Treatment of chronic myeloid leukemia in the imatinib era

Perspective from a developing country




There is paucity of data from developing countries on the efficacy and safety of imatinib mesylate in chronic myeloid leukemia (CML). The primary objective of this study was to document complete and partial cytogenetic responses to imatinib in all phases of CML. Secondary objectives included evaluations of complete hematologic response, safety, time to progression, and survival.


Two hundred seventy-five patients in all phases of CML who received treatment with imatinib from January 2001 to December 2005 were included in the study. All patients had on bone marrow or BCR-ABL positive in peripheral blood by polymerase chain reaction.


After a median follow-up of 18 months, major cytogenetic responses (Ph <35%) in chronic phase (CP), accelerated phase (AP), and blastic phase (BP) were documented in 61%, 57%, and 28% of patients, respectively. A complete cytogenetic response was observed in 39.4%, 35.7%, and 14.3% of patients in CP, AP, and BP, respectively; and a complete hematologic response was observed in 90%, 86%, and 30%, respectively. The median time to progression at 18 months was 91% in CP and 68% in AP. The overall survivals in CP, AP, and BP at 18 months was 92%, 74%, and 38%, respectively.


Impressive hematologic, cytogenetic, and molecular responses to imatinib were observed, similar to the responses reported in patients from Western countries. Patients had good compliance, toxicity was limited, and overall quality of life was improved markedly. The results indicated that the biology of CML is not different in patients from developing countries. Cancer 2007 © 2007 American Cancer Society.

Chronic myelogenous leukemia (CML) is a pluripotent stem cell disorder that is characterized by myeloid hyperplasia in the bone marrow, leukocytosis in the peripheral blood, and splenomegaly.1–3 The disease progresses through 3 distinct phases—initial chronic phase (CP), progressing to accelerated phase (AP), and terminating in blastic phase (BP) and death.4

The hallmark of CML is the Philadelphia (Ph) chromosome and its molecular counterpart, the BCR-ABL fusion gene.5, 6 This fusion oncogene generates a protein with tyrosine kinase activity, resulting in deregulated signal-transduction pathways and causing abnormal cell cycling, the inhibition of apoptosis, and increased proliferation of cells.7, 8 In CML, Ph chromosome-positive cells constitute the major portion of all hematopoietic cells. However, there is also a pool of normal Ph-negative cells.2, 9, 10 The primary objective of treatment for CML is eradication of the Ph chromosome-positive clone and restoration of normal Ph-negative hematopoiesis.

Hydroxyurea previously was the cornerstone of treatment for CML in CP. It resulted in complete hematologic response in 50% to 75% of patients and improved the quality of life, but it did not eliminate the Ph chromosome-positive clone.11 The median survival in these patients was from 3 years to 6 years.

Interferon α (IFN-α) was used later either alone or in conjunction with low-dose cytarabine for patients with CML in CP. It produced major cytogenetic responses in 15% to 25% patients and complete cytogenetic responses in 10% of patients.12–16 The median survival of patients who received IFN-α was prolonged to approximately 7 years.17–19 Recent updates from several studies reported 9- or 10-year overall survival (OS) rates that ranged from 27% to 35%.20–24 The superiority of IFN-α over hydroxyurea and busulphan made it the standard of care.12–14, 25–27 However, treatment with IFN was associated with severe side effects.13, 19, 28 Allogeneic stem cell transplantation (SCT) is the only potentially curative treatment for this disease but is associated with significant morbidity and mortality (10–40%).2, 13, 29 SCT is recommended in young patients with poor Sokal scores who have a human leucocyte antigen (HLA)-matched sibling. Data on 10-year follow-up of patients who receiving allogeneic-hematopoietic SCT (allo-HSCT) in first CP had an OS rate of 60% and an event-free survival rate of 50%.21, 30, 31

Imatinib mesylate is a potent, selective, oral BCR-ABL kinase inhibitor that has shown substantial activity in all phases of CML32–36 It functions through competitive inhibition at the adenosine triphosphate binding site of the enzyme, which leads to the inhibition of tyrosine phosphorylation of proteins involved in BCR-ABL signal transduction.37, 38 This results in elimination of the Ph-positive clone. Imatinib has highly specific biochemical activity and a safe toxicity profile. Several studies have now confirmed the superiority of imatinib over IFN-α,33, 39 and it has become the initial standard of care for all phases of CML in the developed countries.

