- © 2011 by American Society of Clinical Oncology
Imatinib Is Effective in Children With Previously Untreated Chronic Myelogenous Leukemia in Early Chronic Phase: Results of the French National Phase IV Trial
- Frédéric Millot⇓,
- André Baruchel,
- Joelle Guilhot,
- Arnaud Petit,
- Thierry Leblanc,
- Yves Bertrand,
- Françoise Mazingue,
- Patrick Lutz,
- Cécile Vérité,
- Christian Berthou,
- Claire Galambrun,
- Frédéric Bernard,
- Karima Yacouben,
- Pierre Bordigoni,
- Christine Edan,
- Yves Reguerre,
- Gérard Couillault,
- Françoise Méchinaud,
- Jean-Michel Cayuela and
- François Guilhot
- Frédéric Millot, Joelle Guilhot, and François Guilhot, Centre d'Investigation Clinique 802, Institut National de la Santé et de la Recherche Médicale (INSERM), University Hospital Poitiers, Poitiers; André Baruchel and Thierry Leblanc, Assistance Publique–Hôpitaux de Paris and University Paris Diderot, Assistance Publique–Hôpital Saint Louis and Robert Debré; Arnaud Petit, Hôpital Trousseau; Karima Yacouben, Hôpital Robert Debré; Jean-Michel Cayuela, INSERM U944, Hôpital Saint Louis, Paris; Yves Bertrand, Institut d'Hémato-Oncologie Pédiatrique, Hospices Civils de Lyon, Lyon; Françoise Mazingue, University Hospital, Lille; Patrick Lutz, University Hospital, Strasbourg; Cécile Vérité, University Hospital, Bordeaux; Christian Berthoux, University Hospital, Brest; Claire Galambrun, University Hospital, Marseille; Frédéric Bernard, University Hospital, Montpellier; Pierre Bordigoni, University Hospital, Nancy; Christine Edan, University Hospital, Rennes; Yves Reguerre, University Hospital, Angers; Gérard Couillault, University Hospital, Dijon; and Françoise Méchinaud, University Hospital, Nantes, France.
- Corresponding author: Frédéric Millot, MD, CIC 802 INSERM, 2 rue de la Miletrie, Poitiers, France; e-mail: f.millot{at}chu-poitiers.fr.
-
Presented in part at the 51st Annual Meeting of the American Society of Hematology, December 5-8, 2009, New Orleans, LA.
Abstract
Purpose Imatinib is the standard of care in adults with chronic myeloid leukemia (CML) in chronic phase (CP). Only a few studies to assess efficacy in children have been performed. We report on the results of the French prospective trial (ClinicalTrials.gov identifier NCT00845221) conducted in children and adolescents with newly diagnosed CML in CP.
Patients and Methods A total of 44 patients from age 10 months to 17 years with newly diagnosed CML in CP received daily imatinib 260 mg/m2. Progression-free survival, responses, and tolerance were evaluated.
Results With a median follow-up times of 31 months (range, 11 to 64 months), the estimated progression-free survival rate at 36 months was 98% (95% CI, 85% to 100%). A complete hematologic response was achieved in 98% of the patients. The rates of complete cytogenetic response (CCyR) and major molecular response (MMR) were 61% and 31% at 12 months, respectively. During follow-up, CCyR and MMR were achieved in 36 children (77%) and 25 children (57%), respectively. Overall, 30% of the patients discontinued imatinib, mainly because of unsatisfactory response. The most common adverse events were neutropenia and musculoskeletal events.
Conclusion Imatinib is effective in children with CML in CP with response rates similar to rates reported in adults. The adverse effects are acceptable, but longer follow-up studies are required to fully assess the long-term impact.
INTRODUCTION
Chronic myeloid leukemia (CML) is a myeloproliferative disorder driven by the constitutively activated BCR-ABL tyrosine kinase oncoprotein. Imatinib, a tyrosine kinase inhibitor, has been recommended as first-line treatment for patients newly diagnosed in chronic phase (CP) of the disease.1 CML is a rare disease in children, accounting for 2% to 3% of leukemias in this age group.2 Two studies conducted in children and adolescents with CML at various stages of the disease suggest efficacy of this drug in this age group.3,4
Given these preliminary results, we conducted a prospective study testing imatinib in children with newly diagnosed CML in CP to determine its efficacy and tolerability. The purposes of this article are to report on the results of this study and to compare the results with adult standards and outcomes.
