PRIMARY: - To determine in the context of a randomized trial whether the EFS of patients with newly diagnosed high-risk NBL is improved with the addition of 131I-MIBG during Induction, prior to tandem autologous stem cell transplantation (ASCT). - To determine whether the addition of crizotinib to intensive multimodality therapy for patients with high-risk NBL whose tumors harbor activating point mutations in or amplification of the ALK gene results in superior EFS compared to a contemporaneously treated cohort of patients whose tumors lack these ALK aberrations. SECONDARY: - To describe the toxicities associated with treatment for high-risk NBL with and without the addition of 131I-MIBG or crizotinib. - To estimate EFS and describe toxicity in patients with newly diagnosed high risk NBL randomized to treatment with an 131I-MIBG-containing Induction prior to BuMel ASCT. - To describe the OS and response rates (evaluated per INRC criteria prior to ASCT and prior to post-Consolidation therapy) for patients with high-risk neuroblastoma treated with or without 131I-MIBG or crizotinib. - To prospectively evaluate the relationship of response rate per revised International Neuroblastoma Response Criteria (INRC) to EFS and OS in patients with high-risk NBL treated with and without the addition of 131I-MIBG or crizotinib. EXPLORATORY: - To evaluate whole body radiation dose, tumor factors, and host factors as potential predictors of efficacy and/or toxicity associated with 131I-MIBG therapy and transplant conditioning. - To determine whether the efficacy (end-Induction response, EFS, and OS) of crizotinib is associated with specific ALK mutations or ALK amplification. - To characterize changes in tumor markers [circulating tumor DNA, including ALK and other tumor specific genetic aberrations, and circulating GD2] over time in response to protocol therapy. - To correlate results of tumor and host profiling with end-Induction response and EFS. - To prospectively evaluate EFS for patients with MIBG non-avid high-risk NBL compared to patients with MIBG-avid high-risk NBL who are randomized to treatment without 131I-MIBG. - To correlate Curie scores calculated from 131I-MIBG post-treatment scans with end-Induction response, EFS and OS. - To define patterns of failure at time of first relapse or progression in patients with high-risk NBL. - To determine the feasibility of prospectively monitoring adverse events using electronic health records. - To compare local, central, and computer assisted Curie score assignment at baseline and during therapy in patients with MIBG-avid high-risk NBL. - To compare late toxicities (including impaired organ function and secondary tumor occurrence) in patients treated with 131I-MIBG or crizotinib to late toxicities in patients who have not received these therapies.