Fax: (011) 49-621-383-3833
Response and resistance in 300 patients with BCR-ABL–positive leukemias treated with imatinib in a single center
A 4.5-year follow-up
Article first published online: 3 MAR 2005
Copyright © 2005 American Cancer Society
Volume 103, Issue 8, pages 1659–1669, 15 April 2005
How to Cite
Lahaye, T., Riehm, B., Berger, U., Paschka, P., Müller, M. C., Kreil, S., Merx, K., Schwindel, U., Schoch, C., Hehlmann, R. and Hochhaus, A. (2005), Response and resistance in 300 patients with BCR-ABL–positive leukemias treated with imatinib in a single center. Cancer, 103: 1659–1669. doi: 10.1002/cncr.20922
- Issue published online: 4 APR 2005
- Article first published online: 3 MAR 2005
- Manuscript Accepted: 3 DEC 2004
- Manuscript Revised: 21 NOV 2004
- Manuscript Received: 13 OCT 2004
- Novartis Pharma, Nürnberg, Germany
- German Bundesministerium für Bildung und Forschung. Grant Number: DLR e.V.-01 GI9980/6
- European LeukemiaNet
- chronic myeloid leukemia;
- clinical strategies;
- hematologic resistance
The advent of imatinib has considerably changed the treatment of chronic myeloid leukemia (CML). Early studies demonstrated high rates of hematologic and cytogenetic responses in all phases of the disease after limited observation periods.
The authors evaluated long-term outcome, rates of response, and resistance in 300 patients with BCR-ABL–positive leukemias (CML in chronic phase after failure to respond to interferon-alpha [CP], n = 139; accelerated phase [AP], n = 80; myeloid blast crisis [BC], n = 76; lymphoid BC and Philadelphia chromosome-positive acute lymphoblastic leukemia, n = 5) who entered clinical trials with imatinib in a single center after an observation time of 4.5 years.
In CP, hematologic remission was achieved in 97% and major (MCR) and complete cytogenetic remission (CCR) in 61% and 49% of patients, respectively. The chance to achieve MCR was higher in patients commencing imatinib earlier in the course of CML. In AP, the median survival period after the start of imatinib was 44 months, and MCR and CCR were observed in 31% and 26% of patients, respectively. In myeloid BC, the median survival period after the start of imatinib and after diagnosis of BC was 6 and 9 months, respectively. Hematologic resistance occurred in 25%, 41%, and 92% of patients in CP, AP, and myeloid BC, respectively, and was associated with BCR-ABL mutations in 45% of patients and with clonal evolution in 58% of patients.
The data emphasized the need for a prolonged follow-up of patients treated with imatinib to define the clinical potential of the drug and to establish methods to optimize therapy. Cancer 2005. © 2005 American Cancer Society.
Imatinib mesylate (Gleevec, Glivec, Novartis Pharmaceuticals, Basel, Switzerland) is a selective BCR-ABL tyrosine kinase inhibitor that has changed considerably the therapy of chronic myeloid leukemia (CML) in recent years. Imatinib has demonstrated significant activity in all phases of the disease.1–4 Despite promising high rates of hematologic and cytogenetic responses in multinational Phase II clinical trials, refractory disease and resistance were reported in a proportion of patients receiving imatinib monotherapy, mainly in patients with advanced-stage disease.5–7 Although hematologic and cytogenetic response data were reported early after initiation of therapy, detailed long-term evaluations of response rates, and causes of resistance are missing.
We report on the experience with 300 consecutive patients suffering from all stages of Philadelphia (Ph) and/or BCR-ABL–positive CML and acute lymphoblastic leukemia (ALL). Imatinib therapy has been conducted at a single institution in close cooperation with 118 referring centers over a 4.5-year follow-up period.
The objectives of the current study were 1) to analyze the long-term efficacy and tolerability of imatinib; 2) to evaluate the rate of imatinib refractory disease and the frequency and pathogenesis of secondary resistance; and 3) to assess the efficacy of additional drugs as therapeutic strategy for treatment optimization. The single-center approach offered the advantage of a uniform disease management. Patients have been seen regularly by the same group of hematologists and were managed via a telephone counseling hotline in cooperation with the referring physicians.
