Mechanisms of Resistance to Imatinib and Second-GenerationTyrosine Inhibitors in Chronic Myeloid LeukemiaDragana Milojkovic1 and Jane Apperley2

Abstract Targeted therapy in the form of selective tyrosine kinase inhibitors (TKI) has trans-formed the approach to management of chronic myeloid leukemia (CML) and dramat-ically improved patient outcome to the extent that imatinib is currently accepted as thefirst-line agent for nearly all patients presenting with CML, regardless of the phase ofthe disease. Impressive clinical responses are obtained in the majority of patients inchronic phase; however, not all patients experience an optimal response to imatinib,and furthermore, the clinical response in a number of patients will not be sustained.The process by which the leukemic cells prove resistant to TKIs and the restorationof BCR-ABL1 signal transduction from previous inhibition has initiated the pursuit forthe causal mechanisms of resistance and strategies by which to surmount resistance totherapeutic intervention. ABL kinase domain mutations have been extensively implicat-ed in the pathogenesis of TKI resistance, however, it is increasingly evident that thepresence of mutations does not explain all cases of resistance and does not accountfor the failure of TKIs to eliminate minimal residual disease in patients who respondoptimally. The focus of exploring TKI resistance has expanded to include the mecha-nism by which the drug is delivered to its target and the impact of drug influx and effluxproteins on TKI bioavailability. The limitations of imatinib have inspired the develop-ment of second generation TKIs in order to overcome the effect of resistance to thisprimary therapy. (Clin Cancer Res 2009;15(24):7519–27)

Chronic myeloid leukemia (CML) results from the balancedtranslocation of c-ABL from chromosome 9 and BCR on chro-mosome 22 leading to the formation of BCR-ABL1 chimericoncoprotein, the product of the BCR-ABL1 hybrid gene, withconstitutive tyrosine kinase activity (1, 2). Deregulated BCR-ABL1 activity results in enhanced cellular proliferation, and re-sistance to apoptosis and oncogenesis (3, 4). CML naturallyprogresses through distinct phases from early chronic phaseto an intermediate accelerated phase followed by a terminalblast phase. Imatinib, the first tyrosine kinase inhibitor (TKI)approved for the treatment of CML (5), is a phenylaminopyri-dimine, which principally targets the tyrosine kinase activity ofBCR-ABL1, exclusively binding to BCR-ABL1 in the inactiveconformation in addition to inhibitory effects on KIT, ARG,and PDGFR kinases (6). The recent update of the phase III ran-domized IRIS study (International Randomized Study of Inter-feron-α plus Ara-C versus STI571) prospectively comparingimatinib with interferon-α and cytarabine in previously untreat-

ed patients in first chronic phase showed the best observed ratefor a complete cytogenetic response [CCyR; or an undetectablenumber of Philadelphia chromosome positive (Ph+) chromo-somes by conventional metaphase analysis] on imatinib of82% at 6 years (7), with a declining annual rate of progressionas the molecular response improved with time.

Clinical Resistance to TKIs

In order to best determine an individual's response to therapy,an operational set of goals, defined within specific time periodshave been established for all patients (Table 1; ref. 8). An initialrequirement is the achievement of a complete hematological re-sponse (CHR), accepted as a normal peripheral blood countwithin 3 months of imatinib. Further response to treatment issubsequently monitored by sequential cytogenetic assessmentsof the bone marrow with the aim to achieve a CCyR by 18months. Subsequent evaluation of the therapeutic response isrecommended by means of molecular analysis, with reverse-transcriptase polymerase chain reaction (RT-PCR). Patients thatachieve a major molecular response (MMR) equivalent to a re-duction in BCR-ABL1 transcripts to less than 0.1% as defined onthe international scale (9), are predicted to have a remarkablylow risk of disease progression. Within the framework of recom-mendations, proposals for the definition of failure and subopti-mal response are now recognized (8). Resistance to imatinibencompasses failure to reach CHR, CCyR, and MMR within anallocated duration of time (primary resistance). A number of

Authors' Affiliations: 1Department of Haematology, HammersmithHospital, 2Department of Haematology, Imperial College London, London,United KingdomReceived 7/24/09; revised 10/5/09; accepted 10/13/09; published online12/15/09.Requests for reprints: Jane F. Apperley, Imperial College London, Hammer-smith Campus, Du Cane Road, London, W12 0NN United Kingdom. Phone:44-0-20-8383-4017; Fax: 44-0-20-8742-9335; E-mail: j.apperley@imperial.ac.uk.

F 2009 American Association for Cancer Research.doi:10.1158/1078-0432.CCR-09-1068

7519 Clin Cancer Res 2009;15(24) December 15, 2009www.aacrjournals.org

CCR FOCUS

Research. on June 22, 2014. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

AdministratorTypewriter1

http://clincancerres.aacrjournals.org/

more rapidly yield imatinib-resistant mutant subclones thancells with low BCR-ABL1 expression levels (75). Similarly, re-sistance to nilotinib in vitro has also been found consequentto BCR-ABL1 overexpression in vitro (34).

ABL Kinase Domain Mutations

The emergence of mutations within the kinase domain ofBCR-ABL1 is regularly associated with resistance to TKI ther-apy. The most frequently described mechanism of acquiredresistance to imatinib is the occurrence of point mutations,representing a single aa substitution in the kinase domain,which impair drug binding by affecting essential residuesfor direct contact with the TKI or by preventing BCR-ABL1from assuming the inactive conformation appropriate for im-atinib binding. The published incidence of mutations remainsvariable and in the order of 40 to 90% as a consequence ofdifferent methods of detection, nature of resistance, and dis-ease phase examined (76). Mutations were first identified in2001, in which restoration of BCR-ABL1 signal transductionon imatinib therapy was associated with a T315I mutation(72). Thr315 forms a fundamental hydrogen bond with im-atinib, disrupted by a single aa change with a bulkier isoleu-cine, which prevents imatinib localization within the ATPbinding pocket by consequent stearic hindrance. The T315Imutation is one of the most frequent mutations arising inpatients on imatinib therapy, occurring between 4 to 19%of resistant cases (55, 77, 78) and is resistant to all ABL ki-nase inhibitors. Although the T315I mutation is generally ac-cepted as conferring a poor outcome (median survival 12.6months; refs. 79, 80), sustained cytogenetic responses despiteaccelerated phase and during therapy with a second TKI haverecently been reported (78).Four categories of mutations have been recognized to cor-

relate with clinical resistance to imatinib affecting the: (i) im-atinib binding site, (ii) P-loop (ATP binding site), (iii)catalytic (C) domain, and (iv) activation (A) loop (2). Muta-tions in the phosphate (P-loop; residues 244-255 of ABL),which account for up to 48% of all mutations in imatinibresistant cases (81), destabilize the conformation requiredfor imatinib binding, and have been associated with an in-creased transforming potential (82) and a worse prognosisregardless of their sensitivity to imatinib (77, 81, 83, 84).P-loop mutations have been reported to be associated witha worse prognosis in comparison with other categories ofmutations (81, 83), however, other observers have not con-firmed these findings (55), perhaps because of the nature ofthe criteria used to select patients for mutation screening. An-other potential explanation for this inconsistency may be onaccount of the M244V mutation, which may not confer apoor outcome and has been variably included in the P-loopcategories of mutations (84). A series of mutations are locat-ed in the catalytic domain (residues 350-363 of ABL) andcan also affect imatinib binding. The activation loop of theABL kinase is the major regulatory component of the kinasedomain and can adopt an open and/or active or closed and/or inactive conformation. Mutations in the activation loopinstigate the open and/or active configuration, and as the in-

active and/or closed configuration is required for imatinib ac-tivity, resistance occurs. Nevertheless, aa substitutions at onlyseven residues [M244V, G250E, Y253F/H, E255K/V (P-loop),T315I (imatinib binding site), M351T, and F359V (catalyticdomain)] account for 85% of all resistance-associated muta-tions (80).Although point mutations have been more frequently de-

scribed in TKI resistance and advanced-phase CML (Table3), they have also been documented prior to TKI therapy(85), inherently suggesting that pre-existing mutations donot acquire a survival advantage until subjected to a TKI.In addition, investigators have found no difference in muta-tional status in those patients who have relapsed (74). Therelevance of these observations remains unclear, specificallyabout whether certain mutations are responsible for diseaseprogression or whether they occur as a consequence of theunderlying genomic instability linked with advanced phasedisease (86). It would seem that gain-of-function mutationsmay independently contribute to disease progression, whereas

Table 3. Frequency of ABL-kinase domainmutations by disease phase

KDMutation

No. ofMutations*

No. of CP(%)†

No. of AP(%)*

No. of BP(%)*

P-loop‡

M244 47 33 (70) 1 (2) 13 (28)L248 13 10 (77) 2 (15) 1 (8)G250 63 31 (49) 6 (10) 26 (41)Q252 14 3 (21) 3 (21) 8 (58)Y253 68 23 (34) 9 (13) 36 (53)E255 63 17 (27) 12 (19) 34 (54)

IM binding siteD276 12 6 (50) 2 (17) 4 (33)F311 5 2 (40) 1 (20) 2 (40)T315 56 9 (16) 12 (23) 35 (63)F317 15 10 (67) 2 (13) 3 (20)

Catalytic domainM351 62 33 (53) 12 (19) 17 (28)E355 22 13 (59) 4 (18) 5 (23)F359 35 21 (60) 5 (14) 9 (26)

Activation loopH396 29 21 (72) 2 (7) 6 (21)

C-terminal lobeS417 3 2 (67) 1 (33) 0 (72)E459 6 2 (33) 0 (72) 4 (67)F486 8 0 (0) 1 (13) 7 (88)

Note: Adapted from Apperley (100).Abbreviations: KD, kinase domain; CP, chronic phase; AP, acceler-ated phase; BP, blast phase; IM, imatinib.*Number of mutations detected in a pool of patients reviewed inApperley (100). Infrequently an individual patient harbored morethan one KD mutation; any detected- mutation is included in thetable.†Percentage of all KD mutations detected related to disease phase.‡P-loop mutations have been inconsistently reported to be associ-ated with a worse prognosis in comparison with other categories ofmutations (55, 81, 83). Furthermore, the M244V mutation maynot confer a poor outcome and has been variably included in theP-loop categories of mutations (84).

