Chronic Leukemia

Allogeneic hematopoietic SCT in combination with tyrosine kinase inhibitor treatment compared with TKI treatment alone in CML blast crisis

Article metrics


CML treatment with tyrosine kinase inhibitors (TKIs) has improved many patients’ prognosis, but during the disease’s terminal phase, the blast crisis (CML-BC), has been disappointing. Allo-HSCT is another treatment, but survival rates are still disappointing. Currently, a combination of these two is suggested but with little evidence. This retrospective comparison reports on this combination and TKI alone for treatment of CML-BC. Of the 83 CML-BC patients, 45 received TKIs (imatinib; nilotinb or dasatinib after imatinib resistance; TKIs group) and 38 were treated with allo-HSCT after TKI (TKIs+allo-HSCT group). Treatment success was measured in terms of the hematologic, cytogenic and molecular responses, and subject outcome. Follow-up was 30–126 months or until death. Univariate and multivariate analyses determined EFS and OS predictors. Allo-HSCT significantly improved the 4-year OS (46.7 vs 9.7%, P<0.001) and EFS (47.1 vs 6.7%, P<0.001) compared to TKI treatment alone. Hemoglobin <100 g/L, non-return to chronic phase after TKI therapy and TKI treatment alone are independent adverse predictors of OS and EFS. Allo-HSCT with individualized intervention after TKI therapy is superior to TKI alone for CML-BC.


Blast crisis is the terminal phase of CML. Patients with CML in blast crisis (CML-BC) have a poor prognosis.1 For these patients, the median survival time is only 3–6 months if untreated, and the effects of conventional chemotherapies are poor.2, 3, 4

Tyrosine kinase inhibitors (TKIs), such as imatinib have revolutionized the treatment of patients with CML. Median survival in chronic phase (CP) is estimated at 25–30 years. But in CML-BC, despite the relative benefits of imatinib compared with other drugs, long-term survival remains low (median, 6–10 months).5 Furthermore, similar responses were reported in patients with imatinib-resistant CML-BC after the treatment with second generation TKIs (nilotinib and dasatinib), the 24-month OS is 21–27%.6, 7, 8 Thus, TKI treatment alone has not improved long-term survival in patients with CML-BC.

Allogeneic hematopoietic SCT (allo-HSCT) is currently the only strategy with the potential for curing patients in any phase of CML. However, the outcome of allo-HSCT for patients in BC is very disappointing, the 2-year survival rate with allo-HSCT treatment is no more than the 20% achieved in the pre-TKI era.2,9 Poor survival is associated with a high risk of relapse, which is the major obstacle limiting the success of allo-HSCT in CML-BC.

An increased OS rate has been achieved in the advanced stage of the disease by using the combination of imatinib and allo-HSCT.10, 11, 12 According to the European Leukemia Net guidelines, patients in advanced phases of CML are usually treated with imatinib or other TKIs before allo-HSCT.13,14 In spite of this, there is limited published data available on the efficacy of pretransplant and post-transplant TKIs. Hitherto, there are no reports available providing evidence in patients with CML-BC that the outcome from using a combination of TKI and allo-HSCT is superior to TKI treatment alone. Therefore, we undertook a retrospective analysis of this combination to provide more information on the success of the treatments.

Materials and methods


This was a retrospective, single-center study conducted at the Institute of Hematology, Peking University, Beijing, China. Between April 2001 and December 2010, a total of 98 diagnosed CML-BC patients who received TKIs treatment were registered in our center. The treatment regimens dependent on whether the patient were treated with TKI alone or with allo-HSCT and TKI were decided according to each patient’s preferences. Patients were allowed to select treatment with TKI alone or with allo-HSCT and TKI after discussing the treatment options with their doctors. Discussions included fear of TRM, a desire for a radical cure of the disease, concerns regarding donors, and cost considerations.

In this study, the World Health Organization (WHO) classification15 was used for definitions. BC is defined as at least 20% blasts in the BM or PB. CP 2 is defined as a return to CP (RCP) after treatment for BC disease.

