Overcoming Antigen-Low Tumor Escape in Non Small Cell Lung Cancer with Chimeric TCR T Cell Therapy

ABSTRACT CAR-T cell therapy has revolutionized the treatment of hematological cancers but has yet to achieve durable efficacy in more common solid tumors, such as non-small cell lung cancer (NSCLC). In a phase I trial evaluating CAR-Ts targeting the tumor-associated antigen ROR1, we observed limited efficacy due to CAR-T exhaustion and heterogeneous ROR1 expression on tumors, which led to the escape of antigenlow/null tumors even in transient responders. To address these challenges, we developed an autochthonous model of ROR1+ NSCLC (KPROR1), which recapitulates the suppressive tumor microenvironment (TME), heterogeneous ROR1 expression, and CAR-T exhaustion observed in patients. We found that enhancing T cell function by overexpressing the AP-1 transcription factor c-Jun improved response to PD-L1 blockade but failed to prevent escape of ROR1low/null tumors, emphasizing the need for strategies that both preserve CAR-T function and control escape of antigenlow/null tumors. To enhance CAR-T sensitivity, we engineered chimeric TCRs (ChTCRs), which fuse a CAR single-chain variable fragment (scFv) to a TCR constant chain. ChTCRs show superior sensitivity to antigenlow tumors and outperform CAR-Ts in vivo in models of B cell malignancies. Moreover, bi-specific ChTCRs targeting two antigens outperform optimized bi-specific CAR-Ts, suggesting they may be more effective than CAR-Ts at controlling heterogeneous solid tumors. However, while ChTCRs do not require synthetic costimulation for efficacy in B cell malignancies, their activity in solid tumors may be limited by insufficient co-stimulation in the suppressive TME. To overcome this barrier, we designed novel co-stimulatory receptors that convert inhibitory PVR-TIGIT signals into activating signals and enhance ChTCR-T function in vitro. Since co-stimulation partially relies on AP-1 activation, overexpression of c-Jun could further amplify the favorable effects of co-stimulation. I hypothesize that ChTCR-Ts will be more sensitive to ROR1low tumors in vivo than CAR-Ts, and that co-expression of co-stimulatory receptors targeting the PVR/TIGIT axis will synergize with c-Jun overexpression to enhance ChTCR-T activity, enabling more effective control of tumors with heterogeneous ROR1 expression. In this proposal, I will: 1) assess ChTCR-T activity against ROR1low tumors in the KPROR1 model relative to CAR-Ts and identify barriers to efficacy, and 2) determine whether the co-stimulatory receptors I designed synergize with c-Jun to enhance ChTCR-T function and response to PD-L1 blockade. These studies will reveal the barriers to ChTCR-T efficacy in vivo and potentially identify strategies to control escape of heterogeneous lung tumors. Through this fellowship, I will gain expertise in tumor immunology, computational biology, and preclinical mouse models, equipping me to advance immunotherapies by integrating fundamental research with clinical practice.