BCR-ABL Splice Variants In Three Phases Of CML: Association With Disease Biology and Response To Imatinib

Philadelphia chromosome, a characteristic chromosomal translocation (breakpoint cluster region-Abelson [BCR–ABL]; t [9; 22] [q34; q11]) in chronic myeloid leukemia (CML) encodes a product, which is the target of tyrosine kinase inhibitors (TKIs). In most CML patients, during translocation, the break takes place within a 5.8-kb region on both sides of BCR exons12-16 (originally referred to as exons b1-b5), described as themajor breakpoint cluster region (M-bcr)¹. It can be either between exons b2 and b3, or between exons b3 and b4. The ABL exon which is mostly inserted in break points of BCR is a2. As a result of alternativesplicing, fusion transcripts with either b2a2 or b3a2 splicing variantcan be produced. Occasionally, breakpoints occur in unusual places, resulting in generation of rare BCR-ABL transcripts (b2a3, b3a3, e1a2, e8a2, e19a2). Association of BCR-ABL splice variants with clinical outcome in different phases of CML has rarely been studied². We studied BCR/ABL splice variants and their association with disease biology and response to imatinib in 3 phases of CML in 70 patients.

Total RNA was extracted and reverse transcribed. PCR primers and nested PCR protocol for the detection of BCR-ABL splice variants in CML patients were adopted from van Dongen et al., 1999 ³. Data was analyzed using SPSS Software version 17.

Median age of the patients was 34 years; 10 patients were 18 or younger and there were 40 females. 26 patients were in chronic phase (CP), 26 in accelerated phase (AP) and 18 in blast phase (BP). Overall, the frequencies of BCR/ABL splice variants b3a2, b2a2 and b3a2+b2a2 were found in 49 (70%), 15 (21.4%) and 6 (8.6%) patients, respectively. No rare BCR-ABL transcripts were identified. Percentage of BCR-ABL splice variant b2a2 was similar in patients with CP (31) and BP (28) of CML and low in AP (7.75). While the percentage of BCR-ABL splice variant b3a2 was more common in AP (84.5) and BP (72) as compared to CP (54). The combined expression of both transcripts (b3a2 and b2a2) was found in 4 patients with CP and 2 in AP of CML but none in BP of CML (Table 1). The co-existence of b2a2 and b3a2 transcripts was observed only in male patients. The BCR-ABL transcripts had no correlation with the age. Patients with the splice variant b2a2 had a better response to imatinib treatment and better survival as compared to patients with b3a2. The patients expressing both transcripts (b3a2 and b2a2) also showed better survival as compared to patients with only b3a2 transcript (figure 1).

Figure 1

View largeDownload slide

Survival Probability of Patients with different BCR-ABL splice variants

Figure 1

View largeDownload slide

Survival Probability of Patients with different BCR-ABL splice variants

Close modal

We found a higher percentage of BCR-ABL splice variant b3a2 in AP & BP as compared to CP. Patients with the BCR-ABL splice variant b2a2 had a better response to imatinib and better survival, which is in contrast to previous reports ⁴. Co-existence of two transcripts b3a2 and b2a2 was observed mostly in CP, and in some patients in AP; it was also associated with a better response to imatinib. Larger studies are needed to further define the role of BCR-ABL splice variants in different phases of CML and their association with response to TKIs and prognosis. Studies of BCR-ABL splice variants can provide further insights into CML biology and new targets for BCR-ABL positive leukemia treatment ^{1, 5}.

References

1. Adamia S, Pilarski PM, Natan MB, Stone RM, Griffin JD. Curr Cancer Drug Targets. 2013 Jul 30. [Epub ahead of print].

2. Gruber FX, Lundán T, Goll R, Silye A, Mikkola I, Rekvig OP et al. Med Oncol. 2012 Mar; 29(1):219-26.

3. van Dongen JJ, Macintyre EA, Gabert JA, Delabesse E, Rossi V, Saglio G et al. Leukemia. 1999 Dec; 13(12):1901-28.

4. Lucas CM, Harris RJ, Giannoudis A, Davies A, Knight K, Watmough SJ, et al. Haematologica. 2009 Oct;94(10):1362-7

5. Chiarella P, Summa V, De Santis S, Signori E, Picardi E, Pesole G et al. Curr Mol Med. 2012 Jun;12(5):547-65.

No relevant conflicts of interest to declare.