Judicious use of precise fluorescence in situ hybridisation panels guided by population prevalence may assist pragmatic detection of clinically targetable Philadelphia chromosome-like acute lymphoblastic leukaemia fusions: a systematic review.

Diagnosis of Philadelphia chromosome-like acute lymphoblastic leukaemia (Ph-like ALL) in the real-world remains challenging because of definitional complexities, the diverse diagnostic techniques available and the cost, expertise and time involved. We summarise evidence for diagnosis of clinically important Ph-like ALL related genomic lesions using fluorescence in situ hybridisation (FISH) targeting only clinically important and actionable lesions, an accessible and cost-effective diagnostic technique. Electronic databases were interrogated using broad MeSH terms for articles reporting a detailed FISH strategy for diagnosis of Ph-like ALL published since 2014, yielding 653 full text articles and abstracts. We searched the National Library of Medicine Databases including PubMed, Medline, Embase, Cochrane and relevant abstracts. We included studies with a primary aim of determining the utility of FISH for Ph-like ALL diagnosis and studies with broader aims demonstrating Ph-like ALL diagnostic algorithms which partially involved FISH. Nineteen studies met inclusion criteria. Evidence for FISH to detect CRLF2 rearrangements in Ph-like ALL is strongly established and evidence for FISH to detect non-CRLF2 lesions is evolving rapidly. We documented 1620 cases of non-CRLF2 Ph-like lesions diagnosed by FISH. Confirmatory side-by-side methods were applied in six studies (246 samples), four of which demonstrated 100% concordance of FISH results with alternative methods, while two studies demonstrated over 70% sensitivity and specificity. Additional studies demonstrated wide utilisation of FISH in Ph-like ALL classification across diverse geographies and ethnicities, with contrasting prevalence, implicating a need for targeted FISH strategies. In real-world cohorts, it may be clinically useful to prioritise limited early FISH in B-cell ALL (B-ALL) diagnostic algorithms to identify Ph-like abnormalities that respond to locally available kinase inhibitors to promote and prioritise broad access to effective targeted treatment. Additional studies are required to provide adequately powered validations and verifications of targeted Ph-like FISH panels to confirm sensitivity and specificity against side-by-side gold standard methods, and to define optimal local approaches.