Listen to the podcast by Dr Irwin at www.jco.org/podcastsThe International Neuroblastoma Response Criteria (INRC) require serial measurements of primary tumors in three dimensions, whereas the Response Evaluation Criteria in Solid Tumors (RECIST) require measurement in one dimension. This study was conducted to identify the preferred method of primary tumor response assessment for use in revised INRC. Patients and MethodsPatients younger than 20 years with high-risk neuroblastoma were eligible if they were diagnosed between 2000 and 2012 and if three primary tumor measurements (antero-posterior, width, craniocaudal) were recorded at least twice before resection. Responses were defined as $ 30% reduction in longest dimension as per RECIST, $ 50% reduction in volume as per INRC, or $ 65% reduction in volume.Results Three-year event-free survival for all patients (N = 229) was 44% and overall survival was 58%. The sensitivity of both volume response measures (ability to detect responses in patients who survived) exceeded the sensitivity of the single dimension measure, but the specificity of all response measures (ability to identify lack of response in patients who later died) was low. In multivariable analyses, none of the response measures studied was predictive of outcome, and none was predictive of the extent of resection. ConclusionNone of the methods of primary tumor response assessment was predictive of outcome. Measurement of three dimensions followed by calculation of resultant volume is more complex than measurement of a single dimension. Primary tumor response in children with high-risk neuroblastoma should therefore be evaluated in accordance with RECIST criteria, using the single longest dimension.

Our study showed overall high correlation for SUV measurements obtained from MR attenuation correction compared with CT attenuation correction, although PET/MRI underestimated SUV compared with PET/CT. SUVs measured from PET/MRI indicated good intrapatient reliability.