Gaston Garcia-Orueta 1,2, Anna Dari 3, Douglas Steinbach 4, Taylor Kathryne 5, Douglas Yamada 6, Nahor Haddish-Berhane 4, Juan José Pérez Ruixo 3, Zinnia Parra-Guillén 1,2, Iñaki F Troconiz 1,2,7
1 Department of Pharmaceutical Sciences, School of Pharmacy and Nutrition, University of Navarra (Pamplona, Spain), 2 Navarra Institute for Health Research (IdiSNA) (Pamplona, Spain), 3 Johnson & Johnson (Beerse, Belgium), 4 Johnson & Johnson Innovative Medicine (Spring House, USA), 5 Lung Disease Area Stronghold, Oncology TA; Johnson & Johnson Innovative Medicine (Rockville, USA), 6 Prostate Disease Area & Immunotherapy Stronghold, Oncology TA, Johnson & Johnson Innovative Medicine (Spring House, USA), 7 Institute of Data Science and Artificial Intelligence, DATAI, University of Navarra (Pamplona, Spain)
Introduction.
Oncolytic viruses (OVs) are a promising therapeutic strategy in oncology due to their dual mechanism of action: direct oncolytic effect and activation of systemic antitumor immunity [1,2]. Several OVs, including T-VEC (talimogene laherparepvec), have been approved for clinical use, and others are under evaluation worldwide [1]. Combining OVs with immune checkpoint inhibitors, such as anti-PD-1 antibodies, is currently an important research area aimed at improving treatment outcomes [3]. In this sense, mathematical and semi-mechanistic models in immuno-oncology can be considered as valuable tools supporting the design of treatment combinations and dosing regimens [4]. A recent modelling study showed that it is possible to describe how an oncolytic virus behaves and affects tumor growth using a semi-mechanistic approach [5].
Objective.
The objective of this project is to develop a semi-mechanistic model to describe tumor volume dynamics in the injected and non-injected lesions when administering OV monotherapy, anti-PD-1 monotherapy or the combination treatment, evaluating different dosing schemes.
Methods.
Longitudinal tumor volume data from 160 syngeneic mice inoculated with MC38-5AG tumor cells in both flanks were used in this analysis. Treatments included OV monotherapy at three doses (10⁴, 10⁵, 10⁶ plaque forming units, PFU) injected in one of the tumors (right), anti–PD-1 200 µg monotherapy under three different schedules administered intraperitoneally (IP), and OV + anti-PD-1 combinations across all virus doses (the highest virus dose also tested with all anti-PD-1 schedules). Tumor volume was measured in both tumor lesions. Tumor volume data were modelled following a sequential and integrated approach using NONMEM 7.5. State-of-the-art model evaluation techniques were employed to assess model performance.
Results.
The previously developed kinetic-pharmacodynamic (KPD) model for the OV monotherapy including (i) 3 transit compartments to delay the direct and non-linear OV effect on the injected tumor and (ii) the generation of tumor-associated antigens (TAA) responsible of activating an indirect CD8 response capable to (iii) induce immune killing in both tumor lesions was used as starting point. The model was then expanded including also an additional KPD model with 3 transit compartments to describe the disposition of anti-PD-1, capable to further increase the immune CD8 response to ultimately kill the tumor with a share immune killing rate constant. The mean transit times associated with anti-PD-1 and OV disposition were 5.41 and 2.02 days, respectively. Responses in both tumor lesions were not correlated, and overall weaker in the (left) non-injected lesion after OV administration. However, response rates in the non-injected lesion increased with combination therapy, especially at the highest dose level, resulting in a change in the responder rate from 17.5% with OV monotherapy to 70% with combination therapy.
Conclusions.
While clinical studies are ultimately required to fully understand and develop human therapeutics, this semi-mechanistic model captures the interaction between viral oncolysis and immune activation, thus providing a quantitative tool to support the development of OV and anti-PD-1 therapies.
References:
[1] Shalhout et al., 2023, Nat Rev Clin Oncol, 20(3):160–177.
[2] Yan et al., 2024, Cancer Cell Int, 24(1):242.
[3] Santos Apolonio et al., 2021, World J Virol, 10(5):229–255.
[4] Sancho-Araiz et al., 2021, Pharmaceutics, 13(7):1016.
Reference: PAGE 34 (2026) Abstr 12085 [www.page-meeting.org/?abstract=12085]
Poster: Drug/Disease Modelling - Oncology