Intravenous immunoglobulin with prednisone and risk-adapted chemotherapy for children with opsoclonus myoclonus ataxia syndrome associated with neuroblastoma (ANBL00P3): a randomised, open-label, phase 3 trial

No clinical trial has been done for patients with neuroblastoma-associated opsoclonus myoclonus ataxia syndrome (OMA) and existing treatment is based on case reports and small retrospective studies. We aimed to assess OMA response to prednisone and risk-adapted chemotherapy and to establish whether the addition of intravenous immunoglobulin (IVIG) further improves response. We did a randomised, open-label, phase 3 trial in 92 Children's Oncology Group institutions in North America, Australia, and New Zealand with patients younger than 8 years with biopsy-proven, newly diagnosed neuroblastoma or ganglioneuroblastoma associated with OMA. Randomisation was stratified according to the clinical stage of neuroblastoma. We randomly assigned eligible patients through use of computer generated randomisation to receive 12 cycles of IVIG (1 g/kg, day 0 and 1, cycle one; day 0, cycles two to six; day 0, cycles eight, ten, and 12; each cycle lasting 28 days) or no IVIG, in addition to prednisone (2 mg/kg per day divided twice a day for a minimum of two cycles) and neuroblastoma risk-adapted chemotherapy (cyclophosphamide 25 mg/kg for children ≤20 kg and 750 mg/m2 for children >20 kg, day 0 for the first six cycles in patients with low-risk disease). The primary outcome was OMA response, defined as the best score of the three neurological assessments assessed at months 2, 6, and 12 with a scale developed by Mitchell and Pike; baseline versus best response scores were compared. Analysis was by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00033293. Of the 53 patients enrolled in the study between May 15, 2004, and Feb 4, 2013, 33 (62%) were female. 44 patients had low-risk, seven had intermediate-risk, and two had high-risk neuroblastoma. 26 patients were randomly assigned to IVIG and 27 to no IVIG. One patient did not have a neurological assessment and was excluded from OMA response analysis. 21 (81%) of 26 patients in the IVIG group had an OMA response, compared with 11 [41%] of 27 in the no IVIG group (odds ratio 6·1, 95% CI 1·5–25·9, p=0·0029). Median follow-up for patients who were OMA progression-free was 6·1 years (IQR 3·6–8·7, range 1 day to 10·3 years) for both groups. For most patients, the IVIG regimen combined with cyclophosphamide or other risk-based chemotherapy was well tolerated. 28 patients reported grade 3 or higher adverse events, with the most frequent ones being anaemia, neutropenia, vomiting, infectious diseases, ny