Clinical Significance of Clonal Cytogenetic Heterogeneity (CCH) at Diagnosis in Adult Patients with Acute Lymphoblastic Leukemia (ALL)

Background:Although cytogenetic abnormalities at diagnosis are recognized as one of the most potent prognostic factors in acute leukemia patients, CCH acquisition at diagnosis, which are considered as a result of clonal evolution of leukemia cells, is not taken into account in prognostic classifications. Recent studies reported that CCH acquisition was observed in 24 - 32% of adult AML patients with abnormal karyotype, was more likely to occur in patients with older age and complex karyotype, and showed adverse prognostic impact. However, the clinical significance of CCH acquisition has not been investigated in adult ALL patients to date. Patients and methods: Of the 238 adult ALL patients diagnosed between 1990 and 2016, 120 patients with abnormal karyotype at diagnosis, who underwent intensive chemotherapy, were included in this study. CCH was defined as presence of two or more cytogenetically abnormal clones. A defined ancestral clonal evolution included either mother-daughter and/or branched patterns. In the mother-daughter pattern, a daughter clone showed all cytogenetic abnormalities of a mother clone plus additional abnormality(s), which define a distinct subclone. In a branch pattern, all subclones possessed common cytogenetic abnormalities suggesting presence of a common ancestry, but each subclone acquires unique additional abnormality(s), which define them as distinctive subclones. Both patterns of cytogenetic clonal evolution were sometimes seen in a patient. Composite karyotypes were applied to patients where a common ancestry could not be clearly determined because of too complicated cytogenetic findings. Fisher's exact test was used to compare binary variables. The logistic regression model was used for multivariate analysis of predisposing factors. Overall survival (OS) was estimated with the Kaplan-Meier method and compared using the log-rank test. The Cox proportional hazard model was used for multivariate analysis of prognostic factors. Values of p < 0.05 were considered to indicate statistical significance. Results:Of the 120 patients included in this study, 64 patients were male, and 56 were female. The median age was 50 years (range, 16-79 years). Karyotypes at diagnosis were Philadelphia chromosome (Ph) in 56 patients, complex in 15, and t(8;14) in seven. According to the definition described above, 47 patients (39%) showed CCH at diagnosis, and two (4%) among them were categorized as composite karyotype. Of the 45 patients harboring a defined ancestral clonal evolution, numbers of subclones were two, three, and four in 32 patients (68%), 11 (24%), and 2 (4%), respectively. Mother-daughter pattern, branched pattern, and both were seen in 34 patients (76%), 5 (11%), and 6 (13%). In univariate analysis for predisposing factors of CCH acquisition, only younger age was significantly associated with CCH acquisition (48% in age <= 50 vs. 29% in age > 50; p = 0.04), but not karyotype. This statistical significance was confirmed with multivariate analysis (odds ratio = 0.44; p = 0.03). When investigating the prognostic impact of CCH acquisition, patients were divided into Ph-negative or Ph-positive ALL groups. In the 64 Ph-negative ALL patients, the CR rates were not significantly different between patients with or without CCH (78% vs. 78%, respectively; p = 1.00). The OS rates were similar between two groups (26% vs. 39% at five years, respectively; p = 0.56). Multivariate analysis for OS revealed that complex karyotype and t(8;14) were independent prognostic factors, but not CCH acquisition. Likewise, in the 56 Ph-positive ALL patients, CCH acquisition was not significantly associated with the CR rates (92% vs. 78%, respectively; p = 0.27), and the OS rates did not significantly differ between the two groups (34% vs. 40% at five years, respectively; p = 0.90). In multivariate analysis for OS, no independent prognostic factor was identified. Conclusion: Adult ALL patients with abnormal karyotype acquired CCH at diagnosis with a frequency comparable to that of AML patients. However, unlike AML patients, CCH acquisition was more frequently observed in younger population and did not show any prognostic impact in ALL patients. These findings suggested that biological backgrounds of CCH acquisition at diagnosis were possibly different between in patients with ALL and AML. So, to confirm these important findings, clinical studies with larger study subjects are warranted.