Phase I/II Trial of Adding SemisyntheticHomoharringtonine in Chronic Myeloid LeukemiaPatients Who Have Achieved Partial or CompleteCytogenetic Response on Imatinib

David Marin, M.D.

Jaspal S. Kaeda, Ph.D.

Catharina AndreassonSue M. SaundersMarco Bua, M.D.

Eduardo Olavarria, M.D.

John M. Goldman, D.M.

Jane F. Apperley, M.D.

Department of Haematology, Imperial College Lon-don at Hammersmith Hospital, London, UnitedKingdom.

Address for reprints: John M. Goldman, D.M., De-partment of Haematology, Imperial College Londonat Hammersmith Hospital, Du Cane Road, LondonW12 0NN, UK. E-mail: jgoldman@imperial.ac.uk

Received September 26, 2004; revision receivedDecember 21, 2004; accepted December 21,2004.

BACKGROUND. A Phase I/II study was designed to show whether the addition of

semisynthetic homoharringtonine (sHHT) would reduce the level of residual dis-

ease in patients with Ph-positive chronic myeloid leukemia who appeared to have

achieved a suboptimal response to imatinib alone.

METHODS. Patients with CML who had achieved � 35% Ph-negativity on imatinib

were included. All patients had been treated with imatinib at � 400 mg/day for at

least 2 years and had achieved a plateau in BCR-ABL transcripts defined by

measuring BCR-ABL transcripts on at least 4 occasions over a minimum period of

1 year with the latest value not lower than the previous minimum value. Initially

sHHT was given subcutaneously at a dose of 1.25 mg/m2 twice daily for 1 day.

Courses were repeated every 28 days. The dosage of sHHT was escalated by adding

one day of treatment every two days. Efficacy was assessed by serial monitoring of

blood levels of BCR-ABL transcripts.

RESULTS. Of 10 evaluable patients, 7 had an appreciable decline in BCR-ABL

transcript levels; in 5 cases the reduction was greater than 1 log. Asthenia (n � 10)

and cytopenias (n � 3) were prominent side-effects, but the drug was generally

well tolerated. Mutations in the P-loop of the BCR-ABL kinase domain were found

in 2 of the patients who responded to the addition of sHHT.

CONCLUSIONS. The addition of sHHT should be considered for patients on imatinib

who fail to obtain low levels of minimal residual disease. Cancer 2005;103:1850 –5.

© 2005 American Cancer Society.

KEYWORDS: imatinib, chronic myeloid leukemia, homoharringtonine, BCR-ABLmutations.

Imatinib induces complete cytogenetic responses (CCyR) in approx-imately 40% of patients with chronic myeloid leukemia (CML) in

chronic phase treated after failure of interferon-�1 and in at least 70%of patients who start treatment soon after diagnosis.2 Recent datasuggest that imatinib prolongs survival in at least some patient sub-groups.3

Homoharringtonine is a plant alkaloid derived from an evergreentree belonging to the genus Cephalotaxus that inhibits the synthesis ofproteins leading to apoptosis.4,5 Homoharringtonine is active in allphases of CML.6 –11 In combination with imatinib, it has synergistic oradditive effects in CML cell lines12–14 and in blast cells from advancedstage patients with CML.12 For these reasons, a Phase I–II trial wasdesigned to involve the addition to imatinib of semisynthetic homo-harringtonine (sHHT) for patients with CML who had achieved a

1850

© 2005 American Cancer SocietyDOI 10.1002/cncr.20975Published online 22 March 2005 in Wiley InterScience (www.interscience.wiley.com).

partial cytogenetic response or a CCyR with an appar-ent plateau of BCR-ABL transcript numbers at sub-optimal levels.These patients were classified as par-tially sensitive to imatinib such that they might benefitfrom the addition of a second drug.

