• chronic myeloid leukemia;
  • granulocyte-macrophage colony-stimulating factor;
  • imatinib;
  • immune modulation;
  • interferon alpha-2b


  1. Top of page
  2. Abstract


Most patients with chronic myelogenous leukemia (CML) harbor residual disease, as evidenced by molecular techniques even after treatment with high-dose imatinib (ie, 800 mg/d). Interferon alpha (IFN α) is efficacious in CML likely due to its immunomodulatory properties, and is synergistic in vitro with imatinib and granulocyte macrophage-colony stimulating factor (GM-CSF).


A study was undertaken to determine whether adding pegylated (PEG) IFN α-2b and GM-CSF to high-dose imatinib may improve the complete molecular response rate in patients with CML in chronic phase. Ninety-four patients were treated with imatinib 800 mg/d for the first 6 months, then randomly assigned to continue high-dose imatinib alone (n = 49) or in combination with PEG IFN α-2b 0.5 μg/kg/wk and GM-CSF 125 mg/m2 3× weekly (n = 45).


The median follow-up for all patients was 54 months (range, 7-70 months). There were no differences in the rates of complete cytogenetic response (87% vs 90%; P = 1.0), or of major (77% vs 77%; P = 1.0) or complete (11% vs 13%; P = 1.0) molecular response (on the international scale) at 12 months between the 2 arms, or at any time during the study. Adverse events led to PEG IFN α-2b discontinuation in all patients.


The addition of PEG IFN α-2b and GM-CSF to high-dose imatinib therapy does not improve significantly the cytogenetic or molecular response rates compared with high-dose imatinib alone. The high dropout rate in the PEG IFN α-2b arm may have compromised its potential immunomodulatory benefit. Cancer 2011. © 2010 American Cancer Society.

Standard dose imatinib mesylate therapy (ie, 400 mg/d) is associated with complete hematologic response (CHR) and complete cytogenetic response rates of 98% and 82%, respectively, in patients with chronic myeloid leukemia (CML) in chronic phase.1 High-dose imatinib (ie, 800 mg/d) renders complete cytogenetic response rates of 89% in chronic phase CML after interferon-alpha (IFN α) failure,2 and of 90% in newly diagnosed chronic phase CML.3 Achieving a complete molecular response (ie, undetectable BCR-ABL) may further improve event-free survival (EFS) and overall survival (OS).4 Alternatively, patients who fail to achieve a major molecular response (ie, 3-log reduction or ≤0.1%) may have an increased risk of losing their cytogenetic response,5 and of developing ABL1 kinase domain mutations.6 Most patients treated with imatinib harbor residual molecular disease, which supports the notion that imatinib fails to eradicate quiescent BCR-ABL1-positive stem cells.7-9

IFN α produces complete cytogenetic response rates ranging from 5% to 27%,10-13 and many of those who achieved a major molecular response have not progressed after long-term follow-up (even after IFN α discontinuation),14 which has been linked to the immunomodulatory properties of IFN α.15 IFN α synergizes in vitro with imatinib,16 which suggests that IFN α may enhance the clinical activity of imatinib. Pegylated (PEG) IFN α-2b, a more tolerable form of IFN α with a longer half-life, in combination with imatinib rendered major cytogenetic response rates of 80% at 6 months in the frontline setting, and of 60% in patients after IFN α failure, suggesting in vivo synergism.17 The combination of IFN α and granulocyte-macrophage colony-stimulating factor (GM-CSF) is a potent stimulus of dendritic cell differentiation of CML mononuclear cells.18 The addition of GM-CSF to IFN α has been reported to be well tolerated and has resulted in excellent results compared with other IFN α-based regimens in early chronic phase CML.19, 20 Based on these data, we designed a randomized phase 2 study to investigate whether adding PEG IFN α-2b and GM-CSF to high-dose imatinib would increase the rate of major molecular response and prolong remission duration in patients with early chronic phase CML compared with high-dose imatinib therapy.


  1. Top of page
  2. Abstract

Eligibility Criteria

Patients aged ≥18 years with CML in early chronic phase (<12 months from diagnosis) were eligible. Except for hydroxyurea, patients should have received no or minimal prior therapy (<1 month of IFN α and/or standard-dose imatinib). Other eligibility criteria were: 1) performance status ≤2 by the Eastern Cooperative Oncology Group scale; 2) serum creatinine and total bilirubin <1.5× the upper limit of normal; 3) patients with clonal evolution were eligible in the absence of other signs of accelerated phase (AP); 4) women of childbearing potential were required to practice effective contraception methods; and 5) signed informed consent.