7523 Clin Cancer Res 2009;15(24) December 15, 2009www.aacrjournals.org

Resistance to Imatinib and TKIs in CML

Research. on June 22, 2014. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

http://clincancerres.aacrjournals.org/

loss-of-function mutations are more often subject to selectivepressure by imatinib (82, 87). Specific mutations consider-ably affect the transformation potency of BCR-ABL1, andin vitro studies have indicated relative transformation poten-cies of mutations from distinct sections of the kinase domainto be: Y253F > E255K (P-loop) > unmutated BCR-ABL1 ≥ T315I(imatinib binding site) > H396P (activation loop) > M351T(catalytic domain; ref. 82). Particular mutations, as in thecase of E255K, are noted to have increased oncogenic potencydespite reduced kinase activity compared with unmutatedBCR-ABL1 (88). The proliferative advantage of a given mutantseems multifactorial and determined by intrinsic kinase activity,substrate specificity, and extrinsic factors including growth fac-tors and cytokines.Although most of the clinically relevant mutations are in-

hibited by dasatinib and nilotinib, with the exception ofT315I (Fig. 1; ref. 89), the presence of existing mutations af-ter imatinib failure, as well as development of new mutationson a subsequent second TKI is naturally a potential source ofresistance to successive TKI (90–93). The influence of base-line BCR-ABL1 mutations on response to nilotinib in patientswith imatinib-resistant CML in chronic phase has shown aninferior outcome in patients who harbored mutations thatwere less sensitive to nilotinib in vitro (Y253H, E255V/K,F359V/C; ref. 94). Recently, the selective pressure of sequentialTKI therapy has been assessed in the outcome of imatinib-resistant patients already harboring imatinib-resistant kinasedomain mutations subsequently treated with an alternativeTKI on a second or even third occasion and showed that 83%

of cases of relapse after an initial response were associated withthe emergence of newly acquired mutations (95). The T315Imutation was most commonly implicated with a frequency of36% (95). The inability to achieve a sustained cytogenetic re-sponse could in part be as a consequence of the developmentof new therapy-resistant kinase domain mutations as patientsare exposed to sequential TKIs, although some of the arising mu-tations were reported as having a relatively good in vitro sensitiv-ity to the concurrent TKI (96).In summary, the consequence of identifying a mutation re-

mains unclear and seems relevant only according to the dis-ease phase and response, with a greater impact in advancedphase CML in which the mutated clone may be responsiblefor disease progression, but less certain in cases of on-goingresponse to TKI therapy. Resistance mechanisms may be over-come with imatinib dose escalation (97), alternative therapywith a 2G-TKI (98) to which the mutant has documentedsensitivity, withdrawing TKI therapy to allow the mutantclone to recede (99), as well as non-BCR-ABL1-dependenttherapies.

Conclusions

Targeted molecular therapy has afforded exceptional clini-cal responses in the majority of patients with CML to theextent that therapeutic regimens have centered on theachievement of a MMR, early within the start of therapy.As most will continue on imatinib in CCyR, the emphasishas diverted to overcoming imatinib resistance and the

Fig. 1. Although equivalentexperimental systems have beenemployed in a variety of assessmentsto determine BCR-ABL1 kinase domainmutation sensitivity on the basis of IC50values (89, 98), different incubationtimes and TKI concentration rangeshave been used as well as varyingmethods to measure cell viability andproliferation. Color-coded schemes toindicate TKI sensitivity based on in vitroanalyses should be interpreted withclinical caution as in vitro findingscannot be directly extrapolated to theclinical setting. (Figure adapted fromO'Hare et al. 89, © the American Societyof Hematology).

7524Clin Cancer Res 2009;15(24) December 15, 2009 www.aacrjournals.org

CCR FOCUS

Research. on June 22, 2014. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

http://clincancerres.aacrjournals.org/

have excellent responses on either dose: a TOPScorrelative study. Blood 2008;112:3187a.

46. White DL, Saunders VA, Dang P, et al. OCT-1-mediated influx is a key determinant of the intra-cellular uptake of imatinib but not nilotinib(AMN107): reduced OCT-1 activity is the causeof low in vitro sensitivity to imatinib. Blood2006;108:697–704.

47. Dulucq S, Bouchet S, Turcq B, et al. Multidrugresistance gene (MDR1) polymorphisms are as-sociated with major molecular responsesto standard-dose imatinib in chronic myeloidleukemia. Blood 2008;112:2024–7.

48. Kim DH, Sriharsha L, Xu W, et al. Clinical rel-evance of a pharmacogenetic approach usingmultiple candidate genes to predict responseand resistance to imatinib therapy in chronicmyeloid leukemia. Clin Cancer Res 2009;15:4750–8.

49. Majlis A, Smith TL, Talpaz M, O'Brien S, RiosMB, Kantarjian HM. Significance of cytogeneticclonal evolution in chronic myelogenous leuke-mia. J Clin Oncol 1996;14:196–203.

50. Johansson B, Fioretos T, Mitelman F. Cyto-genetic and molecular genetic evolution ofchronic myeloid leukemia. Acta Haematol2002;107:76–94.

51. O'Dwyer ME, Mauro MJ, Blasdel C, et al. Clon-al evolution and lack of cytogenetic response areadverse prognostic factors for hematologic re-lapse of chronic phase CML patients treated withimatinib mesylate. Blood 2004;103:451–5.

52. Cortes JE, Talpaz M, Giles F, et al. Prognosticsignificance of cytogenetic clonal evolution inpatients with chronic myelogenous leukemia onimatinib mesylate therapy. Blood 2003;101:3794–800.

53. Cortes J, O'Dwyer ME. Clonal evolution inchronic myelogenous leukemia. Hematol OncolClin North Am 2004;18:671–84 [x.].

54. Lahaye T, Riehm B, Berger U, et al. Responseand resistance in 300 patients with BCR-ABL-positive leukemias treated with imatinib in a sin-gle center: a 4.5-year follow-up. Cancer 2005;103:1659–69.

55. Jabbour E, Kantarjian H, Jones D, et al. Fre-quency and clinical significance of BCR-ABL mu-tations in patients with chronic myeloid leukemiatreated with imatinib mesylate. Leukemia 2006;20:1767–73.

56. Geahlen RL, Handley MD, Harrison ML. Molec-ular interdiction of Src-family kinase signaling inhematopoietic cells. Oncogene 2004;23:8024–32.

57. Danhauser-Riedl S, Warmuth M, Druker BJ,Emmerich B, Hallek M. Activation of Src kinasesp53/56lyn and p59hck by p210bcr/abl in myeloidcells. Cancer Res 1996;56:3589–96.

58. Donato NJ, Wu JY, Stapley J, et al. BCR-ABLindependence and LYN kinase overexpression inchronic myelogenous leukemia cells selected forresistance to STI571. Blood 2003;101:690–8.

59. Hu Y, Swerdlow S, Duffy TM, Weinmann R,Lee FY, Li S. Targeting multiple kinase pathwaysin leukemic progenitors and stem cells isessential for improved treatment of Ph+ leuke-mia in mice. Proc Natl Acad Sci U S A 2006;103:16870–5.

60. Dai Y, Rahmani M, Corey SJ, Dent P, Grant SA.Bcr/Abl-independent, Lyn-dependent form of im-atinib mesylate (STI-571) resistance is associat-ed with altered expression of Bcl-2. J BiolChem 2004;279:34227–39.

61. Schindler T, Bornmann W, Pellicena P, MillerWT, Clarkson B, Kuriyan J. Structural mecha-nism for STI-571 inhibition of abelson tyrosinekinase. Science 2000;289:1938–42.

62. O'Hare T, Eide CA, Deininger MW. PersistentLYN signaling in imatinib-resistant, BCR-ABL-

independent chronic myelogenous leukemia. JNatl Cancer Inst 2008;100:908–9.

63. Copland M, Hamilton A, Elrick LJ, et al. Dasa-tinib (BMS-354825) targets an earlier progenitorpopulation than imatinib in primary CML butdoes not eliminate the quiescent fraction. Blood2006;107:4532–9.

64. Jiang X, Zhao Y, Smith C, et al. Chronic mye-loid leukemia stem cells possess multiple uniquefeatures of resistance to BCR-ABL targeted ther-apies. Leukemia 2007;21:926–35.

65. Jin L, Tabe Y, Konoplev S, et al. CXCR4 up-regulation by imatinib induces chronic mye-logenous leukemia (CML) cell migrationto bone marrow stroma and promotes surviv-al of quiescent CML cells. Mol Cancer Ther2008;7:48–58.

66. Konig H, Copland M, Chu S, Jove R, HolyoakeTL, Bhatia R. Effects of dasatinib on SRC ki-nase activity and downstream intracellularsignaling in primitive chronic myelogenous leu-kemia hematopoietic cells. Cancer Res 2008;68:9624–33.

67. Jorgensen HG, Allan EK, Jordanides NE,Mountford JC, Holyoake TL. Nilotinib exertsequipotent antiproliferative effects to imatiniband does not induce apoptosis in CD34+ CMLcells. Blood 2007;109:4016–9.

