Abstract B095: Mapping the genetic features of immune checkpoint responsiveness using AAV-CRISPR mediated in vivo screen

Immune checkpoint blockade has achieved tremendous clinical success across many tumor types, but fails to induce clinical responses in many patients. The mechanisms underlying checkpoint resistance remain poorly characterized. Recent studies have applied next generation sequencing techniques to catalog the mutational burden of patient tumors, which provides a wealth of data to determine common mutations. To map the genetic features of response to checkpoint blockade immunotherapy as well as correlating the clinical efficacy with certain mutations, we developed a novel direct in vivo CRISPR screening approach for high-throughput profiling of functional cancer drivers in an autochthonous manner by injecting AAVs carrying an sgRNA library targeting the top 50 TCGA pan-cancer recurrently mutated tumor suppressor genes (mTSG) into the immunocompetent Cas9 transgenic mice. All mice that received the AAV-mTSG library developed liver cancer and died within four months. We then utilized MIP sequencing of sgRNA target sites to chart the mutational landscape of these tumors, revealing the functional consequence of multiple variants in driving liver tumorigenesis as well as identifying specific gene pairs that were co-occurring across mice. Using this approach, we also mapped the mutation landscape changes under the pressures of immune checkpoint inhibitors, anti-PD1 or anti-CTLA4. We monitored liver tumor growth in AAV-mTSG injected LSL-Cas9;LSL-Fluc mice by using intravital bioluminescent imaging system (IVIS) in combination with dissection check before drug administration. Using IVIS data, we grouped them into 3 size-matched cohorts to receive anti-PD1 or anti-CTLA4 treatments or PBS control. According to the survival data, the mice with mTSG-induced liver tumor benefit from anti-PD1 or anti-CLTA4 treatment. By comparing the mutation frequencies of liver tumors in the mice receiving either checkpoint inhibitors or PBS treatment, we mapped the mutation landscape changes associated with anti-PD1 or anti-CTLA4 treatment. We are performing validation studies on top targets such as Arid1a, Stk11, and B2M. Using this approach, we systematically mapped the correlation of these top 50 driver mutations with cancer immune evasion and immunotherapy responsiveness, providing a valuable reference for patient stratification when considering immunotherapy as well as novel targets for synergistic interventions. Citation Format: Guangchuan Wang, Ryan Chow, Zhigang Bai, Lupeng Ye, Sidi Chen. Mapping the genetic features of immune checkpoint responsiveness using AAV-CRISPR mediated in vivo screen [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr B095.