See how mechanistic QSP modeling enables accurate first-in-human dose selection for next-generation CAR-T therapies in multiple myeloma.
CAR-T therapies have transformed treatment for hematologic malignancies, but current approaches come with significant challenges. Ex vivo manufacturing introduces delays, logistical complexity, and risks that can impact patient outcomes. In vivo CAR-T therapies aim to overcome these barriers, but introduce new uncertainty in dose selection and clinical translation.
In this poster, we demonstrate how a mechanistic QSP model can support first-in-human (FIH) dose selection for an in vivo BCMA-targeting CAR-T therapy, bridging preclinical and clinical data to improve decision-making.
What you’ll learn
Download this poster to explore how QSP modeling can:
- Enable model-informed first-in-human dose selection: Predict clinically relevant dosing more accurately than traditional scaling approaches
- Capture complex in vivo CAR-T biology: Model viral delivery, CAR expression, T-cell expansion, and tumor cell killing within a single framework
- Integrate preclinical and clinical datasets: Combine mouse data and clinical Cilta-cel data to inform human translation
- Reduce uncertainty in emerging therapy modalities: Provide a mechanistic foundation for developing in vivo CAR-T therapies
Why it matters
Dose selection is one of the highest-risk decisions in early clinical development—especially for novel modalities like in vivo CAR-T.
This work shows how a mechanistic QSP approach can:
- Improve prediction accuracy compared to allometric scaling and empirical methods
- Provide insight into the biological drivers of efficacy
- Enable more confident and efficient progression into clinical trials
Instead of relying on simplified scaling assumptions, teams can use modeling to directly link mechanism to clinical outcome.
Authors:
Christopher J Morris, Vasiliki Kostiou, Viji Chelliah, Piet van der Graaf, Shannon Grande Contrastano, Sarah Tannehill-Gregg, M Travis Quigley
