IMMU-02. CH14.18 IN THE TREATMENT OF HIGH-RISK NEUROBLASTOMA: A META-ANALYSIS

High-risk neuroblastoma (HRNB) is a rare malignancy that mainly affects young children. Long-term survival remains low despite aggressive, multimodal treatment regimens. The aim is to assess the effect of Ch14.18, a novel anti-GD2 antibody, on survival in HRNB. In this meta-analysis, Pubmed/MEDLINE, EMBASE, Cochrane CENTRAL and Clinicaltrials.gov bibliographic databases were searched from inception to 1(st) July 2018. Prospective or retrospective interventional clinical trials with at least two arms or observational cohort or case-control studies were eligible. Participants had to have HRNB. Ch14.18 was the intervention of choice. Any control, comparator treatments were accepted. Studies were appraised for inclusion by two independent reviewers. Data was extracted from published reports. Primary outcomes assessed were Event-Free Survival (EFS) and Overall Survival (OS) and presented as Weighted Mean EFS (%) ± Standard Error (SE) and Weighted Mean OS (%) ± Standard Error (SE), respectively. Combined effect size was elicited for EFS±SE and OS±SE. The search yielded 367 results of which 7 studies conducted on 1727 patients were eligible for inclusion. The weighted mean EFS±SE for Ch14.18 regimens (n=779) and control regimens (n=787) were 55.6±2.0% and 41.2±1.7%, respectively. The weighted mean OS±SE for Ch14.18 regimens (n=430) and control regimens (n=348) were 63.4±2.3% and 53.6±2.7%, respectively. Results of the meta analysis yielded statistical power >94%. The combined effect size of Ch14.18 regimens for EFS±SE was 0.2907±0.05 [95% CI: 0.19–0.39, p<0.001] and the combined effect size for OS±SE was 0.26±0.07 [95% CI: 0.11- 0.4, p<0.001]. Significant superior survival outcomes were achieved with the use of Ch14.18 in the treatment of HRNB. These findings support using Ch14.18 as an adjunct in maintenance therapy in high-risk neuroblastoma. This meta analysis is the first of its kind. Survival outcomes are to be interpreted with caution due to confounding bias.