Serrano-Alcaide, Alejandro (1); Zalba, Sara (1); Sancho Araiz, Aymara (1); Casares, Noelia (2); Lasarte, Juan José (2); Garrido, MarÃa J (1); Trocóniz, Iñaki F (1)
(1) Pharmacometrics & Systems Pharmacology Group, Department of Pharmaceutical Technology and Chemistry, School of Pharmacy and Nutrition, University of Navarra, Pamplona, Spain. (2) Immunology & Immunotherapy Research Program, Applied Medical Research Centre, University of Navarra, Pamplona, Spain
Introduction: The majority of the tumor growth inhibition models described in
literature evaluate the interplay between three major components: the growth of tumor
cells in the absence of treatment; the activation of immune cells, mainly effector CD8
cells, to recognize and kill the tumor; and lastly the development of resistance
mechanisms to block this immunological response (1). Regulatory T cells (Treg) play
a crucial role in immune homeostasis, infiltrating tumor tissue and blocking CD8
activity, which correlates with a poor prognosis in multiple cancer types.
However, the evaluation of the Treg activation state and its influence on the antitumor
response is scarce. In this study, we monitored several Treg surface biomarkers
associated with its suppressive function and evaluated the antitumor response after the
administration of a linear peptide P60, capable of inhibiting Treg activity without
compromising cell homeostasis (2). To improve the PK characteristics of P60 and
increase its intracellular availability, the peptide has been encapsulated into lipid
nanoparticles (liposomes) selectively targeted against CD25, a surface receptor
constitutively overexpressed at the Treg cell surface.
Objectives: Description of the Treg activation state due to tumor growth and its
modulation and antitumor response using different immune therapies. Monitoring of
treatment kinetics and tissue distribution to evolve the model in a physiologically
based manner.
Methods: This study was performed based on a structural tumor growth model
previously developed (3). Our experimental data compile 80 mice s.c. inoculated with
MC38 murine cancer cells. These mice were divided into seven groups: control
(n=16); free P60 peptide (n=16); non-targeted P60 liposomes (n=8); CD25-targeted
empty liposomes (n=8); CD25-targeted P60 liposomes (IL) (n=16); anti-PD1 (n=8);
and combination of IL with anti-PD1 (n=8). For the Treg activation process, control
mice (n=12) were sacrificed at different time points, and tumor samples were taken to
monitor the expression of activation biomarkers (PD1, ICOS, LAG3, and CD103).
Lastly, for treatment tissue distribution, a total of 66 mice were treated with a single
dose of either free peptide labelled with carboxyfluorescein (CF-P60; n=22); nontargeted P60 liposomes labelled with DiR lipid dye (L-DiR; n=22); and labelled IL
liposomes (IL-DiR; n=22).
Data were analysed using the nonlinear mixed effect modelling software program
Monolix version 2023R1 (https://lixoft.com/). Parameters were estimated by computing
the maximum likelihood estimator using the stochastic approximation expectation
maximization algorithm (SAEM), assuming a normal distribution and a constant error
model for residual variability. Observations of tumor growth were combined with overall
survival data (OS) to adjust the individual predictions. Visual predictive check (VPC)
of the 5th, 50th and 95th percentiles, were utilized to assess the predictive performance
of the model in comparison to the observed concentrations. Goodnes of Fit (GoF)
plots and relative standard errors (RSE) of the estimated parameters were also utilized
for model selection. Finally, simulations were performed to evaluate the agreement
between simulated and experimental clinical outcomes.
Results: The tumor model has been initially characterized using a K-PD approach,
due to the lack of pharmacokinetic data. Tumor growth in the absence of any
therapeutic agent was characterized with an exponential model based on the
parameters of the tumor size at baseline (TS0 = 4.43 mm3) and the proliferation
constant (λ = 0.23 1/day). Then, parameters were fixed and the effects of the different
treatments were simulated. CD103 was selected as the target biomarker to monitor
Treg activation, highly correlated with ICOS (R: 0.57) and LAG3 (R: 0.89)
expression. Treg activation state was included in the model with an activation constant
(Kact = 0.057 1/day) directly influenced by the tumor size (TS), and limited by the
amount of total Treg cells in the tumor tissue (EMAX = TotTreg).
Conclusions: Application of the published platform for describing the antitumor effect
of immune therapy in a preclinical model successfully characterized the activity of Treg
immune modulators, identifying the role of Treg cells in the response and inferring
optimal therapeutic combinations.
References:
1. Arabameri A, Asemani D, Hadjati J. A structural methodology for modeling immunetumor
interactions including pro- and anti-tumor factors for clinical applications. Math Biosci.
2018;304(July):48–61. Available from: https://doi.org/10.1016/j.mbs.2018.07.006
2. Lozano T, Gorraiz M, Lasarte-Cía A, Ruiz M, Rabal O, Oyarzabal J, et al. Blockage of FOXP3
transcription factor dimerization and FOXP3/AML1 interaction inhibits T regulatory cell
activity: sequence optimization of a peptide inhibitor. Oncotarget. 2017;8(42):71709–24.
Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5641083/pdf/oncotarget-08-
71709.pdf
3. Sancho-Araiz A, Zalba S, Garrido MJ, Berraondo P, Topp B, de Alwis D, et al.
SemiMechanistic Model for the Antitumor Response of a Combination Cocktail of
ImmunoModulators in Non-Inflamed (Cold) Tumors. Cancers (Basel). 2021 Oct
9;13(20):5049.
Reference: PAGE 32 (2024) Abstr 11021 [www.page-meeting.org/?abstract=11021]
Poster: Drug/Disease Modelling - Oncology