There is paucity of data on the clinicoepidemiologic features, response to treatment, and survival in patients from countries with limited resources who receive imatinib. Two hundred seventy-five consecutive patients with CML in all disease phases who participated in our expanded-access imatinib program were included in the current study. The primary objective of the study was to evaluate complete and partial cytogenetic responses in patients who received imatinib. Secondary objectives included evaluations of complete hematologic response (CHR), safety, time to progression (TTP), and survival.


Study Design

All patients who presented from January 2001 to December 2005 were included in the study. Written informed consent was obtained prior to the start of the treatment with imatinib. The study was approved by the institutional ethics committee.

Detection of the Ph chromosome in bone marrow aspirate or a positive BCR-ABL result from a nested polymerase chain reaction (PCR) analysis of peripheral blood or bone marrow was essential prior to inclusion in the study. Eligibility criteria included all phases of CML: CP, AP, and BP. All patients who were diagnosed within 18 months of starting on imatinib were evaluable. Patients were eligible if they had received prior treatment with hydroxyurea or IFN/low-dose cytarabine.

All patients underwent a complete history and physical examination. Laboratory data included complete blood counts with differential, hepatic and renal profiles, uric acid, and lactate dehydrogenase (LDH). Bone marrow for Ph chromosome or BCR-ABL by nested PCR analysis of peripheral blood or both were performed at a single laboratory both initially and on follow-up. Cytogenetics were repeated at 6 months and after 1 year. If the patient had a major cytogenetic response, then a repeat cytogenetic evaluation was done after12 to 18 months. Radiology examination included abdominal ultrasound or computed tomography scanning to determine spleen size.

The patients were followed at 4-week intervals with complete blood counts. Women of childbearing age were required to have a negative pregnancy test. All patients were required to use barrier contraception during treatment.

CP was defined by the presence of <5% blasts, 15% to 29% basophils, <30% blasts and promyelocytes in peripheral blood, and no extramedullary blastic disease. AP was defined as 15% to 20% blasts in peripheral blood or bone marrow, 20% basophils in peripheral blood, thrombocytopenia <100,000/mm3, and ≥30% blasts and promyelocytes in bone marrow. BP was defined as ≥30% blasts in the bone marrow or extramedullary disease.


Imatinib was started with single, daily dose of 400 mg orally in CP. Dose escalation to between 600 mg and 800 mg daily was carried out in nonresponders or in patients who had a minor or minimal cytogenetic response after ≥6 months. In patients who presented with CML in AP or BP, the initial starting dose was 600 mg daily up to a maximum of 800 mg. Response to the treatment was evaluated by physical examinations and complete blood counts every 4 weeks. Bone marrow aspirates and biopsies for cytogenetics and morphology were performed initially after 6 months and from 12 months to 18 months thereafter.

Primary Endpoint

The primary study endpoint was the rate of complete/major cytogenetic response. A complete cytogenetic response was defined as 0% Ph chromosome-positive cells in metaphase in bone marrow, and a partial cytogenetic response was defined as from 0% to 35% Ph chromosome-positive cells in bone marrow. Other categories included minor responses (36–65% Ph chromosome-positive cells in bone marrow) and minimal responses (66–95% Ph chromosome-positive cells in bone marrow).

Secondary Endpoints

Secondary study endpoints included CHR, TTP, and OS. A CHR required the disappearance of all signs and symptoms related to disease, including normalization of blasts counts in the peripheral blood and bone marrow (≤5% bone marrow blasts); leucocytes counts <10,000; normal differential counts without blasts, promyelocytes, or myelocytes; platelet counts from 150,00/mm3 to 450,000/mm3; and no evidence of extramedullary disease.

TTP was defined by any of the following events, which ever occurred first, from the start of treatment with imatinib to either death from any cause or disease progression in either AP or BP of CML. Survival (OS) was calculated from the beginning of treatment with imatinib to death from any cause.

Hematologic and nonhematologic grades 1 to 4 toxicities were evaluated at each follow-up. Quality of life was measured by a very basic questionnaire that included the initial performance status (PS) and the patient's ability to go back to the original workplace. It was filled out before the start of imatinib, then at 6 months, 12 months, and 18 months.

Statistical Analysis

SPSS software (version 13.0) was used to perform the statistical analysis. Kaplan-Meier estimates were used to calculate the survival curves in all phases of CML, and the curves were compared by using the log-rank test. A Sokal score calculator was used to subcategorize the patients into groups with low, intermediate, and high scores. Linear regression by Pearson correlation was performed to evaluate the following risk factors in relation to major cytogenetic response and survival in patients with CML in CP prior to the start of imatinib: age, white blood cells >100,000/mm3, platelets <100,000/mm3, spleen size, initial hemoglobin <10 g/dL, the presence of peripheral blasts, basophil level, prior treatment, and Sokal score.


