Drug Resistance and Bcr-Abl Kinase Domain Mutations In Philadelphia-Positive Acute Lymphoblastic Leukemia From the Imatinib to the 2^nd-Generation Tyrosine Kinase Inhibitor Era: The Main Changes Are In the Type of Mutations, but Not In the Frequency of Mutation Involvement

Abstract 575

Incorporation of the tyrosine kinase inhibitor (TKI) imatinib in the frontline treatment of Philadelphia chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL) patients (pts) has significantly improved the anti-leukemic efficacy of induction therapy. In contrast to chronic myeloid leukemia (CML), however, responses are short-lived and relapse is frequently associated with the selection of Bcr-Abl kinase domain (KD) mutations, fostered by the high genetic instability of Ph+ ALL cells. The advent of the 2nd-generation TKIs dasatinib and nilotinib has brought additional treatment options both for newly diagnosed and for imatinib-resistant pts. To analyze the changes they have brought in mutation frequency and type, we have reviewed the database recording the results of BCR-ABL KD mutation analyses done in our laboratory from January 2004 to June 2011. Overall, 781 tests on 258 Ph+ ALL pts (number of tests per pt, range: 1–15) were performed by denaturing high-performance liquid chromatography (D-HPLC) followed by direct sequencing of D-HPLC-positive cases.

One hundred and fourty-three pts were analyzed because of imatinib resistance. One hundred and one out of 143 (71%) pts scored positive for one or more KD mutations. Similarly to what is know to occur in CML, hematologic and cytogenetic resistance were by far more frequently associated with mutations than molecular resistance (Bcr-Abl transcript increase as assessed by RT-Q-PCR). Overall, mutations at thirteen residues were detected. In contrast to what can be observed in CML, three mutations were by far the most frequent, accounting for almost 75% of the mutated cases: T315I (n=38 pts, 37%), E255K/V (n=19 pts, 18%) and Y253H (n=19 pts, 18%). The other mutations were, in order of frequency: F359V/I, M244V, M351T, F317L, G250E, Q252H, L387M, D276G, L248R, E279K. Nine out of 103 (9%) pts had two mutations, in the same (2 pts) or in different (7 pts) subclones.

In 84 pts who were analyzed because they were reported to have developed resistance to dasatinib (n=72) or nilotinib (n=12) as 2^nd- or 3^rd-line TKIs, 65 (77%) had newly acquired mutations (57/72 dasatinib-resistant pts and 8/12 nilotinib-resistant pts). The most frequent newly acquired mutation in this setting was the T315I, detected in 35/57 (61%) cases acquiring mutations on dasatinib and in 2/8 cases acquiring mutations on nilotinib. Other recurrent newly acquired mutations were F317L, V299L, T315A in dasatinib-resistant pts and Y253H and E255K in nilotinib-resistant pts. Thirty out of 65 pts (46%) were positive for multiple mutations (2 to 4 mutations, in the same or in different subclones or both) that emerged under the same TKI in 11 cases (37%) and accumulated as a consequence of multiple lines of TKI therapy in the remaining 19 (63%) cases.

Mutation analysis was also performed in 15 resistant pts enrolled in a clinical trial of dasatinib as first-line treatment for Ph+ ALL. Twelve pts were positive for mutations; 11/12 had a T315I.

Sixty-one pts were analyzed at the time of diagnosis in order to assess whether TKI-resistant mutations could already be detectable. Only two pts (3%) were positive for mutations: one patient had an F311L that disappeared after one month of nilotinib treatment; an additional patient was positive only by D-HPLC, but not by the less sensitive direct sequencing – most likely for the T315I mutation that shortly after the start of dasatinib treatment outgrew and led to resistance. Taking advantage of the recent availability of a next-generation sequencing platform (Roche 454), allowing high sensitivity (0.01%) mutation scanning of the KD, samples collected at the time of diagnosis and during follow-up from selected Ph+ ALL cases who developed mutations and resistance to TKI therapy were retrospectively analyzed – but the mutations were not always already detectable at diagnosis.

In conclusion: a) although 2^nd generation TKIs may ensure a more rapid debulking of the neoplastic clone and have much fewer insensitive mutations, long-term disease control remains a problem and the T315I becomes an even tougher enemy; b) the clinical usefulness of mutation screening of Ph+ ALL pts at diagnosis before TKI start, even with highly sensitive approaches is low – not all mutations pre-exist since genetic instability remains high and fosters mutational events anytime during treatment.

Supported by PRIN, FIRB, AIL and AIRC.