Stratification of patients with neuroblastoma for targeted ALK inhibitor therapy.

Abstract

9514 Background: Mutations in the ALK proto-oncogene are the major cause of hereditary neuroblastoma (NB), can be somatically acquired, and may offer a tractable therapeutic target for a subset of patients. Methods: We identified 1,604 paired constitutional and tumor DNA samples from patients representative of the natural spectrum of NB for comprehensive evaluation of ALK genomic status. This set has complete annotation of clinical covariates and patient outcome, and provides sufficient power to identify clinically relevant ALK aberrations within each NB risk group. We performed DNA sequencing of the ALK kinase domain and defined ALK copy number (CN) status using quantitative PCR. For each tumor where a somatic mutation was identified, we sequenced matched germline DNA, allowing us to understand the frequency of hereditary predisposition to NB. Results: Mutations were discovered in 126/1,604 (7.9%) primary NB tumor samples distributed across the spectrum of phenotypes. Mutations at R1275 were most common (n=54) followed by F1174 (n=39) and F1245 (n=15), with those 3 locations accounting for 86% of all mutations. I1170 and I1171 were identified twice and 17 other previously unidentified mutations were seen only once. Work is ongoing to determine the functional significance of these mutations, as is evaluation of matched germline DNA. We found a statistically significant higher number of ALK mutations in high-risk versus low-risk cases (p=0.018) and in cases with MYCN amplification (p=0.031). There was no significant difference between other risk stratification covariates including age (< or > 1 year, p=0.066), histology (p=0.103) and ploidy (0.922). ALK CN gain was found in 197/1345 (14.6%) samples, while those with amplification were identified in 24/1345 (1.8%) and almost exclusively seen concurrently with MYCN amplification. CN gain was seen more frequently in high-risk, as compared with low- and intermediate-risk, disease (p<0.01). Conclusions: Combining ALK mutation and CN data overall and within each risk group and correlating this with other genetic factors, clinical phenotype, and patient outcome will allow us to determine the clinical significance of ALK mutations in NB and develop strategies for future clinical trials.