Two hundred seventy-five patients with CML in CP, AP, and BP were included as part of our expanded-access imatinib program from January 2001 to December 2005. The patients' demographic data are provided in Table 1. The median age of our patients at presentation was 35 years. The ratio of men to women was approximately 2:1. Two hundred forty patients presented with CML in CP. Prior treatment with IFN and hydroxyurea had been received by 141 patients. The median duration of follow-up was 18 months (range, 2–45 months). The initial PS ranged between 2 and 3 in 78% patients and was 4 in 20% of patients.

Table 1. Patient Characteristics
Characteristic, n = 275No. of patients (%)
  1. PB indicates peripheral blood; IFN, interferon.

Age, y
Male gender184 (66.9)
Splenic enlargement236 (85.8)
Hemoglobin <10 g/dL149 (54.2)
White cell count, /mm3
 50,000–100,00037 (13.5)
 >100,000201 (73.1)
Platelet count, /mm3
 100,000–450,000201 (73.1)
 >450,00052 (18.9)
Blasts in PB <5/mm3219 (79.6)
Basophils in PB <7/mm3264 (96)
Sokal score
 Low105 (38.2)
 Intermediate136 (49.5)
 High34 (12.4)
Phase at presentation
 Chronic240 (87.3)
 Accelerated22 (8)
 Blastic13 (4.7)
Mode of diagnosis
Philadelphia chromosome218 (79.3)
Philadelphia chromosome/BCR-ABL33 (12)
Bcr/Abl24 (8.7)
Previous treatment (>6 mo)
 Hydroxyurea118 (42.9)
 Hydroxyurea/INF23 (8.4)

In 66 patients, we were unable to determine the cytogenetic response because of financial constraints. These patients were continued on imatinib, because they had achieved complete hematologic and clinical responses.

Currently, 237 patients (86.1%) remain alive and are receiving imatinib. Thirty-three patients have died, including 22 deaths from progressive disease and 11 deaths from infections, bleeding, and other causes.

Figure 1 illustrates the survival of patients with CML in CP, AP, and BP. The OS rates at 12 months and 18 months were 95% and 92%, respectively, for patients in CP and 85% and 74%, respectively, for patients in AP; 38% of patients in BP remained alive at 18 months.

Figure 1.

Overall survival by phase in patients with chronic myeloid leukemia (CML). CP indicates chronic phase; AP, accelerated phase; BP, blastic phase.

Chronic Phase

Two hundred forty patients presented with CML in CP. They were divided in 2 groups: One group had received prior treatment with hydroxyurea and/or IFN, and the second group was comprised of newly diagnosed patients. Table 2 shows major cytogenetic responses (complete and partial responses) and CHR in the 2 groups. We were able to perform cytogenetic analysis on 188 patients. Figure 2 shows the survival of all patients in CP according to cytogenetic response. Major cytogenetic responses were observed in 78.9% patients with newly diagnosed CML compared with only 44.9% of patients who had received prior treatment. Similarly, in patients who had lower Sokal scores, higher major cytogenetic response rates (70.3%) were observed compared with patients who had intermediate (56.1%) and high (43.8%) Sokal scores. Only 10 patients who had a complete cytogenetic response underwent testing for BCR-ABL by nested PCR, and 9 of those patients had a complete molecular response.

Figure 2.

Overall survival in chronic-phase CML according to cytogenetic response. CR indicates cytogenetic response.

Table 2. Response to Imatinib by Phase of Disease
Phase of CMLResponse: No. of patients (%)None
  1. CML indicates chronic myeloid leukemia.

Cytogenetic response, n = 209
 Chronic phase, n = 188
  Previously treated patients, n = 9824 (24.5)20 (20.4)7 (07.1)47 (48) 
  Newly diagnosed patients, n = 9050 (55.6)21 (23.3)9 (10)10 (11.1) 
 Sokal score
  Low, n = 8136 (44.4)21 (25.9)7 (08.6)17 (21) 
  Intermediate, n = 9133 (36.3)18 (19.8)6 (06.6)34 (37.4) 
  High, n = 165 (31.3)2 (12.5)3 (18.8)6 (37.5) 
 Accelerated phase, n = 145 (35.7)3 (21.4)2 (14.3)4 (28.6) 
 Blastic phase, n = 071 (14.3)1 (14.3)1 (14.3)4 (57.1) 
Hematologic response, n = 275
 Chronic phase, n = 240
  Previously treated patients, n = 119105 (88.2)6 (5)  8 (6.7)
  Newly diagnosed patients, n = 121111 (91.7)6 (05)  4 (3.3)
 Accelerated phase, n = 2219 (86.4)0 (0)  3 (13.6)
 Blastic phase, n = 134 (30.8)4 (30.8)  5 (38.5)

Figure 2 compares the survival of 209 patients in all phases of CML who underwent cytogenetic evaluation. Survival was improved significantly in patients who achieved complete and partial cytogenetic responses (P = .001). At 18 months, 92% of patients remained alive in CP.

