Bcr-Abl Kinase Domain Mutations in Imatinib and in Second-Generation Tyrosine Kinase Inhibitor Eras: Seven Years of Mutation Analysis, a Report by the GIMEMA CML Working Party

Abstract 2279

Over the years, Bcr-Abl kinase domain (KD) mutation analysis has been more and more extensively applied in Philadelphia-positive (Ph+) patients (pts) resistant to tyrosine kinase inhibitors (TKIs) to assist clinicians in therapeutic decisions. We reviewed the database recording the results of mutation analyses performed in our laboratory from January 2004 to June 2010. Overall, 2996 Bcr-Abl KD mutation screening tests were successfully performed by denaturing-high performance liquid chromatography (D-HPLC) and/or direct sequencing of the Bcr-Abl KD (residues 206–524). The total pts analyzed were 1139 (CML, n=1005; Ph+ ALL, n=134); the number of tests per patient ranged from 1 to 14. One hundred and ninety-one tests on 148 pts were performed in 2004, 391 tests on 214 pts in 2005, 469 tests on 217 pts in 2006, 521 tests on 241 in 2007, 536 tests on 301 pts in 2008 and 576 tests on 311pts in 2009. Overall, 869/2996 tests (29%) yielded a positive result. In 91/869 (10.5%) cases, D-HPLC showed evidence of sequence variations below the lower detection limit of sequencing; in the remaining 778 cases mutations could be characterized – a single mutation was detected in 646 (83%) cases, 2 mutations in 95 (12%) cases, 3 mutations in 25 (3%) cases, 4 or more mutations in 12 (2%) cases. In those pts for whom a longitudinal analysis was performed, 2 or more mutations accumulated as a consequence of multiple lines of therapy in 69% of cases, while in the remaining cases they concomitantly emerged under the same TKI. In 8 (10%) cases, small insertions or deletions were detected (Δ248-274 in 4 cases). Silent mutations were detected in 32 cases, either alone (11 cases) or in association with missense mutations. The K247R polymorphism was detected in 3 pts only. Five hundred and seventy-seven pts were analyzed at the time of resistance to imatinib (IM), 94 at the time of resistance to a 2^nd TKI, 34 at the time of resistance to a 3^rd TKI. In the IM-resistant setting, the ten most frequently detected mutations were F359V (13.4% of pts), M351T (12.3%), M244V (12.3%), H396R (10.3%), G250E (8.2%), E355G/D (8.2%), E255K/V (6.1%), Y253F/H (6.1%), T315I (4.1%), F317L (4.1%) in chronic phase (CP) CML pts; T315I (15.6%), G250E (13.7%), M351T (9.8%), E255K (7.8%), Y253F/H (7.8%), M244V (7.8%), Q252R/H (5.8%), H396R/P (5.8%), L384M (3.9%), F359V (3.9%) in myeloid blast crisis (BC) CML pts; and E255K/V (16.6%), T315I (16.6%), Y253F/H (15.2%), G250E (12.5%), M244V (8.3%), Q252R/H (5.5%), M351T (4.1%), L248V (2.7%), F359V (2.7%), D276G (2.7%) in lymphoid BC CML/Ph+ ALL pts. In the CP CML setting, 102 pts were analyzed because of strictly defined failure to 1^st-line IM according to the 2006 ELN recommendations. Thirty-two (31%) were positive for mutations; only one had a T315I. Ninety-nine out of 128 (77%) CML and Ph+ ALL pts who were reported to be resistant to a 2^nd or a 3^rd TKI (either nilotinib or dasatinib) were positive for one or more mutations. The ten most frequent ones were T315I (30.3% of pts), F317L (16.2%), Y253H (16.2%), F359V (7.1%), V299L (7.1%), E255K (6.1%), E255V (5.1%), F359I (4%), T315A (3%), F359C (2%) – either alone (56% of pts), combined (29%), or together with other mutations (15%). Preferential associations between mutations were observed. Eighty-five CP CML pts on IM were analyzed because of increasing Bcr-Abl transcript levels, including 61 pts who experienced ≥1-log increase without loss of major molecular response (MMR) and 24 pts who experienced ≥1-log increase leading to loss of MMR (but not of complete cytogenetic response). Mutations were identified in 2/61 (3%) and 3/24 (12.5%) pts, respectively. Forty-four CP CML pts (Low Sokal, n=14; Intermediate Sokal, n=15; High Sokal, n=15) were screened for mutations at the time of diagnosis, including 21 pts who later relapsed with evidence of mutations. Only 1 High Sokal risk patient scored positive for a mutation at diagnosis (Y342C); at the time of relapse, however, the mutation had disappeared and an M244V was instead detectable. Fifty-five Ph+ ALL pts were analyzed at the time of diagnosis. D-HPLC showed evidence of mutations in 3 (5%) pts, but they were all below the lower detection limit of sequencing. All 3 pts later relapsed with KD mutations (T315I). Additional sub-analyses will be presented. Our seven-year experience in a large series of pts sheds further light on the frequency and clinical relevance of Bcr-Abl KD mutations in the IM and in the 2^nd generation TKI era. Supported by ELN, AIL and PRIN.