Acquired resistance to imatinib is an important issue in CML. It was correlated to mutations in the tyrosine kinase domain (TKD) of the BCR-ABL fusion gene in 30–80% of cases. However, the mechanisms of resistance in cases without TKD mutations are still insufficiently understood. One potential further mechanism are additional chromosomal abnormalities (ACA) secondary to the Philadelphia translocation. We hypothezised that patients without TKD mutations might have a higher incidence of ACAs. Thus the aim of our analysis was to evaluate correlations between these two genetic mechanisms. In total 88 CML patients with acquired imatinib resistance were analyzed in parallel for additional chromosomal abnormalities by karyotyping and for resistance mutations in the TKD of the BCR-ABL fusion gene by sequence analysis. In 40 of the 88 pts (45.5%) at least one resistance mutation was detected. Of the 40 mutated cases 6 (15%) revealed two different resistance mutations whereas in 48 pts (54.5%) no mutation was detected with a sensitivity of 10–20% that is typical for conventional sequencing techniques. In the group not affected by TKD mutations 15 of 48 (31.3%) cases showed ACAs (−Y: n=2, t(3q26): n=4; +8: n=4; +Ph: n=5; del(17p): n=2; nonrecurrent reciprocal translocations: n=4; complex aberrant: n=1; others: n=2). 8 of these 15 cases had more than one ACA (median 3; range 2–7). In the group with resistance mutations a nearly equal number of 15/40 (37.5%) cases revealed ACAs and the spectrum of aberrations was very similar: (−Y: n=1, t(3q26): n=3; +8: n=5; +Ph: n=3; del(17p): n=2; complex aberrant: n=2; others: n=5). In this group 7 cases had more than one ACA (median:3; range: 2–3). Thus the total amount of pts with ACAs, the spectrum of ACA as well as the number of ACAs per patient is nearly equal in the mutated and the unmutated cohort. In the 40 TKD mutated cases 16 different mutations were detected (M244V: n=2 ; L248V: N=1; G250E: n=6; Y253H: n=3; E255K: n=2; D276G: n=1; E279K: n=2; L298V: n=1; L298V: n=1; F311I: n=1; F311L: n=1; T315I: n=11; F317L: n=1; F359C: n=3; F359I: n=1; M351T: n=6; H396R: n=4;). Most of the recurrent mutations are distributed equally in the groups with or without ACAs. Solely in cases with M244V, G250E or Y253H we never observed ACAs. Thus among 12 cases within the region of amino acid 244–253 there is no case with any ACA indicating that mutations within this region are strong enough to cause high resistance without further need of any ACA or alternatively are caused by different mechanisms that are not based on further genetic instability. In comparison 5 of 11 cases with T315I and 3 of 4 with H396R mutations are associated with multiple or complex chromosomal aberrations (p=0.014). Five pts were analyzed with both methods at 2–4 time points under dose escalation of imatinib. Two cases had stable G250E mutations without ACA for 2 and 10 months. One pt first revealed a T315I mutation and developed two reciprocal tranlocations t(5;10) (q31;q24) and t(6;17)(q26;q12) in addition to the T315I within 6 months. A second pt first had an E279K mutation and within one year developed an additional Y253H mutation and a +8. A third pt first had a t(3;21), after 2 months developed an additional T315I, and after further 5 months an additional F359C mutation. These few cases show that ACAs can preceed TDK mutations and vice versa. Although evolvement of resistance mainly is believed to be a process of selection some degree of genetic instability should be underlying to create some genetic diversity as a pool for any selection mechanism. Mutations within the TP53 gene have been associated with genetic instability and complex aberrant karyotypes in AML as well as in CLL. To analyze whether TP53 may also be involved in the genetic instability that leads to imatinib resistance we sequenced exons 4–9 of TP53 of 16 TKD mutated cases from this cohort (8 with complex aberrations and 8 without ACAs). No TP53 mutation was detected in all 16 cases indicating that TP53 is unlikely a major player in the development of resistance to imatinib. In conclusion,

  1. ACA are equally distributed between TKD mutated and unmutated cases.

  2. amino acid exchanges in the region 244–253 are not observed together with ACA whereas T315I and F359C are associated with multiple aberrations.

  3. TP53 mutations probably do not play a role in the development of genetic changes that are associated with imatinib resistance.

Disclosures: Schnittger:MLL Munich Leukemia Lab: Employment, Equity Ownership; Novartis: Research Funding. Dicker:MLL Munich Leukemia Lab: Employment. Kern:MLL Munich Leukemia Lab: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Lab: Employment, Equity Ownership; Novartis: Honoraria, Research Funding. Haferlach:MLL Munich Leukemia Lab: Employment, Equity Ownership.

Author notes

Corresponding author