All patients had the following conditions: between 18 and 50 years old, WHO Performance Status score of 0–2, normal serum electrolytes and adequate hepatic, renal and pancreatic function. Exclusion criteria: concurrent severe and/or uncontrolled medical conditions, abnormal cardiac function, clinically significant heart disease and central nervous system leukemia infiltration.

Therapeutic clinical management scheme


TKI before allo-HSCT: Before allo-HSCT, imatinib was administered to the patients at a dose of 600 mg daily for <5 months. For imatinib resistance or imatinib intolerance, a short-term nilotinib (400 mg bid) or dasatinib (70 mg bid) therapy was used instead. The protocol and condition of transplantation were as previously reported.16, 17, 18, 19, 20, 21, 22, 23

HLA typing procedures: A HLA-matched sibling donor was the first choice for allo-HSCT. If a matched sibling donor was unavailable as a first treatment option, patients without a suitable closely HLA-matched unrelated donor, that is, with more than eight of 10 matching HLA-A, B, C, DR and DQ loci and at least five of six matching HLA-A, B and DR loci, or whose disease status left insufficient time for an unrelated donor search, were eligible for haploidentical HSCT. Family members were assessed for degree of mismatch. HLA-A and HLA-B typing was performed by intermediate resolution DNA typing, whereas HLA-DRB1 typing was performed using high-resolution DNA techniques. Each patient received stem cells from a family member who shared one HLA haplotype with the patient but differed to a variable degree for the HLA-A, B and D antigens of the haplotype not shared.

Conditioning regimen: Conditioning was performed as follows. HLA)-matched sibling transplant patients received a regimen consisting of 80 mg/kg hydroxyurea orally on day −10, 2 g/m2/day cytarabine intravenously on day −9, 4 mg/kg/day BU orally if treated before 2008 and 3.2 mg/kg/day BU intravenously if treated after 2008 on days −8 to −6, 1.8 g/m2/day CY intravenously on days −5 to −4, and 250 mg/m2 of methyl- N-(2-chloroethyl)-N-cyclohexyl-N-nitrosourea orally on day −3. In cases of HLA-mismatched/haploidentical sibling or unrelated donor transplants, patients received a regimen similar to that for HLA-matched patients, except for the addition of 4 g/m2/day cytarabine on days −10 to −9 and 2.5 mg/kg/day antithymocyte globulin (SangStat, Lyon, France) intravenously on days −5 to −2.

Donors: Donors were treated with G-CSF (recombinant human G-CSF) before their BM stem cells or PBSCs were collected. The fresh and unmanipulated BM and PBSCs were infused into the recipient on the day of collection.

GVHD prophylaxis and treatment: Combinations of CYA, mycophenolate mofetil and short-term MTX were given for acute GVHD prophylaxis. MTX was administered intravenously at 15 mg/m2 on day +1 and then at 10 mg/m2 on days +3 and +6 in HLA-matched sibling transplants. An additional 10 mg/m2 was administered on day +11 in HLA-mismatched/haploidentical familial or unrelated donor transplants. CYA (1.25 mg/kg twice a day) was started intravenously on day −9 and was continued until patients could tolerate oral medication. Thereafter, CYA (3.25 mg/kg twice a day) was given orally with trough levels targeted to 150–250 ng/mL, then tapered based on the presence or absence of severe GVHD. It was a standard practice to taper CYA gradually beginning at 3 months, completing withdrawal by 6 months after HLA-matched sibling or unrelated donor transplants and by 9 months after HLA-mismatched/haploidentical familial transplants. Mycophenolate mofetil (1 g daily) orally was begun on day −9 and was discontinued after engraftment in HLA-matched sibling transplants. In unrelated donor transplants, mycophenolate mofetil was tapered from 1–0.5 g daily after engraftment and was discontinued on day +30. In HLA-mismatched/haploidentical familial donor transplants, mycophenolate mofetil was tapered from day +30 and was discontinued until day +60. Steroids and/or second-line immunosuppressants were used for GVHD management. Prophylactic drugs were administered to prevent infection by bacteria, fungi and viruses.16, 17, 18, 19, 20, 21, 22

Evaluation of engraftment: Engraftment was defined as the maintenance of an absolute neutrophil count above 0.5 × 109/L for three consecutive days after the neutrophil nadir. Platelet engraftment was defined as the maintenance of an absolute platelet count above 20 × 109/L for seven consecutive days without platelet transfusion after the platelet nadir. Follow-up assessments were performed by cytogenetic analysis of BM aspiration in the first, second and third month after transplantation.