PATIENTS AND METHODS
Study Design
Children age 0 to 18 years with previously untreated Philadelphia (Ph) chromosome–positive CML in CP diagnosed within 2 months were eligible. The Sokal risk score, based on sex, spleen size, hematocrit level, peripheral-blood platelet count, and blast count, was determined at diagnosis.5 Prior therapy with hydroxyurea (HU) was permitted. Patients received treatment with imatinib 260 mg/m2 (maximum, 400 mg) orally once daily. Dose escalation of imatinib to 360 mg/m2 was permitted if any of the following criteria were met: complete hematologic response (CHR) not achieved within 3 months, bone marrow containing more than 65% of Ph-positive metaphases cells at 12 months, loss of complete cytogenetic response (CCyR), or loss of major molecular response (MMR). Dose reductions of imatinib were allowed in children with grade 2 to 4 toxic effects according to the National Cancer Institute Common Toxicity Criteria. This trial is registered with ClinicalTrials.gov (identifier NCT00845221).
Patient Evaluation
Bone marrow morphology and cytogenetics were assessed at diagnosis, at 6 and 12 months after the start of imatinib, and annually thereafter when CCyR was achieved. BCR-ABL transcripts were measured in the blood at 3-month intervals using real-time quantitative polymerase chain reaction.
Response criteria were as previously described.6 In brief, CHR was defined as WBC count of less than 10 × 109/L, platelet count of less than 450 × 109/L, no immature cells in the blood, and disappearance of all symptoms and signs related to leukemia lasting for at least 4 weeks. The cytogenetic response was categorized by the percentage of Ph-negative metaphases among at least 20 analyzed metaphases as CCyR (0% Ph-positive cells), partial cytogenetic response (≤ 35% Ph-positive cells) or minor cytogenetic response (36% to 90% Ph-positive cells). MMR was defined as a BCR-ABL/BCR ratio of 0.1% or less according to the international scale.7 Undetectable BCR-ABL transcript level was defined as a BCR-ABL/BCR ratio of less than 0.001% (according to international scale). BCR-ABL kinase domain mutation analysis was performed using direct sequencing.8 Transcript type was determined as previously reported.9
From 2006 on, the criteria of the European Leukemia Net regarding imatinib failure were adopted.10,11 In brief, treatment failure was defined as loss of a CCyR or CHR, failure to achieve a CHR after 3 months of therapy, persistence of 100% Ph-positive metaphases after 6 months of therapy, 35% or more Ph-positive metaphases after 12 months, or 5% or more Ph-positive metaphases after 18 months.
End Points
The primary objective of the study was to determine the progression-free survival rate. Progression was defined as evolution to accelerated phase/blast crisis or death as a result of any cause. The secondary objectives were toxicity, rate and duration of CHR, cytogenetic response 6 and 12 months after the start of treatment, molecular response, and overall survival.
Statistical Analysis
The study is a multicenter, open-label, single-arm phase IV clinical trial. Patients were assigned to receive imatinib and were observed for analysis of safety and efficacy until imatinib discontinuation. Time to response and cumulative incidence of response were estimated using the Kaplan-Meier method. Estimates at specific time points are provided with 95% CIs. The rate of MMR was estimated overall first, with patients who reached MMR being censored at last molecular examination, whatever the reason for the lack of subsequent follow-up and irrespective of the stop of imatinib. To assess the efficiency of imatinib, the cumulative incidence of MMR was also analyzed using censoring for patients with insufficient follow-up and still on imatinib only; patients who stopped imatinib without reaching MMR were considered nonresponders and were not censored. After discontinuing study treatment, patients were observed only for progression to accelerated phase or blast crisis (major end point) and overall survival until death, loss to follow-up, or withdrawal of consent. Survival analyses were performed on an intent-to-treat basis. Univariate analyses to identify prognostic factors for CCyR at 6 and 12 months (ie, age, sex, Sokal risk score, transcript type, leukocyte count, and hemoglobin level) were carried out using the Fisher's exact test. The significance level was set at P = .10 for univariate analyses and P = .05 otherwise. Analyses were performed using SAS version 9.0 (SAS Institute, Cary, NC).
Study Conduct
The protocol was reviewed and approved by the Agence Française de Sécurité Sanitaire des Produits de Santé and the relevant ethical committee. The study was conducted in accordance with the Declaration of Helsinki. Parents provided written informed consent. Children also provided written consent after an age-appropriate written explanation was given.