MATERIALS AND METHODS
Three-hundred patients with Ph and/or BCR-ABL–positive CML (n = 298) or refractory or recurrent Ph-positive ALL (n = 2) were recruited into Novartis-sponsored multinational Phase II studies (protocols 102, 109, or 110) or expanded access protocols (113, 114, or 115) after informed consent was obtained.1–3
Chronic phase (CP) was defined as the presence in peripheral blood (PB) or bone marrow (BM) of blasts < 10%, basophils < 20%, blasts and promyelocytes < 30%, and platelets ≥ 100 × 109/L. Hematologic and cytogenetic failure during interferon-alpha (IFN-α) therapy and intolerance to IFN-α were defined as described previously.1
Accelerated phase (AP) was characterized by a proportion of PB or BM blasts of 10–30%, or blasts and promyelocytes ≥ 30%, or basophils ≥ 20%, or platelets < 100 × 109/L unrelated to therapy, or chromosomal abnormalities other than the Ph chromosome, or progressive splenomegaly.2
Blast crisis (BC) was defined as the presence of > 30% blasts in PB or BM or as evidence for an extramedullary blast infiltration (except liver, spleen, or lymph nodes). Blasts were considered to be myeloid if ≥ 3% blasts were myeloperoxidase positive by cytochemistry, and lymphoid if myeloperoxidase staining was negative, lymphoid surface markers were expressed, and terminal deoxynucleotide transferase staining was positive in ≥ 40% of blasts. If all stains were negative and immunophenotyping demonstrated any myeloid marker (myeloperoxidase, CD13 or 33), blasts were classified as myeloid.3
Acute Lymphoblastic Leukemia
Recurrent ALL was defined as hematologic recurrence after chemotherapy, or after autologous or allogeneic stem cell transplantation (SCT).4 Clinical features of lymphoid BC resembled Ph-positive ALL and therefore these diseases were analyzed together.
Imatinib was given as an oral dose of 400 mg daily for all 139 patients in CP and for the first cohort of patients in AP (n = 6), myeloid BC (n = 8), and ALL (n = 2). Subsequent patients with advanced-stage disease were treated with 600 mg imatinib per day. The studies were approved by the institutional ethics committee and conducted according to the Declaration of Helsinki.
End points were sustained (lasting ≥ 4 weeks) hematologic and cytogenetic response. Hematologic response was defined as described previously.1–3 Return to CP was considered in patients with < 10% blasts in PB and BM.
Cytogenetic analysis of BM metaphases was performed every three months according to standard protocols.8 Cytogenetic response (CR) was based on the proportion of Ph-positive metaphases among ≥ 20 metaphases, and was defined as complete (0% Ph-positive metaphases [CCR]), partial (1–35%), major (0–35% [MCR]), minor (36–65%), minimal (66–95%), and none (> 95%). Clonal cytogenetic evolution was defined as the appearance of additional chromosomal aberrations in at least two metaphases.
Primary hematologic resistance was considered if patients failed to achieve any hematologic response. Hematologic recurrence was defined as loss of hematologic remission. Cytogenetic recurrence was considered as loss of MCR or increase of Ph-positive cells to ≥ 65% confirmed in a subsequent analysis or by a contemporaneous increase of the ratio BCR-ABL/ABL to > 14% in quantitative reverse transcription-polymerase chain reaction (RT-PCR).9
The type of BCR-ABL fusion transcripts was identified by multiplex RT-PCR.10 BCR-ABL transcripts were quantified by real-time RT-PCR as described previously.11–13. In the case of negative real-time PCR, nested PCR was performed.9 Major molecular response (MMR) was defined as ratios BCR-ABL/ABL < 0.1%, a threshold level with prognostic impact in patients with CML receiving first-line imatinib therapy.14 Undetectable BCR-ABL was defined according to previous studies.14 Screening for mutations at the ATP binding site of the BCR-ABL tyrosine kinase domain was performed as described.5
Adverse events were graded using the National Cancer Institute/National Institutes of Health Common Toxicity Criteria, Version 2.0. Grade 3/4 hematologic toxicity was assessed regarding the time of first occurrence and duration. Nonhematologic side effects were analyzed according to time of appearance after start of imatinib therapy and outcome.
Survival analysis was performed using the Kaplan–Meier method. Duration of hematologic or MCR was assessed for responding patients as the time from the first observation of response to disease recurrence, death, or last visit. For patients in CP, time to disease progression was assessed for all patients as the interval from the start of therapy to the onset of advanced-stage disease, disease recurrence, or death from any cause. Overall survival was calculated from treatment initiation to death from any cause and was censored at the time of treatment discontinuation for preparation of SCT, or at the date of the last visit. For comparison of cohorts, the Mann–Whitney test was employed.
Three hundred patients (171 males, 129 females) with a median age of 56.2 years (range, 14.6–79.6 years) who received imatinib were analyzed. Patients were referred by 112 German and 6 non-German centers (in Russia, Finland, Belgium, Austria, Italy, and Poland) and were enrolled between August 1999 and April 2002. One hundred thirty-nine patients had CML in CP, 80 in AP, 76 in myeloid BC, 3 in lymphoid BC, and 2 had Ph-positive ALL.