Minimal residual disease monitoring: Serial measurements of BCR-ABL transcripts in BM were performed at 1, 2, 3, 6, 9, 12, 18 and 24 months after transplantation, and once a year thereafter.

TKI therapy and DLI: Modified DLI (DLI) and/or TKI therapy were given to patients who had a disease relapse after HSCT for therapeutic use or without RTC pre-HSCT for prophylactic use, following immunosuppressant withdrawal, as previously documented.19,20 Modified DLI was administered only in case of no response to discontinuation of immunosuppressants and without severe GVHD. The protocol of modified DLI included two elements; G-CSF-primed peripheral blood progenitor cells instead of unprimed donor lymphocyte harvests were used; and short-term immunosuppressive agents were used for the prevention of GVHD after DLI, which was administered according to previous reports.22,23


Imatinib: Patients treated with imatinib were enrolled in the Novartis Expanded Access Study (protocol 115) before 2003 and in the Gleevec International Patient Assistance Program (GIPAP) in China after 2003. Imatinib was administered at an initial dose of 600 mg (protocol 115) and 600 mg (GIPAP) daily. The dosage was then adjusted according to the response and/or toxicity. Hematologic response was assessed weekly for the first 3 months and once a month thereafter. Cytogenetic and molecular responses were evaluated every 3 months for the first 6 months and every 6–12 months thereafter.

Nilotinib: Patients who received nilotinib were enrolled in the Novartis expanded nilotinib access phase IIIb study (ENACT) between February and December 2007. All of them were given 400 mg of niliotinib orally twice a day. Dosage over 400 mg twice daily was not permitted, but dose reduction or discontinuation was allowed for drug toxicity. Cautious readministration of 400 mg twice daily was permitted when adverse events abated to grade 1 severity, and there was a lack of response or persistent disease.

Dasatinib: CML patients in the advanced stage of the disease were enrolled in the Bristol–Myers Squibb phase II study to evaluate efficacy of dasatinib (clinical protocal CA180160) between February and November 2008. These patients were imatinib resistant or imatinib intolerant and were administered orally with dasatinib at a dose of 70 mg twice daily. For progressive disease despite an initial dose of dasatinib, an increased dosage of 100 mg twice a day was permitted. Dose reduction or interruption was allowed in the case of toxicity.

Cytogenetic and molecular analysis

Cytogenetic analysis was performed by the G-banding technique. BM specimens were examined on direct short-term (24 h) cultures, and at least 20 metaphases were analyzed. BCR-ABL transcripts were detected by analyzing BM with nested reverse transcriptase PCR before 2005 and with quantitative real-time PCR after 2005. PCR reactions and fluorescence measurements were performed with an ABI PRISM 7500 real-time PCR system (PE Applied Biosystems, Foster City, CA, USA). We selected ABL as a control gene to compensate for variations in quality and quantity of RNA. ABL primers and probes were used according to the report of the Europe Against Cancer Program.24 The normalization ratios of BCR-ABLIS transcript in the quantitative PCR analysis were obtained through the comparison with ABL transcript, as reported previously.25 The number of the control gene transcripts (normal ABL) in our laboratory required at least 32 000 transcripts. Our laboratory acquired CF (conversion factor) through a sample exchange with an international reference laboratory in Adelaide, Australia. After determination and validation, we successfully acquired our international scale CF of 0.65. Therefore, 0.15% of BCR-ABL mRNA levels in our center is equal to 0.1% (MMR) of BCR-ABLIS.26

Response criteria

The hematologic response criteria were defined as follows.

  1. 1

    CHR was defined as a myeloblast count 5% in BM, no myeloblasts in PB, neutrophil and platelet counts of at least 1.5 × 109/L and 100 × 109/L, respectively, and no evidence of extramedullary involvement.

