How Mechanistic Modeling Advances CAR T-Cell Therapy Development

In an interview with BioPharm International®, Armin Sepp, PhD, senior principal scientist and head of Biologics Modeling at Certara UK, discusses the growing role of mechanistic modeling in cell and gene therapy development. Presenting his talk, “Quantitative Mechanistic Pharmacology of Cell and Gene Therapies,” at the 2026 AAPS National Biotechnology Conference, Dr. Sepp emphasizes how quantitative approaches may help address the high attrition rates that continue to challenge advanced therapeutics.

Chimeric antigen receptor (CAR) T-cell therapies differ fundamentally from traditional small molecules and biologics because they function as “living drugs,” Dr. Sepp says. After infusion, CAR T cells can rapidly proliferate in response to cancer cells, leading to drug concentrations that may increase by hundreds or even thousands of times beyond the initial administered dose. A subset of these engineered immune cells can also persist long-term within the patient, introducing dynamics not seen with conventional therapies, he explains.

These complex behaviors create challenges in predicting efficacy, toxicity, and patient response, Dr. Sepp notes. He explains that quantitative systems pharmacology (QSP) models are increasingly being used to mechanistically characterize processes such as tissue distribution, target engagement, immune activation, proliferation, differentiation, cytokine release, and eventual cell elimination.

How can QSP improve CAR T-cell therapy development?

QSP frameworks aim to integrate biological and clinical variables, including tumor burden, antigen density, tumor location, and CAR–antigen affinity, to better predict therapeutic outcomes, according to Dr. Sepp. He notes that these models may also help researchers balance treatment efficacy with immune-related toxicities like excessive cytokine release.

“This all creates a very intricate and complex environment with many forward and reverse regulation loops that make it highly, highly sensitive,” Dr. Sepp says.

Although uncertainties remain, particularly in solid tumors and heterogeneous cancers, Dr. Sepp believes advances in systems immunology and population pharmacokinetic modeling are improving predictive capabilities. These approaches may ultimately support both safer dosing strategies and more individualized treatment optimization.

While predictive models may never fully eliminate uncertainty, they can improve the odds of clinical success by providing a more mechanistic understanding of complex cell therapy behaviors and patient-specific responses, Dr. Sepp adds.

The AAPS National Biotechnology Conference runs May 11–14, 2026, in San Diego, Calif.

Click here for more conference coverage.

About the speaker

Armin Sepp, PhD, Senior Principal Scientist, Head of Biologics Modeling, Certara UK

Within Certara’s Simcyp Science team, Dr. Sepp specializes in biologics. Before joining Certara, he was a scientific leader and fellow at the DMPK Modelling group at GlaxoSmithKline where he provided pre-clinical to phase 1 pharmacokinetic/pharmacodynamic support for monoclonal antibodies, antibody fragments, and other engineered proteins. His background is in bio-organic chemistry with post-doctoral training in protein engineering and in vitro evolution at the MRC Immunochemistry Unit in Oxford and MRC Laboratory of Molecular Biology in Cambridge, UK.