Chronic-Phase Chronic Myeloid Leukemia
Seventy-three of the 139 patients in CP were treated with imatinib after IFN-α intolerance, 24 after hematologic, and 35 after cytogenetic resistance to IFN-α. Seven patients had contraindications to IFN-α therapy, and two had disease recurrence after allogeneic SCT (Table 1). In all patients, imatinib was started with a daily oral dose of 400 mg. The median duration of imatinib therapy was 31 months (range, 2–49 months) and the median observation time of living patients was 34 months (range, 19–49 months). One hundred nine patients (78%) were still receiving imatinib therapy at the time of analysis. Imatinib was stopped due to progressive disease in 20 patients (14%), before allogeneic SCT in 2 patients (1%), and due to nonhematologic adverse events in 2 patients (Grade 3 nausea and dermatitis, 1%). Six patients (4%) died while receiving imatinib, 3 from myeloid BC, 1 from advanced-stage breast carcinoma, and 1 each from acute myocardial infarction and cerebrovascular insufficiency. In total, 17 patients (12%) died after a median period of 18 months (range, 6–49 months), 13 from progression of CML and 4 from CML-independent causes.
|Characteristics||CP (n = 139)|
|Median age (yrs) at diagnosis of CML (range)||55.9 (18.5–76.6)|
|Median age (yrs) at start of imatinib (range)||59.2 (21.4–81.6)|
|No. of males/females (%)||78 (56)/ 61 (44)|
|Alive (%)||122 (88)|
|ECOG performance score (%)|
|Grade 0–1||134 (96)|
|Grade 2||5 (4)|
|Interval diagnosis to start of imatinib (mos)|
|Median (range)||31 (2–312)|
|Previous therapy of CML|
|HU and/or IFN||86 (62%)|
|HU and/or IFN + ara-C||18 (13%)|
|Autologous SCT||3 (2%)|
|Allogeneic SCT||4 (3%)|
|Median follow-up of all patients, mos (range)||33 (6–49)|
|Median follow-up of living patients, mos (range)||34 (19–49)|
|Median duration of therapy, mos (range)||31 (2–49)|
|Leukocyte count (× 109/L), median (range)||10.4 (2.1–296)|
|Hemoglobin level (g/dL), median (range)||12.3 (8.9–17.8)|
|Platelet count (× 1012/L), median (range)||332 (46–1318)|
|Blasts in PB (%), median (range)||0 (0–8)|
|Basophils in PB (%), median (range)||3 (0–19)|
|Blasts in BM (%), median (range)||2 (0–12)|
|Splenomegaly n, (%)||33 (24)|
|BCR-ABL transcript type|
Complete hematologic response (CHR) was achieved in 135 of 139 patients (97%; Table 2). At 30 months, the estimated rate of disease progression-free survival was 83% and of CHR, 79% (Fig. 1). MCR and CCR were achieved in 85 (61%) and 68 patients (49%), respectively (Table 2). MCR was sustained in 86% of patients after 24 months, and in 83% of patients after 30 months. MCR after 3 and 6 months of therapy was associated with prolonged disease progression-free survival (P = 0.009 for MCR after 3 months and P = 0.004 for MCR after 6 months; Fig. 2). MCR after 12 months was associated with improved overall survival (P = 0.021) and disease progression-free survival (P = 0.0001). For patients lacking any cytogenetic response after 6 months of therapy, the probability to achieve a CCR during the subsequent 18 months was 8% and the risk of disease progression was 32% (P = 0.045; Fig. 3).
|Characteristics||CP (n = 139) (%)||AP (n = 80) (%)||Myeloid BC (n = 76) (%)|
|Major (Ph+ ≤ 35%)||85 (61%)||25 (31%)||9 (12%)|
|Complete (Ph+ = 0%)||68 (49%)||21 (26%)||6 (8%)|
|Partial (Ph+ = 1–35%)||17 (12%)||4 (5%)||3 (4%)|
|Minor (Ph+ 36–66%)||5 (4%)||7 (9%)||2 (3%)|
|Minimal/no (Ph+ > 66%)||37(27%)||29 (36%)||26 (34%)|
|Ratio BCR-ABL/ABL < 0.1%||43 (31%)||9 (11%)||2 (3%)|
|Undetectable BCR-ABL||1 (1%)||—||—|
|Complete||135 (97%)||49 (61%)||14 (18%)|
|Hematologic response without recovery of normal hematopoiesis||—||5 (6%)||2 (3%)|
|Return to CP||—||19 (24%)||21 (28%)|
|Overall||139 (97%)||73 (91%)||37 (49%)|
The probability of MCR was better for patients starting imatinib within 24 months after diagnosis (P = 0.018) than for patients with a longer pretreatment. Overall survival, disease progression-free survival, and MCR were not influenced by low leukocyte counts (cutoff value, 2.5 × 109/L), low neutrophil counts (cutoff value, 1.0 × 109/L), or low platelet counts (cutoff value, 50 × 109/L or 75 × 109/L) after 3 and 6 months of therapy, respectively. Estimated overall survival after the start of imatinib therapy was 88% at 30 months (Fig. 4). MMR was achieved in 43 of 139 patients (31%), and BCR-ABL became repeatedly undetectable in 1 patient (1%, Table 2).