  2. 2

    Marrow response was similar to CHR, but neutrophil and platelet counts were at least 1.0 × 109/L and 20 × 109/L, respectively.

  3. 3

    RTC was defined as 10% myeloblasts in PB and BM, <20% peripheral basophils and no extramedullary involvement other than in liver or spleen. Sustained responses were required to last at least 4 weeks. Non-RTC was defined as a patient who did not achieve RTC after treatment.

The cytogenetic response was defined as complete (0% Ph-positive cells, complete cytogenic response (CCyR)), partial (1%–35% Ph+ cells), minor (36–65% Ph+ cells, mCyR), minimal (66%–95% Ph+ cells) or none (95% Ph+ cells). MCR was defined as either a complete or partial response. MMR was defined as a 3-log reduction in BCR-ABLIS transcript levels (0.1% BCR-ABLIS ) compared with the standard baseline levels of BCR-ABLIS transcript, which was 31% (range 10–87%), the median level of BM samples calculated from 42 newly diagnosed CML-CP patients at our hospital before imatinib therapy. These were as previously reported.27 The terms MR4, MR4.5 and MR5 were as defined as 0.01% BCR-ABLIS (4-log reduction from baseline), 0.0032% BCR-ABLIS (4.5-log reduction from baseline) and 0.001% BCR-ABLIS (5-log reduction from baseline), respectively.28

Outcome definitions

Follow-up information was collected until August 2012 or until death if this occurred before that date. OS was defined as the time from the start of treatment to all-cause death. EFS was defined as the time between commencement of treatment (TKIs or allo-HSCT) and the appearance of any of the following events: absence of hematologic response; loss of primary CHR, MCyR or CCyR; post-transplant molecular relapse; relapse in AP or BC; or death from any cause. Molecular relapse was defined as positive BCR-ABLIS transcripts confirmed by two consecutive assays after previous complete molecular response or a persistent BCR-ABL transcript increase of more than 1-log. In the TKIs+allo-HSCT group, relapse was defined as relapse in any forms (hematologic, cytogenetic or molecular).

Statistical analysis

The log-rank test was used to identify prognostic factors. Comparison between the groups was analyzed using the Mann–Whitney U-test, χ2-test(for continuous variables) and Fisher exact test (for categorical variables) test. Kaplan–Meier method was adopted in the time-to-event analysis. Univariate analyses and COX multivariate analyses were used to determine predictors of EFS and OS. SPSS Version 13.0 software (SPSS, Inc., Chicago, IL, USA) was used in all our analyses. The R statistical software was used to analyze the cumulative incidence of TRM, relapse and OS (Lucent technologies P-values less than 0.05 indicated statistical significance (P<0.05).


Patient characteristics

We retrospectively included 83 patients (18–50 years) with CML-BC during the same study period; this was from the 98 (12.3%) patients registered in the center with CML-BC, including 38 in the TKIs+allo-HSCT group (patients who received allo-HSCT and TKI in combination) and 45 in the TKIs group (patients who received TKI alone). Fifteen patients with BC were excluded, including 11 patients who were older than 50 years (two in the TKIs+allo-HSCT group and nine in the TKIs group) and four patients with a score greater than 2 on the WHO Performance Status score. Clinical characteristics before treatment are listed in Table 1. The management of patients is illustrated by Figure 1.

Table 1 Patient characteristics
Figure 1

Flow diagram showing the treatment process of the blast crisis (BC) patients. Patients were either treated with stem cell therapy (SCT) and TKI or TKI alone. Treatment was altered depending on patient responses.


Response to pretransplant TKI

In the TKIs+allo-HSCT group, all 38 patients in blast crisis received TKIs before allo-HSCT, including 11 (29%) who had developed BC despite prior imatinib therapy, five of the 11 patients were detected with BCR-ABL kinase domain mutations, which were M244V(n=1), M351I(n=1), Y253H(n=1) and T351I(n=2). Twenty-one of the 27 (77.8%) patients who received imatinib for blast crisis returned to CP. Among the 11 patients who failed to respond to imatinib therapy, seven returned to CP (RTC) after short-term (<2 months) dasatinib treatment, one patient had no response to nilotinib, two of the three patients treated with chemotherapy returned to CP and one patient had no response before transplantation. Therefore, in total there were 30 (78%) patients who returned to CP and eight (22%) patients who remained in BC before transplantation (Figure 1). Among the 38 patients, three (7.9% ) patients achieved mCyR after short-term (<5 months) TKI treatment before allo-HSCT, and no one achieved any molecular response.

