Introduction
The Philadelphia chromosome (Ph) is the most common cytogenetic subgroup in adult acute lymphoblastic leukemia (ALL), comprising 25%–30% of all cases [
]. Historically, patients with adult Ph-positive ALL had poorer outcomes compared to other ALL subgroups. With the advent of tyrosine kinase inhibitors (TKIs), the clinical outcomes for Ph-positive ALL significantly improved over the last decade, with complete remission (CR) rates exceeding 90% and long-term survival of approximately 50% in most trials [
–
]. Currently, the backbone therapy of Ph-positive ALL is a combination of TKIs (imatinib, dasatinib, and ponatinib) with multiagent chemotherapy followed by allogeneic hematopoietic stem cell transplant (allo-HSCT) for eligible candidates [
–
].
However, the approximately 50% long-term survival rate in this group of patients is still less satisfactory. In addition, for patients that lack a suitable human leukocyte antigen (HLA)-matched donors, the choice of best postremission therapy is still not properly established. Thus, innovative approaches are required to further improve survival rates.
Moreover, a large-scale data on the outcomes and risk factors in adult patients with ALL who are treated with a uniform protocol in China is required. The outcomes of adult ALL in China were worse than in western countries [
]. In 2009, the Chinese Acute Lymphoblastic Leukemia Cooperative Group (CALLG) conducted a multiple center trial to evaluate the efficacy of a new treatment protocol (CALLG2008) on the therapeutic outcome in adult ALL (The clinical trial registry at www.chictr.org.cn as ChiCTR-TNC-09000397). Imatinib was incorporated into the regimen of Ph-positive ALL. Early analysis from the CALLG2008 network demonstrated the effectiveness and manageable toxicity of CALLG2008 regimen in adult ALL [
]. However, limited data are available for the efficacy of the combination of imatinib with CALLG2008 backbone in Ph-positive ALL.
Herein, we describe the clinical characteristics and analyze the prognostic risk factors for the survival of a large cohort of Ph-positive ALL patients treated with imatinib plus CALLG2008 protocol.
Patients and methods
Patients
Between May 2009 and December 2015, consecutive patients enrolled in the imatinib combined CALLG2008 protocol were collected through our database. All cases of ALL were diagnosed based on the bone marrow cytology, immunological analyses, cytogenetic, and molecular test according to the WHO2008 classification criteria. Patients with Ph-positive ALL were confirmed at presentation for BCR-ABL1 rearrangement. Written informed consent was obtained in accordance to theDeclaration of Helsinki, and the protocol was approved by the hospital institutional ethics review board.
Treatment schema
The schedules of CALLG2008 comprised an induction course, consolidation, maintenance, and central nervous system (CNS) prophylaxis. Details of the protocol of CALLG2008 for Ph-positive ALL are summarized in Table 1. Imatinib (400 mg daily) was administered concurrent with chemotherapy as soon as the positive result for BCR-ABL1 was confirmed. Imatinib interruption was allowed until neutrophils recovered to 109/L or higher when severe neutropenia (less than 0.2 × 109/L,>7 d) occurred. For grade III/IV non-hematologic toxicity, imatinib was interrupted until recovery to grade I or normal. All patients received convention supportive measures based on the institutional guidelines.
Following achievement of CR, allo-HSCT (HLA-matched donor or haploidentical related donor) was recommended to all eligible patients. Conditioning and graft versus host disease (GVHD) prophylaxis for allo-HSCT was performed according to the institutional and Chinese consensus [
–
]. Restarting imatinib was planned to follow allo-HSCT, and to be continued for 2 years if tolerated. Patients with contraindications to allo-HSCT, or unwilling to proceed with this medication shall continue with consolidation and maintenance chemotherapy. Salvage treatment after relapse was based on the discretion of the investigators and the patients.