Four of 139 patients (3%) did not achieve CHR, and 31 patients (22%) experienced hematologic recurrence. Cytogenetic recurrence during CHR was observed in 6 patients (4%). Of these 41 resistant patients, 19 (46%) were still in CP, 11 (27%) progressed to AP, and 11 (26%) to BC (Table 3). In response to resistance, imatinib dosage was increased to 800 mg per day in 22 patients (54%), and/or combined with other drugs (e.g., cytosine arabinoside [ara-C], homoharringtonine, hydroxyurea) in 19 patients (46%). Imatinib was withdrawn and alternative treatment was started in 16 patients (39%). Four patients (10%) died during imatinib therapy (Table 3). Of 35 patients with hematologic resistance or recurrence, 20 were examined for BCR-ABL mutations as a potential cause of resistance (Table 4). In 10 patients (50%), a BCR-ABL mutation was found. Clonal cytogenetic evolution was observed in 15 of 29 patients (52%) at the time of resistance (Table 4).
|Characteristics||CP (n = 139) (%)||AP (n = 80) (%)||Myeloid BC (n = 76) (%)|
|Overall||41 (29%)||36 (45%)||70 (92%)|
|Primary hematologic resistance||4 (3%)||7 (9%)||39 (51%)|
|Hematologic recurrence||31 (22%)||26 (32%)||31 (41%)|
|Cytogenetic recurrence during continuous hematologic remission||6 (4%)||3 (4%)||0|
|CML phase at resistance|
|CP||19 (46%)||6 (17%)||0|
|AP||11 (27%)||14 (39%)||9 (13%)|
|Myeloid BC||10 (24%)||15 (42%)||59 (84%)|
|Extramedullary myeloid BC||1 (2%)||0||1 (1%)|
|Lymphatic BC||0||1 (3%)||1 (1%)|
|No consequence||10 (24%)||3 (8%)||1 (2%)|
|Dose escalation (max. 800 mg/day)||22 (54%)||12 (33%)||11 (17%)|
|Additional therapy||19 (46%)||14 (39%)||30 (45%)|
|Stop of imatinib and alternative therapy||16 (39%)||22 (61%)||35 (53%)|
|Death during imatinib therapy||4 (10%)||10 (28%)||21 (32%)|
|Patients with hematologic resistance/recurrence||CP (n = 35) (%)||AP (n = 33) (%)||Myeloid BC (n = 66) (%)||All patients (n = 134) (%)|
|BCR-ABL mutations, no. observed/examined||10/20 (50)||13/21 (62)||10/33 (30)||33/74 (45)|
|Additional cytogenetic aberrations, no. observed/examined||15/29 (52)||8/16 (50)||16/22 (73)||39/67 (58)|
|Mutations and additional cytogenetic aberrations, no. observed/examined||5/17 (29)||2/9 (22)||4/17 (24)||11/43 (26)|
The prognostic model proposed by Marin et al.15 was applied to 139 patients in CP. The score is based on the variables of poor cytogenetic response (Ph-positive metaphases > 65%) after 3 months and neutropenia (< 1.0 × 109/L) after 45–90 days of therapy. Of 124 evaluable patients, 41 were allocated to the low-risk group, 70 to the intermediate-risk group (1 parameter was apparent), and 13 to the high-risk group (2 parameters were apparent). The probability of disease progression-free survival at 24 months was 95% for the low-risk group, 86% for the intermediate-risk group, and 92% for the high-risk group, and the probability of overall survival was 98%, 91%, and 100% for the 3 groups, respectively. Thus, we were not able to confirm the Marin model. We further applied a model recently reported by The University of Texas M. D. Anderson Cancer Center.16 This model is based on the variables loss of hematologic response to IFN-α, splenomegaly, and lack of any cytogenetic response after 3 months of imatinib therapy. There was no survival difference comparing patients with 0, 1, and 2–3 adverse factors (3-year survival rates of 96%, 88%, and 90%, respectively), or comparing patients with 0 and 1–3 adverse factors (3-year survival rates of 96% and 89%, respectively).