HLA type

Twelve patients (31.6%) underwent allo-HSCT from an HLA-matched sibling donor; 25 (65.8%) received HLA mismatched/haploidentical familial donor allografts, including both 2-HLA mismatches (n=10) and 3-HLA mismatches (n=15). One patient (2.6%) underwent allograft from an unrelated donor with 10 HLA antigen matches (Figure 1).


The number of hematopoietic mononucleated cells was 6.94 (range 2.02–10.98) × 108/kg and CD34+ cells were approximately 2.20 (range 0.32–7.90) × 106/kg, respectively. All but one patient (96.4%) engrafted successfully. The median time for myeloid engraftment was 15 days (range 9–25 days) and for platelet 13 days (range 10–80 days).


Twenty-four (64.9%) of the 37 engrafted patients developed acute GVHD (15 grades I or II; 9 grades III or IV) and 13 (40.6%) of the 32 patients that survived over 3 months developed chronic GVHD (seven limited GVHD; six extensive GVHD). Acute GVHD did not have an impact on TRM, relapse or survival (Table 2).

Table 2 Univariate analysis of factors associated with the TRM, relapse and OS in allo-HSCT group


TRM occurred in 12 patients. The 4-year estimated TRM rate was 31.6±7.7%. Early TRM (<100 days) occurred in six patients. Among these six patients, two patients died from pulmonary infection on day +61 and +76 while in continuous complete remission. A patient died of pulmonary infection on day +30 without engraftment. A patient died from diffuse alveolar hemorrhage on day +60. The other two patients died from sepsis and septic shock on day +30 and +35.

Late TRM (>100 days) occurred in six patients within the duration of follow-up. The median time for late TRM was 240 days (range: 4–728) after transplantation. The causes of transplant-related death included GVHD (one case), pulmonary infection (two cases), central nervous system infection (two cases) and septic shock (one case).

Efficacy of transplantation and post transplant TKI

Cytogenetic and molecular response analyses were performed for the 32 patients who survived over 3 months. Among these 32 patients, 14 patients received DLI; 27 (84.4%) had achieved a MR4.5. Among the 37 engrafted patients, nine patients (24.3%) experienced molecular or hematologic relapse in a median period of 12 months (1.5–51 months) after transplantation and two other patients (5.4%) presented extramedullary disease at 1.5 and 22 months after transplantation.

Response of DLI and post transplant TKI

Fourteen of the 38 patients (36.8%) received DLIs. Among them, six of the eight patients without RTC pre-HSCT (except two patients who underwent early TRM) received prophylactic DLIs. Three of the six patients who underwent a prophylactic DLI and eight6 of the 26 patients (except four patients who underwent early TRM) without a prophylactic DLI had relapsed post HSCT. After relapse, the eight patients who relapsed without a prophylactic DLI received therapeutic DLI. Three of these eight patients who underwent a therapeutic DLI died from failure to respond.

Among the 37 engrafted patients, 16 of the 37 patients (43%) were treated with TKI post HSCT, including five patients following DLI (two for prophylactic therapy and three for therapeutic use) and the other six patients received TKIs alone after allo-HSCT for prophylactic use.

Despite DLI and TKI therapy, eight of the 11 patients who relapsed after HSCT failed to respond and died (including two T315I mutations), two patients achieved a 43-month and 46-month MR4.5, respectively, and one patient who underwent a second allo-HSCT achieved a 66-month MR4.5.

Survival and relapse

Among the 38 patients in the TKIs+allo-HSCT group, 20 patients (52.6%) died within 1–55 months (median, 6 months): 12 (31.6%) died of TRM within 1–24 months (median, 2.5 months); eight (21.1%) died of relapse. The follow-up time of the 18 patients who survived was 30–126 months (median, 52 months). The estimated 4-year OS and EFS rates were 46.7 and 47.1%, respectively.

