The detection of structural variants (SVs) represents a critical component in the diagnostic evaluation and treatment of many hematologic malignancies. Although clinical SV testing mainly consists of traditional cytogenetic methodologies, technological innovations have led to alternative approaches with improved resolution. In this study, we sought to characterize the clinical impact of targeted RNA sequencing on the diagnosis of myeloid and immature lymphoid malignancies. Across a cohort (n = 380) of myeloid (87%) and immature lymphoid (13%) tumors, we compared SVs detected by chromosome banding analysis (CBA) and fusion events detected by anchored multiplex polymerase chain reaction (AMP)-targeted RNA sequencing. Variants detected by either assay were categorized using a 5-tier system: tier 1 (established clinical significance)
tier 2 (possible clinical significance)
tier 3 (unknown significance)
tier 4 (known germ line variants), and tier 5 (no variants detected). The combined use of AMP and CBA improved the detection of clinically relevant (tier 1 or 2) findings in 10% of cases. Unexpectedly, in 1% (3/380) of the patients in our study, CBA appeared to detect a defining SV, for example, t(9
22)(q34
q11.2), that was not confirmed by AMP fusion studies. Subsequent evaluation by orthogonal approaches confirmed breakpoints on the expected chromosomes but did not involve the anticipated genes. Our study indicates that "chromosomal mimicry," a phenomenon in which chromosome morphology resembles a known SV but lacks the expected gene-level rearrangement, is an infrequent but recurrent finding with the potential to confound clinical management. Our study highlights the need for assays with gene-level resolution in the diagnostic evaluation of hematologic malignancies.