Accelerated-Phase Chronic Myeloid Leukemia
Eighty patients were recruited with CML in AP (Table 5). In 6 patients, treatment was started at 400 mg imatinib per day, 74 patients at 600 mg imatinib per day. At the time of evaluation, median follow-up for all patients was 28 months (range, 0.4–50 months), and for living patients 30 months (range, 8–50 months). Median duration of imatinib therapy was 24 months (range, 0.4–50 months). Forty patients (50%) were still receiving therapy with imatinib at the time of evaluation. Imatinib was stopped due to adverse effects in 6 patients (8%). Five patients (6%) underwent allogeneic SCT, 4 of them after progressive disease while receiving imatinib therapy. Thirty-one patients (39%) died, 28 from CML progression and 3 from CML-independent causes: 1 from lung carcinoma, 1 from acute myocardial infarction, and 1 from acute renal failure independent of imatinib therapy.
|Characteristics||AP (n = 80)||Myeloid BC (n = 76)|
|Median age (yrs) at diagnosis of CML (range)||56.2 (27.7–78.2)||56.2 (20.9–79.6)|
|Median age (yrs) at start of imatinib (range)||60.9 (30.9–81.8)||60.4 (24.9–79.8)|
|No. of males/females (%)||53 (66)/ 27(34)||37 (49)/ 39 (51)|
|Alive (%)||49 (61)||13 (17)|
|ECOG performance score (%)|
|Grade 0–1||60 (75)||38 (50)|
|Grade 2||20 (25)||38 (50)|
|Interval diagnosis to start of imatinib (mos)|
|Median (range)||49 (0.9–216)||32 (0.3–222)|
|Interval diagnosis of AP/BC to start of imatinib (mos)|
|Median (range)||30 (1.0–826)||1 (0–51)|
|Previous therapy of CML (%)|
|None||2 (3)||6 (8)|
|HU and/or IFN||37 (46)||45 (59)|
|HU and/or IFN + ara-C||9 (11)||7 (9)|
|Others||30 (38)||15 (20)|
|Autologous SCT||2 (3)||1 (1)|
|Allogeneic SCT||0||2 (3)|
|Median follow-up of all patients, mos (range)||28 (0.4–50)||6 (0.1–52)|
|Median follow-up of living patients, mos (range)||30 (8–50)||31 (5–52)|
|Median duration of therapy, mos (range)||24 (0.4–50)||4 (0.1–52)|
|Leukocyte count (× 109/L), median (range)||17.9 (2.4–159)||14.6 (1.3–504)|
|Hemoglobin level (g/dL), median (range)||10.8 (5.1–17.4)||9.9 (6.0–14.6)|
|Platelet count (× 1012/L), median (range)||319 (46–1135)||121 (9–2114)|
|Blasts in PB (%), median (range)||2 (0–26)||17 (0–87)|
|Blasts in BM (%), median (range)||11 (0–28)||50 (0–90)|
|BCR-ABL transcript type|
|b3a2||42 (53)||51 (67)|
|b2a2||36 (45)||18 (24)|
|b3a2/b2a2||2 (3)||5 (7)|
Hematologic response was achieved in 73 of 80 patients (91%; Table 2), being complete in 61% of patients. Median duration of hematologic response was 38 months (range, 1–49 months; Fig. 1). MCR was observed in 25 patients (31%), CCR in 21 patients (26%; Table 2). Median duration of MCR was 38 months (range, 2–36 months). The achievement of MCR after 3, 6, and 12 months of therapy was not predictive of overall survival and disease progression-free survival. MMR was achieved in 9 of 80 patients (11%), and BCR-ABL mRNA remained detectable in all patients. Median overall survival after start of imatinib was 44 months (range, 0.4–50 months; Fig. 4), after diagnosis of AP 65 months (range, 1–65 months), and after diagnosis of CML 115 months (range, 15–230 months).
Of 80 patients, 7 (9%) demonstrated primary hematologic resistance, and 26 patients (32%) experienced hematologic recurrence. Cytogenetic recurrence during continuous hematologic remission was observed in 3 patients (4%). Of these 36 resistant patients, 15 patients (42%) progressed to BC during the observation period. In 6 patients (17%), a second CP is ongoing (Table 3). In response to resistance, imatinib dosage was increased in 12 patients (33%), and ara-C, homoharringtonine, or other drugs were added in 14 patients (39%). Imatinib therapy was stopped in 22 patients (61%), and 10 patients (28%) died while receiving imatinib. A mutation of the BCR-ABL tyrosine kinase domain was detected in 13 of 21 patients (62%), and clonal cytogenetic evolution was observed in 8 of 16 resistant patients (50%; Table 4).