MATERIALS AND METHODSPatientsPatients with Philadelphia (Ph) chromosome-positiveCML aged 18 years or older were eligible for the study.Informed consent was obtained from all patients to beenrolled in the study. Patients were eligible if they metall the following criteria: 1) treatment with imatinib ata minimum dose of 400 mg per day for at least 2 years;2) the achievement of at least a minor cytogeneticresponse (defined as at least 35% of Ph-negative mar-row metaphases); 3) achievement of a plateau in BCR-ABL transcripts defined by measuring BCR-ABL tran-scripts on at least 4 occasions over a minimum periodof time of 1 year, with the latest value not lower thanthe previous minimum value. Patients were excludedfrom the study if their imatinib dosage had been mod-ified over the preceding 12 months or if they had othersignificant concomitant disease. The definitions ofchronic and accelerated phases of CML, the classifi-cation of cytogenetic responses, and criteria for failureof interferon-� were based on previous publica-tions.1,15

Treatment ScheduleImatinib (Gleevec�) (Novartis Pharma, Basel, Switzer-land) was supplied in 100 mg capsules. sHHT (My-

elostat�) (originally Oncopharm, Houston, TX, andmore recently Stragen Pharma, Lyon, France) wassupplied by the manufacturer as a lyophilized powdercontaining 5 mg of drug; it was reconstituted withnormal saline at a concentration of 2 mg/mL. WhensHHT was introduced, imatinib was continued at theprevious dosage (Table 1). Patients were taught toself-administer the drug. For the first 3 patients, theinitial dose was 1.25 mg/m2 subcutaneously twicedaily for 3 consecutive days, but this led to prolongedcytopenias (see below). For the subsequent 10 pa-tients, the protocol was therefore amended as follows:

Dosage Level 1a: 1.25 mg/m2 twice daily for 1 day,Dosage Level 1b: 1.25 mg/m2 twice daily for 1.5 days (3

doses),Dosage Level 2a: 1.25 mg/m2 twice daily for 2 consec-

utive days,Dosage Level 2b: 1.25 mg/m2 twice daily for 2.5 con-

secutive days (5 doses),Dosage Level 3: 1.25 mg/m2 twice daily for 3 consecu-

tive days (6 doses).

Courses were repeated every 4 weeks. The dose ofsHHT was increased every 2 courses by 1 dosage level(e.g., from Dosage Level 1a to Dosage Level 2a) if theANC was � 2.5 � 109/l and the platelet count � 200� 109/l by Day �28 of the previous course, or by halfa dosage level (e.g., from Dosage Level 1a to DosageLevel 1b) if the ANC was � 1.5 � 109/l and the plateletcount � 100 � 109/l. The sHHT dose was maintainedat the same level if the ANC was � 1.5 � 109/l and theplatelet count � 75 � 109/l by Day �28. Patients were

TABLE 1Patient Features and Responses to Imatinib

PatientAge(sex)

Sokal20 riskgroup

Time fromdiagnosis tostart ofimatinibtherapy (mos)

Clinical status atthe onset ofimatinib

Dose of IMover theyear beforesHHT(mg/day)

Time fromonset ofimatinib tosHHTtherapy(mos)

Median Q–PCRover the yearprecedingsHHT therapy

Minimal Q–PCRduring the yearpreceding sHHTtherapy

Q–PCR atscreeningfor sHHT

Peripheral bloodcounts atscreening forsHHT: WBC;ANC; Hb; Pla

1 70 (F) High 3.8 Newly diagnosed 600 27.2 7.3 5.4 16.9 5.3; 4.0; 12.6; 1962 57 (M) Intermediate 74.7 IFN–�failure 600 35.9 2.2 0.04 4.5 2.1; 1.3; 12.8; 1493 44 (M) Low 40.1 IFN–�failure 400 34.1 0.012 0.01 0.028 6.4; 3.2; 14.7; 1484 50 (M) Low 94.5 IFN–�failure 400 29.9 4.35 3.08 4.35 5.1; 3.1; 12.4; 1535 52 (M) High 38.5 IFN–�failure 600 29.3 4.9 2.1 9 4.5; 2.8; 12.1; 2276 54 (F) Low 24.6 IFN–�failure 400 25.9 0.12 0.075 0.15 2.4; 1.2; 11.5; 1577 45 (M) Low 121 Accelerated phase 400 32 1.72 0.8 5.77 3.1; 2.2; 11.9; 1408 41 (F) High 5.6 IFN–�failure � CE 600 26.3 23.9 14.1 25 3.3; 1.6; 10.5; 1929 41 (F) High 13.8 IFN–�failure 400 46.7 1.28 1.02 1.2 3.3; 1.7; 12.8; 19010 41 (M) Low 2.8 Newly diagnosed 600 28.1 0.21 0.15 0.38 7.0; 5.4; 10.4; 399