68. Copland M, Pellicano F, Richmond L, et al.BMS-214662 potently induces apoptosis ofchronic myeloid leukemia stem and progenitorcells and synergizes with tyrosine kinase inhibi-tors. Blood 2008;111:2843–53.

69. Holtz M, Forman SJ, Bhatia R. Growth factorstimulation reduces residual quiescent chronicmyelogenous leukemia progenitors remainingafter imatinib treatment. Cancer Res 2007;67:1113–20.

70. Copland M, Fraser AR, Harrison SJ, HolyoakeTL. Targeting the silent minority: emergingimmunotherapeutic strategies for eradicationof malignant stem cells in chronic myeloid leu-kaemia. Cancer Immunol Immunother 2005;54:297–306.

71. Jamieson CH, Ailles LE, Dylla SJ, et al.Granulocyte-macrophage progenitors as candi-date leukemic stem cells in blast-crisis CML. NEngl J Med 2004;351:657–67.

72. GorreME,MohammedM, Ellwood K, et al. Clin-ical resistance to STI-571 cancer therapy causedby BCR-ABL gene mutation or amplification.Science 2001;293:876–80.

73. Modi H, McDonald T, Chu S, Yee JK, FormanSJ, Bhatia R. Role of BCR/ABL gene-expressionlevels in determining the phenotype and imati-nib sensitivity of transformed human hemato-poietic cells. Blood 2007;109:5411–21.

74. Hochhaus A, Kreil S, Corbin AS, et al. Molecu-lar and chromosomal mechanisms of resistanceto imatinib (STI571) therapy. Leukemia 2002;16:2190–6.

75. Barnes DJ, Palaiologou D, Panousopoulou E,et al. Bcr-Abl expression levels determine the rateof development of resistance to imatinibmesylatein chronic myeloid leukemia. Cancer Res 2005;65:8912–9.

76. Quintas-Cardama A, Kantarjian HM, Cortes JE.Mechanisms of primary and secondary resistanceto imatinib in chronic myeloid leukemia. CancerControl 2009;16:122–31.

77. Nicolini FE, Corm S, Le QH, et al. Mutation sta-tus and clinical outcome of 89 imatinib mesylate-resistant chronicmyelogenous leukemia patients:a retrospective analysis from the French inter-group of CML (Fi(phi)-LMC GROUP). Leukemia2006;20:1061–6.

78. Jabbour E, Kantarjian H, Jones D, et al. Char-acteristics and outcomes of patients with chronicmyeloid leukemia and T315I mutation following

failure of imatinib mesylate therapy. Blood 2008;112:53–5.

79. Nicolini FE, Hayette S, Corm S, et al. Clinicaloutcome of 27 imatinib mesylate-resistantchronic myelogenous leukemia patients harbor-ing a T315I BCR-ABL mutation. Haematologica2007;92:1238–41.

80. Soverini S, Colarossi S, Gnani A, et al. Contribu-tion of ABL kinase domain mutations to imatinibresistance in different subsets of Philadelphia-positive patients: by the GIMEMA Working Partyon Chronic Myeloid Leukemia. Clin Cancer Res2006;12:7374–9.

81. Branford S, Rudzki Z, Walsh S, et al. Detec-tion of BCR-ABL mutations in patients withCML treated with imatinib is virtually alwaysaccompanied by clinical resistance, and muta-tions in the ATP phosphate-binding loop (P-loop) are associated with a poor prognosis.Blood 2003;102:276–83.

82. Griswold IJ, MacPartlin M, Bumm T, et al.Kinase domain mutants of Bcr-Abl exhibit al-tered transformation potency, kinase activity,and substrate utilization, irrespective of sensitiv-ity to imatinib. Mol Cell Biol 2006;26:6082–93.

83. Soverini S, Martinelli G, Rosti G, et al. ABL mu-tations in late chronic phase chronic myeloidleukemia patients with up-front cytogenetic re-sistance to imatinib are associated with a greaterlikelihood of progression to blast crisis andshorter survival: a study by the GIMEMA Work-ing Party on Chronic Myeloid Leukemia. J ClinOncol 2005;23:4100–9.

84. Khorashad JS, de Lavallade H, Apperley JF,et al. Finding of kinase domain mutations inpatients with chronic phase chronic myeloidleukemia responding to imatinib may identifythose at high risk of disease progression. J ClinOncol 2008;26:4806–13.

85. Roche-Lestienne C, Soenen-Cornu V, Grardel-Duflos N, et al. Several types of mutations of theAbl gene can be found in chronic myeloid leuke-mia patients resistant to STI571, and they canpre-exist to the onset of treatment. Blood 2002;100:1014–8.

86. Khorashad JS, Anand M, Marin D, et al. Thepresence of a BCR-ABL mutant allele in CMLdoes not always explain clinical resistance to im-atinib. Leukemia 2006;20:658–63.

87. Willis SG, Lange T, Demehri S, et al. High-sensitivity detection of BCR-ABL kinase domainmutations in imatinib-naive patients: correlationwith clonal cytogenetic evolution but not re-sponse to therapy. Blood 2005;106:2128–37.

88. Skaggs BJ, Gorre ME, Ryvkin A, et al. Phos-phorylation of the ATP-binding loop directs on-cogenicity of drug-resistant BCR-ABL mutants.Proc Natl Acad Sci U S A 2006;103:19466–71.

89. O'Hare T, Eide CA, Deininger MW. Bcr-Abl ki-nase domain mutations, drug resistance, and theroad to a cure for chronic myeloid leukemia.Blood 2007;110:2242–9.

90. Shah NP, Skaggs BJ, Branford S, et al. Se-quential ABL kinase inhibitor therapy selectsfor compound drug-resistant BCR-ABL muta-tions with altered oncogenic potency. J Clin In-vest 2007;117:2562–9.

91. Cortes J, Jabbour E, Kantarjian H, et al. Dynam-ics of BCR-ABL kinase domainmutations in chron-ic myeloid leukemia after sequential treatmentwith multiple tyrosine kinase inhibitors. Blood2007;110:4005–11.

92. Khorashad JS, Milojkovic D, Mehta P, et al.In vivo kinetics of kinase domain mutations inCML patients treated with dasatinib after failingimatinib. Blood 2008;111:2378–81.

93. Stagno F, Stella S, Berretta S, et al. Sequentialmutations causing resistance to both Imatinib

7526Clin Cancer Res 2009;15(24) December 15, 2009 www.aacrjournals.org

CCR FOCUS

Research. on June 22, 2014. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

AdministratorTypewriterfailure of imatinib mesylate therapy. Blood 2008; 112:53–5.

userrHighlight

AdministratorRectangle

AdministratorCalloutthis row was moved from right column

http://clincancerres.aacrjournals.org/

have excellent responses on either dose: a TOPScorrelative study. Blood 2008;112:3187a.

46. White DL, Saunders VA, Dang P, et al. OCT-1-mediated influx is a key determinant of the intra-cellular uptake of imatinib but not nilotinib(AMN107): reduced OCT-1 activity is the causeof low in vitro sensitivity to imatinib. Blood2006;108:697–704.

47. Dulucq S, Bouchet S, Turcq B, et al. Multidrugresistance gene (MDR1) polymorphisms are as-sociated with major molecular responsesto standard-dose imatinib in chronic myeloidleukemia. Blood 2008;112:2024–7.

48. Kim DH, Sriharsha L, Xu W, et al. Clinical rel-evance of a pharmacogenetic approach usingmultiple candidate genes to predict responseand resistance to imatinib therapy in chronicmyeloid leukemia. Clin Cancer Res 2009;15:4750–8.

49. Majlis A, Smith TL, Talpaz M, O'Brien S, RiosMB, Kantarjian HM. Significance of cytogeneticclonal evolution in chronic myelogenous leuke-mia. J Clin Oncol 1996;14:196–203.

50. Johansson B, Fioretos T, Mitelman F. Cyto-genetic and molecular genetic evolution ofchronic myeloid leukemia. Acta Haematol2002;107:76–94.

51. O'Dwyer ME, Mauro MJ, Blasdel C, et al. Clon-al evolution and lack of cytogenetic response areadverse prognostic factors for hematologic re-lapse of chronic phase CML patients treated withimatinib mesylate. Blood 2004;103:451–5.

52. Cortes JE, Talpaz M, Giles F, et al. Prognosticsignificance of cytogenetic clonal evolution inpatients with chronic myelogenous leukemia onimatinib mesylate therapy. Blood 2003;101:3794–800.

53. Cortes J, O'Dwyer ME. Clonal evolution inchronic myelogenous leukemia. Hematol OncolClin North Am 2004;18:671–84 [x.].

54. Lahaye T, Riehm B, Berger U, et al. Responseand resistance in 300 patients with BCR-ABL-positive leukemias treated with imatinib in a sin-gle center: a 4.5-year follow-up. Cancer 2005;103:1659–69.

55. Jabbour E, Kantarjian H, Jones D, et al. Fre-quency and clinical significance of BCR-ABL mu-tations in patients with chronic myeloid leukemiatreated with imatinib mesylate. Leukemia 2006;20:1767–73.

56. Geahlen RL, Handley MD, Harrison ML. Molec-ular interdiction of Src-family kinase signaling inhematopoietic cells. Oncogene 2004;23:8024–32.

57. Danhauser-Riedl S, Warmuth M, Druker BJ,Emmerich B, Hallek M. Activation of Src kinasesp53/56lyn and p59hck by p210bcr/abl in myeloidcells. Cancer Res 1996;56:3589–96.

58. Donato NJ, Wu JY, Stapley J, et al. BCR-ABLindependence and LYN kinase overexpression inchronic myelogenous leukemia cells selected forresistance to STI571. Blood 2003;101:690–8.

