To clarify whether karyotype aberrations (KA) involving regions not covered by the standard fluorescence in situ hybridization (FISH) panel have independent prognostic relevance, we evaluated KA by conventional cytogenetics in a learning cohort (LC; n = 166) and a validation cohort (VC; n = 250) of untreated chronic lymphocytic leukemia (CLL) patients. In the VC, novel mitogens were used to improve metaphase generation and TP53, NOTCH1, and SF3B1 mutations were assessed. KA undetected by FISH were found in 35 and 35% of the cases in the LC and VC, respectively. In addition to FISH, KA allowed reclassification of 23 and 26% of cases in the LC and VC, respectively, into a higher cytogenetic risk group. By multivariate analysis, both in the LC and VC, KA other than isolated 13q deletion correlated with a shorter time to first treatment (TFT; P < 0.001 and 0.003, respectively), while a complex karyotype predicted a worse overall survival (OS, P = 0.015 and 0.010, respectively). In the VC, where a comprehensive biologic assessment was performed, a shorter TFT was also predicted by stage (P < 0.001), IGHV mutational status (P = 0.05), and del(17p)/TP53 mutations (P = 0.033) while stage (P = 0.023) and del(17p)/TP53 mutations (P = 0.024) independently predicted a shorter OS. FISH results did not independently impact on TFT and OS, in the LC and VC cohorts; this was also the case for NOTCH1 and SF3B1 mutations in the VC. We suggest that in CLL, conventional karyotyping with novel mitogens could be more effective than FISH for the detection of KA allowing for a more precise refinement of prognosis.
© 2015 Wiley Periodicals, Inc.