BCR-ABL1 minimal residual disease (MRD) monitor and mutation assessment
MRD monitoring for the BCR-ABL1 transcript was evaluated by reverse-transcription quantitative polymerase chain reaction (RT-qPCR) using bone marrow samples according to standard methods [
]. The amount of BCR-ABL1 mRNA was expressed relative to the amount of ABL1 mRNA. Major molecular response (MMR) is the ratio of BCR-ABL1/ABL1 less than 0.1%, and molecular CR is the absence of detectable MRD (minimum level of sensitivity 10
-4) [
,
]. BCR-ABL1 kinase domain mutation analysis was performed on cDNA with a nested PCR strategy.
Definitions and statistical analysis
CR is≤5% marrow blasts with peripheral blood (PB) neutrophil count≥1 × 109/L and platelet count≥100 × 109/L, with no evidence of extra medullary leukemia. Recurrence is the presence of leukemia cells in the bone marrow aspirate (>5%) or extramedullary sites in patients with previously documented CR. Event-free survival (EFS) is time from the date of diagnosis until an event. Overall survival (OS) was calculated from the date of diagnosis to the last follow-up or death. Patients who did not experience an outcome event were censored at the date of last follow-up. Outcome was updated on April 30, 2016.
Categorical variables were compared using the Fisher exact or Chi-square test, and continuous variables were analyzed using the non-parametric Mann–Whitney U-test. The estimated median follow-up time was calculated by the reverse Kaplan–Meier. Survival analyses were computed by the Kaplan–Meier method and compared with the log-rank test. Univariate analysis was performed using Cox regression models. Multivariate analysis was performed using Cox proportional hazard regression model, in which the variables with prognostic potential from univariate analysis were entered. A limited backward selection procedure was used to exclude redundant variates. Results were expressed as hazard ratios (HR)±95% confidence intervals (95% CI). Analyses were performed using GraphPad Prism software version 5.0 (GraphPad software, San Diego, CA, USA) and/or SPSS version 21.0 (SPSS, Inc., Chicago, IL, USA).P<0.05 was considered statistically significant.
Results
Patient characteristics
Among the 592 consecutive adult patients diagnosed with ALL during the study period, 177 (29.9%) showed positive for BCR-ABL1. A total of 153 patients were enrolled in the imatinib plus CALLG2008 schedule. The patients include 81/153 (52.9%) males and 72/153 (47.1%) females. The median age of the patients was 40 years (range, 18–68 years), with 19 patients older than 60 years. The main demographic and clinical characteristics of the study cohort are presented in Table 2.
Hematologic and molecular responses
Imatinib was started at a median of 4 days (range, 0–8 d) after the initiation of chemotherapy. Following induction therapy, five patients showed induction failure. Among them, three patients died during induction due to sepsis (n = 2) or pulmonary hemorrhage (n = 1). Two patients were refractory to primary induction. Overall, 148 patients (96.7%) achieved CR.
A total of 87 patients had MRD assessment for BCR-ABL1 by quantitative PCR, both at CR and at two cycles of consolidation. After induction, 32 of 87 evaluable patients (36.8%) achieved MMR, including 20 patients (23.0%) in molecular CR after induction. In addition, during two cycles of consolidation, 65 of 87 evaluable patients (74.7%) achieved MMR; 29 patients (33.3%) in molecular CR.
Survival outcomes analysis
Median follow-up duration of patients was 24.2 months (range 0.5–76.3 months). Among the 148 patients in CR1 after induction, 64 (43.2%) patients had recurrent leukemia. At the time of relapse, 92.2% of patients (n = 59) had systemic relapse, whereas 7.8% (n = 5) had isolated extramedullary relapse. Fifty-four patients (35.3%) died during follow-up. A total of 99 patients remain alive in CR1 for the final follow-up.
Using the Kaplan–Meier method, we found that the estimated 3-year OS was 49.5% (95% CI: 38.5%–59.5%; Fig. 1A). The estimated 3-year EFS was 49.2% (95% CI: 38.3%–59.2%; Fig. 1B). In all patients, the median EFS was 29 months, and the median OS was 33 months.