Thirty of the total 38 patients had returned to CP (RTC) before transplantation, eight patients had no response (non-RTC). The estimated 4-year survivals were similar (OS: 50.0 vs 37.5%, P=0.32; EFS: 50.0 vs 33.3%; P=0.35) whether they achieved RTC or not before transplantation. At the last follow-up, 18 patients were still alive and in MR4.5.

In the TKIs+allo-HSCT group, univariate analyses were performed to identify predictors of TRM, relapse rate and OS (Table 2). The relapse rate was higher for patients who had non-RTC pretransplant than the RTC patients, but there was no statistical difference.

TKI therapy

Among the 45 patients in the TKIs group, 38 received imatinib alone, seven were administered with a second generation TKI (four received nilotinib and three received dasatinib) when they failed to respond to imatinib. Thirty patients (67.8%) achieved a RTC (27 with imatinib, one with nilotinib and two with dasatinib), and 15 patients showed (32.2%) no hematologic response. Five patients (four with imatinib and one with nilotinib) achieved CCyR and MMR and they were still alive: one was lost to follow-up after 5 years, the other four were event free at 21, 30, 88 and 120 months.

Comparison of survival between the TKIs group and the TKIs+allo-HSCT group

Patients in the TKIs+allo-HSCT group had significantly longer OS (P<0.001) and EFS (P<0.001) than in the TKIs group (Figure 2).

Figure 2

Effect of different therapies on survival in CML-BC patients. (a) The median OS in the TKIs+allo-HSCT group (21 months, 95% CI 1–61.7 months) was significantly longer than that in the TKIs group (6 months, 95% CI 3.9–8.1 months). Likewise, the 4-year OS rates in the TKIs+allo-HSCT group were significantly higher than those in the TKIs group (46.7 vs 9.7%, P<0.001). (b) The median EFS in TKIs+allo-HSCT group (18 months, 95%CI 1–58.8 months) was significantly longer than that in the TKIs group (4 months, 95% CI 2.4–5.6 months). Similarly, the 4-year EFS rates in TKIs+allo-HSCT group were significantly higher than those in TKIs group (47.1 vs 10.0%, P<0.001).

Among the 59 patients with MBC-CML, allo-HSCT induced a significant improvement in both 4-year OS (44.9%, P=0.008) and EFS (45.0%, P=0.007) compared with the TKIs group (9.1 and 9.5%; Figure 3).

Figure 3

Effect of different therapies on survival in myeloid blast crisis patients. (a) The median OS in the TKIs+allo-HSCT group (21 months, 95% CI 4.5–37.5 months) was significantly longer than that in the TKIs group (7 months, 95% CI 3.5–10.5 months). Likewise, the 4-year OS rates in TKIs+allo-HSCT group were 44.9%, significantly higher than those in TKIs group (44.9 vs 9.1%, P=0.008). (b) The median EFS in the TKIs+allo-HSCT group (13 months, 95% CI 11–18.2 months) was significantly longer than that in the TKIs group (5 months, 95% CI 3.3–6.7 months). Similarly, the 4-year EFS rates in the TKIs+allo-HSCT group were significantly higher than those in the TKIs group (45.0 vs 9.5%, P=0.007).

Likewise, among the 24 LBC-CML patients, 4-year OS and EFS rates were 50.0% (P=0.016) and 50.0% (P=0.006) in the TKIs+allo-HSCT group, much higher than those in the TKIs group (10 and 10%; Figure 4).

Figure 4

Effect of different therapies on survival in lymphoid blast crisis patients. (a) The median OS in the TKIs+allo-HSCT group (20 months, 95% CI 1–74 months) was significantly longer than that in the TKIs group (4.5 months; 95% CI 3.5–5.5 months). Similarly, the 4-year OS rates in the TKIs+allo-HSCT group were significantly higher than those in the TKI group (50.0 vs 10.0%, P=0.016). (b) The median EFS in the TKIs+allo-HSCT group (18 months, 95% CI 1–72.7 months) was also significantly longer than that in the TKIs group (3 months; 95% CI 1.9–4.0 months). The 4-year EFS rates in the TKIs+allo-HSCT group were significantly higher than those in the TKIs group (50.0 vs 10.0%, P=0.002).

