Ph+ ALL: resistance seeds sown early

Timothy Hughes and Susan Branford

Pfeifer and colleagues establish for the first time that BCR-ABL kinase domain mutations that are detectable at very low levels in some imatinib-naive patients with Ph+ ALL emerge as the predominant clones at relapse.

The ABL tyrosine kinase inhibitor imatinib has improved the complete remission rate for patients with Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ ALL). However, imatinib resistance frequently leads to early relapse, mainly due to the outgrowth of leukemic clones with kinase domain mutations. Such mutations are also a common cause of imatinib resistance in chronic myeloid leukemia (CML), mainly for patients in the acute phase. One key question is whether the mutant clones that lead to relapse are already present prior to imatinib therapy, or whether they arise under the selective pressure of imatinib. Detection prior to therapy in this and other studies using sensitive molecular techniques including denaturing high-performance liquid chromatography and allele-specific polymerase chain reaction suggests that these mutant clones have already significantly expanded. This is consistent with recent studies suggesting that some of these mutants have a selective advantage over the native BCR-ABL clone, and that this is not imatinib-dependent, perhaps because they have a greater transforming potency and proliferative capacity than native BCR-ABL.1,2 This has been demonstrated most convincingly for selected mutants including the P-loop mutants Y253F and E255K.1 Consistent with this observation, in 10 of the 12 Pfeifer study patients with a pre-imatinib mutation, the mutation was located within the P-loop. In the other 2 cases, the T315I mutation was detected, which has also been shown to have an increase in potency relative to unmutated BCR-ABL.2 What is harder to explain is the apparent inability of these mutant clones with greater transforming potential to become predominant prior to the selective pressure of imatinib. Their capacity to rapidly expand at the time of relapse, after a period of apparent dormancy, suggests that they may need to acquire additional genetic changes before they are capable of sufficient expansion to cause relapse.

Previous studies in late chronic and advanced-phase CML have provided little evidence that the identification of a mutant clone prior to imatinib using these highly sensitive assays is predictive of the emergence of that mutant at the time of relapse.3 This study is the first to find a close linkage in patients with the more aggressive disease, Ph+ ALL. Furthermore, in this population early mutant selection and outgrowth was evident, with mutations detectable in 50% of patients tested after 4 weeks of imatinib treatment.

If confirmed in subsequent studies, sensitive screening in Ph+ ALL patients prior to therapy or in the early phases of therapy, followed by adjusted treatment according to the mutant profile, may be an approach worthy of further assessment. For instance, patients with evidence of low levels of the highly imatinib-resistant E255K mutation might be more effectively treated with a second-generation kinase inhibitor with greater activity against this mutant clone or rapid procession to allograft. It is also worth noting in this study that several patients with no detectable mutation before imatinib treatment relapsed with predominant mutations. Consequently, a negative finding is no guarantee that mutations will not emerge and should not be used to justify less intensive therapy or an early cessation of therapy.


  • Conflict-of-interest disclosure: The authors declare that they receive research funding and honoraria from Novartis and Bristol-Myers Squibb. ■