LAUSR

Large segmental chromosomal alterations are common to cancer and a feature of high-risk neuroblastoma (NBL). Though they are early or initiating events in cancer, including often being found in precancerous lesions, their contribution to tumorigenesis is poorly understood. An unproven model is that these changes promote cancer through the cumulative effect of multiple dosage-sensitive genes. Loss of heterozygosity (LOH) at 1p36 is a frequent structural rearrangement in a broad range of human cancers including NBL. Approximately 70% of MYCN amplified NBL have 1p36 LOH with both mutations independently contributing to tumor aggressiveness in NBL. Two tumor suppressor regions have been proposed to exist in 1p36, a distal and proximal region, which have MYCN-independent and dependent roles. Through CRISPR/Cas9 genome editing of primary mouse neural crest cells (NCCs), a source for NBL, we found that loss of the chromatin remodeling factor Chd5 conferred most of the tumor-suppressor effects of 1p36 LOH in in vitro cell transformation assays in cells with endogenous levels of N-Myc. In contrast, when N-Myc was overexpressed, tumor evolution of NCCs genome edited to have randomly sized 1p36 deletions showed a reduction in tumor latency that significantly correlated with deletion of Arid1a. The Arid1a deletions that were selected for ranged from small indels up to large 1p36 deletions, indicating that large deletions are tolerated to achieve loss of a single critical tumor suppressor. Further, using lentiviral Cre-induced deletion of floxed Arid1a in isolated NCCs, we found that Arid1a is a haploinsufficient tumor suppressor in MYCN-driven transformation of NCCs. As Arid1a is a subunit of the chromatin remodeling complex SWI/SNF, which is mutated in 20% of cancers, Arid1a synthetic lethal therapies are being developed for adult malignancies. We are currently verifying those proposed therapies in NBL and are using small-molecule screening of cells lines derived from Arid1a wild-type, het, or null tumors to identify additional synthetic lethalities. Our findings indicate that context, such as the status of MYCN as an oncogene, dictates which 1p36 gene is the critical tumor suppressor, establishing 1p36 LOH as multifaceted not cumulative, as was previously believed. Citation Format: Kirby Wallace, Jesus Garcia-Lopez, Joel Otero, Rachelle Olsen, Chelsea DeVaux, Ashton King, Andrew Davidoff, Kevin Freeman. ARID1A is a haploinsufficient tumor suppressor for N-Myc transformation of neural crest cells [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr B30.