Cancer genomes were long assumed to reach their disordered state by increments, one lesion at a time over many cell generations (Stratton et al., 2009). That picture was revised when Stephens et al. (2011) reported tumors in which tens to hundreds of rearrangements were confined to a single chromosome or chromosome arm, with copy number switching back and forth between only two states. They argued that such a configuration is very unlikely to arise by sequential events and named the process chromothripsis, from the Greek for chromosome and shattering. The idea of a one-off catastrophe has since been supported by direct experiment and by surveys of thousands of tumors.

The topic has particular weight in pediatric oncology. Tumors of childhood have low burdens of point mutation, and large sequencing efforts have shown that structural variants and copy number changes account for a large share of their driver events (Gröbner et al., 2018; Ma et al., 2018). A mechanism that can delete tumor suppressors, amplify oncogenes and create fusion genes in one step is therefore of direct interest, both for understanding how these tumors begin and for the practical business of classifying them.

Almost everything known about chromothripsis in children comes from short-read whole-genome sequencing or, before that, from single-nucleotide polymorphism arrays. Optical genome mapping is a different kind of instrument. It does not read bases. It records the positions of fluorescent labels along very long single DNA molecules and infers structure from the label pattern. Because it is now used as a routine assay in cytogenetic laboratories, the question of how well it reveals a chromothripsis signature has become a practical one. This review addresses that question for the solid tumors of childhood.

This is a narrative review of the published literature. It draws on the original descriptions of chromothripsis and related phenomena, on the mechanistic studies that followed, on genome-wide studies of pediatric tumors, and on technical and validation papers concerning optical genome mapping. No formal search protocol was applied and no pooled estimates are offered. The aim is to connect two bodies of work that have developed largely apart, and to say plainly where the evidence for their combination is strong, where it is indirect and where it is missing.

Chromothripsis is one member of a family of complex rearrangements for which Holland and Cleveland (2012) proposed the collective term chromoanagenesis. The family matters here because its members leave different traces, and a method that sees only part of the trace may confuse them. Figure 1 places the main findings discussed in this review in order of publication.

Medulloblastoma provided the first clear pediatric association. Rausch et al. (2012) found chromothripsis in Sonic hedgehog subgroup tumors from children with germline TP53 mutations and proposed that loss of p53 function either permits the catastrophe or allows the cell to survive it. Later whole-genome analysis across the medulloblastoma subgroups confirmed the concentration of the pattern in TP53-mutant Sonic hedgehog disease (Northcott et al., 2017). The observation has clinical consequences beyond the tumor, because the finding of chromothripsis in a child's medulloblastoma is a reason to consider germline testing.

In neuroblastoma, Molenaar et al. (2012) detected chromothripsis in a subset of high-stage tumors, where it was associated with poor outcome and with structural changes affecting genes such as MYCN. Experimental work has since shown that shattered fragments can circularize into extrachromosomal DNA and that repeated rounds of breakage drive the evolution of gene amplification (Shoshani et al., 2021), a mechanism plainly relevant to an amplification-driven tumor.

Bone tumors stood out in the original report, in which chromothripsis was notably common (Stephens et al., 2011). Pediatric osteosarcoma genomes are dominated by structural variation, including rearrangements that inactivate TP53 through its first intron (Chen et al., 2014), and a pattern of chromothripsis followed by amplification has been described in the same disease (Behjati et al., 2017). Ewing sarcoma illustrates a neighboring category: in a substantial fraction of cases the defining EWSR1 fusion is produced by chromoplexy, a looped chain of rearrangements across several chromosomes, and not by a simple reciprocal translocation (Anderson et al., 2018). Pan-cancer studies of childhood tumors place these findings in a wider frame, in which complex rearrangement accompanies TP53 disruption across histologies (Gröbner et al., 2018).

The method descends from the nanochannel genome mapping described by Lam et al. (2012). Ultra-high molecular weight DNA is extracted so as to preserve molecules hundreds of kilobases long. A labeling enzyme attaches a fluorophore at each occurrence of a six-base recognition motif, the molecules are driven electrophoretically into nanochannels that hold them stretched, and a camera records the label positions. The resulting pattern acts as a low-resolution barcode. Molecules are assembled into maps or aligned directly to an in-silico digest of the reference genome, and discordances are called as insertions, deletions, inversions, duplications and inter- or intrachromosomal translocations. Copy number is estimated separately from the depth of molecule coverage.

Clinical validation has so far been strongest outside solid tumors. Mantere et al. (2021) reported concordance with standard cytogenetic tests for constitutional aberrations whose breakpoints lie outside centromeric regions, and Neveling et al. (2021) reached a similar conclusion for hematological malignancies while also uncovering additional rearrangements that karyotyping, fluorescence in situ hybridization and arrays had missed. A study of acute lymphoblastic leukemia, which included pediatric patients, came to a comparable view (Lühmann et al., 2021). These results establish that the method reliably detects structural variants of the size and kind that make up a chromothripsis cluster.

The central gap is simple to state. Published use of optical genome mapping in solid tumors of childhood is limited to small series and individual case descriptions, and there is no large comparison in which the same pediatric tumors were assessed for chromothripsis by both optical mapping and whole-genome sequencing. The claim that the method detects the signature in these tumors therefore rests on inference from its validated performance for individual structural variants, not on direct demonstration.

Three kinds of work would close the gap. Paired analyses of the same specimens would show how often the two methods agree on the presence, extent and chromosome of a chromothripsis event. Calling rules adapted to map data are needed, since existing tools assume sequencing input and thresholds chosen for base-pair breakpoints cannot be carried over unchanged. Finally, reporting standards should require that a laboratory state which criteria were assessed and by which method. A report of chromothripsis based on clustering and oscillation alone is a weaker statement than one supported by allelic data, and the difference ought to be visible to the reader. Complementary long-read sequencing, which adds junction sequence and phasing, is the obvious partner where the distinction changes management.

Chromothripsis is an established feature of several pediatric solid tumors and carries diagnostic, prognostic and, through its link with germline TP53 mutation, familial significance. Optical genome mapping is well suited to showing the parts of its signature that depend on large-scale structure: clustered breakpoints, alternating copy number and the physical connection of rearranged fragments. It cannot show allelic state or junction sequence, and so cannot by itself satisfy every accepted criterion or separate chromothripsis cleanly from its relatives. Until direct comparisons in childhood tumors are published, an optical mapping result is best reported as a pattern consistent with chromothripsis, with the criteria examined set out explicitly.

To be declared by the author(s) before publication.

To be declared by the author(s) before publication.