59. Hu Y, Swerdlow S, Duffy TM, Weinmann R,Lee FY, Li S. Targeting multiple kinase pathwaysin leukemic progenitors and stem cells isessential for improved treatment of Ph+ leuke-mia in mice. Proc Natl Acad Sci U S A 2006;103:16870–5.

60. Dai Y, Rahmani M, Corey SJ, Dent P, Grant SA.Bcr/Abl-independent, Lyn-dependent form of im-atinib mesylate (STI-571) resistance is associat-ed with altered expression of Bcl-2. J BiolChem 2004;279:34227–39.

61. Schindler T, Bornmann W, Pellicena P, MillerWT, Clarkson B, Kuriyan J. Structural mecha-nism for STI-571 inhibition of abelson tyrosinekinase. Science 2000;289:1938–42.

62. O'Hare T, Eide CA, Deininger MW. PersistentLYN signaling in imatinib-resistant, BCR-ABL-

independent chronic myelogenous leukemia. JNatl Cancer Inst 2008;100:908–9.

63. Copland M, Hamilton A, Elrick LJ, et al. Dasa-tinib (BMS-354825) targets an earlier progenitorpopulation than imatinib in primary CML butdoes not eliminate the quiescent fraction. Blood2006;107:4532–9.

64. Jiang X, Zhao Y, Smith C, et al. Chronic mye-loid leukemia stem cells possess multiple uniquefeatures of resistance to BCR-ABL targeted ther-apies. Leukemia 2007;21:926–35.

65. Jin L, Tabe Y, Konoplev S, et al. CXCR4 up-regulation by imatinib induces chronic mye-logenous leukemia (CML) cell migrationto bone marrow stroma and promotes surviv-al of quiescent CML cells. Mol Cancer Ther2008;7:48–58.

66. Konig H, Copland M, Chu S, Jove R, HolyoakeTL, Bhatia R. Effects of dasatinib on SRC ki-nase activity and downstream intracellularsignaling in primitive chronic myelogenous leu-kemia hematopoietic cells. Cancer Res 2008;68:9624–33.

67. Jorgensen HG, Allan EK, Jordanides NE,Mountford JC, Holyoake TL. Nilotinib exertsequipotent antiproliferative effects to imatiniband does not induce apoptosis in CD34+ CMLcells. Blood 2007;109:4016–9.

68. Copland M, Pellicano F, Richmond L, et al.BMS-214662 potently induces apoptosis ofchronic myeloid leukemia stem and progenitorcells and synergizes with tyrosine kinase inhibi-tors. Blood 2008;111:2843–53.

69. Holtz M, Forman SJ, Bhatia R. Growth factorstimulation reduces residual quiescent chronicmyelogenous leukemia progenitors remainingafter imatinib treatment. Cancer Res 2007;67:1113–20.

70. Copland M, Fraser AR, Harrison SJ, HolyoakeTL. Targeting the silent minority: emergingimmunotherapeutic strategies for eradicationof malignant stem cells in chronic myeloid leu-kaemia. Cancer Immunol Immunother 2005;54:297–306.

71. Jamieson CH, Ailles LE, Dylla SJ, et al.Granulocyte-macrophage progenitors as candi-date leukemic stem cells in blast-crisis CML. NEngl J Med 2004;351:657–67.

72. GorreME,MohammedM, Ellwood K, et al. Clin-ical resistance to STI-571 cancer therapy causedby BCR-ABL gene mutation or amplification.Science 2001;293:876–80.

73. Modi H, McDonald T, Chu S, Yee JK, FormanSJ, Bhatia R. Role of BCR/ABL gene-expressionlevels in determining the phenotype and imati-nib sensitivity of transformed human hemato-poietic cells. Blood 2007;109:5411–21.

74. Hochhaus A, Kreil S, Corbin AS, et al. Molecu-lar and chromosomal mechanisms of resistanceto imatinib (STI571) therapy. Leukemia 2002;16:2190–6.

75. Barnes DJ, Palaiologou D, Panousopoulou E,et al. Bcr-Abl expression levels determine the rateof development of resistance to imatinibmesylatein chronic myeloid leukemia. Cancer Res 2005;65:8912–9.

76. Quintas-Cardama A, Kantarjian HM, Cortes JE.Mechanisms of primary and secondary resistanceto imatinib in chronic myeloid leukemia. CancerControl 2009;16:122–31.

77. Nicolini FE, Corm S, Le QH, et al. Mutation sta-tus and clinical outcome of 89 imatinib mesylate-resistant chronicmyelogenous leukemia patients:a retrospective analysis from the French inter-group of CML (Fi(phi)-LMC GROUP). Leukemia2006;20:1061–6.

78. Jabbour E, Kantarjian H, Jones D, et al. Char-acteristics and outcomes of patients with chronicmyeloid leukemia and T315I mutation following

failure of imatinib mesylate therapy. Blood 2008;112:53–5.

79. Nicolini FE, Hayette S, Corm S, et al. Clinicaloutcome of 27 imatinib mesylate-resistantchronic myelogenous leukemia patients harbor-ing a T315I BCR-ABL mutation. Haematologica2007;92:1238–41.

80. Soverini S, Colarossi S, Gnani A, et al. Contribu-tion of ABL kinase domain mutations to imatinibresistance in different subsets of Philadelphia-positive patients: by the GIMEMA Working Partyon Chronic Myeloid Leukemia. Clin Cancer Res2006;12:7374–9.

81. Branford S, Rudzki Z, Walsh S, et al. Detec-tion of BCR-ABL mutations in patients withCML treated with imatinib is virtually alwaysaccompanied by clinical resistance, and muta-tions in the ATP phosphate-binding loop (P-loop) are associated with a poor prognosis.Blood 2003;102:276–83.

82. Griswold IJ, MacPartlin M, Bumm T, et al.Kinase domain mutants of Bcr-Abl exhibit al-tered transformation potency, kinase activity,and substrate utilization, irrespective of sensitiv-ity to imatinib. Mol Cell Biol 2006;26:6082–93.

83. Soverini S, Martinelli G, Rosti G, et al. ABL mu-tations in late chronic phase chronic myeloidleukemia patients with up-front cytogenetic re-sistance to imatinib are associated with a greaterlikelihood of progression to blast crisis andshorter survival: a study by the GIMEMA Work-ing Party on Chronic Myeloid Leukemia. J ClinOncol 2005;23:4100–9.

84. Khorashad JS, de Lavallade H, Apperley JF,et al. Finding of kinase domain mutations inpatients with chronic phase chronic myeloidleukemia responding to imatinib may identifythose at high risk of disease progression. J ClinOncol 2008;26:4806–13.

85. Roche-Lestienne C, Soenen-Cornu V, Grardel-Duflos N, et al. Several types of mutations of theAbl gene can be found in chronic myeloid leuke-mia patients resistant to STI571, and they canpre-exist to the onset of treatment. Blood 2002;100:1014–8.

86. Khorashad JS, Anand M, Marin D, et al. Thepresence of a BCR-ABL mutant allele in CMLdoes not always explain clinical resistance to im-atinib. Leukemia 2006;20:658–63.

87. Willis SG, Lange T, Demehri S, et al. High-sensitivity detection of BCR-ABL kinase domainmutations in imatinib-naive patients: correlationwith clonal cytogenetic evolution but not re-sponse to therapy. Blood 2005;106:2128–37.

88. Skaggs BJ, Gorre ME, Ryvkin A, et al. Phos-phorylation of the ATP-binding loop directs on-cogenicity of drug-resistant BCR-ABL mutants.Proc Natl Acad Sci U S A 2006;103:19466–71.

89. O'Hare T, Eide CA, Deininger MW. Bcr-Abl ki-nase domain mutations, drug resistance, and theroad to a cure for chronic myeloid leukemia.Blood 2007;110:2242–9.

90. Shah NP, Skaggs BJ, Branford S, et al. Se-quential ABL kinase inhibitor therapy selectsfor compound drug-resistant BCR-ABL muta-tions with altered oncogenic potency. J Clin In-vest 2007;117:2562–9.

91. Cortes J, Jabbour E, Kantarjian H, et al. Dynam-ics of BCR-ABL kinase domainmutations in chron-ic myeloid leukemia after sequential treatmentwith multiple tyrosine kinase inhibitors. Blood2007;110:4005–11.

92. Khorashad JS, Milojkovic D, Mehta P, et al.In vivo kinetics of kinase domain mutations inCML patients treated with dasatinib after failingimatinib. Blood 2008;111:2378–81.

93. Stagno F, Stella S, Berretta S, et al. Sequentialmutations causing resistance to both Imatinib

7526Clin Cancer Res 2009;15(24) December 15, 2009 www.aacrjournals.org

CCR FOCUS

Research. on June 22, 2014. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

http://clincancerres.aacrjournals.org/

Perspectives

Selecting optimal second-line tyrosine kinase inhibitor therapy for chronicmyeloid leukemia patients after imatinib failure: does the BCR-ABL mutationstatus really matter?

