Study Detects New ALK Mutations at Neuroblastoma Relapse


Key Points

  • New ALK mutations were identified at relapse of neuroblastoma.
  • Subclonal ALK mutations can be present at diagnosis with subsequent clonal expansion detected at relapse.

In a study reported in the Journal of Clinical Oncology, Schleiermacher et al found ALK mutations at relapse of neuroblastoma that were not present at diagnosis. They also found that subclonal mutations may be present at diagnosis with clonal expansion observed at relapse.

The study involved analysis of 54 paired diagnosis and relapse neuroblastoma samples for ALK mutations using Sanger sequencing. When Sanger sequencing showed an ALK mutation in only one of the paired samples, a minor mutated component of the other was searched using > 100,000× deep sequencing of the relevant hotspot, a procedure with a sensitivity of 0.17%.

When Mutations Were Detected

An ALK mutation was detected by Sanger sequencing in the diagnosis sample in nine patients (17%), with the same mutation being present at relapse in all cases. In one of these cases, the mutation was present at relapse in only one of several tumor nodules and not in the stroma-rich tissue surrounding the nodules. A germline ALK mutation was observed in only one case.

ALK mutations were detected by Sanger sequencing at relapse alone in five cases (9%), consisting of F1174L mutation (exon 23) in two cases, F1174S mutation (exon 23) in one case, and Y1278S mutation and R1275Q mutation (exon 25) in two cases.

Deep sequencing in four of the patients with mutation detected only at relapse revealed no mutation in the diagnosis samples of two patients. In one case, the mutated allele found at relapse was detected in the diagnosis sample in 0.798% of reads, indicating the presence of an ALK-mutated subclone at diagnosis.

In the remaining case, two different mutations resulting in an identical amino acid change were detected at diagnosis and at relapse. The mutation at diagnosis was present in 8.15% of reads (below the Sanger sequencing limit) and was not observed at relapse; the second mutation was not observed in diagnosis samples and was found in 19.13% of reads at relapse. As noted by the investigators, this latter finding indicates the presence of a mutation switch between diagnosis and relapse that would be expected to yield an identical functional consequence.


The investigators concluded: “[T]he observation of five of 54 new ALK mutations at relapse suggests that the frequency of ALK mutations may be higher at relapse than at diagnosis… [S]ubclonal ALK mutations can be present at diagnosis with subsequent clonal expansion at relapse. Given the potential of ALK-targeted therapy, the significant spatiotemporal variation of ALK mutations is of utmost importance, highlighting the potential of deep sequencing for detection of subclonal mutations with a sensitivity 100-fold that of Sanger sequencing and the importance of serial samplings for therapeutic decisions.”

Gudrun Schleiermacher, MD, PhD, of Institut Curie, is the corresponding author for the Journal of Clinical Oncology article. Olivier Delattre, MD, PhD, of INSERM, and Tommy Martinsson, MD, PhD, of University of Gothenburg, contributed equally as senior authors of the study.

The study was supported by the Annenberg Foundation and many others.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.