Comparison between the survival of patients with and without allo-HSCT
For patients in CR1, 58 out of 148 patients (39.2%) received allo-HSCT (36 with haploidentical-related donors, 12 sibling donors, and 10 matched unrelated donors). The median time from diagnose to allo-HSCT was 7.2 months (range, 3.3–18.2 months), with a median of 4 treatment cycles. Thirteen patients (13/58) died after allo-HSCT. The causes of death included relapse (n = 5), GVHD (n = 5), infection (n = 2), and interstitial pneumonitis (n = 3). From the remaining 90 patients receiving imatinib plus chemotherapy as post-remission therapy, three patients died of infection complications during CR1 that followed the consolidation courses. Thirty-two patients died from ALL relapse and disease progress.
Furthermore, we compared the characteristics and outcome between patients who were given allo-HSCT (HSCT group, n = 58) and those without allo-HSCT (non-HSCT group, n = 90) (Table 2). The patients in the allo-HSCT group were significantly younger. The median OS of patients who underwent allo-HSCT was not determined, whereas the median OS was 14 months for the non-HSCT group. In addition, the difference in the OS (73.2%, 95% CI: 58.3%–83.5% vs. 22.2%, 95% CI: 8.7%–39.6%,P<0.001) and EFS (66.5%, 50.7%–78.2% vs. 16.1%, 95% CI: 5.1%–32.7%, P<0.001) between the two groups were significant (Fig. 2). Next, according to different age group stratification, much detailed observation of the OS and EFS (Fig. 3) revealed the favorable effect of allo-HSCT.
Identification of prognostic factors and subgroup analyses
Cox-regression analyses were performed to identify factors that influenced the EFS and OS. These patient characteristics included the white blood cell (WBC) counts, hemoglobin level, platelet counts at diagnosis, age, sex, additional cytogenetic abnormality (yes vs. no), the type of BCR-ABL1 transcript (major vs. minor), the MRD status after 2 cycles of consolidation (MMR vs. no MMR), and the performance of allo-HSCT (yes vs. no). Results of the univariate and multivariate with respect to EFS and OS are shown in Table 3. Both the administration of allo-HSCT and the achievement of MMR had favorable effects on EFS and OS, as revealed by the univariate analysis and confirmed by the multivariate analysis.
We further focused on the subgroup of patients with allo-HSCT. Only a higher WBC count portended a worse EFS in univariate analysis. However, achieving MMR and donor type had no significant impact on EFS or OS (Table 4). In patients without allo-HSCT subgroup, MMR was confirmed as the only independent risk factors for poor EFS and OS in the multivariate analysis (Table 5).
Relapses and ABL1 mutations
Kinase mutation analysis was available at relapse for 21 patients; T315I mutation was detected in 8 (38.1%), E255K in 4, and Y253H in 2 patients. G250E, F317L, and F359V appeared in one patient each, and one patient harbored with compound mutation Y253H plus G250E. Mutations were not detectable in three patients. The result of the follow-up mutation analysis showed that two patients with E255K developed T315I after receiving the dasatinib based salvage therapy.
Discussion
This report describes the results of 153 patients with de novo Ph-positive ALL treated with a combination of imatinib plus CALLG2008 protocol. With a median follow-up of 24.2 months, OS and EFS at 3 years were 49.5% and 49.2%, respectively. Interestingly, the 3-year OS and DFS rates in the allo-HSCT group were 73.2% and 66.5%, respectively. The results were comparable to most published results of Ph-positive ALL (Table 6). To the best of our knowledge, this is the first report of a large cohort of patients treated homogenously by the imatinib plus CALLG2008 regimen.