Treatment Schedule

Patients were randomized to 1 of 2 arms at the time of study entry. Both arms received imatinib 400 mg twice daily for the first 6 months as a single agent. After 6 months, patients randomized to Arm A continued therapy with single-agent imatinib 400 mg twice daily or at the same dose they were receiving at the 6-month mark if dose reductions were required during the first 6 months of therapy. Those randomized to Arm B continued imatinib 400 mg twice daily (or current dose at the 6-month mark) plus PEG IFN α-2b (Schering-Plough, Kenilworth, NJ) 0.5 μg/kg subcutaneously weekly and GM-CSF (Berlex, Seattle, Wash) 125 μg/m2 3× per week. Hydroxyurea and/or anagrelide were allowed before and during the first 6 weeks of therapy for white blood debulking.

Dose Modifications

Dose adjustments were done as per Table 1. An individual drug could be discontinued in the event of excessive toxicity after appropriate dose reductions and management. Toxicities were evaluated using the National Cancer Institute Common Terminology Criteria version 3.0. Drugs were dose-adjusted as follows (Table 1).

Table 1. Dose Levels of Imatinib, PEG IFN α-2b, and GM-CSF Therapy
Dose LevelPEG IFN α-2b, μg/kg/wk scGM-CSF, μg/kg sc tiwImatinib, Total Daily Dose, mg
  • PEG indicates pegylated; IFN, interferon, GM-CSF, granulocyte-macrophage colony-stimulating factor; sc, subcutaneously; tiw, 3× per week; NA, not applicable; qow, every other week; biw, 2× per week.

  • a

    No further imatinib dose escalation beyond 800 mg daily.

11125 Monday-FridayNAa
−30.125 qow50 biw300
PEG IFN α-2b

For grade 3-4 nonhematologic toxicities, therapy was held until toxicity resolved to grade ≤1 and then resumed with 1 dose level reduction; for persistent grade 2 toxicity not responsive to supportive measures, therapy was reduced by 1 dose level.


For grade 2 nonhematologic toxicity unresponsive to therapy, imatinib was withheld until resolution to grade ≤1 and then resumed at the same dose or at the next lower dose level if the toxicity was recurrent; for grade 3-4 toxicity, imatinib was withheld until resolution to grade ≤1 and then resumed at the next lower dose level. Imatinib was also held when the absolute neutrophil count (ANC) was <0.5 × 109/L or platelets were <40 × 109/L. Therapy was then resumed at the same dose level if recovery (granulocytes >1 × 109/L or platelets >60 × 109/L) occurred within 2 weeks or at the next lower dose level if this occurred beyond 2 weeks. The lowest and highest imatinib dose allowed were 300 mg/d and 800 mg/d. In patients with prior dose reductions, imatinib could be dose-escalated if the patient remained free of toxicity requiring dose adjustments for at least 1 month. Patients with dose reductions who did not achieve a molecular remission 6 months after randomization and had no toxicities could have their dose escalated back to 400 mg twice daily.


GM-CSF dose could be reduced by 1 dose level if white blood cell (WBC) count was ≥20 × 109/L or escalated by 1 dose level if ANC was <0.5 × 109/L. For grade 3-4 nonhematologic toxicity related to GM-CSF, this was discontinued until it resolved to grade 1 and then resumed at the next lower dose level.

Patient Evaluation

Before the start of study treatment, all patients had a complete history taken and received a complete physical examination, a complete blood count (CBC), a comprehensive biochemistry panel, a bone marrow (BM) aspirate with cytogenetics, and fluorescence in situ hybridization (FISH). A CBC with differential was obtained every 1 to 2 weeks for 4 weeks, every 4 to 6 weeks for the first 12 months of therapy, and every 3 to 4 months thereafter. Total bilirubin, alanine aminotransferase or aspartate aminotransferase, and creatinine were obtained every 2 to 4 weeks for 1 month, then every 4 to 6 weeks for the first 12 months of therapy, and every 3 to 4 months thereafter. BM aspirate and cytogenetics (FISH, when cytogenetic studies were unevaluable) were performed every 3 to 4 months for 1 year, then every 6 to 12 months during therapy. Real-time polymerase chain reaction (PCR) was performed every 3 for the first 12 months and every 6 months thereafter. BM and/or peripheral blood samples with undetectable levels of BCR-ABL1 transcripts were confirmed by nested PCR as previously reported.5 Results were expressed according to the international scale (IS). Response criteria were as previously described.10 A major molecular response was defined as a BCR-ABL1 IS transcript levels of ≤0.1%, and a complete molecular response as undetectable BCR-ABL1 transcript levels confirmed by nested PCR.