RESULTS
From March 2004 until December 2008, 44 patients were enrolled in 16 academic hospitals in France. The 34 patients younger than 15 years of age in our cohort represent 81% of potentially eligible children with CML diagnosed in France during the same period of time according to the French national cancer registries of children younger than 15 years old (J. Clavel, personal communication, May 2010).
Follow-Up
As of May 2010, the median follow-up from the time imatinib was started was 31 months (range, 11 to 64 months). Eighty-four percent and 75% of patients had more than 18 months and 24 months of observation, respectively. All patients but one are alive.
Patients and Treatment
A total of 44 children and teenagers with newly diagnosed CML were analyzed. Patients characteristics are listed in Table 1. The median age was 11.5 years (range, 10 months to 17 years) at enrollment. The majority of our patients presented with a high Sokal risk score. Among the 44 patients, 29 (66%) had received treatment with HU before the start of imatinib for a median time of 2 weeks (range, 6 days to 2.5 months). All patients but one stopped HU at the onset of imatinib. One patient received the combination of HU and imatinib during 7 months; however, he was not excluded from analysis. The median average daily dose of imatinib received by the patients was 259 mg/m2/d (range, 188 to 403 mg/m2/d) during the study period for a median duration of 27 months (range, 2 to 77 months).
Demographics and Clinical Characteristics of the Study Group at Diagnosis
Progression-Free Survival
Of the 44 patients, only one patient experienced progression to lymphoblastic crisis at 11 months from the onset of imatinib and died 21 months after the start of imatinib. As a result, the estimated progression-free survival rate at 36 months was 98% (95% CI, 85% to 100%).
Hematologic, Cytogenetic, and Molecular Responses
Of the 44 patients, 43 (98%) achieved a CHR. The remaining child discontinued the treatment because of occurrence of a severe adverse event 2 months after the start of imatinib, still showing signs of the disease. After 3 months of imatinib treatment, 38 patients (86%) reached a CHR, and five patients (12%) still showed minor signs of disease (palpable spleen, n = 4; leukocyte count just > 10 × 109/L, n = 1). All of these patients achieved a CHR with a median time of 4 months (range, 3.5 to 7 months) after the start of imatinib. One patient (minor cytogenetic response as best response) lost the CHR 11 months after the start of imatinib with occurrence of mutation and experienced progression to lymphoblastic phase.
Cytogenetic and molecular responses to imatinib are listed in Table 2. At 12 months, 27 patients (61%) had achieved CCyR. Six additional patients obtained the first CCyR after 12 months (median, 18 months; range, 16 to 20 months). During follow-up, 36 children (77%) achieved CCyR (median time, 6 months; range, 6 to 20 months). Pretherapy factors such as age, sex, Sokal risk score, transcript type, leukocyte count, and hemoglobin level were not predictive for the achievement of CCyR at 6 or 12 months. We did not find any association between the dose of imatinib administered during the first 12 months of treatment and the probability of achieving CCyR.
Cytogenetic and Molecular Responses to Imatinib Therapy
During follow-up, 25 patients (57%) achieved an MMR. On an intent-to-treat basis, with patients who did not reach MMR being censored at last contact, the estimated rate of MMR at 36 months was 71% (95% CI, 53% to 87%). When patients who discontinued imatinib without reaching MMR were considered as experiencing treatment failure instead of being censored, the cumulative incidence of MMR on imatinib therapy at 36 months was 60% (95% CI, 46% to 75%). Among the 25 patients who achieved an MMR, 14 reached a BCR-ABL/BCR ratio of 0.01% or less, including 10 patients in whom the transcript was undetectable. At the time of analysis, three patients had lost their MMR. Cumulative incidences of hematologic, cytogenetic, and molecular responses are reported in Figure 1.
Cumulative incidences of hematologic, cytogenetic, and molecular responses to imatinib. At 36 months, the complete hematologic response (CHR) rate was 98% (95% CI, 90% to 100%), the major cytogenetic response (MCyR) rate was 84% (95% CI, 71% to 93%), the complete cytogenetic response (CCyR) rate was 83% (95% CI, 69% to 92%), the major molecular response (MMR) rate was 60% (95% CI, 46% to 75%), the rate of 4-log reduction in transcript levels (CMR4) was 35% (95% CI, 21% to 55%), and the rate of undetectable transcript (UND) was 17% (95% CI, 8% to 35%).
Patients Remaining on Trial
Of 44 patients treated with first-line imatinib, 31 (70%) continued to receive imatinib treatment. At the time of data analysis, 13 patients (30%) had permanently discontinued imatinib after a median time of 13.5 months (range, 10 weeks to 33 months). Reasons for discontinuation are provided in Figure 2.