Susan Branford,1 Junia V. Melo,1 and Timothy P. Hughes1

1Centre for Cancer Biology, Departments of Molecular Pathology and Haematology, SA Pathology, Adelaide, Australia

Preclinical studies of BCR-ABL mutationsensitivity to nilotinib or dasatinib sug-gested that the majority would be sensitive.Correspondingly, the initial clinical trialsdemonstrated similar response rates forCML patients after imatinib failure, irrespec-tive of the mutation status. However, oncloser examination, clinical evidence nowindicates that some mutations are less sen-sitive to nilotinib (Y253H, E255K/V, andF359V/C) or dasatinib (F317L and V299L).T315I is insensitive to both. Novel mutations

(F317I/V/C and T315A) are less sensitive/insensitive to dasatinib. We refer to thesecollectively as second-generation inhibitor(SGI) clinically relevant mutations. By invitro analysis, other mutations confer a de-gree of insensitivity; however, clinical evi-dence is currently insufficient to define themas SGI clinically relevant. Here we examinethe mutations that are clearly SGI clinicallyrelevant, those with minimal impact on re-sponse, and those for which more data areneeded. In our series of patients with muta-

tions at imatinib cessation and/or at nilotinibor dasatinib commencement, 43% had SGIclinically relevant mutations, including 14%with T315I. The frequency of SGI clinicallyrelevant mutations was dependent on thedisease phase at imatinib failure. The clini-cal data suggest that a mutation will often bedetectable after imatinib failure for whichthere is compelling clinical evidence thatone SGI should be preferred. (Blood. 2009;114:5426-5435)

Introduction

The outcome for patients with chronic myeloid leukemia (CML) whofail imatinib has improved since the availability of second-generationBCR-ABL kinase inhibitors (SGIs). The most common mechanisms ofimatinib resistance are mutations within the BCR-ABL kinase domainand protein overexpression by gene amplification.1-12 Resistance is alsoassociated with other genetic events, as indicated by the detection ofcytogenetic abnormalities in the Philadelphia chromosome–positive(Ph�) clone in more than 50% of imatinib-resistant patients.13 It hasbeen suggested that some BCR-ABL mutations play no causal role inresistance.14-16 However, approximately half of the patients who com-mence SGIs after imatinib therapy have detectable imatinib-resistantBCR-ABL mutations. Imatinib binds to the inactive conformation ofBCR-ABL, leading to disruption of the adenosine triphosphate (ATP)binding site and blockade of the catalytic activity.17,18 BCR-ABLmutations that impair imatinib binding while still enabling ATP binding,or that alter the specific protein conformation required for imatinibbinding, are selected in the presence of imatinib.19-21 In the absence ofimatinib, these mutations do not confer a growth advantage.22

For patients commencing nilotinib or dasatinib after imatinib cessa-tion, clinical trials have demonstrated similar responses for patients withor without mutations, except for T315I for which neither drug isactive.23-31 This mutation demonstrates cross-resistance to imatinib,nilotinib, and dasatinib.32-34 However, a closer examination of responsesto SGI therapy for individual mutations has identified a limited number,other than T315I, that are less sensitive to either nilotinib or dasat-inib.35-37 Furthermore, in vitro studies have identified mutations thatconfer a degree of insensitivity38 or resistance.39

How well do the problematic mutations identified by in vitrostudies correlate with those identified by clinical studies? More-over, does the in vitro sensitivity of mutations provide a reliable

indication of the probable response to SGIs? Undoubtedly, in vitrosensitivity of imatinib-resistant mutations can be a useful guidewhen considering an increased imatinib dose.40 Here we assessBCR-ABL mutations in the context of their impact on responseafter a change to SGI therapy by an examination of the availableclinical data. The mutation status may contribute to therapeuticdecisions after imatinib failure or indeed after failure of an SGI. Weassess the frequency that mutations conferring a degree of clinicalinsensitivity to SGIs are detectable at the time of imatinibcessation. These are collectively referred to as SGI clinicallyrelevant mutations. We also examine whether the disease phaseinfluences their frequency. Last, we examine the occurrence ofmultiple mutations in imatinib-treated patients and the extent towhich disease phase influences their detection.

BCR-ABL mutations in the era of SGIs:type still matters

Mutant sensitivity assessed by in vitro studies

Preclinical studies of nilotinib against 33 BCR-ABL mutants predictedthat the inhibitor would have clinical activity in patients harboring thesemutations, except for T315I.33,34,41 Similarly, among 19 imatinib-resistant mutants tested against dasatinib, T315I was the only clearlyresistant mutation.32,33 The in vitro results were similar to the earlier invitro studies of imatinib, in that mutants displayed various degrees ofsensitivity.15,42 SGI sensitivity was assessed by various methods, includ-ing the degree of inhibition of BCR-ABL autophosphorylation or cellproliferation after transfection of mutants into Ba/F3 cells.32-34Although

Submitted August 10, 2009; accepted October 5, 2009. Prepublished online asBlood First Edition paper, October 30, 2009; DOI 10.1182/blood-2009-08-215939.

© 2009 by The American Society of Hematology

5426 BLOOD, 24 DECEMBER 2009 � VOLUME 114, NUMBER 27

userrTextbox2. Branford2009 (truncated)

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

AdministratorTypewriterBlood, 2009, Vol. 114, pp. 5426-5435)

AdministratorTypewriter2

Perspectives

Selecting optimal second-line tyrosine kinase inhibitor therapy for chronicmyeloid leukemia patients after imatinib failure: does the BCR-ABL mutationstatus really matter?

Susan Branford,1 Junia V. Melo,1 and Timothy P. Hughes1

1Centre for Cancer Biology, Departments of Molecular Pathology and Haematology, SA Pathology, Adelaide, Australia

Preclinical studies of BCR-ABL mutationsensitivity to nilotinib or dasatinib sug-gested that the majority would be sensitive.Correspondingly, the initial clinical trialsdemonstrated similar response rates forCML patients after imatinib failure, irrespec-tive of the mutation status. However, oncloser examination, clinical evidence nowindicates that some mutations are less sen-sitive to nilotinib (Y253H, E255K/V, andF359V/C) or dasatinib (F317L and V299L).T315I is insensitive to both. Novel mutations

(F317I/V/C and T315A) are less sensitive/insensitive to dasatinib. We refer to thesecollectively as second-generation inhibitor(SGI) clinically relevant mutations. By invitro analysis, other mutations confer a de-gree of insensitivity; however, clinical evi-dence is currently insufficient to define themas SGI clinically relevant. Here we examinethe mutations that are clearly SGI clinicallyrelevant, those with minimal impact on re-sponse, and those for which more data areneeded. In our series of patients with muta-

tions at imatinib cessation and/or at nilotinibor dasatinib commencement, 43% had SGIclinically relevant mutations, including 14%with T315I. The frequency of SGI clinicallyrelevant mutations was dependent on thedisease phase at imatinib failure. The clini-cal data suggest that a mutation will often bedetectable after imatinib failure for whichthere is compelling clinical evidence thatone SGI should be preferred. (Blood. 2009;114:5426-5435)

Introduction

The outcome for patients with chronic myeloid leukemia (CML) whofail imatinib has improved since the availability of second-generationBCR-ABL kinase inhibitors (SGIs). The most common mechanisms ofimatinib resistance are mutations within the BCR-ABL kinase domainand protein overexpression by gene amplification.1-12 Resistance is alsoassociated with other genetic events, as indicated by the detection ofcytogenetic abnormalities in the Philadelphia chromosome–positive(Ph�) clone in more than 50% of imatinib-resistant patients.13 It hasbeen suggested that some BCR-ABL mutations play no causal role inresistance.14-16 However, approximately half of the patients who com-mence SGIs after imatinib therapy have detectable imatinib-resistantBCR-ABL mutations. Imatinib binds to the inactive conformation ofBCR-ABL, leading to disruption of the adenosine triphosphate (ATP)binding site and blockade of the catalytic activity.17,18 BCR-ABLmutations that impair imatinib binding while still enabling ATP binding,or that alter the specific protein conformation required for imatinibbinding, are selected in the presence of imatinib.19-21 In the absence ofimatinib, these mutations do not confer a growth advantage.22

For patients commencing nilotinib or dasatinib after imatinib cessa-tion, clinical trials have demonstrated similar responses for patients withor without mutations, except for T315I for which neither drug isactive.23-31 This mutation demonstrates cross-resistance to imatinib,nilotinib, and dasatinib.32-34 However, a closer examination of responsesto SGI therapy for individual mutations has identified a limited number,other than T315I, that are less sensitive to either nilotinib or dasat-inib.35-37 Furthermore, in vitro studies have identified mutations thatconfer a degree of insensitivity38 or resistance.39

How well do the problematic mutations identified by in vitrostudies correlate with those identified by clinical studies? More-over, does the in vitro sensitivity of mutations provide a reliable

indication of the probable response to SGIs? Undoubtedly, in vitrosensitivity of imatinib-resistant mutations can be a useful guidewhen considering an increased imatinib dose.40 Here we assessBCR-ABL mutations in the context of their impact on responseafter a change to SGI therapy by an examination of the availableclinical data. The mutation status may contribute to therapeuticdecisions after imatinib failure or indeed after failure of an SGI. Weassess the frequency that mutations conferring a degree of clinicalinsensitivity to SGIs are detectable at the time of imatinibcessation. These are collectively referred to as SGI clinicallyrelevant mutations. We also examine whether the disease phaseinfluences their frequency. Last, we examine the occurrence ofmultiple mutations in imatinib-treated patients and the extent towhich disease phase influences their detection.

BCR-ABL mutations in the era of SGIs:type still matters

Mutant sensitivity assessed by in vitro studies

Preclinical studies of nilotinib against 33 BCR-ABL mutants predictedthat the inhibitor would have clinical activity in patients harboring thesemutations, except for T315I.33,34,41 Similarly, among 19 imatinib-resistant mutants tested against dasatinib, T315I was the only clearlyresistant mutation.32,33 The in vitro results were similar to the earlier invitro studies of imatinib, in that mutants displayed various degrees ofsensitivity.15,42 SGI sensitivity was assessed by various methods, includ-ing the degree of inhibition of BCR-ABL autophosphorylation or cellproliferation after transfection of mutants into Ba/F3 cells.32-34Although

Submitted August 10, 2009; accepted October 5, 2009. Prepublished online asBlood First Edition paper, October 30, 2009; DOI 10.1182/blood-2009-08-215939.

