The type, frequency and timing of ACA emergence predicts response to tyrosine kinase inhibitor (TKI) treatment and overall survival, say Shimin Hu, from the University of Texas MD Anderson Cancer Center in Houston, USA, and co-authors.

 

Karyotyping revealed ACAs in 30% of 2013 CML patients, with sufficient information to determine exact time of ACA emergence in 88% of these cases. At time of initial emergence, over half (54%) of the patients had just one ACA, while 20% had two and 26% had more than two ACAs. 

 

Patients with two or more than two ACAs had comparable overall survival (OS), and both had significantly poorer OS than those with one ACA. And all patients with at least one ACA had significantly poorer OS than patients without any additional aberrations. 

 

Trisomy 8 and a second Philadelphia chromosome were the most common abnormalities in the 292 patients with just one ACA, each occurring in 12% of cases. The Y chromosome was deleted in 10% (15% of 188 males), and 3q26.2, i(17)(q10) and -7/del (7q) rearrangements were detected in 8%, 6% and 5% of patients, respectively. 

 

The remaining 47% of chromosomal changes were made up of mutations occurring in small frequencies, such as +21 in almost 2% of patients and trisomy 19 in 1%. 

 

Analysis of patients who received TKI treatment after the emergence of one of the six most common ACAs revealed that those with trisomy 8, –Y and an extra Philadelphia chromosome – group 1 – had a “relatively good response”, with 55–65% achieving a complete cytogenetic response (CCyR) and 50–62% achieving a major molecular response (MMR). 

 

By comparison, group 2 patients who had 3q26.2, i(17)(q10) or –7/del (7q) alterations had a “relatively poor response” with CCyR and MMR rates of 0–25% and 0–7%, respectively.

 

Similarly, group 1 patients had better OS than their group 2 counterparts. Patients with other ACAs had heterogeneous survival outcomes, but as a group was comparable to patients with an extra Philadelphia chromosome, the researchers say.

 

Patients with –Y had comparable survival to those without ACAs, whereas the presence of trisomy 8 or an extra Philadelphia chromosome was associated with worse OS. And the survival of patients with two or more ACAs was closer to that of group 2 patients with a single ACA than group 1 individuals, the researchers add. 

 

Further analysis indicated that ACAs in group 1 patients tended to emerge during chronic phase while group 2 ACAs appeared for the first time during accelerated phase or blast phase. 

 

“Although the number of patients who developed ACAs in [accelerated phase] and [blast phase] is relatively small, these patients appeared to have a very poor CCyR and MMR”, Hu et al observe. 

 

After adjusting for timing of ACAs, the survival difference between group 1 and group 2 was no longer significant “presumably due to the presence of other risk factors associated with [accelerated phase] and [blast phase], such as increased blasts which contributed to the prognosis”, they write. 

 

In addition, group 1 patients with ACAs at CML diagnosis had better treatment responses than those whose ACAs appeared during the course of treatment, and comparable survival to patients without ACAs. By contrast, group 2 patient response was poor regardless of timing and these individuals had poorer survival than both group 1 patients and those without ACAs. 

 

“In contrast to the poor prognostic group, ACAs in the good prognostic group showed no adverse impact on survival when they emerged from chronic phase (in other words, no other concurrent [accelerated phase] features) or at the initial CML diagnosis”, the researchers summarise. 

 

“This classification system may be useful to guide the clinical management and assess prognosis of CML patients who develop ACAs as a manifestation of clonal evolution.”

 

By Lynda Williams, Senior medwireNews Reporter

Free abstract

{rscomments off}

 

medwireNews is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2016