Q–PCR: transcript level expressed as a BCR–ABL/ABL, ratio percentage; CE: clonal evolution; WBC: white blood cell count expressed as number of cells � 109/1; ANC: absolute neutrophil count expressed as number

of cells � 109/1; Hb: hemoglobin level expressed in g/dL; Pla: platelet count expressed as number of cells � 109/1.

Homoharringtonine with Imatinib in CML/Marin et al. 1851

considered to have reached the maximal tolerabledose (MTD) at a given dosage level if the dose couldnot safely be escalated further. The dose of sHHT wasreduced by half a dosage level if Grade II thrombocy-topenia or Grade III neutropenia occurred and by onedosage level in the presence of Grade III thrombocy-topenia or Grade IV neutropenia. Hematopoieticgrowth factors were given if required. Patients withGrades III and IV nonhematologic toxicity were with-drawn from the study. Treatment with sHHT was dis-continued 1) if, having reached the MTD, a patientachieved a complete molecular response, or 2) if, hav-ing reached the MTD, a patient did not achieve areduction in the transcript level of at least 0.5 log after6 courses. For patients whose BCR-ABL transcript lev-els continued to decline but who had not yet reachedmolecular remission, treatment was designed to con-tinue for 1 or more further 6-course blocks. The tran-script level in a given patient was defined as “stilldeclining” if, at the end of a block of 6 courses, it wasat least 0.5 log below the lowest result obtained in theprevious block.

Quantitative Realtime Reverse Transcriptase PolymeraseChain Reaction (Q-PCR)The numbers of BCR-ABL transcripts in the peripheralblood were measured serially while imatinib was ad-ministered as a single agent, at the time of startingsHHT and serially thereafter as described previously.16

Expression of the ABL gene was used as control andresults were expressed as the BCR-ABL/ABL ratio per-centage. Occasional blood specimens were invalidatedfor technical reasons; for example, samples with con-trol � 1 � 104 ABL transcripts were considered sub-optimal and were excluded from the analysis. In caseswhere BCR-ABL transcripts were undetectable by Q-PCR, the results were confirmed by nested primerpolymerase chain reaction..

BCR-ABL Kinase Mutation ScreeningPatients were screened for BCR-ABL kinase domainmutations at the time enrollment in the trial. TotalRNA extracted from cellular pellet isolated from pe-ripheral blood and/or bone marrow aspirates was re-verse transcribed as previously reported. The cDNAsynthesized was subjected to polymerase chain reac-tion (PCR) using primers B2B (5�-ACAGCATTCCGCT-GACCATCAATAAG-3�) and NTPE� (5�-CTTCGTCT-GAGATACTGGATT-3�) that map to exon 13 of the BCRgene and exon 9 of the ABL gene, respectively, therebyexcluding the ABL allele not involved in the t(9;22)translocation. The amplicons from the first roundwere then subjected to seminested PCR using primersNTPB� (5�-AAGCGCAACAAGCCCACTGTCTATGG -3�)

and NTPE�, yielding an amplicon 863 base pairs (bp)in length. The success of the PCR was checked byelectrophoresis through a 2.0% agarose gel. The am-plified products were subjected to magnetic bead-based PCR purification using a commercially availablekit according to manufacturer’s instructions. The pu-rified products were sequenced by Sanger chain ter-mination reaction using Big Dye 3100 ABI DNA anal-yser (Appliedbiosystems, Foster City, CA). Thesequencing was primed by oligonucleotides ABLKF(5�-AACGCCGTGGTGCTGCTGTAC-3�) and NTPB� forthe coding strand and by ITKE� and ABLKR (5�-CAGTTTCGGGCAGCAAGATC-3�) for the noncodingstrand. The data were analyzed using EditView Soft-ware (shareware from Perkin Elmer Biosystems, Bos-ton, MA). Its complementariness to the published wildtype ABL kinase domain sequence (Accession no.AAB60394) was tested using Blast 2 software (BerkeleyLazy Abstraction Software Verification Tool, Berkeley,CA).