Chronic Myeloid Leukemia in Myeloid Blast Crisis
Seventy-six patients were treated in myeloid BC (Table 5). Eight patients were treated initially with 400 mg imatinib per day, and 68 were treated with 600 mg per day. Median follow-up was 6 months (range, 0.1–52 months) for all patients, and 31 months (range, 5–52 months) for living patients. Thirteen patients (17%) were alive, of whom 6 patients were still receiving imatinib therapy (8%). In four patients, imatinib monotherapy is maintained in CHR whereas 1 patient lost CHR and is being treated with a combination of imatinib and anagrelide and another patient never achieved CHR.
Imatinib was stopped because of disease recurrence in 45 patients (59%), nonhematologic side effects in 3 (4%), and allogeneic SCT in second CP in 4 (5%) patients. Eighteen patients (24%) died while receiving imatinib. In total, 63 patients (83%) died, all from progressive disease. Thirty-eight patients (50%) had received previous therapy for BC. In 39 patients (51%), clonal cytogenetic evolution was apparent at the start of imatinib therapy. Aberrations of chromosome 17p were observed in 12 patients (16%), trisomy 8 in 11 patients (14%), and a second Ph chromosome in 4 patients (5%).
Thirty-seven patients (49%) reached any and 14 (18%) achieved CHR (Table 2). Median duration of hematologic response was 6 months (range, 0.8–49 months), and the estimated rate of survival at 12 months was 32%, and at 24 months 18% (Fig. 1). MCR was achieved in 9 patients (12%), and CCR in 6 patients (8%; Table 2). MMR was achieved in 2 of 76 patients (3%), but BCR-ABL mRNA remained detectable in all patients. Median overall survival for all patients was 6 months (range, 0.1–52 months) after start of imatinib (Fig. 4), 46 months (range, 5–249 months) after diagnosis of CML, and 9 months (range, 0.9–61 months) after diagnosis of BC. Of the 76 patients in myeloid BC, 39 patients (51%) showed primary hematologic resistance, and 31 (41%) had hematologic recurrence (Table 3). To restore hematologic control, imatinib dose was increased to a maximum of 800 mg per day in 11 patients (17%), additional drugs (anthracyclins, ara-C, homoharringtonine, mercaptopurin, and others) were added in 30 patients (45%), and alternative treatments were administered in 35 patients (53%). Twenty-one of these patients died while receiving therapy (32%). BCR-ABL tyrosine kinase mutations were found in 10 of 33 patients (30%), and novel clonal cytogenetic aberrations were observed in 16 of 22 resistant patients (73%; Table 4).
Philadelphia-Positive Acute Lymphoblastic Leukemia and Chronic Myeloid Leukemia in Lymphoid Blast Crisis
Three patients with lymphoid blast crisis, and two patients with Ph-positive ALL were treated. One of five patients showed upfront resistance and three had hematologic recurrence. Imatinib was discontinued and alternative treatments were started in all four patients. However, these patients died of progressive disease. One patient with lymphoid BC is alive and free of detectable leukemia by PCR 14 months after allogeneic SCT.
The incidence of Grade 3/4 hematologic side effects increased with the stage of CML. Median duration of neutropenia was 2.6 months in CP, 1.3 months in AP, and 1.1 months in myeloid BC. Dominating nonhematologic adverse events were muscle cramps, nausea, edema, dermatitis, and diarrhea (Table 6). Grade 3 adverse effects were rare and included dermatitis in 7 patients and nausea in 1 patient in CP.
|Adverse events||CP (n = 139) (%)||AP (n = 80) (%)||Myeloid BC (n = 76) (%)|
|Hematologic (Grades 3 and 4)|
|Leukocytopenia||19 (14)||29 (36)||41 (54)|
|Neutropenia||27 (19)||38 (48)||43 (57)|
|Thrombocytopenia||22 (16)||32 (40)||49 (64)|
|Anemia||9 (6)||31 (39)||31 (41)|
|Grades 1 and 2||Grade 3||Grades 1 and 2||Grades 1 and 2|
|Muscle cramps||74 (53)||0||31 (39)||24 (32)|
|Periorbital edema||64 (46)||0||30 (38)||19 (25)|
|Nausea||40 (29)||1 (1)||25 (31)||36 (47)|
|Edema (lower legs)||41 (29)||0||29 (36)||22 (29)|
|Dermatitis||25 (18)||7 (5)||19 (24)||20 (26)|
|Diarrhea||34 (24)||0||16 (20)||22 (29)|
Cytogenetic Abnormalities in Philadelphia-Negative Metaphases
Chromosomal abnormalities in Ph-negative metaphases were identified in 3 of 119 patients with MCR (3%), and trisomy 8 was detected in all patients. All 3 patients received 600 mg imatinib per day for AP CML and did not show any other cytogenetic abnormalities apart from t(9;22)(q34;q11) at start of imatinib. Two patients achieved CCR after 3 and 21 months of therapy, respectively, and 1 patient reached MCR after 12 months. The proportion of trisomy 8 per Ph-negative metaphases was 28%, 42%, and 100%, respectively, for these 3 patients. Trisomy 8 was transient in 1 patient and is a continuing phenomenon in 2 other patients. All patients had been pretreated with hydroxyurea and IFN-α with additional ara-C in one patient. None of the patients has progressed or developed signs of myelodysplastic syndrome up until 22, 27, and 33 months after first observation of additional aberrations.
