [6], the loss of cytokine release was not modeled as a consequence of loss of tumor cells, but rather due to a shift in the T-cell phenotypes. Nevertheless, this has been demonstrated in vitro only and should be confirmed in vivo. The model thus provides new hypotheses on the mechanisms behind cytokine release and tumor cell killing that could be tested in a pre-clinical setting. This would allow to confirm or update the model, to be then used to inform dosing regimen selection in the clinical setting.
References:
[1] Saber et al., An FDA oncology analysis of CD3 bispecific constructs and first-in-human dose selection, Regulatory Toxicology and Pharmacology (2017)
[2] Li et al., CD3 bispecific antibody–induced cytokine release is dispensable for cytotoxic T cell activity, Sci. Transl. Med. 11(2019)
[3] Lefevre et al., A mechanistic PK/PD model to predict cytokine release and tumor cell killing associated to T-cell bispecific therapies, Population Approach Group Europe (2023)
[4] Simeoni M et al., Predictive pharmacokinetic-pharmacodynamic modeling of tumor growth kinetics in xenograft models after administration of anticancer agents. Cancer Res. (2004)
[5] Chen et al., A Modeling Framework to Characterize Cytokine Release upon T-Cell-Engaging Bispecific Antibody Treatment: Methodology and Opportunities. Clin Transl Sci. (2019)
[6] Jiang et al., Development of a minimal physiologically-based pharmacokinetic/pharmacodynamic model to characterize target cell depletion and cytokine release for T cell-redirecting bispecific agents in humans. Eur J Pharm Sci. (2020)
Reference: PAGE 32 (2024) Abstr 11259 [www.page-meeting.org/?abstract=11259]
Poster: Drug/Disease Modelling - Oncology