Related Experiment Video
Updated: Sep 14, 2026
Gene-environment Interaction Models to Unmask Susceptibility Mechanisms in Parkinson's Disease
Published on: January 7, 2014
PLP1-triggered NOT gate to address neurotoxicity risk of PSMA-targeted prostate cancer therapy
Sanam Shafaattalab1, Alexander Partin2, Aaron Winters2
1A2 Biotherapeutics Inc, Agoura Hills, California, USA sshafaattalab@a2biotherapeutics.com.
Background:
On-target, off-tumor toxicity is a recurring challenge for solid tumor therapeutics targeted to antigens preferentially expressed on malignant cells. Prostate-specific membrane antigen (PSMA) is a tumor-associated antigen that is expressed on the healthy prostate and is generally upregulated in prostate cancer. It is also detected via RNA-Seq in the central and peripheral nervous system (CNS and PNS) at levels that can exceed those in the prostate itself. The CNS/PNS expression of PSMA has been largely ignored in the oncology field, and it is possible that therapeutics targeting PSMA have limited access to antigens in the nervous system. However, many PSMA-targeted therapeutics that are active in the clinic induce substantial neurotoxicity, in some cases, even treatment-associated fatalities.
Methods:
To protect PSMA-expressing neural tissues from on-target, off-tumor toxicity, we applied a synthetic cellular NOT gate that uses a LIR-1-based inhibitory receptor. We selected the NOT gate target antigen by searching genomic databases for a surface protein co-expressed with PSMA in the brain.
Results:
Here, we demonstrate that proteolipid protein 1 (PLP1), a protein expressed on the surface of oligodendrocytes and Schwann cells, is co-expressed with PSMA in the CNS and PNS. A NOT gate construct (PSMA | PLP1 Tmod) protects human oligodendrocytes in vitro from the cytotoxicity of a PSMA CAR. Though PSMA is not expressed in mouse brains at high levels, we show that PLP1 CARs induce toxicity in mice, consistent with the possibility that CNS/PNS expression of the target of an immunotherapeutic poses a serious risk. This toxicity is mitigated by a Tmod construct (PLP1 | PLP1), in which a PLP1 CAR substitutes for a PSMA CAR, a proof of principle that circumvents the problematic expression pattern of PSMA in the mouse.
Conclusions:
Together, these findings not only highlight some of the potential risks and benefits of immunotherapy for solid tumors regarding antigens expressed in the brain, but also suggest a specific solution for PSMA CAR-Ts.