Figure 3 shows the TTP in 240 patients with CML in CP. At 12 months and 18 months, 94% and 91% of patients, respectively, were free of disease progression.

Figure 3.

Time to progression to accelerated phase, blastic phase, or death in chronic-phase CML.

Figure 4 shows TTP in chronic phase according to cytogenetic response. At 12 and 18 months, 99% and 98% patients who had complete and major cytogenetic responses were free of disease progression (P = .008).

Figure 4.

Time to progression according to cytogenetic response in chronic-phase CML.

Significant predictive factors in the linear regression analysis for major cytogenetic response were splenomegaly (P = .006), initial hemoglobin (P = .03), white blood count (P = .02), Sokal score (P = .001), and prior treatment (P = .001). For survival in CP, the important predictive factors were prior treatment (P = .05), hemoglobin (P = .007), cytogenetic response (P = .001), and Sokal score (P = .05). For TTP, the significant predictive factors were prior treatment (P = .01), cytogenetic responses (P = .004), and hemoglobin (P = .05).

Quality of life was gauged according to the initial PS and the patient's ability to go back to their workplace. Twelve months after starting imatinib, the PS improved to 0 or 1, and 90% of patients went back to their workplace. No patient underwent allogeneic SCT or was willing to undergo allogeneic SCT.

Accelerated Phase

Twenty-two patients (8.1%) presented with CML in AP, including 19 patients who had a CHR. In 14 patients, a cytogenetic response was determined. Complete and partial cytogenetic responses were observed in 5 patients (35.7%) and 3 patients (21.4%), respectively. At 18 months, 68% of patients still had not progressed to blast crisis, and 74% patients remained alive and still had not reached their median survival.

Blastic Phase

There were 13 patients who presented in blast crisis. Of these, 4 patients achieved a CHR. Cytogenetic analyses were performed in 7 patients, and only 1 patient had a complete cytogenetic response. A partial response was observed in 1 patient. At 18 months, 38% patients remained alive. For patients in blast crisis, the median survival was 10 months (95% confidence interval, 7–12.9 months); at 18 months, 38% of patients remained alive.

Safety Profile

Imatinib was generally well tolerated in our patients. The frequencies of grade 1 through 4 adverse events that were attributable to imatinib are summarized in Table 3, which shows that no grade 3 or 4 hematologic or nonhematologic toxicities were observed in patients with CML in CP who received imatinib at a dose of 400 mg. There were no dose discontinuations or modification in these patients.

Table 3. Drug-related Hematologic and Nonhematologic Toxicity According to the Dose of Imatinib Mesylate
ToxicityNo. of patients (%)
CP 400 mg, n = 240AP/BP 600–800 mg, n = 35
Grade 1/2Grade 3/4Grade 1/2Grade 3/4
  1. CP indicates chronic phase; AP, accelerated phase; BP, blastic phase.

 Anemia4 (1.6)0 (0)6 (17.1)2 (5.7)
 Neutropenia5 (2)0 (0)10 (28.5)5 (4.2)
 Thrombocytopenia7 (2.9)0 (0)12 (34.2)5 (4.2)
 Diarrhea10 (4.1)0 (0)9 (25.7)5 (14.2)
 Myalgias/cramps10 (4.1)0 (0)12 (34.2)9 (25.7)
 Nausea/vomiting7 (2.9)0 (0)10 (28.5)7 (20)
 Edema15 (6.2)0 (0)10 (28.5)5 (14.2)
 Rash/hyperpigmentation3 (1.2)0 (0)4 (11.4)2 (5.7)
 Hepatotoxicity2 (0.8)0 (0)0 (0)4 (11.4)

Among patients in AP/blast crisis and patients who received from 600 mg to 800 mg of imatinib, grade 1 and 2 anemia, neutropenia, and thrombocytopenia were observed in 17.1%, 28.5%, and 34.2% of patients, respectively. Grade 3 and 4 anemia was observed in 5.7% patients, whereas neutropenia and thrombocytopenia were observed in 14.2% of patients each. In 5 patients who received 800 mg of imatinib, treatment had to be withheld and was resumed at 400 mg once the counts improved. Significant grade 3 and 4 nonhematologic toxicities were myalgias/cramps (25.7%) and nausea/vomiting (20.0%). No dose reductions/modifications were required for nonhematologic toxicities.