Statistical Considerations

An accrual of 98 patients was initially planned. Assuming randomization of 92 patients and a 1-sided significance level of .05, the trial provided 80% power to detect a doubling in the rate of major molecular response. Patients were assigned in a 1:1 ratio to Arms A and B. The study was designed to be stopped before 98 patients were accrued if the probability of showing a significant benefit for the combination arm (Arm B) over the high-dose imatinib arm (Arm A) was <5% or >95% by 12 months. The chi-square test and Fisher exact test were used to compare patient characteristics and response rates between both treatment arms. Survival was estimated by the Kaplan-Meier method and compared by the log-rank test. OS was calculated from the start of high-dose imatinib to the date of death from any cause or last follow-up. Progression-free survival (PFS) was calculated from the start of high-dose imatinib to unsatisfactory response (similar to the European Leukemia Net recommendations for definition of failure), loss of complete cytogenetic response or CHR, progression to AP or blast phase, or death. For calculation of EFS, events included failure to achieve major cytogenetic response by 12 months, treatment discontinuation due to toxicity, loss of CHR, loss of major cytogenetic response, increasing WBC count, death resulting from any cause, or progression to AP or blast phase.


  1. Top of page
  2. Abstract

Patient Characteristics

Between May 2003 and July 2005, 94 patients with newly diagnosed Philadelphia-positive chronic phase CML were randomized to high-dose imatinib (Arm A; 49 patients) or to high-dose imatinib with PEG IFN α-2b and GM-CSF (Arm B; 45 patients) (Table 2). The median time from CML diagnosis to randomization was 25 days (range, 2-193 days). Fifteen (16%) patients had received therapy with single-agent imatinib for a median of 16 days (range, 7-30 days). Thirty-two (34%) patients were receiving hydroxyurea or anagrelide but discontinued within 7 days of study entry. The median follow-up for all patients was 54 months (range, 7-70 months).

Table 2. Patient Clinical Characteristics
CharacteristicOverallArm A, n = 49Arm B, n = 45P
  1. PB indicates peripheral blood.

Median age, y (range)48 (19-79)46 (19-73)51 (19-79).08
Median leukocyte count, ×109/L (range)26.9 (2.2-283)29 (4.7-283)25.4 (2.2-239).39
Median hemoglobin level, g/dL (range)12.4 (6.2-15.5)12.5 (9-15.5)12.4 (6.2-14.5).53
Median platelet count, ×109/L (range)323 (58-1324)316 (58-1324)335 (107-998).94
Median PB blasts, % (range)0 (0-12)0 (0-12)0 (0-7).15
Median PB basophils, % (range)3 (0-16)4 (0-13)3 (0-16).04
Median Philadelphia-positive metaphases, % (range)100 (0-100)100 (0-100)100 (0-100).42
Median PB BCR-ABL/ABL ratio, % (range)78.04 (0.03-100)78.04 (1.76-100)75.92 (0.03-100).63
Splenomegaly, No.28 (30%)18 (37%)10 (22%).18
Sokal score
 Low65 (69%)32 (65%)33 (73%).64
 Intermediate23 (24%)13 (26%)10 (22%)
 High6 (6%)4 (8%)2 (4%)
Prior imatinib therapy, No.15 (16%)7 (14%)8 (18%).78

Response and Outcome

Ninety-one (97%) of 94 patients have been followed for ≥3 months. One patient in Arm A (lost to follow-up) and 2 in Arm B (1 for noncompliance, 1 for treatment refusal) were taken off study before the 3-month first cytogenetic assessment. In addition, in Arm B, 11 (24%) patients were noncompliant with PEG IFN α-2b. Table 3 summarizes the reported response rates. CHR was achieved in 89 (95%) patients, whereas complete cytogenetic response and partial cytogenetic response were attained in 82 (90%) and 3 (3%) patients, respectively. Seventy-eight (91%) of 86 evaluable patients achieved major cytogenetic response at 12 months, including 76 (88%) with complete cytogenetic response. The rates of major cytogenetic response, complete cytogenetic response, and partial cytogenetic response were 90%, 89%, and 1% at 24 months (82 patients evaluable) and 89%, 89%, and 0% at 30 months (81 patients evaluable), respectively. Overall cytogenetic responses by arm were as follows: 44 (92%) of 48 patients assessable for response in Arm A achieved a major cytogenetic response, including 43 (90%) with complete cytogenetic response. In Arm B, 41 (95%) of 43 patients had a major cytogenetic response, including 39 (91%) with a complete cytogenetic response (P = 1.0).

Table 3. Cytogenetic and Molecular Responses for All 94 Patients Randomized to Either HDIM (Arm A) or the Combination of HDIM, PEG IFN α-2b, and GM-CSF (Arm B)
ResponseOverallaHDIMaHDIM + PEG IFN α-2b + GM-CSFaP
  • HDIM indicates high-dose imatinib; PEG, pegylated; IFN, interferon; GM-CSF, granulocyte-macrophage colony-stimulating factor; CG, cytogenetic.

  • a

    No./No. evaluable (%).

  • b

    International scale.