RESULTSResponseTable 1 shows the features of the 10 evaluable patientsincluding their responses to imatinib alone. AftersHHT was added, the transcript levels declined by� 0.5 log in 7 cases and by � 1 log in 5 cases (Table 2).The two patients not in CCyR achieved CCyR (Patients1 and 8). Patient 8 had developed a trisomy 8 inaddition to the Ph chromosome in all metaphaseswhen she was receiving interferon; she was thentreated with imatinib, first 400 mg and then 600 mg forthe 18 months before enrollment in the trial, but shefailed to gain a major cytogenetic response. At thetime of entry in the sHHT trial, she was 44% Ph-positive with trisomy 8. On sHHT, she achieved aCCyR; her marrow metaphases contained neither Phchromosome nor trisomy 8. Two patients achievedcomplete molecular responses (Patients 3 and 6).These molecular responses were lost 6 months (Pa-tient 6) and 4 months (Patient 3) respectively afterdiscontinuation of the sHHT; transcript numbers had,by then, risen to levels similar to baseline levels, i.e.,0.011% for Patient 3 and 0.020% for Patient 6.

Patients were screened for the presence of BCR-ABL kinase domain mutations. Mutations were foundin two patients, Patients 7 and 8. In both cases, themutation was a M244V substitution in the ATP phos-phate-binding loop (P-loop), which has been associ-ated previously with resistance to imatinib.17–19 Bothpatients responded to the addition of sHHT: Patient 7had a reduction in the transcript levels of 1.26 logs(from BCR-ABL/ABL ratio of 5.77% to 0.32%), andPatient 8 achieved CCyR (from 44% Ph-positive met-

1852 CANCER May 1, 2005 / Volume 103 / Number 9

aphases) with a reduction in the transcript levels of1.67 logs (BCR-ABL/ABL ratio from 25% to 0.53%).

ToxicityThe initial dose of 1.25 mg/m2 subcutaneously twicedaily for 3 consecutive days led to prolonged cyto-penias in the initial 3 patients; these patients hadboth neutropenia and thrombocytopenia. The neu-tropenia was Grade III in two cases and grade IV inone case. It responded rapidly to treatment withG-CSF. One patient had Grade III thrombopenialasting 7 days, and 2 patients had Grade IV throm-bopenia that needed 14 and 42 days to revert toGrade II. Both these patients needed platelet sup-port. No episodes of infection or bleeding occurred.Grade III asthenia was seen in 1 patient and lastedfor 5 days. Because of this toxicity, the protocol wasamended as described above. In the next 10 pa-tients, the combination was relatively well tolerated;the principal toxicities were asthenia and myelosup-pression. For these 10 patients, the maximum sHHTdose achieved was Dosage Level 2a in 5 patients,

Dosage Level 1b in 3 patients, and Dosage Levels 2band 3 in 1 patient each. Hematologic toxicity wasclearly dose related and prevented further dose in-creases in all cases except for Patient 2. Grade IIIneutropenia was observed on three occasions (twicein Patient 8 and once in Patient 6); Grade III throm-bocytopenia was observed on one occasion in twodifferent patients. No Grade IV neutropenia orthrombocytopenia was observed. No infections orbleeding episodes occurred. Asthenia was observedin all patients during and after the administration ofsHHT; it typically lasted for a further 2–3 days afterthe last dose. It was Grade I in three patients, GradeII in six patients, and Grade III in one patient. Therewas no relation between the severity of the astheniaand the sHHT dosage. All patients had reactions atthe site of injection (Grade I), which needed 3–7days to resolve. Nausea (Grade I) was observed inthree patients in spite of the use of antiemetic drugs.Transient fever, conjunctivitis, diarrhea, and musclepain each occurred on one occasion in differentpatients.