© 2009 by The American Society of Hematology

5426 BLOOD, 24 DECEMBER 2009 � VOLUME 114, NUMBER 27

userrTextbox2. Branford2009 (truncated)

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

AdministratorTypewriterBlood, 2009, Vol. 114, pp. 5426-5435)

AdministratorSquiggly

AdministratorSquiggly

AdministratorSquiggly

bosutinib, which is under clinical trial as SGI therapy.39 However,G250E is described as sensitive to nilotinib and dasatinib in anotherassessment.38 F317L is classified as moderately resistant or resis-tant to the 4 inhibitors in one study39 and sensitive to nilotinib andinsensitive to dasatinib in another.38 These discrepancies may bethe result of methodologic differences or differences in the cutoffvalues for the classification of mutant sensitivity and may causedifficulty for interpretation.

Mutations identified from resistance screens

Resistance screens have identified a limited number of mutationsthat emerge in the presence of increasing doses of nilotinib anddasatinib,43-46 and these correspond, to a degree, to the mutantsensitivity determined in cell proliferation assays.38,39 Mutations atdasatinib contact residues appeared to be particularly relevant,including V299L. In 2 resistance screens, mutations at T315 andF317 accounted for 95% of all mutants recovered, including novelmutations F317V/I/S/C and T315A, which had not been reported inimatinib-treated patients.44,45 In one study, F317V and T315A werethe most frequent to emerge (41% and 30%, respectively) and had40- to 90-fold reduced dasatinib sensitivity compared with unmu-tated BCR-ABL.45 In accord with these results, T315A and F317Vhad the highest IC50 values, except for T315I, in the in vitroassessment of dasatinib by O’Hare et al.33

In 3 in vitro resistance screens of increasing doses of nilotinib,T315I emerged most frequently and represented 49% of allmutations recovered.43,44,46 However, the common imatinib-resistant mutation, Y253H, was also among the most frequent toemerge and had the highest nilotinib IC50 value in each of thesestudies, apart from T315I. E255K, Y253H, and T315I were theonly mutations to emerge in all 3 screens, and E255V and Q252Hemerged in 2 of 3 studies. All other mutations were confined to oneof the screens. With the exception of T315I, all mutations wereeffectively suppressed by nilotinib concentrations of 2000nM,which falls within the peak-trough plasma levels (3600-1700nM)measured in patients treated with 400 mg nilotinib twice daily.24

Over the past few years, clinical studies have identified alimited number of mutations that may be relevant for response tonilotinib and dasatinib as second- or third-line inhibitortherapy35,36,47,48 and are implicated in resistance.47-51 In one report,a new mutation was acquired in 83% of patients who relapsed aftera response.48 The available clinical data present an opportunity toassess how effectively in vitro studies predict the SGI clinicallyrelevant mutations and their validity for determining appropriatetherapy after imatinib failure.

Mutation sensitivity assessed by clinical studies of dasatinib

The majority of imatinib-resistant mutations remain sensitive todasatinib.23,26-29,31 However, consistent with the dasatinib resis-tance screens and cell proliferation assays, clinical reports con-firmed T315I/A, F317L/I/V/C, and V299L as relevant for de-creased clinical efficacy, either as preexisting or as emergingmutations.37,48-51

One of the most frequent mutations to emerge with clinicaldasatinib resistance was V299L.48,50 This mutation was reportedvery rarely in imatinib-treated patients.8,52 V299L and F317L werealso preferentially associated with dasatinib failure in a study ofmutation dynamics after sequential inhibitor therapy.51 Interest-ingly, V299L was only detected at a frequency of 1% in resistancescreens44,45 and only at lower dasatinib concentrations.44

The largest analysis of clinical response to dasatinib afterimatinib failure to date involved 1043 CML patients treated inchronic phase (CP).37 The presence of T315I or F317L at the timeof commencing dasatinib was associated with the least favorableresponses. Furthermore, the most frequently detected new muta-tions were T315I, F317L, and V299L. A conclusion of the studywas that alternative treatment options should be considered forpatients with these mutations.37 Consistent with these findings,other studies reported low response rates for patients with F317L.47,48

In one study, 8 of 16 dasatinib-treated patients after imatinib failureacquired F317L, and this mutation was deemed dasatinib-resistantbut sensitive to other inhibitors.47

Mutant sensitivity assessed by clinical studies of nilotinib

Clinical studies have demonstrated that the majority of imatinib-resistant mutations remain sensitive to nilotinib.24,25,30 Neverthe-less, several mutations are less sensitive, which influences theresponse.36 The mutations that emerged in the nilotinib resistancescreens have corresponded, to a degree, with the clinical findings.In an evaluation of 281 CP patients in the nilotinib phase 2registration study, those with T315I, Y253H, E255K/V, andF359V/C (n � 31) at nilotinib commencement had the leastfavorable responses. These mutations had the highest IC50 values incell proliferation assays as assessed by Weisberg et al( � 150nM).34,41 No patient with these mutations achieved acomplete cytogenetic response (CCyR) by 12 months, 6 (19%)achieved a major cytogenetic response, and 10 (32%) a completehematologic response (CHR).36 In contrast, 32 of 74 patients (43%)with any other mutation and 35 of 87 (40%) with imatinibresistance but no mutation achieved CCyR. These mutations werealso among the most common new mutations during nilotinibtherapy and were associated with a higher risk of progression. Inanother study, 13 of 14 patients who relapsed with new mutationson nilotinib as second- or third-line inhibitor therapy had one ofthese mutations.48

The poor response associated with Y253H, E255K/V, andF359V/C when present at the time of commencing nilotinib wasconfirmed in patients treated in accelerated phase (AP) CML.53

Of 17 of 87 AP patients with these mutations, only 24%achieved a CHR. In contrast, 55% without mutations and 58%with other mutations (excluding T315I) achieved a CHR. Fromthe nilotinib clinical response data, the suggestion was thattherapies other than nilotinib should be considered for patientswith these mutations.36,53 Consistent with these findings, themost frequent mutations detected in patients with nilotinibfailure in the study of Cortes et al51 were at residues 253, 255,359, and 311. A mutation at residue 311 was also detected in anilotinib resistance screen.43

Which mutations are relevant for response and resistance tonilotinib and dasatinib from clinical studies?

The current clinical data suggest the SGI clinically relevantmutations are T315I for both inhibitors: F317L/I/C/V, V299L, andT315A for dasatinib and Y253H, E255K/V, and F359V/C fornilotinib. Sensitive mutation detection of these specific mutationsto aid therapeutic choices may be beneficial. Several techniquesmoderately improve the sensitivity to 1.5% to 10%, includingpyrosequencing,16,51 ligation-dependent competitive polymerasechain reaction (PCR),54 and SEQUENOM MassARRAY.55 Highlysensitive techniques have a detection limit from 0.0003% to 0.1%,including mutation-specific PCR based on the Taqman platform,56

5428 BRANFORD et al BLOOD, 24 DECEMBER 2009 � VOLUME 114, NUMBER 27

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrUnderline

userrHighlight

userrHighlight

userrHighlight

PCR–restriction fragment length polymorphism,57 polymerasecolony assay,58 allele-specific PCR,59 and a nanofluidic platform.60

Whether highly sensitive detection of SGI clinically relevantmutations before SGI therapy will always correlate with theirclonal expansion and resistance is unknown. This was not alwaysthe case using highly sensitive mutation detection before imatinibtherapy.59

Do the SGI clinically relevant mutations correspond to thein vitro data?

Based on current clinical information, the answer to this question isyes, to a degree. Why are some mutations clinically relevant forSGIs and not others that either emerged more frequently in in vitroresistance screens and/or those with greater in vitro insensitivity?In the case of dasatinib, identification of clinically relevantmutations at residues T315 and F317 is consistent with theiremergence in resistance screens. However, despite the high fre-quency of F317V and T315A in a dasatinib resistance screen andtheir significantly reduced sensitivity to dasatinib compared withF317L,45 F317V was not detected in any patient and T315A in only2 patients in the initial reports.49,50 Could this be related to reducedoncogenicity of these mutations? Severely attenuated transformingactivity of T315A was demonstrated relative to wild-type BCR-ABL.61 However, F317L was only marginally more transformingthan T315A. Furthermore, reduced transforming activity does notappear to be related to frequency of detection of imatinib-resistantmutations: M351T displays reduced transforming activity61,62 yet isamong the most commonly detected imatinib-resistant mutations.63

The sensitivity rankings by cell proliferation assays consistentlysuggest that the P-loop mutations Q252H and E255K/V may berelevant for dasatinib (Table 1).38,39 In dasatinib-treated CP pa-tients, the CCyR rates for patients with these mutations rangedfrom 17% to 38%.37 These mutations have rarely been associatedwith clinical dasatinib resistance or as new mutations duringdasatinib therapy.37,48-51 However, E255K and Q252H were amongthe mutations recovered in in vitro dasatinib resistance screens butwere the only noncontact residues.44,45 The BCR-ABL crystalstructure in complex with dasatinib suggested that interactionsbetween the P-loop and dasatinib were less critical for binding.64

Manley et al65 proposed that it is doubtful mutations of Q252 andE255 could cause a change in the structure of the P-loop to interferewith dasatinib binding, without also disturbing binding of ATP,which is critical for BCR-ABL reactivation. Clearly, additionalclinical information is required before the significance for dasatinibresponse of E255K/V and Q252H is elucidated.