CG response at 12 months
 Partial2/86 (2)2/47 (4)0/39 (0).50
 Complete76/86 (88)41/47 (87)35/39 (90)1.0
Best molecular response
 ≤0.1%b80/91 (75)44/48 (92)36/43 (84).34
 Undetectable42/91 (46)24/48 (50)18/43 (42).57
Molecular response at 12 months
 ≤0.1%b66/86 (77)36/47 (77)30/39 (77)1.0
 Undetectable10/86 (12)5/47 (11)5/39 (13)1.0

Figure 1 depicts the dynamics of molecular response. The median BCR-ABL1 IS transcript levels at the time of randomization was 10.92 in Arm A and 10.62 in Arm B (P = .67). BCR-ABL1 IS transcript levels decreased at the same rate in both arms at 3 (0.26 vs 0.28) and 6 (0.025 vs 0.024) months. Although by 12 months a trend towards lower transcript levels was observed among patients in Arm B (0.017 vs 0.008) and maintained by 18 months (0.011 vs 0.007), this was not confirmed at 24 (0.0042 vs 0.0056) or at 30 (0.0028 vs 0.004) months. Major molecular response was achieved by 44 (92%) of 48 assessable patients in Arm A and by 36 (84%) of 43 in Arm B (P = .34). A complete molecular response was observed in 24 (50%) and 18 (42%) patients, respectively (P = .57).

thumbnail image

Figure 1. Molecular response over time is shown for patients receiving either high-dose imatinib (Arm A, black curve) or high-dose imatinib, pegylated interferon α-2b, and granulocyte-macrophage colony-stimulating factor (Arm B, red curve). Values are expressed according to the international standard.

Download figure to PowerPoint

Survival is shown in Figure 2A. Two patients in Arm A died after 43 and 49 months of follow-up (graft-versus-host disease and Alzheimer disease, respectively) and 2 patients, receiving therapy in Arm B, died after 11 and 26 months of follow-up (myocardial infarction and melanoma, respectively). No significant differences in survival were observed between arms (P = .89). Figure 2B depicts the PFS. Five patients progressed in Arm A and 5 in Arm B. There was no significant difference in PFS between the 2 arms (P = .81). Likewise, no differences were observed between arms A and B regarding EFS (P = .97), even when treatment discontinuation was considered an event (Fig. 2C).

thumbnail image

Figure 2. Overall survival is shown for patients receiving either high-dose imatinib (HDIM) alone or HDIM, pegylated (PEG) interferon (IFN) α-2b, and granulocyte-macrophage colony-stimulating factor (GM-CSF). (A) Kaplan-Meier estimates of the proportion of patients remaining alive, (B) the probability of progression-free survival, and (C) event-free survival by treatment arm are shown.

Download figure to PowerPoint


The most frequent nonhematologic toxicities for the total population were fatigue (n = 79, 87%), periorbital edema (n = 67, 74%), diarrhea (n = 65, 71%), nausea (n = 64, 70%), and peripheral edema (n = 61, 67%) (Table 4). Toxicities during the first 6 months of therapy were similar to those reported in prior studies of high-dose imatinib and comparable between the 2 arms.2, 3 In Arm A, high-dose imatinib-related toxicity declined significantly after 6 months of therapy. Higher rates of fatigue (59%), fever (51%), influenza-like syndrome (36%), rash (31%), injection site reaction (31%), myalgia (28%), pruritus (28%), and depression (18%) were reported after the addition of PEG IFN α-2b and GM-CSF in Arm B when compared with Arm A. Grade 3-4 toxicities in Arm B were mainly fatigue (31%), headache (10%), and diarrhea (8%). Only 6 patients developed grade 3-4 toxicity in Arm A, with the most frequent being peripheral edema (4%) and elevated liver enzymes (4%). After 6 months, significant differences were observed in the incidence of neutropenia grade 3-4 (4% vs 12%), and thrombocytopenia of any grade (8% vs 20%) between both arms.

Table 4. Adverse Events Related to Therapy With HDIM (Arm A) and HDIM, PEG IFN α-2b, and GM-CSF (Arm B)
ToxicityNo. (%) of Patients by Arm
Arm AArm BArm A + Arm B
First 6 Months, n=49After 6 Months, n=46First 6 Months, n=45After 6 Months, n=39Overall, n=94
Any gradeGrade 3-4Any GradeGrade 3-4Any GradeGrade 3-4Any GradeGrade 3-4Any GradeGrade 3-4
  1. HDIM indicates high-dose imatinib; PEG, pegylated; IFN, interferon; GM-CSF, granulocyte-macrophage colony-stimulating factor.

  2. Data represent the number of patients who developed any given toxicity in either arm of the study. Only the maximum grade was considered in the event of recurrent toxicities.

