Sudden blastic transformation in patients with chronic myeloid leukemia treated with imatinib mesylate

Philadelphia chromosome (Ph)–positive chronic myeloid leukemia (CML) has a bi- or triphasic clinical course that includes chronic, accelerated, and blastic phases.¹  Imatinib is first-line therapy for CML, with 80% to 90% achieving a complete cytogenetic response (CGCR) and 60% to 70% achieving a major molecular response.^2,3  Patients who do not respond to therapy will ultimately progress to blast phase, usually going through the accelerated phase or a poorly controlled chronic phase.⁴  Sudden blastic transformation (SBT; blastic transformation occurring unexpectedly in patients in complete hematologic remission [CHR]) has been reported after interferon-α (IFN-α) therapy at a rate of 0.5% to 2.5% in the first 3 years.⁵  There is a little information on the occurrence of SBT among patients treated with imatinib. Two isolated instances of SBT have been reported among patients in CGCR after imatinib therapy.^6,7  Herein we report 4 patients who developed SBT after being in CGCR achieved with imatinib. This is the first series to report the frequency and clinical characteristics of SBT after imatinib.

All 541 patients with Ph-positive CML in chronic phase (CP) treated with imatinib at M. D. Anderson Cancer Center since 1999 were reviewed. Approval was obtained from the M. D. Anderson Cancer Center institutional review board for these studies. Informed consent was provided according to the Declaration of Helsinki. Patients were followed with complete blood counts (CBCs) every 6 to 8 weeks and bone marrow aspiration and cytogenetic analysis every 3 months for the first 12 months, and then every 6 months. A complete hematologic remission (CHR) was defined as a white blood cell count of less than 10 × 10⁹/L with a normal differential platelet count less than 450 × 10⁹/L, and no splenomegaly lasting for at least 4 weeks. CHR was further categorized by the best cytogenetic response: CGCR, Ph-positive 0%; partial cytogenetic response (CGPR), Ph-positive 1% to 34%; and minor, Ph-positive 35% to 95%. Major molecular response (MMR) was defined as a BCR-ABL/ABL ratio of less than 0.05%.

Blastic-phase CML was defined by the presence of 30% peripheral or bone marrow blasts or presence of extramedullary disease, and categorized as lymphoid or myeloid by immunohistochemistry and flow cytometry.⁴  A sudden onset of blastic-phase CML was defined as having its onset after a documented CGCR in the immediately preceding bone marrow analysis and within 3 months of a normal CBC.

After a median follow-up of 46 months (range, 7-54 months), 23 of the 541 patients analyzed (4%; 95% confidence interval [CI] 0.01, 0.04) have developed a blast phase. Four of them (17%; 0.7% of all patients, 95% CI 0.00, 0.02) developed SBT. The clinical features of these 4 patients are presented in Table 1.

A 50-year-old woman was initially treated with IFN-α. After loss of major cytogenetic response (MCR), she started imatinib 400 mg daily. She rapidly achieved a CGCR and molecular remission. The patient remained in CGCR for 15 months when she developed lymphoid SBT. The patient achieved a CHR, CGCR, and major molecular response with hyperfractionated cyclophosphamide, vincristine, adriamycin, and dexamethasone (HyperCVAD) and imatinib. She received an allogeneic stem cell transplant and achieved a complete molecular remission. Five months later BCR-ABL transcript levels increased to 0.76% while still in CGCR. Two months later she had a second SBT.

A 54-year-old woman received imatinib 400 mg daily as first-line therapy for CP CML. She achieved CGCR 3 months later but low levels of BCR-ABL transcripts persisted (nadir ratio, 0.661%). Nine months after achieving CGCR the patient had a myeloid SBT. She received acute myeloid leukemia (AML)–type induction chemotherapy followed by a matched unrelated donor transplantation that induced a CGCR.

A 24-year-old man with CP CML started therapy with imatinib 800 mg daily. He achieved a CGCR within 6 months of start of therapy with a nadir BCR-ABL/ABL ratio of 0.0012%. Subsequent evaluations showed progressive increase of transcript levels but remained in CGCR. After 12 months in CGCR he developed lymphoid SBT. The patient received therapy with HyperCVAD plus imatinib and rituximab, achieving a complete molecular response. He received a stem cell transplant from an unrelated donor and has sustained a CGCR and complete molecular remission for 8 months.

A 58-year-old man with CP CML started therapy with 800 mg imatinib daily. He achieved a CGCR after 3 months and was sustained for 22 months. The best BCR-ABL/ABL ratio achieved was 0.8656% and rose progressively to 9.837% (still in CGCR). He then presented with myeloid SBT. He received therapy with AMN 107 (Novartis, Basel, Switzerland) and achieved a transient minor cytogenetic response always with increased blasts.

Pretreatment characteristics for these patients did not reveal significant high-risk features. We then compared whether there were any clinical features that differentiated the 4 patients with SBT from those with a more gradual transformation (Table 2). While all 4 patients achieved CGCR, none of the other 19 achieved CGCR. The median time from diagnosis to treatment was shorter for patients with SBT (2.3 vs 73.2 months). Prior IFN-α treatment was associated with gradual blastic transformation, while clonal evolution (CE) was more frequent in patients with SBT. Patients with SBT had better response to treatment, results similar to what has been observed for SBT after IFN-α.⁷ 

SBT was reported at a rate of 2.2% in patients in CHR after IFN-α.⁵  SBT occurring in patients in CGCR has been reported rarely,^6-8  including 2 patients following treatment with imatinib.^6,7  SBT as defined in this report occurred in 4 (2.7%) of 145 patients in CGCR after IFN-α therapy (3 lymphoid, 1 undifferentiated).⁵  SBT has also been reported in 5 patients after SCT,⁹  and, interestingly, 1 of our patients had a second SBT after SCT. These instances could suggest the selection of an aggressive subclone of Ph-positive cells with a proliferative advantage over the normal hematopoietic cells. It was reported that primitive Ph-positive cells may be insensitive to imatinib and persist despite an apparent optimal clinical response.^10,11  Combination therapies may be required to eliminate these early progenitors and prevent emergence of resistant clones.

Although achieving an MMR is associated with improved progression-free survival,^2,12  2 patients in this series had an SBT after achieving an MMR. The 2 other patients did not achieve a MMR and were having an increase in BCR-ABL transcript levels before the SBT. Interestingly, we found no mutations in abl kinase domain, while all 4 patients developed clonal evolution at the time of transformation. Clonal evolution was also identified at the time of SBT in the 1 patient previously reported evaluable for cytogenetics;⁷  1 of the 2 previously reported patients was assessed for mutations, and none was identified.⁶  Thus, the mechanism of resistance in these patients may be less dependent on abl kinase mutations than among patients with a more “gradual” failure to imatinib.

Finally, the 2 patients with SBT after imatinib previously reported and most of those reported after IFN-α had lymphoid phenotype,^6,7  while 2 of our patients had a myeloid phenotype. Interestingly, all had abnormalities in chromosome 8 at the time of transformation. To investigate whether these abnormalities may have been present before the time of SBT, we did fluorescence in situ hybridization (FISH) for chromosome 8 in the bone marrow immediately preceding the transformation in 3 patients with analyzable samples. None had abnormal signals for chromosome 8.

Although the numbers are small, SBT appears to be an uncommon event after imatinib, possibly less common than after IFN-α. Still, the occurrence of these events underscores the need to develop strategies for detecting and eliminating minimal residual disease.