Nilotinib in vitro sensitivity classifications38,39 correlate veryclosely with clinical data (Table 1). T315I, Y253H, E255K/V, andF359V have the highest IC50 values (F359C was not tested). Themajor inconsistency is for the classification of G250E (sensitive38/resistant39). The IC50 reported by Weisberg et al41 for G250E was145nM, which is close to the cutoff value of 150nM used to definemutations less sensitive to nilotinib in the clinical evaluation of CPpatients.36 There were only 5 CP patients with this mutation atnilotinib start, and 3 (60%) achieved a CCyR.36 G250E was amongthe most common mutations to emerge with nilotinib.36 However, itwas not among the most common mutations associated withprogression. Mutations that are less sensitive, but still responsive,to nilotinib or dasatinib may mistakenly appear to be newlyacquired as more sensitive alleles disappear more rapidly. G250Edid not emerge in another study in which 13 patients with nilotinibresistance acquired new mutations.48 Inconsistency is also apparentfor in vitro sensitivity classification of G250E for dasatinib

(sensitive38/resistant39). This mutation was the most commonlydetected in CP patients at the start of dasatinib, and 20 of 60 (33%)achieved a CCyR.37 G250E did not emerge in the resistance screensof dasatinib.44,45 There is currently no strong clinical evidence tosuggest that the presence of G250E would influence the response tonilotinib or dasatinib. Q252H and Y253F are consistently classifiedas moderately insensitive38 or resistant39 to nilotinib by in vitroassessment, and Q252H emerged in the nilotinib resistancescreens.43,46 Further clinical data are required for adequate assess-ment of their response to nilotinib.

Validation of in vitro sensitivity of different mutations wasrecently demonstrated.35 In vitro sensitivity was predictive ofresponse and long-term outcome for patients treated with nilotinibor dasatinib. Mutations for which there was a discrepancy inreported sensitivity among in vitro studies were classified accord-ing to the highest IC50 to the corresponding inhibitor. However,G250E was classified as a sensitive mutation to both inhibitorsdespite the discrepancy in the in vitro sensitivity classifications.38,39

In Figure 1, CCyR rates of CP patients with various mutations atthe start of dasatinib therapy in the large clinical study of Müller etal37 are plotted according to in vitro sensitivity classifications.CCyR rates were only partially predicted by in vitro sensitivity.

There may be rare imatinib-resistant mutations not included inin vitro studies that could also be less sensitive to SGIs, such asF359I. This mutation emerged in one nilotinib resistance screen.43

Furthermore, rapid progression was observed in a nilotinib-treatedpatient harboring F359I at commencement of nilotinib.66 The IC50value of this mutation is unknown.

From the available studies, we now have a clearer understand-ing of the BCR-ABL mutations for which there is compellingclinical evidence that response could be compromised by treatmentwith one and/or another of the SGIs if present after imatinib failure.These are T315I, F317L, V299L, Y253H, E255K/V, and F359V/C,the finding of which would influence the therapeutic decision.These mutations are classified in Table 2 as either class D (no rolefor SGI therapy) or class C (compelling clinical evidence torecommend an alternative inhibitor). At this stage, the presence ofother mutations should have no impact on clinical decisions.Nevertheless, there are mutations where further clinical evidencemay reveal relevance for an inhibitor (class B, Table 2). However,additional clinical assessment is required before an alternativeinhibitor would be recommended for class B mutations. Table 2lists the frequency of mutations at our institution, whereas theclassifications are based on published literature. Of course, muta-tion status is only one factor that needs to be considered whenselecting SGIs, which include issues of tolerance and the diseasephase.

How frequently will the mutation status be anissue when considering therapeutic optionsafter imatinib failure?

We have performed BCR-ABL mutation analysis at our institutionfor imatinib-treated patients since 2001, which allows an assess-ment of the frequency of SGI clinically relevant mutations at asingle institution. Mutations were detected in 386 patients usingdirect sequencing5,67,68 at the time of imatinib cessation (n � 159)and/or at commencement of nilotinib or dasatinib after imatinibfailure (n � 227). The assay has a mutation sensitivity of 10% to20%, and the approximate percentage of mutant nucleotide relativeto unmutated nucleotide is calculated either by the mutation

BCR-ABL MUTATION STATUS AND THERAPEUTIC DECISIONS 5429BLOOD, 24 DECEMBER 2009 � VOLUME 114, NUMBER 27

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

PCR–restriction fragment length polymorphism,57 polymerasecolony assay,58 allele-specific PCR,59 and a nanofluidic platform.60

Whether highly sensitive detection of SGI clinically relevantmutations before SGI therapy will always correlate with theirclonal expansion and resistance is unknown. This was not alwaysthe case using highly sensitive mutation detection before imatinibtherapy.59

Do the SGI clinically relevant mutations correspond to thein vitro data?

Based on current clinical information, the answer to this question isyes, to a degree. Why are some mutations clinically relevant forSGIs and not others that either emerged more frequently in in vitroresistance screens and/or those with greater in vitro insensitivity?In the case of dasatinib, identification of clinically relevantmutations at residues T315 and F317 is consistent with theiremergence in resistance screens. However, despite the high fre-quency of F317V and T315A in a dasatinib resistance screen andtheir significantly reduced sensitivity to dasatinib compared withF317L,45 F317V was not detected in any patient and T315A in only2 patients in the initial reports.49,50 Could this be related to reducedoncogenicity of these mutations? Severely attenuated transformingactivity of T315A was demonstrated relative to wild-type BCR-ABL.61 However, F317L was only marginally more transformingthan T315A. Furthermore, reduced transforming activity does notappear to be related to frequency of detection of imatinib-resistantmutations: M351T displays reduced transforming activity61,62 yet isamong the most commonly detected imatinib-resistant mutations.63

The sensitivity rankings by cell proliferation assays consistentlysuggest that the P-loop mutations Q252H and E255K/V may berelevant for dasatinib (Table 1).38,39 In dasatinib-treated CP pa-tients, the CCyR rates for patients with these mutations rangedfrom 17% to 38%.37 These mutations have rarely been associatedwith clinical dasatinib resistance or as new mutations duringdasatinib therapy.37,48-51 However, E255K and Q252H were amongthe mutations recovered in in vitro dasatinib resistance screens butwere the only noncontact residues.44,45 The BCR-ABL crystalstructure in complex with dasatinib suggested that interactionsbetween the P-loop and dasatinib were less critical for binding.64

Manley et al65 proposed that it is doubtful mutations of Q252 andE255 could cause a change in the structure of the P-loop to interferewith dasatinib binding, without also disturbing binding of ATP,which is critical for BCR-ABL reactivation. Clearly, additionalclinical information is required before the significance for dasatinibresponse of E255K/V and Q252H is elucidated.

Nilotinib in vitro sensitivity classifications38,39 correlate veryclosely with clinical data (Table 1). T315I, Y253H, E255K/V, andF359V have the highest IC50 values (F359C was not tested). Themajor inconsistency is for the classification of G250E (sensitive38/resistant39). The IC50 reported by Weisberg et al41 for G250E was145nM, which is close to the cutoff value of 150nM used to definemutations less sensitive to nilotinib in the clinical evaluation of CPpatients.36 There were only 5 CP patients with this mutation atnilotinib start, and 3 (60%) achieved a CCyR.36 G250E was amongthe most common mutations to emerge with nilotinib.36 However, itwas not among the most common mutations associated withprogression. Mutations that are less sensitive, but still responsive,to nilotinib or dasatinib may mistakenly appear to be newlyacquired as more sensitive alleles disappear more rapidly. G250Edid not emerge in another study in which 13 patients with nilotinibresistance acquired new mutations.48 Inconsistency is also apparentfor in vitro sensitivity classification of G250E for dasatinib

(sensitive38/resistant39). This mutation was the most commonlydetected in CP patients at the start of dasatinib, and 20 of 60 (33%)achieved a CCyR.37 G250E did not emerge in the resistance screensof dasatinib.44,45 There is currently no strong clinical evidence tosuggest that the presence of G250E would influence the response tonilotinib or dasatinib. Q252H and Y253F are consistently classifiedas moderately insensitive38 or resistant39 to nilotinib by in vitroassessment, and Q252H emerged in the nilotinib resistancescreens.43,46 Further clinical data are required for adequate assess-ment of their response to nilotinib.

Validation of in vitro sensitivity of different mutations wasrecently demonstrated.35 In vitro sensitivity was predictive ofresponse and long-term outcome for patients treated with nilotinibor dasatinib. Mutations for which there was a discrepancy inreported sensitivity among in vitro studies were classified accord-ing to the highest IC50 to the corresponding inhibitor. However,G250E was classified as a sensitive mutation to both inhibitorsdespite the discrepancy in the in vitro sensitivity classifications.38,39

In Figure 1, CCyR rates of CP patients with various mutations atthe start of dasatinib therapy in the large clinical study of Müller etal37 are plotted according to in vitro sensitivity classifications.CCyR rates were only partially predicted by in vitro sensitivity.

There may be rare imatinib-resistant mutations not included inin vitro studies that could also be less sensitive to SGIs, such asF359I. This mutation emerged in one nilotinib resistance screen.43

Furthermore, rapid progression was observed in a nilotinib-treatedpatient harboring F359I at commencement of nilotinib.66 The IC50value of this mutation is unknown.

From the available studies, we now have a clearer understand-ing of the BCR-ABL mutations for which there is compellingclinical evidence that response could be compromised by treatmentwith one and/or another of the SGIs if present after imatinib failure.These are T315I, F317L, V299L, Y253H, E255K/V, and F359V/C,the finding of which would influence the therapeutic decision.These mutations are classified in Table 2 as either class D (no rolefor SGI therapy) or class C (compelling clinical evidence torecommend an alternative inhibitor). At this stage, the presence ofother mutations should have no impact on clinical decisions.Nevertheless, there are mutations where further clinical evidencemay reveal relevance for an inhibitor (class B, Table 2). However,additional clinical assessment is required before an alternativeinhibitor would be recommended for class B mutations. Table 2lists the frequency of mutations at our institution, whereas theclassifications are based on published literature. Of course, muta-tion status is only one factor that needs to be considered whenselecting SGIs, which include issues of tolerance and the diseasephase.