 Anemia41 (45)4 (4)25 (27)0 (0)36 (40)5 (5)27 (30)1 (1)80 (88)10 (11)
 Thrombocytopenia38 (42)13 (14)7 (8)1 (1)31 (34)9 (10)18 (20)3 (3)69 (76)25 (27)
 Neutropenia32 (35)15 (16)25 (27)4 (4)27 (30)11 (12)19 (21)11 (12)66 (73)36 (40)
 Fatigue31 (63)1 (2)11 (24)1 (2)30 (67)3 (6)23 (59)12 (31)79 (87)16 (18)
 Periorbital edema33 (67)0 (0)2 (4)0 (0)30 (67)1 (2)5 (13)0 (0)67 (74)1 (1)
 Diarrhea24 (49)1 (2)14 (30)0 (0)26 (58)0 (0)19 (49)3 (8)65 (71)4 (4)
 Nausea20 (41)1 (2)15 (33)0 (0)28 (62)0 (0)13 (33)1 (3)64 (70)2 (2)
 Peripheral edema24 (49)1 (2)14 (30)2 (4)27 (60)2 (4)12 (31)0 (0)61 (67)4 (4)
 Cramps17 (35)1 (2)10 (22)0 (0)24 (53)0 (0)9 (23)1 (3)53 (58)2 (2)
 Arthralgia22 (45)1 (2)11 (24)0 (0)22 (49)1 (2)10 (26)2 (5)52 (57)4 (4)
 Fever13 (27)0 (0)3 (7)0 (0)11 (24)1 (2)20 (51)0 (0)44 (48)1 (1)
 Myalgia16 (33)0 (0)3 (7)0 (0)20 (44)0 (0)11 (28)1 (3)43 (47)1 (1)
 Headache12 (24)0 (0)9 (20)0 (0)16 (36)0 (0)13 (33)4 (10)43 (47)4 (4)
 Pruritus15 (31)2 (4)5 (11)0 (0)13 (29)4 (8)11 (28)2 (5)37 (41)7 (8)
 Rash11 (22)3 (6)6 (13)0 (0)13 (29)2 (4)12 (31)1 (3)32 (35)6 (7)
 Dyspnea6 (12)0 (0)8 (17)1 (2)8 (18)1 (2)11 (28)1 (3)29 (32)3 (3)
 Influenzalike syndrome0 (0)0 (0)0 (0)0 (0)5 (11)0 (0)14 (36)1 (3)17 (19)1 (1)
 Elevated liver enzymes4 (8)0 (0)4 (9)2 (4)2 (4)0 (0)5 (13)2 (5)14 (15)4 (4)
 Injection site reaction0 (0)0 (0)0 (0)0 (0)0 (0)0 (0)12 (31)0 (0)12 (13)0 (0)
 Ocular hemorrhage0 (0)0 (0)4 (9)0 (0)3 (7)0 (0)3 (8)0 (0)10 (11)0 (0)
 Depression0 (0)0 (0)0 (0)0 (0)3 (7)0 (0)7 (18)0 (0)10 (11)0 (0)

Eleven (24%) and 12 (27%) of the 45 patients treated in Arm B never started therapy with PEG IFN α-2b or GM-CSF, respectively. The main reasons were patient's refusal (n = 5) or being taken off study before the start date (n = 3). PEG IFN α-2b was eventually discontinued in all 34 patients treated with this cytokine, due to influenzalike symptoms (n = 5), fatigue (n = 8), rash (n = 4), myelosuppression (n = 3), neurotoxicity (n = 3), headache/pain (n = 2), progression (n = 2), death from cardiac event (n = 1), increased liver function tests (LFT) (n = 1), injection site reaction (n = 1), comorbidities (n = 1), or other personal reasons (n = 3). GM-CSF was discontinued in all 33 patients treated with this agent due to fatigue (n = 8), rash (n = 7), influenza-like symptoms (n = 5), injection site reactions (n = 1), myelosuppression (n = 1), neurotoxicity (n = 1), headache (n = 1), progression (n = 2), death from cardiac event (n = 1), increased LFT (n = 1), comorbidities (n = 1), or other personal reasons (n = 4). Thirteen patients were taken off study in Arm A due to imatinib resistance (n = 3), increased transaminases (n = 2), renal insufficiency (n = 1), patient's request (n = 1), stem cell transplantation (n = 2; 1 in complete cytogenetic response, 1 in partial cytogenetic response), disease progression (n = 1), death of Alzheimer disease (n = 1), or loss to follow-up (n = 2). Thirteen patients in Arm B were taken off study because of resistance (n = 4), intolerance-rash (n = 1), financial reasons (n = 2), lack of compliance (n = 2), increased LFT (n = 1), metastatic melanoma (n = 1), patient's choice (n = 1), or death due to myocardial infarction (n = 1).