TABLE 2Courses, Response, and Reason for Discontinuation in 10 Evaluable Patients

Patient

BaselineQ-PCR(Cy)

Course number

Logreductionfrombaseline

Reason fordiscontinuation

1 2 3 4 5 6 7 8

Dosagelevel

Dosagelevel

Dosagelevel

Doselevel

Dosagelevel

Dosagelevel

Dosagelevel

Dosagelevel

Q-PCR (Cy) Q-PCR (Cy) Q-PCR (Cy) Q-PCR (Cy) Q-PCR (Cy) Q-PCR (Cy) Q-PCR (Cy) Q-PCR (Cy)

1 16.9 (20) 1a 1a 2a 1b 2a 2a 2a 1b 1.45 Still on trial5.4 4.7 5.7 (10) 5.3 5.0 1.2 (0) 0.6

2 4.5 (0) 1a 1a 1b 1b 0.58 Grade III asthenia3.2 2.4 1.18

3 0.028 (0) 1a 1a 2a 2a 2b 2b 3 1.60 Complete molecularresponse

0.007 0.003 0 0 0 04 4.35 (0) 1a 1a 1b 1b 1a 1a 0.10 Lack of efficacy

2.15 3.3 4.1 3.4 3.65 9 (0) 1a 1a 2a 2a 2a 2a 0.37 Lack of efficacy

2.35 2.9 3.86 0.15 (0) 1a 1a 2a 2a 2b 2b 2a 2.30 Complete molecular

response0.06 0.02 0.008 0 0.004 0.004 0

7 5.77 (0) 1a 1a 2a 2a 1b 2a 2a 1.26 Withdrew consenta

1.7 1.5 2.11 0.83 0.328 25 (44) 1a 1a 1b 1b 1b 1b 1.67 Still on trial

13.6 10.1 12.5 0.53 (0)9 1.2 (0) 1a 1a 1b 1b 2a 0.10 Still on trial

0.8 0.94710 0.38 (0) 1a 1a 1b 1b 2a 0.65 Withdrew consenta

0.18 0.27 0.21 0.087 0.084

Q–PCR: transcript level expressed as a BCR–ABL/ABL ratio percentage; Cy: percentage of Philadelphia positive metaphases in bone marrow. Dose levels 1a, 1b, 2a, 2b are defined in the text.a Patient withdrew consent on account of Grade II asthenia.

Homoharringtonine with Imatinib in CML/Marin et al. 1853

DISCUSSIONHomoharringtonine induces responses alone or incombination in various phases of CML.6 –11 O’Brien etal.6 reported cytogenetic responses in 31% of 71 pa-tients with CML in late chronic phase treated withhomoharringtonine as a constant intravenous infu-sion at an initial dose of 2.5 mg/m2 per day for 14consecutive days. The same research group9 also re-ported cytogenetic responses in 66% of 47 patients inchronic phase using homoharringtonine at a dose of2.5mg/m2 per day by constant infusion over 5 days incombination with interferon-� (intended dose 5MU/m2 per day). Homoharringtonine and imatinibhave synergistic or additive effects in CML celllines12–14 and in blast cells from advanced stage pa-tients with CML.12 To explore further this synergism invivo, the authors of the current study designed a PhaseI/II trial where patients partially sensitive to imatinibreceived escalating doses of sHHT in combinationwith imatinib. Because all patients had already re-ceived imatinib 400 mg or higher (Table 1) for � 2years (median 29.6 mos), because their transcript lev-els had remained unchanged or had increased overthe year preceding enrollment (Table 1), and becausethe dosage of imatinib had remained unchanged forthe year before the addition of sHHT, our group wasconfident that the responses observed (Table 2) couldbe attributed to sHHT. In 5 cases (50%), the transcriptlevel was reduced by � 1 log in relation to the baselinevalue (range 1.2–2.3), and the 2 patients not already inCCyR (1 with clonal evolution) achieved CCyR.