Sixty patients returned to CP after TKI therapy. Among them, the median OS and EFS in the allo-HSCT group (24 months, 95% confidence interval (CI) 1–76 months and 24 months, 95% CI 0–76 months, respectively) were significantly longer than those in the TKIs group (9 months, 95% CI 3.6–14.4 months and 6 months, 95% CI 3.3–8.7 months; P=0.005). The 4-year OS and EFS of RTC patients in the TKIs+allo-HSCT group were 50.0% (P=0.015) and 50.0% (P=0.005), much higher than those in the TKIs group (14.6 and 12.0%, respectively).

Twenty-three patients failed to RCP after short-term TKI therapy (eight in the TKIs+allo-HSCT group; 15 in the TKIs group). Among them, the median survival in the TKIs+allo-HSCT group (OS: 7.5 months, 95% CI 0–27 months; EFS: 3 months, 95% CI 0.9–5.1 months) were significantly longer than in the TKIs group (OS: 3.5 months, 95% CI 3.2–3.8 months, P=0.023; EFS: 2 months, 95% CI 1.4–2.6, P=0.039). The 4-year OS and EFS rates of non-RTC patients in the TKIs+allo-HSCT group were 33.3% (P=0.023) and 37.5% (P=0.039), much higher than those in the TKIs group (0 and 0%, respectively). At the last follow-up, three of the eight RTC patients in the TKIs+allo-HSCT group were still alive, with an EFS of 13, 38 and 84 months, respectively, whereas in the TKIs group, none of the 15 RTC patients survived over 5 months.

Identification of prognostic factors

Univariate and multivariate analyses were performed on all 83 patients to identify the predictors of EFS and OS. Univariate analysis showed that hemoglobin <100 g/L, non-RTC after TKI therapy, previous treatment with chemotherapy for BC and TKI therapy were adverse factors (Table 3). Multivariate analysis of variables, including pretreatment characteristics and therapy choice (TKIs or TKIs+allo-HSCT), revealed that hemoglobin <100 g/L, non-RTC after TKI therapy and TKI therapy were independent adverse predictors of OS and EFS.

Table 3 Univariate analysis of factors associated with OS and EFS for all patients


This retrospective study looked at whether the suggested treatment of CML-BC with a combination of TKI and allo-HSCT was supported by the evidence from two groups of patients; one group had undergone the combined treatment of TKI before allo-HSCT as recommended by the European Leukemia Net,13,14 whereas the other had been treated with TKI alone.

The results presented here clearly show that the combined treatment improved the 4-year OS (46.7 vs 9.7%, P<0.001) and EFS (47.1 vs 10.0%, P<0.001) when compared with TKI treatment alone. These improvements were expected, but nevertheless the large differences are quite striking.

Although patients were not selected into the two different treatment groups but were treated according to the best option available and their own will, there were no significant differences in clinical characteristics before treatment between the groups except for the median WBC count. Data from the TKIs group showed that the estimated 4-year OS and EFS rates (9.7 and 10.0%, respectively) were similar to previous reports.5,29 Our experience also indicated that survival (estimated 4-year OS and EFS rates of 46.7 and 47.1%, respectively) of TKIs combined with allo-HSCT were superior to those reported in other studies (OS rates were less than 20%) from before the era of imatinib use.2,9 When comparing the data from the current study with previous reports, it is worth noting that we used the WHO guidelines and determined that 20% blasts in blood or marrow should be classed as BC.15 Other studies have used the European LeukemiaNet management recommendations of 30%. We have also presented the information split according to these definitions in the tables, and in this study, found that OS and EFS showed no statistical differences between the two definitions, although patients with 20–29% blasts would be expected to have a better overall prognosis.30 The outcome of this study strongly supports the recommendation of the European Leukemia Net guidelines for CML-BC.13,14

Improved outcomes from the combination treatment may be because pretreatment with TKIs reduced the leukemia burden before HSCT, and more importantly, the individualized TKI-based intervention strategy based on TKIs and modified DLI post transplant reduced the risk of relapse.