Although imatinib monotherapy may lead to high CR rates of 90%–100%, combining with multiagent chemotherapy was necessary to prevent early relapse in adults with Ph-positive ALL. Usually, imatinib has been combined with intensive regimens, such as hyper-CVAD regimen. However, the most effective doses and schedules of chemotherapeutic agents are yet to be determined. The regimen of the CALLG2008 protocol consisted of 8 intensive courses with multiagent approach and maintenance therapy. The schedules of asparaginase were deleted to allow for better tolerance. In our own experience, the CALLG2008 regimen was much tolerable compared to hyper-CVAD for Chinese patients. Interestingly, low intensity regime (involving steroids, vincristine) plus TKIs are associated with impressive outcomes, particularly in elder or fragile patients. Chalandon
et al. reported that the low intensity regime reduces early mortality without impairing efficacy in a prospective randomized study (vincristine and steroids vs. hyper-CVAD) [
,
]. However, the long-term outcome of low intensity regime is still unclear. Thus, we suggested that mid-intensive regimens, such as CALLG2008 may be a reasonable choice for eligible patients, especially for those without a donor or patients unsuitable for allo-HSCT.
When TKIs serve as front-line therapy in Ph-positive ALL more patients become eligible for transplantation [
]. Allo-HSCT generally offer the best therapeutic option to cure Ph-positive ALL [
,
]. Haploidentical transplantation is an encouraging alternative option to treat patients with high-risk acute leukemia without HLA–matched donor, particularly in China [
–
]. However, limited data are available on the efficacy of haploidentical transplantation in Ph-positive ALL patients [
]. Huang
et al. reported that the 5-year DFS and OS rates were 65.8% and 74.0%, respectively. Our results further indicated that haploidentical transplantation represents an exciting treatment option for Ph-positive ALL patients who lack suitable HLA-matched donors. In addition, no differences in outcomes between the haploidentical and matched donors were observed by univariate and multivariate analyses. Thus, haploidentical donors may offer an alternative expedient donor pool. Allo-HSCT may be allowed to patients, in a higher proportion.
Accurate assessment of prognosis is critical to the management of Ph-positive ALL in the TKIs era. In the current analysis, two favorable prognostic factors (MMR and allo-HSCT) were identified through univariate and multivariate analyses. Consistent with previous reports, our results suggested that early MMR response correlated well with the overall outcomes. The poor-response group may be the most highly resistant group of patients. MMR loses significance in allo-HSCT subgroup analysis. Therefore, in future MRD-based strategies for Ph-positive ALL, active administration of allo-HSCT should be considered priority for patients that show suboptimal early MRD responses.
Recent studies suggested that in patients achieving deep MMR, outcome was similar after autologous and allogeneic transplantation [
]. Jabbour
et al. reported that sustained survival was achieved in adults with Ph-positive ALL treated by ponatinib combination with hyper-CVAD even without allo-HSCT consolidation [
]. Thus, allo-HSCT may not be indicated in all patients with Ph-positive ALL since the introduction of novel TKIs and the MRD monitoring.
In the TKIs era, development of resistance and relapse remains the major obstacles to Ph-positive ALL treatment. Consistent with results of previous studies, a spectrum of mutations was detected in relapsed patients. A T315I mutation was detected in 8 out of 21 relapsed cases. Novel TKIs such as ponatinib and monoclonal antibodies are options to prevent or overcome drug resistance [
].
Our data has significant strengths that include a large number of patients that received a uniform imatinib combined CALLG2008 regimen, which permitted a much accurate estimation of the end points, and added statistical power to the analyses. However, the present study also suffered from the inherent limitations of retrospective analysis. Detailed, graded data of adverse events were not collected. Patients that did not receive allo-HSCT preferred to not be randomly assigned because of contraindication, early relapse, or lack of donors. In addition, the complications of the transplant (engraftment, immune reconstitution, graft versus host disease (GVHD)) may still offset the benefit of the procedure in many patients.
In summary, we showed that the combination of imatinib with CALLG2008 regimen results in durable responses in newly diagnosed Ph-positive ALL. A haploidentical donor may become a choice in postremission therapy. More studies, particularly randomized trials, are warranted to test the efficacy of similarly intensive regimens followed by a haploidentical donor of Ph-positive ALL. In the future, with the emerging sequencing techniques of the next generation, the precision medicine strategies by integrating a novel genomic profiling into contemporary therapy, may further improve the outcomes.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(229KB)