Multinational Phase II studies involving 32 centers in 6 countries were conducted to evaluate the impact of imatinib therapy in patients with CML in CP resistant or intolerant to IFN-α,1 in AP,2 myeloid BC,3 and lymphoid BC or Ph-positive ALL.4 Consecutively, an expanded access protocol was performed for all stages of the disease until registration of imatinib.17 Updated outcome data were reported in previous studies.17–19
The aim of the current study was to review response to imatinib in a homogeneously managed cohort of patients. The majority of patients were treated in CP after failure of IFN-α therapy. After a median of 31 months of imatinib therapy, rates of CHR (97%) and MCR (61%) are encouraging and similar to those identified in a Phase II study (CHR 95%, MCR 60%1) and other analyses (CHR 98% and MCR 64%20 and CHR 99% and MCR 66%21). Further, rates of disease progression-free and overall survival at 30 months of 83% and 88%, respectively, confirm previously published data1, 16, 20, 21 and demonstrate the continuous benefit of patients treated with imatinib. The data compare favorably with a study reporting much lower rates of disease progression-free (52%) and overall survival (63%) at 24 months.15
The positive prognostic impact of the achievement of MCR within 3 and 6 months with regard to disease progression-free survival confirms earlier observations with IFN-α treatment.22 Moreover, MCR within 12 months of therapy was associated with improvement of both disease progression-free and overall survival. Duration of pretreatment before start of imatinib was identified as an important prognostic factor associated with subsequent cytogenetic response. The rate of molecular response (31% MMR and 1% undetectable BCR-ABL) was comparable to other series.16
Patients with start of imatinib within 24 months after diagnosis had a significantly better probability to reach MCR than patients with a longer pretreatment as observed by others.20 The occurrence of initial cytopenias did not have any significant impact on the attainment of MCR. We were not able to confirm the risk score proposed by Marin et al.15 The combination of neutropenia between Days 45 and 90 of therapy and poor cytogenetic response after 3 months did not predict disease progression-free and overall survival in our cohort of patients. Neither neutropenia < 1.0 × 109/L between Days 45 and 90 nor lack of cytogenetic remission (Ph-positive > 65%) after 3 months of therapy had an impact on overall and disease progression-free survival in our cohort of patients in CP. Reasons for the limitation of the Marin score may be the poor outcome of the patient cohort analyzed in the original population and the overlapping effect of failure to achieve a cytogenetic response and cytopenia due to the lack of recovery of normal hematopoiesis. Neutropenia during imatinib therapy is associated with a lower probability to achieve MCR or CCR.23, 24
The analysis of patients in advanced-stage disease confirmed the effect of hematologic control and cytogenetic response at this disease stage on long-term benefit for patients. In AP, Kantarjian et al.25 demonstrated a survival advantage for patients reaching MCR after 6 months of therapy with imatinib. They concluded that MCR could serve as a prognostic factor and be used for decisions regarding further therapy. In our cohort, no difference in overall and disease progression-free survival was observed in patients with and without MCR after 3, 6, and 12 months. Reasons for the difference could be the prolonged follow-up combined with a longer median treatment duration which was associated with an increased rate of secondary resistance.