Dose Intensity

Because toxicity precluded the administration of the planned treatment in some patients, we determined the intensity of therapy delivered in both arms. The median duration on high-dose imatinib in Arm A was 48 months (range, 1-68 months) and the median number of days off imatinib therapy was 7 (range, 0-200 days). The median imatinib dose intensity in Arm A was 93% (range, 46%-100%). Twenty-five (51%) of 49 patients received 90% to 100% of the planned imatinib dose, 16 (33%) received 70% to 89%, and 8 (16%) received 0% to 69%. Seventeen (35%) of 49 patients had to reduce imatinib dose to ≤400 mg for a median of 392 days (range, 1-1347 days). In contrast, the median time on high-dose imatinib among patients in Arm B was 44 months (range, 1-70 months), and the median time off imatinib was 16 days (range, 0-306 days). Imatinib dose intensity in Arm B was 92% (range, 45%-100%). Twenty-five (56%) of 45 patients received 90% to 100% of the planned imatinib dose, 9 (20%) received 70% to 89%, and 11 (24%) received 0% to 69%. Twenty (44%) of 45 patients had to reduce the dose to 400 mg or lower for a median number of 117 days (range, 2-1653 days).

The median time on PEG IFN α-2b therapy for the 34 (76%) patients in Arm B who received this agent was 335 days (range, 7-1191 days), the median time off PEG IFN α therapy 14 days (range, 0-118 days), and the median time that PEG IFN α-2b had to be dose-reduced was 38 days (range, 0-1105 days). Patients received GM-CSF for a median time of 249 days (range, 1-1234 days), which was interrupted for a median of 10 days (range, 0-96 days).


  1. Top of page
  2. Abstract

The poor efficacy of imatinib in rendering complete molecular response has been linked to the inherent resistance of CML stem cells to tyrosine kinase inhibitors.21 A mathematical tetra-compartmental model postulates that CML cells are hierarchically organized into leukemic stem cells (with limitless ability to self-renew), leukemic progenitor cells (devoid of self-renewal capacity), differentiated leukemic cells (evolving toward specialized cell types), and terminally differentiated cells.22 Imatinib induces an initial dramatic decrease in BCR-ABL transcripts (5% per day), reflecting the rapid depletion of mature differentiated and terminally differentiated cells and a slower (0.8% per day) gradual turnover of leukemic progenitor cells.22 However, CML stem cells remain unaffected, in keeping with the observation that Philadelphia-positive CD34+ cells, colony-forming cells, and long-term culture-initiating cells can be readily isolated even after the achievement of complete cytogenetic response.23, 24BCR-ABL–expressing CD34+/Lin leukemic stem cells obtained from patients with CML in chronic phase remained quiescent but viable even in the presence of growth factors and imatinib,8 providing a reservoir for the generation and expansion of imatinib-resistant mutant clones.25

The mechanism of action of IFN α involves restoration of the adhesion of leukemic stem cells to marrow stroma, down-regulation of BCR-ABL expression, activation of transcription factors that regulate cell proliferation, maturation, and apoptosis,26-30 and generation of dendritic cells that can present CML-specific antigens and enhance the immune recognition and elimination of leukemic cells.18 Importantly, IFN α has been recently shown to activate dormant hematopoietic stem cells by facilitating their transition from a quiescent G0 phase to an active cell cycle state,31 whereas imatinib would act on more differentiated leukemic cells, thus suggesting that IFN α may sensitize (prime) CML cells to imatinib therapy. Therapy with imatinib 400 mg/d in combination with variable weekly doses of PEG IFN α-2b (50 μg, 100 μg, or 150 μg) given to 76 patients with untreated chronic phase CML rendered major cytogenetic response and complete cytogenetic response rates of 83% and 70%, respectively, and 68% of patients with a complete cytogenetic response also achieved major molecular response.32 Forty-five (59%) patients discontinued PEG IFN α-2b therapy, and the median administered dose ranged from 32 μg to 36 μg weekly. The recently reported SPIRIT study randomized patients with CML in early chronic phase to receive standard-dose imatinib, high-dose imatinib (600 mg/d), imatinib + cytarabine, or imatinib + PEG IFN α-2a.33 At 18 months, the PEG IFN α-2a–containing combination rendered the highest rate of major molecular response (62%) compared with standard-dose (41%) or high-dose imatinib (52%), as well as the highest rate of BCR/ABL1 transcript levels (IS) ≤0.1% (36%) compared with 19% to 25% in the other arms.33 In the present study, however, no differences in major molecular response rates were observed on addition of PEG IFN α-2a to high-dose imatinib. Several differences between the 2 studies could explain these discrepancies. First, in the SPIRIT trial, PEG IFN α-2a was initiated from the start of therapy, whereas in our study this was added after 6 months of single-agent high-dose imatinib. Second, PEG IFN α-2a was used in the French study, whereas we employed PEG IFN α-2b. Third, in our study, patients received high-dose imatinib together with PEG IFN α-2b, whereas in the SPIRIT trial patients receiving the combination used standard-dose imatinib. It is possible that when using high-dose imatinib, the addition of PEG IFN α may add little to the response rate. Finally, in our study, GM-CSF was used in addition to PEG IFN α-2b. The possibility that the concomitant use of this growth factor counteracted the potential benefit of interferon should be entertained. However, it is still possible that PEG IFN α-2b may prolong response duration, although such effect has not been seen in our study after a median follow-up of 54 months.

