2ⁿᵈ Edition of the Cancer R&D World Conference 2026

Speakers - CRDWC 2026

Chun Huang, Cancer R&D World Conference, Miami, Florida, USA

Chun Huang

Chun Huang

  • Designation: Johnson & Johnson
  • Country: USA
  • Title: Developing a Next-Generation In Vitro Human BBB Model to Predict TCE-Mediated BBB Disruption and ICANS Risk

Abstract

T cell engager (TCE) therapies represent an established class of cancer immunotherapies that redirect endogenous T cells toward tumor cells through binding to tumor-associated antigen(s) and CD3. This induced proximity and T-cell receptor clustering causes T cell activation and tumor cell killing. Despite showing clinical efficacy, TCE therapies are frequently associated with immune-related toxicity.

TCE-activated T cells release pro-inflammatory cytokines which can activate other immune cells, especially monocytes and macrophages, and amplify inflammatory signaling through mediators such as GM-CSF, IL-1β, and IL-6. This cytokine cascade underlies cytokine release syndrome (CRS), the severity of which correlates with immune-effector cell-associated neurotoxicity syndrome (ICANS). Furthermore, emerging clinical and translational evidence suggests that endothelial activation and blood brain barrier (BBB) disruption contribute to ICANS pathophysiology. Better preclinical models are needed to predict TCE-associated ICANS and BBB disruption.

Existing in vitro BBB models used to evaluate cytokine-mediated neurotoxicity are often limited to brain microvascular endothelial cells (BMECs) and lack other key neurovascular components such as pericytes and astrocytes, which are critical for BBB integrity. In vivo models (i.e., mouse and nonhuman primate) provide additional context but have limited translational relevance due to interspecies differences in CNS physiology, immune responses, species cross-reactivity and biomarker expression.

Here, we aimed to assess TCE-associated immunotoxicity and neurovascular injury using a multicellular human BBB model designed to recapitulate key features of BBB structure and function.

We leveraged a commercially available multicellular BBB model, comprised of three essential human BBB cell types: BMECs, astrocytes, and pericytes. This configuration supports BMEC maturation and formation of a physiologically relevant barrier with transendothelial electrical resistance (TEER) values of ~4,000–5,000 Ω·cm², exceeding many previously reported in vitro models, which typically remain below commonly cited physiological thresholds (>1,500 Ω·cm²).

Exposure to pro-inflammatory cytokines mimicking CRS conditions induced dose-dependent BBB model disruption, characterized by reduced TEER, increased paracellular permeability, and modulation of clinically relevant ICANS- and BBB damage–associated markers. Endothelial activation markers, including ICAM-1 and VCAM-1, were significantly upregulated. Markers associated with ICANS and BBB injury showed differential regulation, including increased S100β and GFAP and decreased ANG1. Together, these findings indicate loss of barrier integrity following CRS-like inflammatory challenge.

This multicellular human BBB platform has potential to serve as a component of a translational framework to assess TCE-associated BBB disruption and ICANS risk. Incorporation of immune-derived inflammatory milieus generated from TCE-driven cytotoxicity assays, combined with functional and molecular readouts, may enable comparative assessment of neurotoxicity liability across TCE candidates and support safety de-risking during development.