Because of the paucity of data from developing countries, we conducted the current study to determine the efficacy and safety of imatinib in our patients with CML. Secondary endpoints of the study were TTP and survival. Impressive responses with imatinib mesylate were documented at hematologic, cytogenetic, and molecular levels in all disease phases.

No reports are available from developing countries on the clinicoepidemiologic features, cytogenetic responses, and survival of patients with CML who receive hydroxyurea or interferon. We could identify only 1 study from Brazil that reported results on imatinib in patients with CML.40

Our patient population is not a selective population, because >95% of patients with CML across our country participate in 4 expanded-access programs. Most of our patients on IFN were unable to tolerate the recommended doses because of severe side effects. None of these patients had a cytogenetic response before they started imatinib.

There were some differences between our patient population and the patients reported in Phase II and III studies from developed countries. Unlike reports from high-resource countries,33, 39 the median age of our patients was only 35 years, which was similar to the median age of patients in the Brazilian study.40 Shorter life expectancy in developing countries41 or under diagnosis in the geriatric population may account for the younger age of our patients.

Major cytogenetic responses on the order of 61.2% were noted in patients with CML in CP, and those results were in concordance with the results of Kantarjian et al.32 Similar to other reports,34, 42 newly diagnosed patients and patients with low Sokal scores had superior major cytogenetic responses (Table 2); however, because of the short duration of follow-up, we were unable to comment on progression-free survival and OS.

Because of the financial constraints of our patients, molecular remissions were documented by BCR-ABL in only 9 of 10 patients who achieved a complete response. Significant predictive factors for major cytogenetic responses were Sokal score, initial hemoglobin splenomegaly, white blood count, and prior treatment. For survival in CP, significant predictive factors were prior treatment, hemoglobin, cytogenetic response, and Sokal score. For TTP, significant predictive factors were prior treatment, cytogenetic responses, and hemoglobin.

In the patients with CML in AP and BP, major cytogenetic responses and superior survival were similar to those reported by others.34, 40, 43 Our higher response rates in patients with blast crisis may have been because of small patient numbers, younger age, early presentation, and less clonally evolved disease.

Consistent with the report by Drucker et al,36 grade 3 and 4 hematologic toxicities were observed in patients who received imatinib at a dose of 800 mg. This probably reflects severely compromised bone marrow function. Frequent grade 3 and 4 nonhematologic toxicity were nausea, vomiting, and edema. Despite these toxicities, we were able to treat most patients on an outpatient basis, which probably contributed to markedly improved quality of life and satisfaction.

Imatinib was tolerated well in our patients with good compliance. We were able to give 400 mg imatinib to all of our patients in CP with no dose interruptions or modifications. No grade 3 or 4 nonhematologic toxicities were observed in CP. Some reasons may be the younger ages of our patients and their limited access to our cancer clinics, because >50% of patients live ≥150 kilometers from our center, which may cause patients to report toxicities to their local physicians. Overall quality of life is improved markedly, as also reported by the International Randomized Interferon versus ST1571 Study Group.44 Greater than 90% of patients with CML in CP have gone back to work. Patients in AP started working once they achieved CHR and a good PS. No patient with CML in BP has gone back to work.

In our patient cohort, no patient underwent or is likely to undergo allo-HSCT because of economic hardship. The significant post-transplantation morbidity and mortality from overcrowding, poor hygienic conditions, access to the treatment center, and limited finances do not allow us to recommend allo-HSCT to the vast majority of our patients, even if an HLA-identical donor is available. The limited availability of treatment options will allow us to document the natural disease progression of our patients who are receiving imatinib over time. In addition, the duration of follow-up is too short for us to comment on the survival of our patients.

We realize that our study has several weaknesses, because we were unable to document molecular remissions in most of our patients. We also were unable to perform PCR on our patients with molecular remissions for levels of BCR-ABL transcripts to evaluate disease prognosis,45 because this test was not available. Cytogenetic studies could not be done in 24% patients to assess response because of financial problems. However, our report is one of the few reports from a developing country on the efficacy and survival of patients with CML on imatinib. Our results show that the biology of this disease is not different in patients from developing countries, and the response to imatinib in these patients is similar to that observed in patients from high-resource countries.


We are grateful to Dr. Timothy Hughes for his valuable comments and advice.