Several factors may have accounted for the nonsuperiority of the combination arm. First, although the imatinib dose intensity between arms was virtually identical (93% vs 92%), the overall treatment intensity in the combination arm was significantly handicapped, because 11 (24%) of the 45 patients did not start PEG IFN α-2b. Second, 32% of patients receiving PEG IFN α-2b discontinued therapy within 12 months due to toxicity. Thus, only 23 (51%) of 45 patients randomized to receive combination therapy at 6 months were actually receiving PEG IFN α-2b at 12 months. In the SPIRIT trial, 45% of patients have also discontinued PEG IFN α-2a by 12 months.33 Third, treatment-related toxicity required the discontinuation of GM-CSF in 25 (56%) patients by 12 months. Thus, in all patients randomized to Arm B, the intended treatment approach eventually failed as determined by the inability to continue the combination therapy (although most continued imatinib as a single agent). However, analyzing only those patients who received ≥6 months of PEG IFN α-2b, we still did not observe a difference in response rate or long-term outcome (data not shown).

In summary, the combination of high-dose imatinib with PEG IFN α-2b and GM-CSF has not impacted significantly the achievement of cytogenetic or major molecular response after 12 months of therapy at the dose schedule employed in the present study. The high dropout rate observed in the combination arm due to toxicity may have compromised any potential benefit from the proposed immune modulation. These results should be considered in the future design of more efficacious and tolerable therapies targeting minimal residual disease that may lead to the cure for CML.


  1. Top of page
  2. Abstract

Jorge Cortes received research support from BMS and Novartis and has served as consultant for BMS, and Hagop Kantarjian received research support from Novartis and BMS.


