Introduction

Chronic myeloid leukemia (CML) is characterized by a reciprocal translocation between chromosome 9 and 22, resulting in a BCR-ABL fusion gene. In CML, over 95% of the breakpoints involve the major breakpoint cluster region (M-bcr) of BCR introns downstream of either exon 13 or 14 and introns downstream of ABL exon 2, resulting in fusion transcripts e13a2 (b2a2) and e14a2 (b3a2). The minor breakpoint cluster region (m-bcr) in BCR gene is a less common breakpoint involved the exon 1, fused to ABL a2, resulting in e1a2 fusion transcript. Other transcripts such as e19a2, e1a3, e6a2, e13a3, and e14a3 occur less frequently.

The e1a2 transcripts exist in the most of acute lymphoblastic leukemia, and CML expressing only e1a2 transcripts is rare. Although anecdotal reports of patients with the e1a2 transcripts had a good prognosis, to our knowledge, the outcome of most of patients with this transcript alone in the era of imatinib is very poor. We performed this study to investigate the frequency of e1a2BCR-ABL CML and mutations in the kinase domain of BCR-ABL after imatinib failure, and to evaluate their treatment prognosis.

Objective

To investigate the frequency of e1a2BCR-ABL CML and mutations in the kinase domain of BCR-ABL after imatinib failure, and to evaluate their treatment prognosis.

Patients and Methods

The records of all CML patients at our hospital from January 2004 to March 2014 were reviewed to identify patients with only e1a2BCR-ABL fusion transcripts consequent to breakpoints in minor BCR. Patients with e1a2 coexisting with e13a2 and/or e14a2 were excluded from this analysis. BCR-ABL transcripts were revealed at diagnosis by quantitative PCR followed by conventional agarose electrophoresis of PCR products. Molecular follow-up of BCR-ABL transcripts throughout treatment was performed by quantitative PCR and cytogenetic follow-up was performed by karyotype analysis following the guidelines of the European Leukemia Net. BCR-ABL kinase domain point mutations have been assessed using Sanger sequencing in patients with imatinib failure and progression.

Event-free survival (EFS) was measured from the start of each therapy until progression to the BP or death from any cause during treatment. Overall survival was defined from date of CML diagnosis to date of death or last follow-up.

Results

Nineteen (0.6%) of the 3172 CML patients, during the study period had e1a2BCR-ABL CML. At the time of diagnosis, 16 patients were in CP, 2 in BP, and 1 in AP. We only retrospectively study 12 CML patients. The median follow-up since the diagnosis of CML is 35 months (range, 5-149 months).

9 of 12 patients received TKI as their initial therapy (7 imatinib, 1 nilotinib;Table 1). Among patients in CP, 6 received imatinib as frontline therapy and 3 received imatinib after Hu or in combination with IFN, and after a median follow-up of 28 months (range, 9-145 months), 2 patients had no response to imatinib, and the best response for the others was complete hematologic response (CHR) in 5, mCyR in 1 and MR4.5 in 1. Notably, only 1 of 9 patients receiving imatinib achieved MR4.5. 2 patients who was not found imatinib-resistant mutation was changed to nilotinib as second line TKI and 2 patients was changed to alloSCT. Only 1 who underwent alloSCT achieved MR4.5 and the others retained a hematologic response. 2 of the patients in CP progressed to BP (including 1 with H396R mutation) after a median of 27 months (range, 12-41 months) and died at 61 and 48 months after diagnosis, respectively.

2 patients in BP (including 1 after CDOP+IA failure) received imatinib . The two patients had no response to imatinib and 1 died after 8 months. 1 in AP received nolitinib as frontline therapy. The patient achieved MR4.5 after 3 months and returned to CP.

Overall, 8 patients (7 CP, 1 AP) were alive at a median of 26 months (range, 5-149 months) after diagnosis: 2 with >MR4.5 on TKI (1 on imatinib, 1 nilotinib) and 1 with >MR4.5 after alloSCT. Median survival was 28 months for patients in CP at the start of therapy, and 5 months for the patient in AP.

Conclusion

We conclude that the frequency of e1a2BCR-ABL CML is only 0.6%. It suggests an inferior outcome to imatinib, and the mutations in the kinase domain of BCR-ABL have been detected in 20% of imatinib-resistant patients. Perhaps nilotinib as first line treatment has a high rate of MR4.5 in newly diagnosed CML and should be considered firstly, but the role of alloSCT for these patients is irreplaceable.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution