chronic myeloid leukemia - clinical 2

Chronic myeloid leukemia - Clinical 2

P881

DO CHRONIC MYELOID LEUKEMIA PATIENTS WITH LATE “WARNING”RESPONSES BENEFIT FROM SWITCHING THERAPY TO A SECONDGENERATION TYROSINE KINASE INHIBITOR?V García-Gutiérrez1,*, JM Puerta2, B Maestro3, LF Casado Montero4, JR MolinaHurtado5, M Perez-Encinas6, MV Moreno Romero7, I Massague8, R SolaGarcia9, R De Paz10, MJ Ramirez Sanchez11, S Osorio12, MI Mata Vazquez13,J Martinez14, JL Sastre15, MDLA Portero16, G Bautista17, MS Duran Nieto18,P Giraldo19, M Jimenez Jambrina20, C Burgaleta21, J Ruiz Aredondo22,MJ Peñarrubia23, MJ Requena24, MDC Fernández Valle25, C Calle26, A PazColl27, JÁ Hernández-Rivas28, R Franco Osorio29, P Cano30, D Tallón Pérez31,M Fernández de la Mata32, P López Garrido2, JL Steegmann331Servicio Hematología y Hemoterapia, Hospital Universitario Ramón y Cajal,Madrid, 2Servicio Hematología y Hemoterapia, Hospital Universitario Virgen delas Nieves, Granada, 3Registro Español de Investigación y Tratamiento deLeucemia Mieloide Crónica (RELMC),, Madrid, 4Hematology Dept, HospitalVirgen de la Salud, Toledo, 5Hematology Dept, Hospital Reina Sofía, Cordoba,6Servicio Hematología y Hemoterapia, Hospital Clinico U. de Santiago deCompostela, Santiago de Compostela, 7Servicio Hematología y Hemoterapia,Hospital Juan Ramón Jiménez, Huelva, 8Servicio Hematología y Hemoterapia,Hospital Vall de Hebron, Barcelona, 9Servicio Hematología y Hemoterapia,Hospital Clínico San Cecilio, Granada, 10Servicio Hematología y Hemoterapia,Hospital Universitario La Paz, Madrid, 11Servicio Hematología y Hemoterapia,Hospital de Jerez, Jerez, 12Servicio Hematología y Hemoterapia, HospitalUniversitario Gregorio Marañon, Madrid, 13Servicio Hematología yHemoterapia, Hospital Costa del Sol, Malaga, 14Servicio Hematología yHemoterapia, Hospital Universitario 12 de Octubre, Madrid, 15HematologyDept, Complejo Hospitalario de Orense, Orense, 16Servicio Hematología yHemoterapia, Hospital Virgen de la Macarena, Sevilla, 17Servicio Hematologíay Hemoterapia, Hospital Universitario Puerta de Hierro, Madrid, 18ServicioHematología y Hemoterapia, Hospital de Jaen, Jaen, 19Servicio Hematologíay Hemoterapia, Hospital Miguel Servet, Zaragoza, 20Servicio Hematología yHemoterapia, Hospital de Rio Tinto, Huelva, 21Servicio Hematología yHemoterapia, Hospital Universitario Principe de Asturias, Madrid, 22ServicioHematología y Hemoterapia, Hospital de Antequera, Malaga, 23ServicioHematología y Hemoterapia, Hospital Clinico Universitario de Valladolid,Valladolid, 24Servicio Hematología y Hemoterapia, Hispital Severo Ochoa,Madrid, 25Servicio Hematología y Hemoterapia, Hospital Puerta del Mar, Cadiz,26Servicio Hematología y Hemoterapia, Hospital General de Ciudad Real,Ciudad Real, 27Servicio Hematología y Hemoterapia, Hospital de Puerto Real,Cadiz, 28Servicio Hematología y Hemoterapia, Hospital Infanta Leonor, Madrid,29Servicio Hematología y Hemoterapia, Hospital Punta de Europa, Cadiz,30Servicio Hematología y Hemoterapia, Hospital La Mancha Centro, CiudadReal, 31Servicio Hematología y Hemoterapia, Hospital San Agustín, Jaén,32Servicio Hematología y Hemoterapia, Hospital Infanta Margarita, Córdoba,33Servicio Hematología y Hemoterapia, Hospital Universitario de la Princesa,Madrid, Spain

Background: In the latest recommendations of the European LeukemiaNetguidelines for the management of chronic-phase Chronic Myeloid Leukemiasuboptimal responses have been reclassified as “warning responses”. Incontrast to previous recommendations current guidance advises closemonitoring without changing therapy. There is little literature available regardingthe potential benefit of changing treatment of patients categorized as latewarning responders (patients with complete cytogenetic response (CCyR)without major molecular response (MMR)), and even more importantly, mostof the published data focuses on patients classified according to previouseditions of ELN recommendations (after 18 months of treatment).Aims: The aim of this study was to describe outcomes of late warning patients,classified by the latest ELN recommendations, in order to explore the potentialbenefit of treatment switching in these patients.Methods: Our registry includes a cohort of 945 CP-CML patients that aretreated with first-line imatinib and are not involved in clinical trials. Registrieswere approved by the corresponding Ethical committees, and all the procedureswere performed in accordance with international legislation. The monitoringand treatment strategies were chosen at the hematologist’s discretion and thusreflect the actual treatment of CML patients outside clinical trials. All data wasrecorded by data managers that operated independently from the physicianstaking care of the patients. From this cohort we identified 198 patients with awarning response after 12 months of treatment (patients with a completecytogenetic response but no major molecular response [MMR]). Patients whounderwent treatment change due to intolerance were identified but notcategorized as late warning responders.Results: One hundred and fourty six patients (group 1) remained on imatinib,while 52 patients (group 2) changed treatment to a Second Generation TyrosineKinase Inhibitor (2GTKI). The overall probabilities of obtaining MMR and MR4.5

by 24 months on an intention-to-treat basis were 10% vs. 21 % respectively,whereas by 48 months the corresponding probabilities were: 21% vs. 44%.

Overall survival (OS) was 97% vs. 92% by 24 and 48 months respectively,whereas progresion free survival (PFS) was 98% vs. 96% for the same periodsof time.Changing therapy resulted in a significant improvement in the probabilityof a MMR: 24% vs. 42% by 12 months and 43% vs. 64% by 24 months (p=.002);as well as the probability of achieving a deep molecular responses (MR4.5): 1%vs. 19% and 7% vs. 23% by 12 and 24 months respectively (p=<0.001).Treatment change was also associated with an increased stability of cytogeneticresponses, reflected by the observation that 17 patients (11%) in group 1 lostCCyR, compared to 2 patients (3%) in group 2 (p=.001). Similarly, treatmentchange significantly influenced the probability of remaining in MMR and MR4.5

at the last follow-up: 36% vs. 49% and 14% vs. 25% implying a relative risk of1.5 (1.2; 1.9) for MMR and a relative risk of 1.8(1.0; 3.3) for MR4.5 in the groupof patients who changed treatment vs. patients continuing imatinib. Theseimprovements did not correlate with an increase in overall survival (OS) orprogression-free survival (PFS).Treatment change was generally well tolerated.However, 10 patients in Group 2 (19%) discontinued treatment due to side effectswith long-term side effects leading to 4 patients (2%) discontinuing imatinib.Summary and Conclusion: To our knowledge our study represents the largeststudy of late warning patients treated with imatinib (including the publisheddata regarding outcomes of late suboptimal responders under the oldclassification scheme). Our study shows how late warning responders had anexcellent prognosis in terms of PFS and OS, and offers an empiricaldemonstration of the adequacy of classifying these responses as a warning, ifsurvival outcomes are the objectives of the treatment. However, if the objectivesinclude lowering (mitigating) the probability of treatment failure as well asobtaining deeper molecular responses, our study demonstrates that switchingto a 2GTKI is the preferred option.

P882

SENSOR INTERIM DATA WITH MUTATION ANALYSIS: SWITCHING TONILOTINIB AFTER MOLECULAR SUBOPTIMAL RESPONSE TO IMATINIBIN PATIENTS WITH CHRONIC MYELOID LEUKEMIA IN CHRONIC PHASEK Yamamoto1,*, K Miyamura2, T Miyamoto3, M Tanimoto4, M Taniwaki5,S Kimura6, K Ohyashiki7, T Kawaguchi8, I Matsumura9, T Hata10, H Tsurumi11,S Saito12, M Hino13, S Tadokoro14, K Meguro15, H Hyodo16, M Yamamoto17,K Kubo18, J Tsukada19, M Kondo20, T Amagasaki20, E Kawahara20, M Yanada211Aichi Cancer Center, 2Japanese Red Cross Nagoya Daiichi Hospital, Nagoya,3Kyushu University Hospital, Fukuoka, 4Okayama University Graduate Schoolof Medicine, Okayama, 5Kyoto Prefectural University of Medicine, Kyoto, 6SagaUniversity, Saga, 7Tokyo Medical University, Tokyo, 8Kumamoto UniversityGraduate School of Medical Sciences, Kumamoto, 9Kinki University Faculty ofMedicine, Osaka-Sayama, 10Nagasaki University Hospital, Nagasaki, 11GifuUniversity Graduate School of Medicine, Gifu, 12Nagoya University GraduateSchool of Medicine, Nagoya, 13Osaka City University, Osaka, 14OsakaUniversity Graduate School of Medicine, Suita, Osaka, 15National HospitalOrganization Sendai Medical Center, Sendai, 16Research Institute for RadiationBiology and Medicine, Hiroshima University, Hiroshima, 17Tokyo Medical andDental University, Tokyo, 18Aomori Prefectural Central Hospital, Aomori,19University of Occupational and Environmental Health, Kitakyushu, 20NovartisPharma KK, Tokyo, 21Fujita Health University School of Medicine, Toyoake,Aichi, Japan

Background: Clinical trials have shown that nilotinib (NIL) is superior to imatinib(IM) as frontline treatment for chronic myeloid leukemia in chronic phase (CML-CP) and can improve responses in pts resistant or intolerant to IM, but the optimaltreatment for patients (pts) with suboptimal response (SoR) to IM is unclear.Aims: The Study to Evaluate Nilotinib in CML pts with SubOptimal Response(SENSOR, NCT0104387) was designed to evaluate the safety and efficacy ofNIL in pts with CML-CP with SoR to frontline IM. Results of the planned 12-mointerim analysis are presented.Methods: SENSOR is a multicenter, phase 4, open-label study in which adultpts with SoR to firstline IM switched to NIL 400 mg twice daily, with planned 24-mo follow-up. Per European LeukemiaNet 2009 criteria, SoR was defined ascomplete cytogenetic response, but no major molecular response (MMR; BCR-ABL ≤ 0.1% on the International Scale [IS]) after at least 18 mo. MMR rate at12 mo was the primary endpoint. BCR-ABLIS transcript levels were assessedby a central laboratory at screening, monthly for 3 mo, and every 3 mothereafter. Mutation analyses (using direct sequencing and the polymerasechain reaction invader method), were conducted centrally at baseline (BL) andend of study in all pts; additional analyses were done every 3 mo for pts withBL mutations and in pts not in MMR with a ≥ 5-fold increase in BCR-ABL fromthe lowest level. The invader method, introduced to detect 25 IM-resistant pointmutations with increased sensitivity, cannot detect other mutations or alternativesplicing abnormalities.Results: Between Dec 2009 and Feb 2012, 45 pts enrolled. At BL, medianBCR-ABL/ABLIS ratio was 0.24% (range, 0.11%>3.49%) after a median of 23.0mo (range, 17.0-103.3) of IM. At the data cutoff, 19 pts (42%) completed the24-mo study, 21 (47%) were ongoing, and 5 (11%) discontinued. At 12 mo, 51%of pts achieved MMR and 4.4% achieved BCR-ABLIS ≤ 0.0032%. Thecumulative rate of MMR by 12 mo was 67%. Most pts with mutations at any time(12/15) achieved MMR. Direct sequencing detected point mutations and

330 | haematologica | 2014; 99(s1)

19th Congress of the European Hematology Association

alternative splicing abnormalities, such as exon 7 deletion (del) and exon 8/935 base pair insertion (bp ins). Twelve pts had newly detected mutations duringthe study; of these, 9 pts achieved MMR (6 with exon 8/9 35 bp ins, and 1 eachwith either E459K, Y393C, or T319A). These mutations were identified by directsequencing, not the invader method. One pt had a newly detected T315Imutation using both methods; this pt progressed at 5.4 mo and died at 9.4 mo,never achieving MMR. This was the only pt who progressed or died. Thus, the12-mo progression-free survival and overall survival (95% CI) were 97.8%(85.3%>99.7%) and 97.7% (84.9%>99.7%), respectively. Safety data wereconsistent with other studies of pts who switched to NIL after prior IM. The mostcommon any-grade drug-related adverse events were hyperbilirubinemia(53.3%), increased ALT level (28.9%), headache (28.9%), hypophosphatemia(26.7%), rash (26.7%), and increased lipase level (24.4%).

Table 1.

Summary and Conclusion: After switch to NIL for 12 mo, 51% of pts with SoRto frontline IM achieved MMR. Additional follow-up is required to evaluate long-term outcomes. These results support data from other international studies (eg,ENESTcmr) showing that switch to NIL improves responses in pts withdetectable disease on IM. Direct sequencing and the invader methodcomplemented each other in detection of certain point mutations, but directsequencing detected additional mutations and alternative splicing abnormalitiesin pts with SoR to frontline IM. Most pts in this study with point mutationsdetected at any time (with any method) achieved MMR with NIL.

P883

TYROSINE KINASE INHIBITOR (TKI) SWITCHING: EXPERIENCE FROMSIMPLICITY, A PROSPECTIVE OBSERVATIONAL STUDY OF CHRONIC-PHASE CHRONIC MYELOID LEUKEMIA (CP-CML) PATIENTS IN CLINICALPRACTICER Hehlmann1,*, J Cortes2, C Gambacorti-Passerini3, S L Goldberg4, HJ Khoury5,M Mauro6, M Michallet7, H Mohamed8, T Powell9, R Paquette10, B Simonsson11,M Subar8, T Zyczynski81Universität Heidelberg, Mannheim, Germany, 2The University of Texas, MDAnderson Cancer Centre, Houston, United States, 3University of Milano Biocca,San Gerardo Hospital, Monza, Italy, 4John Theurer Cancer Center, HackensackUniversity Medical Center, Hackensack, United States, 5Emory University,Atlanta, United States, 6Memorial Sloan-Kettering Cancer Center, New York,United States, 7Centre Hospitalier Lyon-Sud, Lyon, France, 8Bristol-MyersSquibb Co, Princeton, United States, 9ICON Clinical Research, Biostatistics &Programming, San Diego, United States, 10UCLA Medical Center, Los Angeles,United States, 11Uppsala Universitet, Upsala, Sweden

Background: Few data from clinical practice describe treatment switchingpatterns in patients (pts) with CP-CML treated with TKIs.Aims: To investigate patterns of, and reasons for, TKI switching in pts with CP-CML who discontinued first-line treatment.Methods: SIMPLICITY is an ongoing observational cohort study of adult ptswith newly diagnosed CP-CML receiving first-line treatment with imatinib (IM),dasatinib (DAS) or nilotinib (NIL) in Europe (Eu) and the United States (US) andoutside of clinical trials (NCT01244750). The primary study objective is tounderstand TKI management patterns in clinical practice. The study includesthree prospective cohorts of pts treated with IM, DAS or NIL as initial therapysince 2010 (the study opened after first-line approval of all three TKIs) and ahistorical cohort treated with IM since 2008. Data on treatment discontinuationand treatment switching in all pts with ≥12 months of follow-up are presentedfor prospective cohorts.

Results: 949 pts (Eu: 34%, US: 66%) were enrolled through 3 October 2013,initially treated with IM (N=415), DAS (N=275) or NIL (N=259). Median follow-up was 1.6 years. Demographics were consistent across all cohorts (median age:56 years of age at first-line TKI, 56% male). Across all regions, 733 of 949 ptswere followed for ≥12 months after initiation of first-line TKI (IM: n=375, DAS:n=178 or NIL: n=180). Of these, 30.1%, 17.4% and 22.8% of patientsdiscontinued IM, DAS and NIL, respectively, within a year of initiation. The mainreason for TKI discontinuation was physician-reported intolerance (IM: 62.8%,DAS: 80.6%, NIL: 87.8%). Physician-reported primary resistance, leading todiscontinuation, was only noted in IM-treated pts (8.8%). Median time to firstdiscontinuation varied by TKI (IM: 127 days, DAS: 129 days, NIL: 56 days). Aproportion of pts who discontinued first-line TKI within 12 months switched to asecond-line TKI (IM: 70.8%, DAS: 54.8%, NIL: 58.5%), while no further TKItreatment information was available for the remaining pts at the time of data lock(IM: 29.2%, DAS: 45.2%, NIL: 41.5%). Of pts who switched to a second-line TKIwithin 12 months, 55.0% and 45.0% of IM-treated pts switched to DAS and NIL,respectively; most DAS-treated pts switched to IM (70.6% vs. 29.4% to NIL)and 54.2% and 45.8% of NIL-treated pts switched to DAS and IM, respectively.In Eu, the proportion of pts discontinuing was highest in the IM cohort (IM: 27.7%,DAS: 6.1%, NIL: 11.4%); in the US, the proportion of pts discontinuing was moreevenly distributed (IM: 31.6%, DAS: 20.0%, NIL: 26.5%). Few pts, all in the US,discontinued therapy due to financial reasons (IM: 6.8%, DAS: 3.4%, NIL: 2.8%).Median time to first discontinuation was consistently higher in the US, with theexception of median time to discontinuation of IM, which was longer in Eu (130.0vs. 122.5 days in the USA). Among pts in Eu who received a second-generationTKI as second-line therapy, NIL predominated (16/26 NIL vs. 10/26 DAS),whereas, in the US, DAS predominated (47/78 DAS vs. 25/71 NIL).Summary and Conclusion: The proportion of pts discontinuing first-linetreatment for CP-CML, and reasons for discontinuation, vary by TKI. Whileintolerance was the primary reason for treatment discontinuation in all TKIcohorts during the first 12 months, primary resistance was reported in IM-treatedpts only. Additional analyses will be presented that identify further differencesbetween TKI cohorts, as well as types of intolerances leading to first-linetreatment discontinuation.

P884

SAFETY AND TOLERABILITY OF DASATINIB IN PATIENTS WITHCHRONIC MYELOID LEUKEMIA (CML) AND PHILADELPHIACHROMOSOME–POSITIVE ACUTE LYMPHOBLASTIC LEUKEMIA (PH+ALL): POOLED ANALYSIS OF OVER 2400 PATIENTSG Saglio1,*, P le Coutre2, J Cortes 3, J Mayer 4, P Rowlings 5, M Subar 6,J Preston7, N Shah81University of Torino, Ospedale San Luigi Gonzaga, Orbassano-Torino, Italy,2Charité, Campus Virchow Klinikum, Universitätsmedizin Berlin, Berlin,Germany, 3The University of Texas MD Anderson Cancer Center, Houston,Texas, United States, 4Department of Internal Medicine, Hematology andOncology, University Hospital Brno, Brno, Czech Republic, 5Calvary MaterNewcastle Hospital, University of Newcastle, Waratah, New South Wales,Australia, 6Bristol-Myers Squibb, Lawrenceville, New Jersey, 7Bristol-MyersSquibb, Hopewell, New Jersey, 8UCSF School of Medicine, San Francisco,California, United States

Background: CML has become a manageable chronic disease for the majorityof patients (pts). However, several rare, severe, and irreversible adverse events(AEs) have been reported in tyrosine kinase inhibitor (TKI)-treated populations.To better understand the incidence of such events previously reported in thecontext of individual trials, a large pooled dataset from Ph+ leukemic pts treatedwith dasatinib was interrogated for the incidence of reversible or irreversiblenonhematologic AEs.Aims: The objective of this analysis is to report clinical safety data from a largeadult population of dasatinib-treated pts (n=2440) derived from 9 single-arm orcomparative clinical trials of CML or Ph+ ALL resistant or intolerant to imatinib(IM; n=2182) or newly diagnosed CML in chronic phase (CML-CP; n=258).Methods: All pts who received ≥1 dose of dasatinib 100 mg once daily, 140 mgonce daily, 50 mg twice daily, or 70 mg twice daily are included. Duration oftreatment was defined as the interval from the first to the last dose of study drugand may include intermittent dose interruptions. AEs were graded using NCICTCAE Version 3.0, and investigator AE terms were coded and grouped bySystem Organ Class using MedDRA Version 12.1. AEs that were determinedby the investigator to have a certain, probable, or possible relationship to thestudy drug were defined as drug-related. Very common AEs (frequency ≥10%)and AEs of interest (eg, AEs with known association with dasatinib or other TKIs,clinical significance warranting evaluation, or nonclinical data suggesting anassociation) were analyzed.Results: The median duration of therapy was 37 months (range<1 to 50months) in pts with newly diagnosed CML-CP (n=258) and 15 months (range0 to 65.6 months) in pts with IM-resistant or -intolerant CML or Ph+ ALL(n=2182). In this pooled analysis, minimum follow-up was 3 years and 5 yearsin the first-line DASISION (n=258) and second-line CA180-034 (n=662) CML-CP trials, respectively. On-study drug-related AEs with frequency ≥10% (anygrade; n=2440) were diarrhea, pleural effusion, headache, dyspnea, rash,

haematologica | 2014; 99(s1) | 331

Milan, Italy, June 12 – 15, 2014

fatigue, nausea, peripheral edema, musculoskeletal pain, hemorrhage, pyrexia,vomiting, abdominal pain, infection, and cough. Drug-related AEs, such aspleural effusion, thrombocytopenia, or neutropenia, led to discontinuation in16% of the 2440 pts. Fluid-related AEs (drug-related, any grade) were reportedin 44% of pts and included: pleural effusion (30%), superficial edema (22%),pericardial effusion (5%), generalized edema (4%), congestive heartfailure/cardiac dysfunction (3%), pulmonary edema (2%), pulmonaryhypertension (1%), and ascites (0.5%). The single pulmonary arterialhypertension AE reported was not considered drug-related. Cardiac AEs,including ischemic AEs, rhythm abnormalities, and congestion were infrequent(Table), as were other AEs that may be related to vascular thrombosis. All-cause cerebrovascular accidents and transient ischemic attacks (TIAs) werereported in 0.6% and 0.4% of pts, respectively, and were rarely drug-related.Peripheral arterial occlusive disease (PAOD) and PAOD-related AEs have beenreported (le Coutre Blood 2013;122:1489) in 0.3% of pts in a larger, 11-studydataset (n=2705).

Table 1. Drug-related cardiac AEs

Summary and Conclusion: Analysis of a large pooled dataset from dasatinibclinical trials did not identify new safety signals. Although pleural effusions areconsidered common (frequency >10%), they are generally reversible. AEs thatare potentially irreversible, including vascular occlusive AEs, were rarelyreported in dasatinib-treated pts.

P885

DASATINIB PHARMACOKINETICS AND ITS CORRELATION WITHCLINICAL RESPONSE IN CHRONIC MYELOID LEUKEMIA: ASUBANALYSIS OF THE DARIA-01 STUDYK Miyao1,*, M Sawa1, S Mizuta2, K Matsumoto3, H Tsurumi4, T Takahashi5,T Hara4, R Sakemura6, A Kohno7, H Kojima8, K Ohbayashi9, M S Oba10,S Morita11, J Sakamoto12, N Emi21Hematology&Oncology, Anjo Kosei Hospital, Anjo, 2Hematology&MedicalOncology, Fujita Health University Hospital, Toyoake, 3Doshisha Women’sCollege of Liberal Arts, Kyoto, 41st Department of Internal Medicine, GifuUniversity Graduate School of Medicine, 5Hematology, Gifu Municipal Hospital,Gifu, 6Hematology&Oncology, Nagoya University Graduate School of Medicine,Nagoya, 7Hematology&Oncology, Konan Kosei Hospital, Konan, 8Hematology,Daido Hospital, Nagoya, 9Hematology, Nishio Munisipal Hospital, Nishio,10Biostatistics and Epidemiology, Yokohama City University Graduate Schoolof Medicine, Yokohama, 11Biomedical Statistics and Bioinformatics, KyotoUniversity Graduate School of Medicine, Kyoto, 12Tokai Central Hospital,Kakamigahara, Japan

Background: In the International Randomized Study of IFN versus STI571study, imatinib plasma exposure, measured following the first month oftreatment with the standard dose, correlated with therapeutic efficiency.However, there have been a few reports regarding the relationship betweendasatinib plasma concentration and clinical results.Aims: To characterize the relationship between dasatinib pharmacokineticexposure, efficacy, and tolerability with data from the DARIA-01 study. Methods: We conducted a multicenter, open label, phase 2 study of mid-termcontinuation and effectiveness of dasatinib therapy in patients with chronicmyeloid leukemia in chronic phase (CML-CP). Plasma trough levels wereobtained on day 28 (steady state).Results: Thirty-two CML-CP patients were included in the study. The median

age was 52.5 years (range 20–86). Twenty-four (75.0%) previously untreatedCML-CP patients and the remaining eight (25.0%) were switched from othertyrosine-kinase inhibitors (imatinib, five patients; nilotinib, three patients)because of intolerance or resistance to these drugs. Dasatinib treatment wasinitiated at a dose of 100 mg/day. During the initial 28 days, 28 patientscontinued with the 100 mg/day treatment and four received doses of 50 mg/day.Age significantly correlated with Cmin on day 28 (p=0.0138), and was moresignificant after an adjustment in the dasatinib dose according to the weight ofthe patients on day 28 (C/D ratio; p=0.0088; Fig.1). The C/D ratio on day 28(>1.6) significantly correlated with a higher incidence of pleural effusion duringthe first month (p=0.049) and/or dose reduction or interruption during the initial3 months (p=0.009). On the other hand, the higher C/D ratio on day28 was notstatistically-significant increase of major molecular response achievement at 3months.

Figure 1.

Summary and Conclusion: Our finding reveals that the C/D ratio on day 28appeared to have a predictive value for pleural effusion and dose reduction orinterruption during the initial 3 months. Thus, pharmacokinetic monitoring couldbe beneficial in determining optimal therapy conditions for CML treatment,particularly in elderly patients.

P886

LONG-TERM FOLLOW-UP OF PATIENTS WITH CHRONICMYELOGENOUS LEUKEMIA IN COMPLETE MOLECULAR RESPONSEWITH ALPHA-INTERFERON AFTER TREATMENT DISCONTINUATIONR Latagliata1,*, A Romano1, M Mancini1, M Breccia1, I Carmosino1, F Vozella1,C Montagna1, P Volpicelli1, L Petrucci1, A Serao1, M Molica1, A Salaroli1,D Diverio1, A Tafuri1, G Alimena11Cellular Biotechnologies and Hematology, University Sapienza, Rome, Italy

Background: Treatment with alpha-interferon (IFN) has been widely employedbefore the advent of Tyrosine-kinase inhibitors in patients (pts) with ChronicMyelogenous Leukemia (CML), leading to the achievement in few cases of aprolonged Complete Molecular Response (CMolR) and to IFN discontinuation.Aims: To evaluate the long-term follow-up after IFN discontinuation, 23 pts(M/F 11/12) who achieved CMolR at RT- nested PCR with IFN at our Institutionhave been revised.Methods: Median age at diagnosis was 43,3 years [Interquartile Range (IR)35.8 – 51.3], Sokal score was low in 18 pts, Intermediate in 3 and not evaluablein 2, median WBC at diagnosis was 39.9 x 109/l (IR 24.4 – 67.6). IFN wasgiven alone in 16 pts or in association with autologous bone marrowtransplantation (ABMT) in 7 pts (ABMT at onset followed by IFN in 4 pts, IFNfollowed by ABMT in 3 pts). Median time to Complete Cytogenetic Responsewas 21.4 months (IR 14.4 – 37.4), median time to CMolR was 63.7 months (IR30.3 – 106.0).Results: After a median period of IFN treatment of 105.8 months (IR 56.1 –127.3), all pts discontinued IFN due to medical decision for prolonged CMolR(12 pts), intolerance (8 pts) or planned ABMT (3 pts). After 12.5 months fromIFN discontinuation, 1 patient developed a sudden extramedullar lymphoidblast crisis and died from disease progression. Four pts needed to start a newtreatment with imatinib [2 for cytogenetic relapse after 24,8 and 44,0 months,1 for molecular relapse after 39,8 months and 1 for progressive rise in moleculartranscript but still in Major Molecular Response (MMolR) after 39,7 months], allachieving a new molecular response which was complete in 3 out 4. Theremaining 18 pts are still off-therapy after a median time from IFNdiscontinuation of 125.5 months (IR 86.9 – 205.3); among them, 5 pts resultedalways negative at the molecular follow-up, 6 pts presented a sporadic positivitybut always resulted in MMolR with bcr-abl ratio<0.1 and 7 showed a mild rise



of transcript levels with a long-lasting stable positivity (bcr-abl ratio<0,5 withoutfurther increments). At the last molecular evaluation, 11/18 pts were in CMolR,4/18 in MMolR and 3 had bcr-abl ratio between 0.5 and 0.1. The 5-year and 10-year cumulative overall survival was 95.5%. The cumulative 5-year event-freesurvival was 77.4%, with a stable plateau in the subsequent follow-up (lastevent recorded 39.8 months after IFN discontinuation).Summary and Conclusion: Our data show that CML pts who achieved aprolonged CMolR with IFN and discontinued the treatment had a very low riskof disease recurrence; it is worth of note that in many cases the reappearanceof a bcr-abl positivity<0.5 did not precede a disease relapse but was sporadicor stable over long time, suggesting a possible role for immunologicalmechanisms induced by IFN.

P887

FIRST LINE TREATMENT OF CHRONIC PHASE CHRONIC MYELOIDLEUKEMIA PATIENTS WITH THE GENERIC FORMULATIONS OF IMATINIBMESYLATEAE Eskazan1, Z Baslar1,*, M Ayer2, B Kantarcioglu3, D Arica3, N Demirel3,D Aydin3, FF Yalniz1, T Elverdi1, A Salihoglu1, MC Ar1, S Ongoren Aydin1,Y Aydin1, N Tuzuner4, U Ozbek5, T Soysal11Department of Internal Medicine, Division of Hematology, Istanbul UniversityCerrahpasa Faculty of Medicine, 2Department of Hematology, Haseki Trainingand Research Hospital, 3Department of Hematology, Okmeydani Training andResearch Hospital, 4Department of Pathology, Istanbul University CerrahpasaFaculty of Medicine, 5Department of Genetics, Istanbul University Institute ofExperimental Medicine (DETAE), Istanbul, Turkey

Background: The high cost of tyrosine kinase inhibitors (TKIs) developed forchronic myeloid leukemia (CML) is a major concern for the health care payers,especially in countries with restricted resources. Reimbursement policiesencourage generic drug use to lower the prices. It is true that generics lead toconsiderable cost savings but they also give rise to questions associated withtheir efficacy, safety and quality. In Turkey, there are three commerciallyavailable generics of imatinib mesylate (IM) in the market, which are approvedfor the treatment of CML. There is a price difference between the original IM(Glivec) and generics, and due to the reimbursement policy in Turkey, patientswho prefer to receive the original molecule, should pay the price difference.Aims: The aim of this multicenter study was to evaluate the efficacy andtolerability of generics of IM, and to compare these with Glivec when usedamong patients with chronic phase (CP) CML in the upfront setting.Methods: There were two study groups, thirty-six patients who started TKItreatment between January 2010 and June 2013 with Glivec (Group A), and 26patients who were diagnosed between August 2012 (when generics becameavailable in Turkey) and December 2013, in which a generic of IM was initiatedupfront (Group B). Patients’ demographics, imatinib dose, Sokal risk scores,adverse events (AEs), and follow-up periods were noted from the patients’ filesretrospectively. Imatinib response was evaluated according to the criteriarecommended by the European LeukemiaNet. Molecular response (MR) wasclassified based on BCR-ABL1 to control gene transcript ratios, expressed onthe International Scale. Major molecular response (MMR) was defined as ratios≤0.1%, and the MMR and complete cytogenetic response (CCyR) rates of botharms were calculated.Results: The two groups were balanced regarding age, gender, and Sokal riskscores (Table 1). Patients in group A had a longer median follow-up under IMthan patients in group B (20 months vs. 8.5 months). There were five patientsin group B with a follow-up duration less than 6 months. They all have completehematological response (CHR), but while calculating the CCyR and MMR rates,these patients were excluded. The CCyR rates at 6 months for groups A and Bwere 56% and 52%, respectively. There was no significant difference betweenthe groups regarding MMR rates at 6 months of IM treatment (33% vs. 33%).During the follow-up, four patients in group A, and 4 in group B were switchedto 2nd generation TKIs due to resistance (p=0.623). The rates of hematologicaland non-hematological AEs were similar in both groups (Table 1), and thepercent of patients who needed a lower dose of IM (300 mg daily) due to AEswere 14% and 12% for groups A and B, respectively. There was no switch to2nd generation TKIs in the both study groups due to intolerance. After a medianfollow-up of 29 months under the original molecule, eight patients were switchedto a generic due to the reimbursement policy. They were all in MMR at the timeof the switch, and they maintained their responses during the follow-up.Summary and Conclusion: TKIs are now the mainstay of treatment of CML,and patients with CML live close to normal life spans. The current prices of TKIsare high, and the launch of generics might reduce health care costs. Among ourpatient cohort, the generics were at least non-inferior to the original moleculeregarding efficacy and tolerability when used in the upfront setting. Prospectiverandomized trials with larger number of patients are needed to address theefficacy of generics of IM in patients with CML.

Table 1. The characteristic of patients in both groups (AE, adverse event;CCyR, complete cytogenetic response; GI, gastrointestinal; IM, imatinibmesylate; MMR, major molecular responser) *There were 3 patients withmore than one AE (one with both GI symptoms and muscle cramps, andthe other had edema and muscle cramps), so the rates of non-hematolog-ical AEs in this patient group were added up to >100%.

P888

LONG-TERM RESULTS OF A PROSPECTIVE RANDOMIZED TRIAL OFHIGH-DOSE IMATINIB WITH OR WITHOUT INTERFERON-Α IN PATIENTSWITH NEWLY DIAGNOSED CHRONIC MYELOID LEUKEMIA IN CHRONICPHASEK Sasaki1,*, H Kantarjian1, E Jabbour1, F Ravandi1, G Borthakur1, W Wierda1,G Garcia-Manero1, Z Estrov1, S Pierce1, S O’Brien1, J Cortes11Department of Leukemia, MD Anderson Cancer Center, Houston, UnitedStates

Background: Before tyrosine kinase inhibitors (TKIs) became the standardfrontline therapy, interferon-α (IFN-α) achieved durable cytogenetic responsesin occasional patients with chronic myeloid leukemia in chronic phase (CML-CP).Trials of the combination of imatinib and IFN- α have yielded conflicting results.Aims: We report the long-term outcome of a randomized trial of imatinib ± IFN-α, with a minimum follow-up of 104 months (last patient, first visit 06/2005).Methods: We conducted a prospective, randomized clinical trial on newlydiagnosed CML-CP randomized to receive 800 mg of imatinib daily with(IM800+IFN; n=43) or without (IM800; n=48) weekly subcutaneous injectionsof 0.5 µg/kg polyethylene glycol (PEG)–IFN-α-2b. Administration of IFN-α-2bstarted 6 months after the start of imatinib for patient randomized to that arm.Patients aged ≥18 years with CML-CP were eligible if performance status ≤2by the Eastern Cooperative Oncology Group, serum creatinine and totalbilirubin<1.5 x the upper limit of normal. The Kaplan-Meier method was usedto calculate overall survival (OS; dated from the start of imatinib until deathfrom any cause), event-free survival (EFS; from the start of imatinib to loss ofcomplete hematologic response, loss of major cytogenetic response,transformation to accelerated [AP] or blast phase [BP], or death from anycause), transformation-free survival (TFS; from the start of imatinib totransformation to AP or BP or death), and failure-free survival (FFS; from thestart of imatinib to any event defined above, imatinib discontinuation for anyreason, or death.) The chi-square test was used to compare response ratesbetween groups and a log-rank test was used for univariate comparisons.Results: The median follow-up duration was 107 months (range, 6-121months). Cumulative response rates and rates of OS, EFS, TFS, and FFS aregiven in Table 1. The median duration of PEG–IFN-α-2b therapy was 335 days(range, 7-1191 days), and all patients discontinued PEG–IFN-α-2b. Thecomplete cytogenetic response rate (CCyR), major molecular response (MMR),major response ≥4.5-log reduction (MR4.5), and complete molecular responserate(CMR) at 6 months after high-dose imatinib treatment before PEG–IFN-α-2b therapy in IM800 + IFN group were 74%, 60%, 21% and 2% compared tocomparable rates of 83%, 67%, 21% and 6% in IM800 group,respectively(P=0.877). The CCyR, MMR, MR4.5 and CMR at 12 months fromstart of all therapy in IM800 + IFN group were 77%, 65%, 26% and 7%compared to comparable rates of 83%, 73%, 21% and 10% in IM800 group,respectively (P=0.888). The groups’ best response rates and rates of 10-yearFFS, TFS, EFS, and OS did not differ significantly.



Table 1. Patient Characteristics and Outcomes

IM800 + IFN [n=43] IM800 [n=48] P value

Cumulative Response, %

CMR 58 60 0.825MR4.5 67 75 0.426MMR 81 88 0.420CCyR 86 88 0.838

10-y Outcome, %

FFS 60 67 0.436TFS 97 95 0.639EFS 61 88 0.094OS 88 91 0.873

Summary and Conclusion: The addition of PEG–IFN-α-2b to high-doseimatinib as a frontline therapy does not confer a clinical benefit in patients withCML-CP.

P889

FOUR-YEAR FOLLOW-UP OF PATIENTS WITH IMATINIB-RESISTANT OR-INTOLERANT CHRONIC-PHASE CHRONIC MYELOID LEUKEMIARECEIVING DASATINIB: EFFICACY AND SAFETYX Huang1, J Hu2, J Li3, J Jin4, F Meng5, Z Shen6, T Liu7, D Wu8, J Wang9,J Wang10,*1Peking University People’s Hospital, Beijing, 2Fujian Medical University UnionHospital, Fuzhou, 3The First Affiliated Hospital with Nanjing Medical University,Nanjing, 4The First Affiliated Hospital of The College of Medicine, ZhejiangUniversity, Hangzhou, 5Guangzhou Nanfang Hospital, Guangzhou, 6ShanghaiRuijin Hospital, Shanghai, 7West China Hospital, Sichuan University, Chengdu,8The First Affiliated Hospital of Soochow University, Suzhou, 9ChanghaiHospital of Shanghai, Shanghai, 10The Institute of Hematology & BloodDiseases Hospital, Chinese Academy of Medical Science & Peking UnionMedical College, Tianjin, China

Background: The Chinese dasatinib registration study, initiated in 2007,enrolled 119 chronic myeloid leukemia (CML) patients (pts) intolerant of, orresistant to, imatinib, including 59 pts with chronic-phase (CP) CML. After 18months of follow-up, 50.8% and 91.5% of pts with CML-CP achieved a majorcytogenetic response (MCyR) or complete hematologic response (CHR),respectively. Furthermore, none of the CML-CP pts who achieved a MCyRdied or experienced disease progression at 18 months. The safety findingssupported prior Phase 1/2 trials, with common adverse events (AEs) includingneutropenia and thrombocytopenia. Here we present the efficacy and safetyresults from the 4-year follow-up of this pivotal Chinese study.Aims: To evaluate the long-term efficacy and safety of dasatinib for thetreatment of Chinese pts with CML-CP who were resistant or intolerant toimatinib, and to investigate the association between pre/on-treatment factorsand treatment efficacy after 4 years of follow-up.Methods: This was a single-arm, Phase 2 study conducted at 10 centers inChina. Adult, Chinese pts with imatinib-resistant or -intolerant CML-CP receivedoral dasatinib 100 mg/d until disease progression, unacceptable toxicity, or atthe investigator/patient’s discretion. The primary endpoint was the rate of MCyRand secondary endpoints included time to MCyR, rate of progression-freesurvival (PFS), and incidence of AEs. Pts who discontinued treatment were notroutinely followed up, therefore disease progression and survival data are notavailable for these pts post study departure. The association between pre- andon-treatment factors and complete cytogenetic response (CCyR) wasinvestigated by univariate regression. Written, informed consent was obtainedfrom all pts and the study was registered at clinicaltrials.gov (NCT00529763).Results: After 4 years of follow-up, 72.9% (43/59) of pts remained on treatment.The cumulative rate of the primary endpoint, MCyR, was 66.1% (39/59; 95%CI, 52.6–77.9); an increase of 30% from that observed at 18 months (50.8%[30/59]). Median time to MCyR was 12.7 weeks (range, 4.3–206.1).Interestingly, all pts who achieved MCyR also met the criteria for CCyR.The 4-year PFS was 85.7% (95% CI, 77.0–95.4), and rates at 12, 24 and 36months were 94.7%, 91.2% and 85.7%, respectively; no pts experienceddisease progression between 36 and 48 months. After 4 years of follow-upthere had been 5 deaths on study, none of which were considered to betreatment-related. None of the 20 pts who achieved an early response (MCyRwithin 90 days of treatment initiation) had experienced disease progressionafter 4 years (Figure), and these pts had a greater likelihood of achieving aCCyR than those without an early response (odds ratio 42.1, 95% CI2.37–746.84; P=0.01). Dasatinib was generally well tolerated; 11.9% (7/59) ofpts discontinued treatment due to AEs. Drug-related AEs of any grade wereexperienced by 69.5% (41/59) of pts, and Grade 3–4 by 18.6% (11/59).Reported AEs included pleural effusion (23.7% [14/59]; Grade 3–4: 3.4%[2/59]), thrombocytopenia (11.9% [7/59]; Grade 3–4: 11.9% [7/59]), andneutropenia (6.8% [4/59]; Grade 3–4: 6.8% [4/59]). Pulmonary hypertensionwas reported in 6.8% (4/59) of pts (Grade 3–4: 1.7% [1/59]).

Figure 1. Kaplan-Meier PFS curve by early response (MCyR within 90days of treatment initiation; green line) or no early response (red line).

Summary and Conclusion: This 4-year follow-up study supports dasatinibtreatment in Chinese pts with imatinib-resistant or -intolerant CML-CP: 72.9%of pts remained on treatment, 66.1% achieved a MCyR, and high rates of PFSwere sustained from 36 months through 4 years. Dasatinib was well tolerated,with a low rate of discontinuations and a safety profile that supports previousclinical trial data.

P890

ABNORMALITY OF GLUCOSE AND LIPID METABOLISM DURINGNILOTINIB THERAPY – RESULTS OF ENIGMA 2 STUDYL Semerad1,*, Z Racil1, P Belohlavkova2, O Cerna3, P Cetkovsky4, E Faber5,H Klamova4, L Malaskova6, R Minarik1, J Prochazkova1, T Sacha7, J Voglova2,D Zackova1, P Zak2, J Mayer11Department of Internal Medicine, Hematology and Oncology, MasarykUniversity and University Hospital Brno, Brno, 24th Department of InternalMedicine-Hematology, Charles University and University Hospital HradecKralove, Hradec Kralove, 3Department of Clinical Hematology, CharlesUniversity and University Hospital Kralovské Vinohrady, 4Institute ofHematology and Blood Transfusion, Praha, 5Department of Hemato-Oncology,Palacký University and University Hospital Olomouc, Olomouc, 6Departmentof Clinical Biochemistry, University Hospital Brno, Brno, Czech Republic,7Department of Hematology, University Hospital Kraków, Kraków, Poland

Background: Recently we have published results of pilot study on CMLpatients demonstrating fast development of hyperinsulinaemia, peripheralinsulin resistance, hypoadiponectinaemia and hypercholesterolemia duringnilotinib therapy.Aims: To analyze preliminary results from follow up study ENIGMA 2 to confirmor exclude result from the pilot study, as well as to analyze whether theseabnormalities are detected in control group of patients treated with imatinib.Methods: Patients received intensive laboratory workup before the start of TKIand after 3 month of therapy. This included fasting insulin, glucose, adiponectinand lipid serum concentration, HbA1c and oral glucose tolerance test.Results: Between 2/2011-11/2013 twenty two CML patients initiated therapywith nilotinib and 4 with imatinib. Patients treated with nilotinib developedsignificant hypersinulinaemia and hyperglycaemia. We also proved significantdevelopment of insulin resistance using HOMA-2 index, that significantlyincreased after 3 month of nilotinib therapy (medians - 1,8 vs. 2,5; p=0,0098).More over, we have proved significant decrease of adiponectin concentrationas well as significant increase in total cholesterol concentration. Details arepresented in Table. Contrary – none of these abnormalities was detected in thecontrol group of patients treated with imatinib, including any change in insulinresistance measured by HOMA-2 index (medians - 0,5 vs. 0,8; p=0,3750).



Table 1.

Summary and Conclusion: Our preliminary results from presented studyproved fast development of peripheral insulin resistance already during the first3 month of nilotinib therapy as underlying cause of glucose and secondary alsolipid metabolism impairment during nilotinib treatment. Moreover, this was notproved for patients treated with imatinib. Final data from target of 40 patientson nilotinib and 10 patients on imatinib will be presented during the meeting.

P891

HIGHER RATE OF MAJOR MOLECULAR RESPONSE (MMR) WITHNILOTINIB vs IMATINIB IN CHINESE PATIENTS (PTS) WITH NEWLYDIAGNOSED CHRONIC MYELOID LEUKEMIA IN CHRONIC PHASE (CML-CP): ENESTCHINA 24-MO UPDATEZX Shen1,*, X Huang2, M Baccarani3, CE Ortmann4, S Hertle4, Y Yuan5,P Bourquelot4, G Saglio6, J Wang71Shanghai Ruijin Hospital, Shanghai, 2Peking University People’s Hospital,Beijing, China, 3University of Bologna, Bologna, Iraq, 4Novartis Pharma AG,Basel, Switzerland, 5Beijing Novartis Pharma Co., Ltd., Beijing, China,6University of Turin, Orbassano, Italy, 7Institute of Hematology, Tianjin, China

Background: In the global ENESTnd study, nilotinib resulted in superiorefficacy vs imatinib in pts with newly diagnosed CML-CP. ENESTchina is alocal, randomized, multicenter phase 3 study of nilotinib vs imatinib in Chinesepts with newly diagnosed CML-CP. ENESTchina met its primary endpoint witha significantly higher rate of MMR at 12 mo in the nilotinib arm (52.2%) vs theimatinib arm (27.8%; P<0.0001).Aims: This analysis aims to evaluate the safety and efficacy of nilotinib vsimatinib in ENESTchina after all pts completed the mo 24 visit or discontinuedearly.Methods: Adult Chinese pts (N=267) with newly diagnosed (≤ 6 mo) CML-CPwere randomized to nilotinib 300 mg twice daily (n=134) or imatinib 400 mgonce daily (n=133; dose escalation to 600 mg/day was allowed). Randomizationwas stratified by Sokal risk score and prior interferon therapy. Response rateswere compared between arms using a stratified Cochran-Mantel-Haenszel test.Freedom from progression to accelerated phase/blast crisis (AP/BC) and overallsurvival (OS) were estimated by Kaplan-Meier analysis and compared betweenarms using a stratified log-rank test. P values for secondary endpoints were notadjusted for multiple comparisons and are provided for descriptive purposesonly.Results: In the nilotinib and imatinib arms, respectively, the median age was41 y and 39 y, and 67.9% and 60.9% of pts were male. Median dose intensitieswere nilotinib 579.4 mg/day (range, 167-600) and imatinib 399.4 mg/day (range,247-574). By the data cutoff, 18 pts (13.4%) in the nilotinib arm and 15 pts(11.3%) in the imatinib arm discontinued treatment, most commonly due tosuboptimal response/treatment failure (nilotinib, n=6; imatinib, n=4) or diseaseprogression (per investigator’s judgment; nilotinib, n=4; imatinib, n=6).Compared with imatinib, nilotinib resulted in higher rates of MMR by 24 mo(Table; P=0.0096), MMR at 24 mo (P=0.0395), and durable MMR at 24 mo(defined in table footnote d; P=0.0001). Rates of MMR by 24 mo were higheron nilotinib vs imatinib across all Sokal risk groups. Median time to first MMRamong pts who achieved MMR was shorter on nilotinib vs imatinib. Rates of

complete cytogenetic response (CCyR) were high and comparable in botharms. Four pts in the nilotinib arm and 5 pts in the imatinib arm progressed toAP/BC (calculated) on treatment; estimated 2-y freedom from progression toAP/BC was 96.8% on nilotinib and 96.1% on imatinib (P=0.7481). Two pts ineach arm died on study (due to cerebral hemorrhage and disease progression[1 each] in the nilotinib arm and non-Hodgkin’s lymphoma and diseaseprogression [1 each] in the imatinib arm), all after discontinuing treatment.Estimated 2-y OS was 98.5% in each arm. Both drugs were well tolerated, andno new safety signals were identified. No cases of peripheral artery diseasewere reported in either arm; 1 pt (nilotinib arm) with a history of cerebralinfarction had an ischemic cerebrovascular event, and 1 pt (imatinib arm) hadan asymptomatic creatinine kinase-MB elevation with no clinical orelectrocardiographic findings. New or worsening biochemical abnormalitieswere typically grade 1/2, including all cholesterol elevations and all glucoseelevations except 2 (both in the nilotinib arm). New or worsening grade 3/4biochemical abnormalities occurring in ≥ 5% of pts in either arm includedelevated lipase (14.3% and 6.8% of pts on nilotinib and imatinib, respectively),elevated magnesium (6.0% and 5.3%, respectively), and decreased phosphate(4.5% and 5.3%, respectively).

Table 1.

Summary and Conclusion: Through 24 mo of follow-up, MMR rates wereconsistently higher on nilotinib vs imatinib. This study confirms the superiorityof nilotinib over imatinib for the treatment of pts with newly diagnosed CML-CP,as initially demonstrated in ENESTnd.

P892

FOUR-YEAR OUTCOME OF 215 NEWLY DIAGNOSED CHRONIC MYELOIDLEUKEMIA PATIENTS TREATED WITH NILOTINIB FRONTLINE: A GIMEMACML WORKING PARTY ANALYSISG Gugliotta1,*, F Castagnetti1, M Breccia2, L Levato3, C Fava4, L Luciano5,A Zaccaria6, A Russo Rossi7, E Usala8, A Gozzini9, P Vigneri10, M Cedrone11,M Tiribelli12, E Abruzzese13, P Avanzini14, F Palandri1, S Soverini1, G Alimena2,F Pane5, G Martinelli1, M Cavo1, G Saglio4, M Baccarani15, G Rosti11Department of Experimental, Diagnostic and Specialty Medicine, S.Orsola-Malpighi Hospital, University of Bologna, Bologna, 2Department of CellularBiotechnologies and Hematology, Sapienza University, Roma, 3Department ofHematology, Azienda Ospedaliera Pugliese Ciaccio, Catanzaro, 4Division ofInternal Medicine - Hematology, San Luigi Gonzaga Hospital, University ofTorino, Orbassano, Torino, 5Department of Hematology, CEINGE and AdvancedBiotechonolgies, Federico II University, Napoli, 6Hematology Unit, IRST-IRCCS,Meldola, Forlì-Cesena, 7Hematology Unit, Dept. of Emergency and OrganTransplantation, University of Bari, Bari, 8Hematology and Bone MarrowTransplantation Unit, Armando Businco hospital, Cagliari, 9Department ofHematology, Careggi Hospital, University of Firenze, Firenze, 10Division ofHematology, Ferrarotto Hospital, University of Catania, Catania, 11HematologyUnit, San Giovanni Hospital, Roma, 12Division of Hematology and Bone MarrowTransplantation, Department of Experimental and Clinical Medical Sciences,University of Udine, Udine, 13Department of Hematology, S. Eugenio Hospital,Roma, 14Hematology Unit, Santa Maria Nuova Hospital, Reggio Emilia,



15University of Bologna, Bologna, Italy

Background: Nilotinib (NIL) is a potent and selective BCR-ABL inhibitorapproved for the frontline treatment of chronic myeloid leukemia (CML) basedon the results of the ENESTnd study. The sustained superiority of NIL vs.imatinib (IM) was confirmed after 5 years of follow-up (Saglio et al, abs. 92, ASH2013). However, few data are available on pts treated frontline with NIL outsideof company-initiated trials.Aims: To analyze the response rates and outcome in a large, independentcohort of newly diagnosed CML pts treated frontline with NIL-based regimens.Methods: We analyzed 215 early chronic phase pts, enrolled in 2 multicenterphase II studies conducted by the GIMEMA CML WP (ClinicalTrials.gov.NCT00481052 and NCT00769327) or treated at the Bologna UniversityHospital, with NIL 300 mg or 400 mg BID as initial treatment; 123 pts receiveda sequential treatment with NIL and IM, with a 3-months rotation period. Themedian age was 53 years (range 18–86). Ten out of 215 pts (5%) had a highEUTOS score. The median follow-up was 43 months (range 18–69 months).We assessed: the rates of Complete Cytogenetic Response (CCyR) and MajorMolecular Response (MMR); the rates of optimal responders at each milestoneaccording to ELN 2013 recommendations; the overall survival (OS; any deathincluded), progression-free survival (PFS; progression to accelerated/blastphase [AP/BP] and deaths for any cause), failure-free survival (FFS; failuresaccording to ELN 2013 recommendations and deaths for any cause), andevent-free survival (EFS; events: failures, permanent discontinuation of NILfor any cause, including deaths). All analysis was made according to theintention-to-treat principle.Results: The cumulative rates of CCyR and MMR were 93% and 88%,respectively. At 3 months, 82% of the pts were in Partial Cytogenetic Responseand 90% had a BCR-ABL/ABL (IS)<10%; at 6 months, 86% were in CCyR and83% had a BCR-ABL/ABL (IS)<1%; at 12 months, 72% were in MMR; all thesepts were optimal responders according to ELN 2013 recommendations. Overall,64 (30%) pts permanently discontinued NIL: 31 (14%) for adverse events orintolerance; 22 (10%) for failures (figure); 11 (5%) for other reasons. In detail,10 (4.6%) pts discontinued NIL for cardiovascular events, including 3 forperipheral arterial occlusive disease and 2 for acute myocardial infarction. Eight(3.7%) pts progressed to AP/BP, all during the 1st year of therapy, and all ptssubsequently died (after a median of 13 months, range 1- 34 months). NIL-resistant mutations were identified in 5 of these pts (4 T315I; 1 Y253H). Nodifference in the rate of progression to AP/BP was observed between ptsreceiving NIL alone or NIL and IM in sequential schedule. During the follow-up,8 patients developed a secondary resistance (3 loss of CHR, 3 loss of CCyR,and 2 confirmed loss of MMR). Overall, 15 (7%) pts died, in 7 cases for reasonsunrelated to CML progression. The estimated 4-year OS, PFS, FFS, and EFSwere 93%, 93%, 86%, and 69%, respectively.

Figure 1.

Summary and Conclusion: Our National experience confirmed that ptstreated frontline with NIL-based regimens obtained fast and high rates ofcomplete cytogenetic and major molecular response, and most pts wereoptimal responders according to ELN recommendations. Of note, allprogressions to AP/BP occurred early (in the first year). Overall, 93% of the ptswere estimated to be alive and progression-free at 4-years, with 69% of the ptsstill on NIL.Acknowledgements: European LeukemiaNet, COFIN, Bologna University,BolognAIL.

P893

PONATINIB IN PATIENTS (PTS) WITH PHILADELPHIACHROMOSOME–POSITIVE (PH+) LEUKEMIAS RESISTANT ORINTOLERANT TO DASATINIB OR NILOTINIB, OR WITH THE T315IMUTATION: LONGER-TERM FOLLOW-UP OF THE PACE TRIALP D le Coutre1,*, DW Kim2, J Pinilla-Ibarz3, R Paquette4, C Chuah5, F E Nicolini6,J F Apperley7, HJ Khoury8, M Talpaz9, J F DiPersio10, D DeAngelo11,E Abruzzese12, D Rea13, M Baccarani14, M Muller15, C Gambacorti-Passerini16,S Lustgarten17, F G Haluska17, F Guilhot18, M W Deininger19, A Hochhaus20,T P Hughes21, N P Shah22, H M Kantarjian23, J E Cortes231Charité-Universitätsmedizin Berlin, Berlin, Germany, 2Seoul St. Mary’sHospital, The Catholic University of Korea, Seoul, Korea, Republic Of, 3H. LeeMoffitt Cancer Center & Research Institute, Tampa, FL, 4Ronald Reagan UCLAMedical Center, University of California, Los Angeles, CA, United States,5Singapore General Hospital, Duke-NUS Graduate Medical School, Singapore,Singapore, 6Centre Hospitalier Lyon-Sud, Pierre Benite, France, 7ImperialCollege, London, United Kingdom, 8Winship Cancer Institute of EmoryUniversity, Atlanta, GA, 9Comprehensive Cancer Center, University of Michigan,Ann Arbor, MI, 10Washington University School of Medicine, Saint Louis, MO,11Dana-Farber Cancer Institute, Boston, United States, 12S. Eugenio Hospital,Tor Vergata University, Rome, Italy, 13Hopital Saint-Louis, Paris, France,14Orsola-Malpighi University Hospital, Bologna, Italy, Italy, 15III. Med. Klinik,Universitätsmedizin Mannheim, Mannheim, Germany, 16Unità di RicercaClinica - Ematologia, Azienda Ospedaliera San Gerardo/University of MilanoBicocca, Monza, Italy, 17ARIAD Pharmaceuticals, Inc., Cambridge, MA, UnitedStates, 18CHU de Poitiers University Hospital, Poitiers, France, 19HuntsmanCancer Institute, University of Utah, Salt Lake City, UT, United States,20Universitätsklinikum Jena, Jena, Germany, 21Institute of Medicine andVeterinary Science (now SA Pathology), Adelaide, Australia, 22University ofCalifornia, San Francisco, San Francisco, CA, 23The University of Texas M.D.Anderson Cancer Center, Houston, TX, United States

Background: Ponatinib is a potent oral pan–BCR-ABL inhibitor with activityagainst native and mutant forms of BCR-ABL, including the resistant T315Imutant.Aims: The efficacy and safety of ponatinib (45 mg QD) were evaluated in thephase 2 PACE trial.Methods: 449 pts resistant or intolerant (R/I) to dasatinib or nilotinib or with theT315I mutation were enrolled and gave informed consent. Data are as of 3 Sept2013; median (range) follow-up was 24 (0.1-35) months. NCT01207440.Results: Pts were heavily pretreated: 58% received ≥3 prior tyrosine kinaseinhibitors. At analysis, 42% remained on study (55% chronic-phase chronicmyeloid leukemia [CP-CML]). The most common reasons for discontinuationwere progressive disease (21%) and adverse events (AEs) (14%; mostcommon was thrombocytopenia, 4%). The table shows response rates at anytime. In CP-CML, 89% of pts maintained major cytogenetic response (MCyR)for at least 2 yrs; progression-free survival (PFS) and overall survival (OS) at2 yrs were 67% and 86%, respectively. For accelerated-phase (AP) CML, blast-phase (BP) CML, and Ph+ acute lymphoblastic leukemia (ALL), OS at 2 yrs was72%, 18%, and 21%, respectively. The most common treatment-emergent AEs(≥30%) were thrombocytopenia (43%), rash (40%), abdominal pain (40%),headache (36%), constipation (36%), and dry skin (36%). Pancreatitis andpneumonia were the most common serious AEs (SAEs; both 6%). Vascularocclusive AEs [SAEs] were reported as follows: overall 20% [14%], includingcardiovascular 9% [6%], cerebrovascular 6% [4%], peripheral vascular 6%[4%] (collectively, arterial thrombotic events [ATEs]), and venousthromboembolic 5% [3%]. In a multivariate analysis, higher dose intensity, olderage, and cardiovascular risk factors were associated with a higher likelihoodof an ATE; pts with and without cardiovascular risk factors experienced theseevents. OS at 2 yrs was not reduced in pts with an ATE (73%) vs without anATE (69%); MCyR in CP-CML pts with vs without an ATE was 70% vs 51%.Summary and Conclusion: Ponatinib has substantial clinical activity in heavilypretreated pts with Ph+ leukemias. Vascular occlusive events were observed.Ponatinib is an important treatment option for pts in whom the need and benefitoutweigh the risk.

Table 1.

R/I, n (%) T315I, n (%) Total, n (%)

CP-CML N=203 N=64 N=267MCyR 113 (56) 46 (72) 159 (60)CCyR 98 (48) 45 (70) 143 (54)MMR 63 (31) 37 (58) 100 (38)AP-CML N=65 N=18 N=83MaHR 40 (62) 11 (61) 51 (61)BP-CML N=38 N=24 N=62MaHR 12 (32) 7 (29) 19 (31)Ph+ ALL N=10 N=22 N=32MaHR 5 (50) 8 (36) 13 (41)

Responses at any time after first dose



P894

CLINICAL IMPACT OF DOSE MODIFICATION AND DOSE INTENSITY ONRESPONSE TO PONATINIB IN PATIENTS (PTS) WITH PHILADELPHIACHROMOSOME–POSITIVE (PH+) LEUKEMIASF E Nicolini1,*, J E Cortes2, DW Kim3, J Pinilla-Ibarz4, P D le Coutre5,R Paquette6, C Chuah7, J F Apperley8, HJ Khoury9, M Talpaz10, J F DiPersio11,D J DeAngelo12, E Abruzzese13, D Rea14, M Baccarani15, M C Muller16,C Gambacorti-Passerini17, S Lustgarten18, V M Rivera18, F G Haluska18,F Guilhot19, M W Deininger20, T P Hughes21, N P Shah22, H M Kantarjian2,A Hochhaus231Centre Hospitalier Lyon-Sud, Pierre Benite, France, 2The University of TexasMD Anderson Cancer Center, Houston, United States, 3Seoul St. Mary’sHospital, The Catholic University of Korea, Seoul, Korea, Republic Of, 4H. LeeMoffitt Cancer Center, Tampa, United States, 5Charite-Universitatsmedizin,Berlin, Germany, 6Ronald Reagan UCLA Medical Center, University ofCalifornia, Los Angeles, United States, 7Singapore General Hospital, Duke-NUS Graduate Medical School, Singapore, Singapore, 8Imperial College,London, United Kingdom, 9Winship Cancer Institute of Emory University,Atlanta, 10Comprehensive Cancer Center, University of Michigan, Ann Arbor,11Washington University School of Medicine, St Louis, 12Dana-Farber CancerInstitute, Boston, United States, 13S. Eugenio Hospital, Tor Vergata University,Rome, Italy, 14Service des Maladies du Sang and CIC Clinical InvestigationCenter, Hopital Saint-Louis, Paris, France, 15S. Orsola-Malpighi UniversityHospital, Bologna, Italy, 16III. Med. Klinik, Universitatsmedizin, Mannheim,Germany, 17Unita di Ricerca Clinica-Ematologia, Azienda Ospedaliera SanGerardo/University of Milano Bicocca, Monza, Italy, 18ARIAD Pharmaceuticals,Inc., Cambridge, United States, 19CHU de Poitiers, University Hospital, Poiters,France, 20Huntsman Cancer Institute, University of Utah, Salt Lake City, UnitedStates, 21Institute of Medicine and Veterinary Science, Adelaide, Australia,22University of California San Francisco, San Francisco, United States,23Universitatsklinikum Jena, Jena, Germany

Background: Ponatinib is a potent oral pan–BCR-ABL tyrosine kinase inhibitorwith clinical activity in pretreated pts with Ph+ leukemias. Dose modification ofponatinib was used to manage adverse events (AEs).Aims: This post hoc analysis assessed the clinical impact of dose modificationand dose intensity on outcomes of pts in the phase 2 PACE trial.Methods: All pts provided informed consent. Ponatinib starting dose was 45 mgQD. Dose reduction was defined as any reduction below 45 mg/day; doseinterruption was defined as treatment withheld for ≥3 consecutive days. Efficacyanalyses were performed on chronic-phase chronic myeloid leukemia (CP-CML) pts (N=267). Analysis of arterial thrombotic events (ATEs) included all pts(CP-CML, blast-phase CML, accelerated-phase CML, Ph+ acute lymphoblasticleukemia; N=449). Data are as of 3 Sept 2013; median (range) follow-up was24 (0.1-35) months for all pts. NCT01207440.Results: 78% of CP-CML pts had dose modification within the first 12 months(82% at any time). Responses in pts with and pts without modification werecomparable (Table). Of 149 responders, 87 (58%) achieved major cytogeneticresponse (MCyR) at 45 mg/day, 46 (31%) at 30 mg/day, and 16 (11%) at 15mg/day. Most pts who had a dose reduction after achieving a responsemaintained that response: 97% maintained MCyR; 96% maintained completecytogenetic response (CCyR); and 92% maintained major molecular response(MMR). Among pts with a dose reduction lasting ≥6 months after achieving aresponse at a higher dose, 100% (33/33) maintained MCyR (96% [26/27] CCyRand 93% [13/14] MMR). While dose intensity was the most significant predictorof MCyR by 12 months (multivariate analysis), substantial responses occurredat lower doses; estimated response rates were ~75% at 45 mg/day, ~60% at30 mg/day, and ~30% at 15 mg/day. ATEs occurred in 17% of pts; each 15mg/day reduction in average daily dose is predicted, by multivariate analysis,to lead to ~40% reduction in risk of ATE. 2-yr overall survival was similar for CP-CML pts who had dose modifications (86%) and those who did not (86%), andfor pts who had ATEs (85%) and those who did not (87%). Of pts with ATEs,46% had dose modifications.Summary and Conclusion: Most of the CP-CML pts in the PACE trial had dosemodifications. Pts who underwent dose modification still responded totreatment. Dose modification was an effective management tool for CP-CMLpatients. Careful consideration of the potential benefits and risks of ponatinibshould guide treatment decisions.

Table 1.

CP-CML n MCyR, %b CCyR, %b MMR, %aDose modificationa 218 58 47 38No dose modification 49 47 45 35

aAt any time; bBy 12 months

P895

LONGER-TERM FOLLOW-UP OF A PHASE 1 STUDY OF PONATINIB INPATIENTS (PTS) WITH PHILADELPHIA CHROMOSOME-POSITIVE (PH+)LEUKEMIASJ Cortes1,*, M Talpaz2, H M Kantarjian1, N P Shah3, D Bixby2, I W Flinn4,T O’Hare5, S Hu6, V Rivera6, T Clackson6, C Turner6, F Haluska6, B J Druker7,M W Deininger5, M J Mauro81The University of Texas MD Anderson Cancer Center, Houston,2Comprehensive Cancer Center, University of Michigan, Ann Arbor, 3Universityof California San Francisco, San Francisco, 4Sarah Cannon Research Institute,Nashville, 5Huntsman Cancer Institute, University of Utah, Salt Lake City,6ARIAD Pharmaceuticals, Inc., Cambridge, 7Oregon Health & ScienceUniversity, Knight Cancer Institute, 8Oregon Health & Science University KnightCancer Institute, Portland, United States

Background: Ponatinib is a potent oral pan–BCR-ABL tyrosine kinase inhibitor(TKI) that is active against native and mutated forms of BCR-ABL.Aims: The safety and anti-leukemic activity of ponatinib in pts with chronicmyeloid leukemia (CML) or Ph+ acute lymphoblastic leukemia were evaluatedin a phase 1 clinical trial.Methods: Pts (N=81) with resistant/refractory hematologic malignancies wereenrolled and gave informed consent in this ongoing, open-label, dose-escalation, phase 1 study. Ponatinib was dosed once daily (2–60 mg). 65 ptshad Ph+ leukemia and are included in this analysis (data as of 15 Oct 2013).Median (range) follow-up for 25 pts still on treatment was 45 (38–57) months.NCT00660920.Results: The median age of pts was 55 yrs; median time since diagnosis was6.5 yrs. Pts were heavily pretreated (94% had received ≥2 prior TKIs, 62% ≥3).65% had baseline BCR-ABL mutations (29% with T315I). 38% (58% chronicphase [CP] CML) of pts remained on study at the time of analysis. Adverseevents (AEs) and progression were the most common reasons fordiscontinuation (20% and 17%, respectively). The most common treatment-emergent AEs were rash (52%), fatigue (52%), abdominal pain (51%),headache (48%), and arthralgia (46%). Treatment-emergent vascular occlusiveevents were observed in 23% (serious events) and 37% (all events) of pts,including cardiovascular (15% serious, 23% all) peripheral vascular (5%, 9%),cerebrovascular (5%, 8%), and venous thrombotic (0%, 5%) events. Significantanti-leukemic activity was observed: among CP-CML pts, major cytogeneticresponse (MCyR), complete cytogenetic response (CCyR), and majormolecular response (MMR) rates were 72%, 65%, and 51%, respectively; 75%of pts with MCyR, 69% with CCyR, and 53% with MMR are estimated (Kaplan-Meier [KM]) to maintain response for at least 3 yrs (4-yr KM estimates: 75%MCyR, 53% MMR). Of 28 CP-CML pts with CCyR, 23 remained on study atanalysis (17 with continuous CCyR); of 22 pts with MMR, 20 remained on studyat analysis (12 with continuous MMR).Summary and Conclusion: Substantial and durable responses were observedwith ponatinib in heavily pretreated CP-CML pts. Vascular occlusive eventswere observed. Risk and benefit considerations should be evaluated whenutilizing ponatinib in this patient population.

P896

A NATIONAL EXPERIENCE OF THE USE OF PONATINIB IN PATIENTSFAILING MULTIPLE TYROSINE KINASE INHIBITORS CONFIRMSEFFICACY IN A HEAVILY PRE-TREATED COHORT OF PATIENTS WITHPH+ LEUKAEMIASD Milojkovic1,*, A Bakir2, L Whitehead2, J Byrne3, R E Clark4, M Copland5,H de Lavallade6, M Dennis7, J Ewing8, L Foroni1, D Gillett9, D Marin2, A Mead10,P Mehta11, L Mitchell12, H Oakervee13, G Smith14, P Srinivas15, J Tighe16,J Apperley21Haematology, Imperial College Healthcare NHS Trust, 2Haematology, ImperialCollege London, London, 3Haematology, Nottingham University Hospitals Trust,Nottingham, 4Haematology, Royal Liverpool University Hospital, Liverpool,5Haematology, University of Glasgow, Glasgow, 6Haematology, Kings CollegeHospital NHS Foundation Trust, London, 7Haematology, The Christie NHSFoundation Trust, Manchester, 8Heart of England NHS Trust, Birmingham,9Maidstone and Tunbridge Wells NHS Trust, Maidstone, 10University of Oxford,Oxford, 11University Hospital Bristol, Bristol, 12Monklands Hospital NHSLanarkshire, Lanarkshire, 13Barts Health NHS Trust, London, 14Leeds TeachingHospitals NHS Trust, Leeds, 15University Hospital of North Staffordshire, Stoke-on-Trent, 16Aberdeen Royal Infirmary, Aberdeen, United Kingdom

Background: Ponatinib is a highly potent third generation tyrosine kinaseinhibitor(TKI) licensed for use in patients with Ph positive leukaemias who havefailed two or more prior TKI and/or have a documented T315I kinase domainmutation. Ponatinib first became available in the United Kingdom in 2011through the phase II PACE study and subsequently through an expandedaccess scheme.Aims: To examine efficacy of Ponatinib in Patients failing Multiple TyrosineKinase Inhibitors.Methods: From May 2011 to September 2012, 66 patients received ponatinibfor chronic myeloid leukaemia (CML) in chronic phase (CP) (n=57), accelerated



phase (n=3), second or subsequent CP (n=3), blast crisis (n=1) or Ph positiveALL (n=2). Of the patients treated in CP, 10/57 had the T315I mutation. Sixpatients received ponatinib for relapse after allogeneic stem cell transplantationand in 4 of these the T315I mutation had been identified before and/or aftertransplant. In the 51 patients treated without transplant, a T315I mutation wasidentified in 6 and all had received two prior TKI. However the majority of thesepatients (45/51) received ponatinib for poor responses and/or intolerance ofother TKI without evidence of kinase domain mutations, 4 after failing a singlesecond generation TKI, 14 after failing two TKI and 27 after 3-4 prior TKI.Results: Focussing on the 51 patients who received ponatinib in CP andwithout a previous transplant, the median age was 62 years (range 20-78 yrs).The median duration of ponatinib treatment was 9 months (range 0.25-31mths). 21/51 (41%) and 20/51 (39%) patients remained on ponatinib at 12 and24 months respectively. The reasons for discontinuation were resistance onlyin 8, intolerance only in 5 and resistance and intolerance in 15 patients. Patientswere considered evaluable for response if they received at least 3 months oftherapy (n=48). Six patients commenced treatment in complete cytogeneticremission (CCyR) (4 of whom were also in major molecular response - MMR)and all 6 maintained these levels of response. Achievement of CCyR was seenin 15 of the remaining 42 patients including 3 of the 6 patients with a T315Imutation at the start of treatment. CCyR and MMR were seen in 12/36 (33%)and 3/36 (8%) of patients without and 3/6 (50%) and 3/6 (50%) with the T315Imutation.Summary and Conclusion: More precise data collection for the type andgrade of toxicities is on-going. Our national experience reflects the dataemerging from the phase II PACE study confirming useful efficacy of ponatinibin a previously heavily treated patient cohort in whom the need and benefitoutweigh the risk.

P897

OUTCOMES OF THIRD-LINE BCR-ABL1 TYROSINE KINASE INHIBITORSIN THE TREATMENT FAILED CHRONIC PHASE CHRONIC MYELOIDLEUKEMIA PATIENTS WHO HAVE RECEIVED TWO PRIOR TKIS SE Lee1,2,*, SY Choi2, SH Kim2, YJ Oh2, MJ Chae2, HY Song2, HL Yoo2,MY Lee2, EJ Jang2, KH Jeong2, JS Lee2, DW Kim1,21Department of Hematology, Seoul St. Mary’s Hospital, The Catholic Universityof Korea, 2Cancer Research Institute, The Catholic University of Korea, Seoul,Korea, Republic Of

Background: With the availability of several TKIs, in patients who have failedto 2 prior TKIs, anyone of the remaining TKIs can be considered for salvagetherapy. However, further evaluation on the use of a novel TKI as third-linetherapy is needed.Aims: The aim of this study was to evaluate the response rates and outcomesof third-line TKI therapy in the treatment failed patients with chronic phasechronic myeloid leukemia who have received 2 prior TKIs.Methods: We evaluated 100 CP CML patients who failed to 2 prior TKIs. Mainstudy objectives were to evaluate major cytogenetic response (MCyR) by 12months, cumulative incidence (CI) of major molecular response (MMR),progression-free survival, and overall survival. Molecular responses weremonitored using qRT-PCR assay in3-month intervals, and then 6-monthintervals after achieving major molecular response (MMR). MMR was definedas a BCR-ABL1 transcript level of 0.1% or lower on the international scale (IS).Results: A total of 100 patients were received third-line therapy with dasatinib(n=48), radotinib (n=28), nilotinib (n=14), and bosutinib (n=10). 65 men and 35women were included and their median age was 41 years (range, 18-73). Thepercentages of patients with low, intermediate and high Sokal risk scores were19%, 30% and 17%, respectively with unknown risk score in 34%. 93 patientsreceived IM as first-line therapy and 7 patients received 2G-TKI as first-linetherapy. At the start of first-line therapy, 92 patients were in CP, 7 in AP, and 1in BP. As second-line therapy, IM (n=7), dasatinib (n=43), nilotinib (n=24),radotinib (n=22), and bosutinib (n=4) were used. At the start of second-linetherapy, 84 patients were in CP, 14 in AP, and 2 in BP. With a median follow-up of 20.1 months (range, 0.2-81.6 months) since the start of the third-line TKI,55 patients continue on third-line TKI therapy. Forty-five patients werepermanently discontinued from third-line TKI treatment due to intolerance(n=18), failure (n=6), progression (n=14), death (n=3) and others (n=4).Overall,18 patients died. MCyR rate at 12 months after third-line TKI, in 31 patients withcytogenetic assessments at 12 months was 74.2% (23/31 patients). The3-yearOS and PFS after third-line TKI were 81.1%and 76.0%, respectively. 50 patientswith MCyR at the time of third-line TKI had a better 3-year OS (100% vs. 69%,P<0.001) and PFS (95.5% vs. 51.3%, P<0.001) than those of 50 patientswithout MCyR.Summary and Conclusion: Our data showed the favorable outcomes of CPCML patients treated with a third-line TKI. The patients with MCyR at the timeof third-line TKI showed a better outcome. It implies that the use of anyone ofremaining TKI, as third-line therapy may induce feasible outcomes in CPpatients. Further study focusing on selecting the patients who may benefit froma third-line TKI is needed.

P898

PREDICTIVE FACTORS FOR OUTCOME TO NOVEL BCR-ABL1 TYROSINEKINASE INHIBITORS IN IMATINIB FAILED CHRONIC PHASE CHRONICMYELOID LEUKEMIASE Lee1,2,*, SY Choi2, SH Kim2, YJ Oh2, MJ Chae 2, HY Song2, HL Yoo2,MY Lee2, EJ Jang2, KH Jeong2, JS Lee2, DW Kim1,21Department of Hematology, Seoul St. Mary’s Hospital, The Catholic Universityof Korea, 2Cancer Research Institute, The Catholic University of Korea, Seoul,Korea, Republic Of

Background: The first BCR-ABL1tyrosine kinase inhibitor (TKI), imatinibmesylate (IM), has become a first-line therapy for chronic phase (CP) chronicmyeloid leukemia (CML).However, approximately one third of IM-treatedpatients discontinue therapy due to an inadequate response or adverse event.More potent novel TKIs such as dasatinib, nilotinib, radotinib, bosutinib andponatinib have developed and these agents have shown high rates ofhematologic and cytogenetic responses after failure of IM therapy. However,factors associated with long-term benefit from 2nd line therapy are notconfirmed.Aims: The aim of this study was to evaluate the predictive factors for outcometo novel TKIs in IM failed CP CML.Methods: We evaluated 328 imatinib failed CP CML patients who had beentreated with novel TKIs. Predictive factors for failure-free survival (FFS),progression-free survival (PFS), and overall survival (OS) were evaluated. FFSwas measured from the day of 2ndline therapy initiation to death, progressionto AP or BC, treatment failure according to 2013 ELN recommendations orcontact, whichever came first. OS included any death regardless of causes, andPFS included progression to AP or BC as well as death resulting from anyreason. OS and PFS were also collected on patients who were treated withother TKIs after 2nd-line TKI discontinuation.Results: A total of 328 patients were treated with 2nd line TKI, dasatinib(n=131), nilotinib (n=103), radotinib (n=76), bosutinib (n=16), and ponatinib(n=2). 205 men and 123 women were included and their median age was 43years (range, 15-77). The percentages of patients with low, intermediate andhigh Sokal risk scores were 23%, 31% and 24%, respectively with unknownSokal risk scores in 22%. At diagnosis, 315 patients were in CP, 9 in AP, and4 in BP. With a median follow-up of 28.4 months (range, 0.2-108.6 months)since the start of 2nd line TKI, 194 patients continue on therapy. 134 patientswere permanently discontinued due to intolerance (n=74), failure (n=31),warning (n=1), progression (n=18), death (n=3) and others (n=7).The3-yearFFS, PFS, and OS were 79.7%, 85.1%, and 89.1%, respectively. After adjustingfor factors affecting FFS on univariate analyses, multivariate analyses showedthat AP/BP at diagnosis and presence of BCR-ABL1 kinase domainabnormalities (KDA) at baseline were associated with a lower FFS. Less thanPCyR on IM had a lower FFS, compared with those with CCyR on IM (RR of3.25, P=0.016). AP/BP at diagnosis, presence of KDA at baseline and lessthan PCyR on IM were also associated with a lower PFS and OS. Increasingage was a predictive factor for lower OS.Summary and Conclusion: Our data showed the predictive factors foroutcome to secondline TKIs in imatinib failed CP CML. Disease phase atdiagnosis, previously cytogenetic response on IM, and presence of KDA atbaseline had the value of predicting the outcomes. It implies that the patientswith these factors may require more precise monitoring on secondline therapy.Further study focusing on selecting the patients who may benefit from a third-line TKI is needed.



chronic myeloid leukemia - clinical 2

Documents