Abstract B19: ARID1A is a 1p36 tumor suppressor that collaborates with N-Myc in neuroblastoma

Neuroblastoma (NB) is an embryonal cancer of the sympathetic nervous system (SNS) that causes 15% of pediatric cancer deaths. A major barrier to advancing targeted therapy in NB is identifying the unknown oncogenes and tumor suppressors that are buried within large segmental chromosomal gains and losses that characterize high-risk NB. To assist in identifying them, we have established a novel system for studying NB based on the transformation of neural crest cells (NCCs), the progenitor cells of the SNS, isolated from mouse embryonic day 9.5 trunk neural tube explants. Based on pathology and gene expression analysis, we have demonstrated that the enforced expression of N-Myc in wild-type NCCs generates phenotypically and molecularly accurate NB that most closely models MYCN-amplified NB. Using this system we next addressed a longstanding problem in the NB field. Though 70% of MYCN-amplified NBs have 1p36 deletions, representing ~28% of high-risk NBs, the 1p36 tumor suppressor that collaborates with N-Myc is unknown. Using CRISPR/Cas9 technology in primary NCCs, we screened multiple 1p36 candidate tumor suppressors in a tumorsphere-forming assay. We found tumorspheres only formed when two chromatin-remodeling factors, CHD5 and ARID1A, were both lost, though their loss was not sufficient to generate tumors in vivo. Therefore, we next tested whether deletion of Chd5 and Arid1a, Arid1a alone, or deletion of the entire syntenic 1p36 region, as validated by FISH and karyotype analysis, is selected for during N-Myc driven transformation of primary NCCs. To achieve this, we started with low-efficiency deletion of each target in three independently derived NCC lines and then allowed clonal evolution during tumorigenesis to identify the favored deletion for N-Myc overexpressing NCCs. Tumors with CRISPR/Cas9 targeted 1p36 genes formed with a reduced latency of ~50 days over control (p 80% of the reads in the daughter tumors. We also saw selection for indel mutations and deletions of CHD5, but not in every tumor analyzed and no selection for loss of the entire 1p36 locus in any tumor. Using ex vivo adeno-cre mediated deletion of one or both floxed Arid1a alleles in NCCs, we found that loss of one Arid1a allele significantly (p=0.02) decreased latency by ~40 days over control in collaboration with N-Myc. Loss of both alleles conferred no advantage, indicating that when combined with N-Myc, ARID1A is an obligate haplo-insufficient tumor suppressor. In sum, these tumor studies support a model in which loss of one Arid1a, but not the entire syntenic 1p36 locus, is strongly selected for during N-Myc driven NB tumorigenesis, suggesting that Arid1a is the critical 1p36 tumor suppressor. The next step in this longstanding problem is to understand why loss of ARID1A collaborates with MYCN amplification and to exploit this discovery to improve treatment for MYCN amplified/1p36 deleted NBs. Citation Format: Jesus Garcia-Lopez, Kirby Wallace, Joel Otero, Rachelle Olsen, David Finkelstein, Jerold Rehg, Yongdong Wang, Kevin Freeman. ARID1A is a 1p36 tumor suppressor that collaborates with N-Myc in neuroblastoma [abstract]. In: Proceedings of the AACR Special Conference: Pediatric Cancer Research: From Basic Science to the Clinic; 2017 Dec 3-6; Atlanta, Georgia. Philadelphia (PA): AACR; Cancer Res 2018;78(19 Suppl):Abstract nr B19.