Rates and reasons for discontinuation of imatinib by period of time. Reasons for discontinuation included unsatisfactory therapeutic effect (cytogenetic refractoriness, n = 1; loss of complete hematologic response, complete cytogenetic response [CCyR], or molecular response, n = 3; nonachievement of major molecular response [MMR], n = 5), adverse events (n = 2), and allogeneic stem-cell transplantation in MMR (n = 1) or CCyR (n = 1) according to physician choice. The two patients who lost their hematologic response or cytogenetic response were tested for mutation; the first patient had an L248V mutation, and the second patient had an L384M mutation.
Nine patients had their dose escalated because of unsatisfactory therapeutic effect of standard-dose imatinib. The median time to dose escalation was 10 months (range, 7 to 25 months). The median time on imatinib after the initial dose increase was 5 months (range, 1 to 25 months), with a median dose-intensity of 358 mg/m2 daily (range, 278 to 500 mg/m2) after dose escalation. Four of these patients did not achieve or regain a hematologic, cytogenetic, or molecular response, and one patient achieved an MMR after escalation. A longer follow-up is expected to assess the response in the remainder of patients. Nine patients had their doses reduced because of toxicity; dose reduction was from 250 to 110 mg/m2 (median, 167 mg/m2) in eight of the patients and from 400 to 273 mg/m2 in the remaining patient who started the treatment with higher dose.
Adverse Effects
Table 3 lists the most common adverse effects observed. No treatment-related death occurred. All patients experienced at least one adverse effect, and grade 3 or 4 hematologic or nonhematologic toxicity was reported in 20 patients (45%). neutropenia was the most frequently reported hematologic adverse effect related to imatinib therapy. No patients had to discontinue therapy permanently because of myelosuppression. Of the nonhematologic imatinib-related adverse effects, musculoskeletal events were the most frequent. Two patients dropped out of the study at 2 and 28 months after the start of imatinib because of unacceptable severe nonhematologic adverse effects (intolerable diffuse musculoskeletal pain in the first patient and grade 3 liver toxicity in the second patient).
Adverse Effects in the Total Population (N = 44)
Toxicities were generally reversible with temporary treatment discontinuation or dose reduction; six patients (14%) required a temporary interruption of imatinib, and six patients (14%) required a transient treatment withdrawal for hematologic or nonhematologic toxicity of ≥ grade 1. Sixteen episodes of temporary interruption of imatinib were noted in 11 patient for hematologic toxicity (four episodes in three patients) or nonhematologic toxicity (four episodes in three patients) of ≥ grade 1, or for other reasons (eight episodes in seven patients). The median duration of the interruptions was 6 days (range, 1 day to 3 months). Nine patients had their doses reduced because of hematologic toxicity (n = 2) or nonhematologic toxicity (n = 7); dose reduction was from 250 to 110 mg/m2 (median, 167 mg/m2) in eight patients and from 400 to 273 mg/m2 in the remaining patient who started the treatment with a higher dose. The most common adverse effects leading to dose reduction or treatment interruption were musculoskeletal events (grades 1 to 3) in seven patients, thrombocytopenia (grades 1 to 3) in five patients, and neutropenia (grades 2 and 3) in three patients.
DISCUSSION
We prospectively treated 44 children and adolescents with newly diagnosed CML in early CP with imatinib 260 mg/m2. This dosage was selected because it is equivalent in terms of drug exposure to the dose of 400 mg in adults, which is the initial recommended dose in adults with CML in CP.3
Few data are available on the proportion of patients recruited in clinical trials. The participation rate in our clinical trial (81%) seems to be higher than that of 52% to 64% reported in trials of adults with CML.12,13
The milestones for the management of adult patients with CML in CP are achievement of CHR at 3 months, CCyR at 12 months, and MMR at 18 months.1 Achievement of a CCyR is currently accepted as a surrogate marker for survival, and patients who achieve a CCyR and MMR within 12 months after initiation of imatinib have a low risk of long-term progression.1,14,15
On the basis of an intent-to-treat analysis, our prospective study showed CHR, CCyR, and MMR rates of 98%, 61%, and 31% at 12 months, respectively. These results are comparable to those obtained in adults with newly diagnosed CML in CP treated with imatinib 400 mg/d in prospective trials reporting a CHR rate at 3 months of 86% to 96% and CCyR and MMR rates at 12 months of 57% to 69% and 22% to 47%, respectively.14,16–20 A high percentage of patients with a high Sokal risk score was observed in our cohort of patients. It has been reported that patients with higher Sokal risk scores have a lower rate of CCyR.13 However, the usefulness of prognostic scores such as Sokal risk score has not been well established in children and adolescents, limiting their use in younger patients.5 Similar to adults, a predominance of the b3a2 transcript was observed in our patients.21 The number of our patients was relatively small, limiting statistical analysis of transcript type as a prognostic factor.