Despite encouraging rates of hematologic and cytogenetic remission, primary hematologic resistance was observed in 3% of patients in CP, 9% of patients in AP, and 51% of patients in myeloid BC. Hematologic recurrence occurred in 22% of patients in CP, 32% of patients in AP, and 41% of patients in myeloid BC. Median duration of hematologic remission was 38 months in AP and 6 months in myeloid BC. Strategies to overcome resistance are dose escalation, combination with other drugs (e.g., ara-C or homoharringtonine), withdrawal of imatinib with initiation of a conventional therapy like hydroxyurea, or treatment with novel tyrosine kinase inhibitors like AMN107 or BMS354825.7
In the current study, imatinib was escalated in 22 of 41 (54%) patients with hematologic resistance or cytogenetic recurrence in CP, in 12 of 36 (33%) in AP, and in 11 of 66 (17%) patients in myeloid BC. However, additional therapy was given simultaneously in most patients so that the effect of dose escalation was not independently evaluable. The combination of imatinib with other agents was chosen if patients had loss of hematologic control. Despite conflicting data with regard to survival, the combination of ara-C and IFN-α resulted in higher cytogenetic remission rates in two independent studies.26, 27 Several in vitro studies showed the additive or synergistic effect of ara-C in combination with imatinib.28, 29 According to Phase I data, the combination of imatinib and ara-C is feasible and successful regarding hematologic and cytogenetic response.30, 31 However, there is general agreement that after hematologic recurrence, dose escalation and/or combination therapy are only temporarily successful.7
In the our study, BCR-ABL mutations were identified5 in 33 of 74 (45%) resistant patients (50% in CP, 62% in AP, and 30% in myeloid BC). New additional chromosomal aberrations were observed in 58% of resistant patients (52% in CP, 50% in AP, and 73% in myeloid BC), supporting the notion of the intrinsic evolution of CML in the advanced stages. Thus, resistance to imatinib cannot be attributed to one mechanism.
Cytogenetic abnormalities have been reported to occur even in Ph-negative cells during therapy with imatinib.32–34 In our cohort, novel cytogenetic abnormalities in Ph-negative cells occurred in 3% of patients in MCR. The incidence was lower than reported by others (15%32 and 6%34). Trisomy 8 was identified simultaneously in Ph-positive and Ph-negative clones in 2 patients. A similar observation was reported for patients receiving IFN-α therapy.35 Trisomy 8 in the Ph-positive clone disappeared in both patients during continuous imatinib therapy, and only the trisomy 8 in the Ph-negative clone persisted. In 1 patient, trisomy 8 of the Ph-negative clone was transient.
We report the high efficacy of imatinib in all phases of CML. Prognostic models predicting long-term outcome should be considered with caution because the observation time is still limited. Consequent cytogenetic and molecular analyses are necessary for patients with CML to identify resistance and adapt therapy as soon as possible. The strategies to overcome resistance like dose increases or combination therapies are limited. The objectives of further studies are to prevent or decrease occurrence of resistance. It has been proposed to combine tyrosine kinase inhibitors with agents with different modes of action. Such clinical trials were initiated and are ongoing. One example is the German CML Study IV, which compares imatinib monotherapy with combination therapies.36, 37
The authors thank the members of the German chronic myeloid leukemia study group, colleagues from 118 referring institutions, and the nursing and research staff for their excellent cooperation.
- 6High frequency of point mutations clustered within the adenosine triphosphate-binding region of BCR/ABL in patients with chronic myeloid leukemia or Ph-positive acute lymphoblastic leukemia who develop imatinib (STI571) resistance. Blood. 2002; 99: 3472–3475., , , et al.
- 17Gathmann I. Imatinib expanded access program (EAP): results of treatment in >7000 patients with chronic myeloid leukemia (CML) [abstract]. Blood. 2003; 102(Suppl.1): 906a., , , , ,
- 18Imatinib mesylate therapy improves survival in patients with newly diagnosed Philadelphia chromosome-positive chronic myelogenous leukemia in the chronic phase: comparison with historic data. Cancer. 2003; 98: 2636–2642., , , et al.
- 19High dose imatinib (STI571, Gleevec®) provides durable long-term outcomes for patients (pts) with chronic myeloid leukemia (CML) in accelerated phase (AP) or myeloid blast crisis (BC): follow-up of the phase II studies [abstract]. Blood. 2003; 102(Suppl. 1): 905a–906a., , , et al.
- 30A phase I study of Gleevec (imatinib mesylate) administered concomitantly with cytosine arabinoside (Ara-C) in patients with Philadelphia positive chronic myeloid leukemia (CML). Blood. 2001; 98(Supp.1): 845a–846a., , , et al.
- 34Chromosomal abnormalities in Philadelphia chromosome-negative metaphases appearing during imatinib mesylate therapy in patients with Philadelphia chromosome-positive chronic myelogenous leukemia in chronic phase. Cancer. 2003; 98: 1905–1911., , , et al.
- 36Imatinib and beyond—the new CML study IV. A randomized controlled comparison of imatinib vs imatinib/interferon-alpha vs imatinib/low-dose AraC vs imatinib after interferon-alpha failure in newly diagnosed chronic phase chronic myeloid leukemia. Ann Hematol. 2004; 83: 258–264., , , , .