  1. Top of page
  2. Abstract
  • 1
    Druker BJ, Guilhot F, O'Brien SG, et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med. 2006; 355: 2408-2417.
  • 2
    Cortes J, Giles F, O'Brien S, et al. Result of high-dose imatinib mesylate in patients with Philadelphia chromosome-positive chronic myeloid leukemia after failure of interferon-alpha. Blood. 2003; 102: 83-86.
  • 3
    Kantarjian H, Talpaz M, O'Brien S, et al. High-dose imatinib mesylate therapy in newly diagnosed Philadelphia chromosome-positive chronic phase chronic myeloid leukemia. Blood. 2004; 103: 2873-2878.
  • 4
    Press RD, Galderisi C, Yang R, et al. A half-log increase in BCR-ABL RNA predicts a higher risk of relapse in patients with chronic myeloid leukemia with an imatinib-induced complete cytogenetic response. Clin Cancer Res. 2007; 13: 6136-6143.
  • 5
    Cortes J, Talpaz M, O'Brien S, et al. Molecular responses in patients with chronic myelogenous leukemia in chronic phase treated with imatinib mesylate. Clin Cancer Res. 2005; 11: 3425-3432.
  • 6
    Branford S, Rudzki Z, Parkinson I, et al. Real-time quantitative PCR analysis can be used as a primary screen to identify patients with CML treated with imatinib who have BCR-ABL kinase domain mutations. Blood. 2004; 104: 2926-2932.
  • 7
    Lowenberg B. Minimal residual disease in chronic myeloid leukemia. N Engl J Med. 2003; 349: 1399-1401.
  • 8
    Graham S, Jorgensen HG, Allan E, et al. Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood. 2002; 99: 319-325.
  • 9
    Elrick LJ, Jorgensen HG, Mountford JC, Holyoake TL. Punish the parent not the progeny. Blood. 2005; 105: 1862-1866.
  • 10
    Kantarjian HM, Smith TL, O'Brien S, Beran M, Pierce S, Talpaz M. Prolonged survival in chronic myelogenous leukemia after cytogenetic response to interferon-alpha therapy. The Leukemia Service. Ann Intern Med. 1995; 122: 254-261.
  • 11
    Kantarjian HM, O'Brien S, Anderlini P, Talpaz M. Treatment of myelogenous leukemia: current status and investigational options. Blood. 1996; 87: 3069-3081.
  • 12
    Ozer H, George SL, Schiffer CA, et al. Prolonged subcutaneous administration of recombinant alpha 2b interferon in patients with previously untreated Philadelphia chromosome-positive chronic-phase chronic myelogenous leukemia: effect on remission duration and survival. Cancer and Leukemia Group B study 8583. Blood. 1993; 82: 2975-2984.
  • 13
    Kantarjian HM, O'Brien S, Cortes JE, et al. Complete cytogenetic and molecular responses to interferon-alpha-based therapy for chronic myelogenous leukemia are associated with excellent long-term prognosis. Cancer. 2003; 97: 1033-1041.
  • 14
    Bonifazi F, de Vivo A, Rosti G, et al. Chronic myeloid leukemia and interferon-alpha: a study of complete cytogenetic responders. Blood. 2001; 98: 3074-3081.
  • 15
    Guilhot F, Lacotte-Thierry L. Interferon-alpha: mechanisms of action in chronic myelogenous leukemia in chronic phase. Hematol Cell Ther. 1998; 40: 237-239.
  • 16
    Kano Y, Akutsu M, Tsunoda S, et al. In vitro cytotoxic effects of a tyrosine kinase inhibitor STI571 in combination with commonly used antileukemic agents. Blood. 2001; 97: 1999-2007.
  • 17
    O'Brien S VS, Craddock C, Holyoake TL, Goldman JM, the UK PISCES Group. PEG-Intron and STI571 combination evaluation study (PISCES) in chronic phase chronic myeloid leukaemia [abstract]. Blood. 2001; 98: 846a. Abstract 3512.
  • 18
    Paquette RL, Hsu N, Said J, et al. Interferon-alpha induces dendritic cell differentiation of CML mononuclear cells in vitro and in vivo. Leukemia. 2002; 16: 1484-1489.
  • 19
    Cortes J, Kantarjian H, O'Brien S, Kurzrock R, Keating M, Talpaz M. GM-CSF can improve the cytogenetic response obtained with interferon-alpha therapy in patients with chronic myelogenous leukemia. Leukemia. 1998; 12: 860-864.
  • 20
    Smith B, Matsui, WH, Miller, CB, et al. GM-CSF and interferon rapidly induce cytogenetic remissions in chronic myeloid leukemia [abstract]. Blood. 2000; 96: 544a. Abstract 2339.
  • 21
    Bhatia R, Holtz M, Niu N, et al. Persistence of malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete cytogenetic remission following imatinib mesylate treatment. Blood. 2003; 101: 4701-4707.
  • 22
    Michor F, Hughes TP, Iwasa Y, et al. Dynamics of chronic myeloid leukaemia. Nature. 2005; 435: 1267-1270.
  • 23
    Deng M, Daley GQ. Expression of interferon consensus sequence binding protein induces potent immunity against BCR/ABL-induced leukemia. Blood. 2001; 97: 3491-3497.
  • 24
    Cortes J, Fayad L, Kantarjian H, O'Brien S, Lee MS, Talpaz M. Association of HLA phenotype and response to interferon-alpha in patients with chronic myelogenous leukemia. Leukemia. 1998; 12: 455-462.
  • 25
    Holtz MS, Slovak ML, Zhang F, Sawyers CL, Forman SJ, Bhatia R. Imatinib mesylate (STI571) inhibits growth of primitive malignant progenitors in chronic myelogenous leukemia through reversal of abnormally increased proliferation. Blood. 2002; 99: 3792-3800.
  • 26
    Andrews DF III, Singer JW, Collins SJ. Effect of recombinant alpha-interferon on the expression of the bcr-abl fusion gene in human chronic myelogenous human leukemia cell lines. Cancer Res. 1987; 47: 6629-6632.
  • 27
    Dowding C, Guo AP, Osterholz J, Siczkowski M, Goldman J, Gordon M. Interferon-alpha overrides the deficient adhesion of chronic myeloid leukemia primitive progenitor cells to bone marrow stromal cells. Blood. 1991; 78: 499-505.
  • 28
    Bhatia R, McCarthy JB, Verfaillie CM. Interferon-alpha restores normal beta 1 integrin-mediated inhibition of hematopoietic progenitor proliferation by the marrow microenvironment in chronic myelogenous leukemia. Blood. 1996; 87: 3883-3891.
  • 29
    Hochhaus A, Yan XH, Willer A, et al. Expression of interferon regulatory factor (IRF) genes and response to interferon-alpha in chronic myeloid leukaemia. Leukemia. 1997; 11: 933-939.
  • 30
    Pane F, Mostarda I, Selleri C, et al. BCR/ABL mRNA and the P210(BCR/ABL) protein are downmodulated by interferon-alpha in chronic myeloid leukemia patients. Blood. 1999; 94: 2200-2207.
  • 31
    Essers MA, Offner S, Blanco-Bose WE, et al. IFNalpha activates dormant haematopoietic stem cells in vivo. Nature. 2009; 458: 904-908.
  • 32
    Baccarani M, Martinelli G, Rosti G, et al. Imatinib and pegylated human recombinant interferon-alpha2b in early chronic-phase chronic myeloid leukemia. Blood. 2004; 104: 4245-4251.
  • 33
    Guilhot F, Preudhomme, C, Guilhot, J, et al. Significantly higher rates of undetectable molecular residual disease and molecular responses with pegylated form of interferon a2a in combination with imatinib (IM) for the treatment of newly diagnosed chronic phase (CP) chronic myeloid leukaemia (CML) patients (pts): confirmatory results at 18 months of pt 1 of the Spirit Phase III Randomized Trial of the French CML Group (FI LMC) [abstract]. Blood. 2009; 114. Abstract 340.