How frequently will the mutation status be anissue when considering therapeutic optionsafter imatinib failure?

We have performed BCR-ABL mutation analysis at our institutionfor imatinib-treated patients since 2001, which allows an assess-ment of the frequency of SGI clinically relevant mutations at asingle institution. Mutations were detected in 386 patients usingdirect sequencing5,67,68 at the time of imatinib cessation (n � 159)and/or at commencement of nilotinib or dasatinib after imatinibfailure (n � 227). The assay has a mutation sensitivity of 10% to20%, and the approximate percentage of mutant nucleotide relativeto unmutated nucleotide is calculated either by the mutation

BCR-ABL MUTATION STATUS AND THERAPEUTIC DECISIONS 5429BLOOD, 24 DECEMBER 2009 � VOLUME 114, NUMBER 27

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

irrespective of the disease phase at imatinib start.5-7,9,12 Therefore,the higher rate of detection of these mutations in patients whoprogressed to advanced phases in our study is highly consistent.However, there are some inconsistencies in our mutation fre-quency. In AP, some mutation frequencies were intermediatebetween CP and BC (Figure 2), which could be associated withtransition between the phases. However, F359V was the mostfrequent mutation detected in AP but detected at a relatively lowfrequency in BC. The frequencies of individual mutations inspecific disease phases should be considered with caution, andvalidation is required.

The differences evident in BC phenotype, if validated, raiseseveral questions. Why were these particular mutations related toeither a lymphoid or myeloid phenotype? Could the BCR-ABLmutant genotype drive the lineage phenotype in some cases or viceversa? A known genetic determinant of LBC is homozygousdeletion of the p16INK4A/p19ARF gene locus that occurs in asubstantial proportion of LBC patients, but not MBC.70-74 It waspostulated that p16INK4A and/or p19ARF mutations could induce theselective expansion of B-cell progenitors by favoring their cellcycle entry, rather than myeloid progenitors.75

In the case of BCR-ABL mutations, in vitro transformationassays have demonstrated a gain or loss of function relative tounmutated BCR-ABL.61,62 Mutations can alter the substrates thatbind to BCR-ABL and activate alternate signal transductionpathways that influence disease progression. Interestingly, E255Kand Y253F mutations showed a pronounced increase of transforma-tion potency in primary B-lymphoid progenitor cells.62 However,the transformation potency of these mutations was not significantlyincreased over unmutated BCR-ABL in the myeloid lineage. In ourcohort, the frequency of E255K was markedly increased inLBC/Ph� ALL (19.2%) compared with MBC (2.0%). However,Y253F was detected at a low frequency in all disease phases. TheSRC kinases, LYN, HCK, and FGR, have also been implicated inthe molecular pathogenesis of Ph� ALL76 and are critical for

transition of CML to LBC.77 Whether BCR-ABL mutations that arecharacteristic of BCR-ABL� lymphoid leukemia in our cohortenhance the activation of SRC kinases and hence the transition to alymphoid phenotype is unknown.

Frequency of mutations that would influence the therapeuticdecision

In our cohort of patients with mutations, 166 of 386 (43%) had oneor more SGI clinically relevant mutations. T315I was detected in53 of 386 patients (14%). In 110 of 386 patients (28%), one or moreof their mutations were clinically relevant for either nilotinib ordasatinib, but not to both. For these patients, there may be anadvantage for one or other inhibitor. The remaining 3 of 386patients (0.8%) had 2 mutations, one of which was clinicallyrelevant for dasatinib and the other for nilotinib (neither wasT315I).

Among the disease phases, there were significant differences infrequency of SGI clinically relevant mutations: 63% LBC/Ph�

ALL, 32% MBC, 49% AP, and 35% CP (P � .001, �2; Figure 3).When patients with LBC/Ph� ALL were subdivided, the frequencywas 59% for LBC and 67% for Ph� ALL.

Our analysis was performed in patients with detectable muta-tions; and from this, we can estimate the percentage of patients withan SGI clinically relevant mutation among all imatinib-resistantpatients. For imatinib-resistant CP patients commencing nilotinibor dasatinib, 48% to 55% had a mutation.36,37 Similarly, forimatinib-resistant patients in AP who commenced nilotinib ordasatinib, the frequency of mutations was 62% to 64%.27,53

Mutations in patients with LBC/Ph� ALL were detected in 62% to83%9,28,78 and up to 75% of patients with MBC.9 From thesemutation frequencies, the estimated percentage of all imatinib-resistant patients with an SGI clinically relevant mutation is 18%for CP, 31% for AP, up to 29% for MBC, and 39% to 52% forLBC/Ph� ALL. For imatinib-intolerant CP patients, the frequency

Table 2. Most frequent mutations detected at a single institution, which accounted for 88% of all mutations

Mutation No. detected Percentage of patients with mutations (n � 386) Percentage of all mutations (n � 503)

Mutation class for therapeutic decision*

Nilotinib Dasatinib

T315I 53 13.7 10.6 D D

M351T 47 12.2 9.4 A A

G250E 46 11.9 9.2 A A

F359V 35 9.1 7.0 C A

M244V 33 8.5 6.6 A A

Y253H 32 8.3 6.4 C A

E255K 27 7.0 5.4 C B

H396R 26 6.7 5.2 A A

F317L 22 5.7 4.4 A C

E355G 16 4.1 3.2 A A

Q252H 15 3.9 3.0 B B

E255V 14 3.6 2.8 C B

E459K 14 3.6 2.8 A A

F486S 13 3.4 2.6 A A

L248V 10 2.6 2.0 A A

D276G 10 2.6 2.0 A A

E279K 10 2.6 2.0 A A

Y253F 6 1.6 1.2 B A

F359C 6 1.6 1.2 C A

F359I 6 1.6 1.2 B A

*Class A indicates currently no compelling clinical evidence to suggest that the mutation would not respond to the inhibitor. Class B, In vitro assessment consistentlyindicates that the mutation may confer intermediate insensitivity38/resistance39 to the inhibitor, or clinical evidence may be suggestive of reduced sensitivity. At this stage, thepresence of these mutations should have no impact on clinical decisions and additional clinical assessment is required before an alternative inhibitor would be recommended.Class C, Compelling clinical evidence to recommend an alternative inhibitor; V299L, which is very rarely detected in imatinib-treated patients, is a dasatinib class C mutation.Class D, No role for SGI therapy.

BCR-ABL MUTATION STATUS AND THERAPEUTIC DECISIONS 5431BLOOD, 24 DECEMBER 2009 � VOLUME 114, NUMBER 27

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrHighlight

userrUnderline

userrHighlight

userrHighlight

userrHighlight

irrespective of the disease phase at imatinib start.5-7,9,12 Therefore,the higher rate of detection of these mutations in patients whoprogressed to advanced phases in our study is highly consistent.However, there are some inconsistencies in our mutation fre-quency. In AP, some mutation frequencies were intermediatebetween CP and BC (Figure 2), which could be associated withtransition between the phases. However, F359V was the mostfrequent mutation detected in AP but detected at a relatively lowfrequency in BC. The frequencies of individual mutations inspecific disease phases should be considered with caution, andvalidation is required.

The differences evident in BC phenotype, if validated, raiseseveral questions. Why were these particular mutations related toeither a lymphoid or myeloid phenotype? Could the BCR-ABLmutant genotype drive the lineage phenotype in some cases or viceversa? A known genetic determinant of LBC is homozygousdeletion of the p16INK4A/p19ARF gene locus that occurs in asubstantial proportion of LBC patients, but not MBC.70-74 It waspostulated that p16INK4A and/or p19ARF mutations could induce theselective expansion of B-cell progenitors by favoring their cellcycle entry, rather than myeloid progenitors.75

In the case of BCR-ABL mutations, in vitro transformationassays have demonstrated a gain or loss of function relative tounmutated BCR-ABL.61,62 Mutations can alter the substrates thatbind to BCR-ABL and activate alternate signal transductionpathways that influence disease progression. Interestingly, E255Kand Y253F mutations showed a pronounced increase of transforma-tion potency in primary B-lymphoid progenitor cells.62 However,the transformation potency of these mutations was not significantlyincreased over unmutated BCR-ABL in the myeloid lineage. In ourcohort, the frequency of E255K was markedly increased inLBC/Ph� ALL (19.2%) compared with MBC (2.0%). However,Y253F was detected at a low frequency in all disease phases. TheSRC kinases, LYN, HCK, and FGR, have also been implicated inthe molecular pathogenesis of Ph� ALL76 and are critical for

transition of CML to LBC.77 Whether BCR-ABL mutations that arecharacteristic of BCR-ABL� lymphoid leukemia in our cohortenhance the activation of SRC kinases and hence the transition to alymphoid phenotype is unknown.

Frequency of mutations that would influence the therapeuticdecision

In our cohort of patients with mutations, 166 of 386 (43%) had oneor more SGI clinically relevant mutations. T315I was detected in53 of 386 patients (14%). In 110 of 386 patients (28%), one or moreof their mutations were clinically relevant for either nilotinib ordasatinib, but not to both. For these patients, there may be anadvantage for one or other inhibitor. The remaining 3 of 386patients (0.8%) had 2 mutations, one of which was clinicallyrelevant for dasatinib and the other for nilotinib (neither