Researchers Identify Cancer Gene Mutations Associated with Poor Outcomes in CML

Despite the good outcomes experienced by most patients with chronic-phase chronic myeloid leukemia (CP-CML) treated with tyrosine kinase inhibitors (TKIs), a subset do not respond to therapy. A next-generation sequencing study published in Blood suggests that patients with CML and consistently poor treatment outcomes are more likely to have mutations that predated the leukemia-initiating BCR-ABL1 fusion. Based on these findings, identification of a mutated cancer gene at CML diagnosis, such as the ASXL1 gene, may help clinicians stratify patients based on their risk of treatment failure.

“Our study has identified a broader spectrum of mutated cancer-associated genes than previously described in CML and included novel gene fusions and copy number variation,” study co-author Susan Branford, PhD, of the University of South Australia, told ASH Clinical News. “Of interest, BCR-ABL1 kinase domain mutations, which are the major known mechanism of TKI resistance, rarely occurred as the sole mutated cancer gene in patients with poor outcome.”

The analysis included 65 patients (median age = 49 years; range = 14-82 years) with CML who underwent either DNA wholeexome sequencing (WES) or RNA whole-transcriptome sequencing (RNA-seq) to identify mutations at diagnosis and at blast crisis.

“We included diagnosis samples of patients at the extremes of TKI response,” Dr. Branford explained, defined as:

  • optimal response, or durable major molecular response (MMR) for a median of 3 months (range = 3-55 months; n=19)
  • poor response, or transformation to blast crisis after a median of 6 months (range = 1-60 months) and/or TKI treatment failure by 18 months (n=27)

Forty-six patients with CP-CML who were treated with firstline TKI were sequenced at diagnosis; 39 patients eventually developed blast crisis. Twenty of these patients also underwent sequencing at diagnosis.

At diagnosis, clinically relevant or possibly relevant variants were observed in 23 of 46 patients (50%). Most of these occurred in cancer genes and were detected more frequently in patients with a poor outcome, compared with patients with subsequent MMR: 19/27 and 4/19, respectively (70% vs. 21%; p=0.002).

In patients with poor outcome, blast crisis occurred earlier in those with clinically relevant variants at diagnosis, compared with patients without these variants (5 months vs. 15 months, respectively; p=0.018).

The most frequently mutated gene at diagnosis was ASXL1, which occurred in nine patients. Six of these patients eventually developed blast crisis; however, they also had a significantly longer time to blast crisis than patients with other mutated genes at diagnosis (median = 21 months vs. 4.5 months [ranges not reported]; p=0.037). “Furthermore, two patients with ASXL1 variants … achieved an MMR at three months of firstline imatinib,” the authors reported. “The variants were not detected at remission, suggesting they were present in a leukemic clone and were reduced or eradicated with effective imatinib treatment.”

Additional mutated genes at diagnosis of CP-CML included the following cancer genes: RUNX1, TP53, and IKZF1.

Approximately 30 percent of patients in the poor-response group who were treated with firstline TKI had primary resistance, defined as failure to achieve a major cytogenetic response. All of these patients had clinically relevant variants at diagnosis, compared with two-thirds of those who achieved a major cytogenetic response (n=6/9) and one-third of those with a complete cytogenetic response or MMR (n=2/6; p=0.027).

In the group of 39 patients who underwent sequencing at blast crisis, all had at least one – and up to six – variants in the following cancer genes:

  • IKZF1
  • RUNX1
  • ASXL1
  • BCORL1
  • IDH1

ABL1 was the most frequently mutated cancer gene in this group (58%). However, ABL1 mutations co-occurred with other mutated cancer genes in 89 percent of patients. These predated ABL1 mutations in 62 percent of evaluable patients.

“Many of the mutated cancer genes discovered in our cohort may be targets of emerging therapy and some patients may benefit from more potent kinase inhibition or combination therapy targeting additional pathways,” Dr. Branford concluded.

The study’s relatively small sample size limits the generalizability of its findings across diverse population groups typically seen in clinical practice. Other questions about the prognostic significance of these mutated genes, according to Dr. Branford, included, “Will some mutated cancer genes confer a higher risk of treatment failure than others? Will high clinical risk at diagnosis modify treatment response when coupled with a mutated cancer gene? Will the presence of multiple mutated cancer genes at diagnosis confer a higher risk of treatment failure? And, importantly, will treatment with more potent TKIs overcome any risk associated with mutated cancer genes at diagnosis?” However, she noted, “the incorporation of additional mutation testing of specific genes at diagnosis or at treatment failure may guide treatment decisions.”

The authors report financial relationships with Novartis, Bristol-Myers Squibb, Qiagen, Cepheid, Ariad, and Celgene.


Branford S, Wang P, Yeung DT, et al. Integrative genomic analysis reveals cancer-associated mutations at diagnosis of CML in patients with high risk disease. Blood. 2018 July 2. [Epub ahead of print]