Our group could not establish a clear relationbetween dosage level and response; in some cases, thetranscript levels started to decline soon after the firstcourse of sHHT, whereas, in others, the decline didnot become apparent until patients had received morethan one course (Table 2).

Mutations in the P-loop mutations of the Abl ki-nase domain have been associated with resistance toimatinib and rapid progression towards accelerated/blastic phase.19 Two patients in this series (Patients 7and 8) had the same P-loop mutation (M244V), butboth responded to the addition of sHHT. Patient 7achieved a reduction in transcript levels of 1.26 log.Patient 8 had a reduction of 1.67 log and achievedCCyR; the originally present trisomy 8 was no longerdetectable. In both cases, the mutated clone was stilldetectable at the time of latest follow up (data notshown).

Phase I trials are usually designed to include suc-cessive cohorts of small numbers of patients treated atescalating doses of the agent under study until prede-termined levels of toxicity are reached. Our group

decided to deviate from this design because we knewfrom our previous experience with sHHT as a singleagent in accelerated phase CML11 that there is greatinterindividual variability in dosage toleration. Be-cause our patients had achieved some degree of cyto-genetic response and could be classified as low risk,we believed a design whereby the dose was escalatedin each patient to be safer and more appropriate. Thesame rationale applies to our definition of MTD.

Because we anticipated that the combination ofimatinib and sHHT could cause significant myelosup-pression, our initial dose level was 1.25 mg/m2 subcu-taneously twice daily for 3 consecutive days. This dosestill proved to be excessive and had to be reduced toan initial dose of 1.25 mg/m2 subcutaneously twicedaily for only 1 day. With this starting dosage level, thetoxicity we encountered was moderate. An MTD wasreached in all patients; it was based on hematologictoxicity in all but one patient. Grade II–III astheniawas observed frequently and was the reason why twopatients withdrew consent (Grade II). We believe thatthe preclinical data summarized above and these pre-liminary clinical observations justify further study ofthese two drugs in combination. For future studies, werecommend starting at a dose of 0.25 mg/m2 twicedaily for 1.5 days (3 doses) and escalating the dosethereafter according to tolerance, though our datasuggest that it would be safe to begin at a slightlyhigher dose, i.e., 0.25 mg/m2 twice daily for 2 days (4doses).

REFERENCES1. Kantarjian H, Sawyers C, Hochhaus A, Guilhot F, Schiffer C,

Gambacorti-Passerini C, et al. Hematologic and cytogeneticresponses to imatinib mesylate in chronic myelogenous leu-kemia. N Engl J Med. 2002;346:645– 652.

2. Hughes TP, Kaeda J, Branford S, Rudzki Z, Hochhaus A,Hensley ML, et al. Frequency of major molecular responsesto imatinib or interferon alfa plus cytarabine in newly diag-nosed chronic myeloid leukemia. N Engl J Med. 2003;349:1423–1432.

3. Marin D, Marktel S, Szydlo R, Klein JP, Bua M, Foot N, et al.Survival of patients with chronic-phase chronic myeloidleukaemia on imatinib after failure on interferon alfa. Lan-cet. 2003;362:617– 619.

4. Fresno M, Jimenez A, Vazquez D. Inhibition of translation ineukaryotic systems by harringtonine. Eur J Biochem. 1977;72:323–330.

5. Visani G, Russo D, Ottaviani E, Tosi P, Damiani D, Mich-elutti A, et al. Effects of homoharringtonine alone and incombination with alpha interferon and cytosine arabinosideon “in vitro” growth and induction of apoptosis in chronicmyeloid leukemia and normal hematopoietic progenitors.Leukemia. 1997;11:624 – 628.

6. O’Brien S, Kantarjian H, Keating M, Beran M, Koller C,Robertson LE, et al. Homoharringtonine therapy inducesresponses in patients with chronic myelogenous leukemiain late chronic phase. Blood. 1995;86:3322–3326.