Despite a short median observation time (31 months), this trial shows clinical benefit, with 70% of the patients with CML in CP remaining on study treatment and 77% being in CCyR with a continuing increase in the rate of MMR. The overall survival and progression-free survival rates are high (only one patient died after progression of disease). A 6-year update of the International Randomized Study of Interferon Versus STI571 (IRIS) trial revealed a cumulative best CCyR rate of 82%, an estimated rate of freedom from progression to accelerated or blast phase of 93%, and an overall survival rate of 95% in adults with CML receiving imatinib.22
Approximately 20% of our patients received imatinib dose escalation because of unsatisfactory therapeutic response with the standard dose of imatinib, which is a proportion similar to that reported in adults.23 However, few of our patients attained a benefit after dose increase instead of a median dose-intensity of 358 mg/m2/d, which provides systemic exposure similar to 600 mg administered daily in adults.
The proportion of our patients who discontinued imatinib (30%) is quite similar to that reported in adults (25% to 28%).14,24 Discontinuation of imatinib in our patients was mainly a result of unsatisfactory responses, which remain the main cause of discontinuation in adults.
In this study, we confirm that a daily imatinib dose of 260 mg/m2 was tolerable. The most common grade 3 to 4 toxicities were hematologic. Musculoskeletal events were the most frequent nonhematologic grade 3 to 4 adverse events. The incidence of reported adverse events is consistent with previous studies in adults receiving imatinib 400 mg.22 In the present study, most of these adverse events were resolved by dose reduction or temporary interruption of imatinib. However, two patients were removed from the study because of severe adverse effects related to imatinib therapy. This is consistent with the results of the IRIS trial, in which 5% of the patients treated with imatinib discontinued imatinib treatment as a result of adverse events at a median follow-up of 6 years.22 In conclusion, the analysis of this study, performed at a median follow-up of 30 months, shows a satisfactory rate of response and an acceptable rate of adverse effects of imatinib as initial therapy in children and adolescents with newly diagnosed CML in CP.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Frédéric Millot, André Baruchel, Joelle Guilhot, Arnaud Petit, Thierry Leblanc, Yves Bertrand, Françoise Mazingue, Patrick Lutz, Cécile Vérité, Christian Berthou, Claire Galambrun, Frédéric Bernard, Karima Yacouben, Pierre Bordigoni, Christine Edan, Yves Reguerre, Gérard Couillault, Françoise Méchinaud, Jean-Michel Cayuela, François Guilhot
Provision of study materials or patients: Frédéric Millot, André Baruchel, Joelle Guilhot, Arnaud Petit, Thierry Leblanc, Yves Bertrand, Françoise Mazingue, Patrick Lutz, Cécile Vérité, Christian Berthou, Claire Galambrun, Frédéric Bernard, Karima Yacouben, Pierre Bordigoni, Christine Edan, Yves Reguerre, Gérard Couillault, Françoise Méchinaud, Jean-Michel Cayuela, François Guilhot
Collection and assembly of data: Frédéric Millot, André Baruchel, Joelle Guilhot, Arnaud Petit, Thierry Leblanc, Yves Bertrand, Françoise Mazingue, Patrick Lutz, Cécile Vérité, Christian Berthou, Claire Galambrun, Frédéric Bernard, Karima Yacouben, Pierre Bordigoni, Christine Edan, Yves Reguerre, Gérard Couillault, Françoise Méchinaud, Jean-Michel Cayuela, François Guilhot
Data analysis and interpretation: Frédéric Millot, Joelle Guilhot, Jean-Michel Cayuela
Manuscript writing: All authors
Final approval of manuscript: All authors
Acknowledgment
We thank Irene Roberts (Professor, Hematology, Imperial College, London, United Kingdom) for helpful comments on the article. We also thank Sophie Zin-Ka-Ieu for the data monitoring.
Footnotes
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Clinical trial information can be found for the following: NCT00845221.
- Received September 21, 2010.
- Accepted May 5, 2011.
