Major Route Additional Chromosomal Aberrations (ACA) Precede Increase of Blasts in CML: An Analysis of the German CML-Studies III and IIIA

During the course of chronic myeloid leukemia (CML) progression to blast crisis (BC) is thought to be caused by genetic instability such as cytogenetic aberrations in addition to the translocation t(9;22)(q34;q11). We have shown previously that major route ACA indicate an unfavorable outcome (Fabarius et al., Blood 2011). We now investigate whether there is a correlation in time between appearance of major route ACA and increase in blast count.

Methods: Cytogenetic data and blast count in the peripheral blood were available from 1,290 CML patients recruited to the German CML-studies III (621 patients) and IIIa (669 patients) from January 1995 to January 2004. Treatments were interferon-alpha-based or related allogeneic stem cell transplantation (HSCT). Presence of ACA and major route ACA was considered as a time-dependent covariate. Multivariate proportional hazards models were estimated taking Euro CML score, study III vs. IIIa and stem cell transplantability into account. Cumulative incidences of blast increases were calculated starting at the date of the first ACA or major route ACA, respectively, regarding death as a competing risk. Patients were censored at the date of HSCT with an unrelated donor.

Results: 1,287 patients were evaluable with median observation times of 13 and 12 years and a 10-year survival of 48% and 61% in CML studies III and IIIa, respectively. 258 patients progressed to BC with a cumulative 10-year incidence of 20%. 195 patients displayed ACA during the course of disease. 45 patients (15.7%) showed ACA already at diagnosis. 44 patients showed unbalanced minor route, 29 balanced minor route aberrations, 23 -Y. 109 patients showed major route aberrations including 10 with other prior ACA.

In a multivariate analysis on 1,257 patients, patients with ACA had a hazard ratio (HR) for a blast increase of between 2.0-2.2 (p<0.001) for blast increases to ≥1%, ≥5%, ≥10%, ≥15%, ≥ 20% and ≥30% compared with patients without ACA (Table). When the same model was performed for major route ACA only at any time during disease, HRs of 2.2-2.7 (p<0.001) were found. For ACA without major route ACA HRs were 1.6-2.1 (p<0.001). In the multivariate analyses of major route ACA vs. no major route ACA a blast increase of 1-5% after diagnosis of major route ACA seems already indicative of progression.

5 years after the diagnosis of any ACA the cumulative incidence for a blast increase was 30% (95%- confidence interval (CI): 23-38%), of a major route ACA 40% (95%- CI: 28-49%). The 6-year probability of death without blast increase was 10%. 14 additional patients received an unrelated transplant of which 6 died.

We conclude that ACA, particularly major route ACA, precede an increase of blasts. Major route ACA have to be considered as a prognostic indicator for disease progression at any time.

 

Blast increase to	HR (univariate): ACA vs. no ACA	HR(multivariate)*: ACA vs. no ACA	HR (univariate): major route ACA vs. no major route ACA	HR (multivariate)*: major route ACA vs. no major route ACA
≥30%	2.409	2.139	2.646	2.203
≥20%	2.413	2.144	2.656	2.211
≥15%	2.415	2.161	2.868	2.426
≥10%	2.416	2.160	2.799	2.357
≥5%	2.286	2.047	2.719	2.278
≥1%	2.209	1.999	3.171	2.684

*adjusted to Euro-Score, study (III vs. IIIa) and transplantability