1854 CANCER May 1, 2005 / Volume 103 / Number 9

7. O’Brien S, Kantarjian H, Koller C, Feldman E, Beran M,Andreeff M, et al. Sequential homoharringtonine and inter-feron-alpha in the treatment of early chronic phase chronicmyelogenous leukemia. Blood. 1999;93:4149 – 4153.

8. Kantarjian HM, Talpaz M, Smith TL, Cortes J, Giles FJ, RiosMB, et al. Homoharringtonine and low-dose cytarabine inthe management of late chronic-phase chronic myeloge-nous leukemia. J Clin Oncol. 2000;18:3513–3521.

9. O’Brien S, Talpaz M, Cortes J, Shan J, Giles FJ, Faderl S, et al.Simultaneous homoharringtonine and interferon-alpha inthe treatment of patients with chronic-phase chronic my-elogenous leukemia. Cancer. 2002;94:2024 –2032.

10. O’Brien S, Giles F, Talpaz M, Cortes J, Rios MB, Shan J, et al.Results of triple therapy with interferon-alpha, cytarabine,and homoharringtonine, and the impact of adding imatinibto the treatment sequence in patients with Philadelphiachromosome-positive chronic myelogenous leukemia inearly chronic phase. Cancer. 2003;98:888 – 893.

11. Burton CH, Ranger A, Nadal E, Avery S, Nathan I, ApperleyJF, et al. Semisynthetic homoharringtonine (Myelostat�) forchronic myeloid leukemia in accelerated phase after ima-tinib failure. Blood. 2003;102:908a.

12. Tipping AJ, Mahon FX, Zafirides G, Lagarde V, Goldman JM,Melo JV. Drug responses of imatinib mesylate-resistantcells: synergism of imatinib with other chemotherapeuticdrugs. Leukemia. 2002;16:2349 –2357.

13. Kano Y, Akutsu M, Tsunoda S, Mano H, Sato Y, Honma Y, etal. In vitro cytotoxic effects of a tyrosine kinase inhibitorSTI571 in combination with commonly used antileukemicagents. Blood. 2001;97:1999 –2007.

14. Scappini B, Onida F, Kantarjian HM, Dong L, Verstovsek S,Keating MJ, et al. In vitro effects of STI 571-containing drugcombinations on the growth of Philadelphia-positivechronic myelogenous leukemia cells. Cancer. 2002;94:2653–2662.

15. Kantarjian HM, Dixon D, Keating MJ, Talpaz M, Walters RS,McCredie KB, et al. Characteristics of accelerated disease inchronic myelogenous leukemia. Cancer. 1988;61:1441–1446.

16. Cross NC, Feng L, Chase A, Bungey J, Hughes TP, GoldmanJM. Competitive polymerase chain reaction to estimate thenumber of BCR-ABL transcripts in chronic myeloid leuke-mia patients after bone marrow transplantation. Blood.1993;82:1929 –1936.

17. Shah N, Nicoll J, Nagar B, Gorre M, Paquette R, Kuriyan J, etal. Multiple BCR-ABL kinase domain mutations confer poly-clonal resistance to the tyrosine kinase inhibitor imatinib(STI571) in chronic phase and blast crisis chronic myeloidleukemia. Cancer Cell. 2002;2:117–223.

18. Azam M, Latek RR, Daley GQ. Mechanisms of autoinhibitionand STI-571/imatinib resistance revealed by mutagenesis ofBCR-ABL. Cell. 2003;112:831– 843.

19. Branford S, Rudzki Z, Walsh S, Parkinson I, Grigg A, Szer J, etal. Detection of BCR-ABL mutations in patients with CMLtreated with imatinib is virtually always accompanied byclinical resistance, and mutations in the ATP phosphate-binding loop (P-loop) are associated with a poor prognosis.Blood. 2003;102:276 –283.

20. Sokal JE, Cox EB, Baccarani M, Tura S, Gomez GA, Robert-son JE, et al. Prognostic discrimination in “good-risk”chronic granulocytic leukemia. Blood. 1984;63:789 –799.

Homoharringtonine with Imatinib in CML/Marin et al. 1855