After transplantation, immunosuppressant withdrawal and either TKI therapy or modified DLI were used to treat disease relapse or non-RTC before HSCT. Our previous data had demonstrated that modified DLI is safe and effective for prophylaxis and treatment of relapse.21,22 The present study also revealed a similar benefit of TKIs before transplantation in non-RTC patients. The 4-year OS and EFS rates of non-RTC patients in the TKIs+allo-HSCT group were 33.3 and 37.5%, respectively, higher than those reported before the introduction of imatinib.2,9 All these data indicate that allo-HSCT with individualized intervention has an important role in the treatment of CML-BC in the TKI era.

In the present study, the 4-year OS and EFS rates of RTC patients in the TKIs+allo-HSCT group were significantly higher than those in the TKIs group. Similar outcomes were observed in subgroups of non-RTC, MBC and LBC patients as well. There was no significant difference in 4-year OS (50.0 and 37.5%, P=0.32) and EFS (50.0 and 33.3%, P=0.35) rates between the RTC and non-RTC patients after TKI treatment in the TKIs+allo-HSCT group. We think that these patients may benefit from individualized intervention along with allo-HSCT and TKI.

According to our study, the patients who had achieved MMR also achieved long-term survival in the TKIs group. However, the MMR rate of BC patients in the imatinib alone group was very low. Second-generation TKIs can rapidly reduce leukemia burden and induce an improved response rate in BC.6,7 Therefore, an initial therapy with second-generation TKIs might induce early and durable CCyR and MMR, and improve long-term survival in CML-BC.31 Whether allo-HSCT is superior to second-generation TKIs for CML-BC not treated with imatinib remains an issue for future study.

This study has some necessary limitations because it was a retrospective study and the number of study subjects was quite small; a randomized, controlled and larger study would be difficult to undertake primarily because it would not be ethical to select one group not to receive transplants if they were available, but would be the ideal basis to provide more evidence for the benefit of the combination treatment. As a result of the limitations of a retrospective study, the usefulness of continuing TKI therapy even after allo-HSCT independently from detection of residual disease or to react only in the case of a positive PCR cannot be sufficiently addressed.

In summary, the present study indicates that allo-HSCT in combination with TKIs is a better option for CML-BC in the TKI era. This combination is superior to TKI treatment alone, and can improve EFS as well as OS. Therefore, early allo-HSCT after short-term TKI therapy should be the treatment of choice for CML-BC.


  1. 1

    Sawyers CL . Chronic myeloid leukemia. New Engl J Med 1999; 340: 1330–1340.

  2. 2

    Faderl S, Talpaz M, Estrov Z, Kantarjian HM . Chronic myelogenous leukemia: biology and therapy. Ann Intern Med 1999; 131: 207–219.

  3. 3

    Derderian PM, Kantarjian HM, Talpaz M, O'Brien S, Cork A, Estey E et al. Chronic myelogenous leukemia in the lymphoid blastic phase: characteristics, treatment response, and prognosis. Am J Med 1993; 94: 69–74.

  4. 4

    Sacchi S, Kantarjian HM, O'Brien S, Cortes J, Rios MB, Giles FJ et al. Chronic myelogenous leukemia in nonlymphoid blastic phase: analysis of the results of first salvage therapy with three different treatment approaches for 162 patients. Cancer 1999; 86: 2632–2641.

  5. 5

    Palandri F, Castagnetti F, Testoni N, Luatti S, Marzocchi G, Bassi S et al. Chronic myeloid leukemia in blast crisis treated with imatinib 600 mg: outcome of the patients alive after a 6-year follow-up. Haematologica 2008; 93: 1792–1796.

  6. 6

    Nicolini FE, Masszi T, Shen Z, Gallagher NJ, Jootar S, Powell BL et al. Expanding Nilotinib Access in Clinical Trials (ENACT), an open-label multicenter study of oral nilotinib in adult patients with imatinib-resistant or -intolerant chronic myeloid leukemia in accelerated phase or blast crisis. Leuk Lymphoma 2012; 53: 907–914.

  7. 7

    Saglio G, Hochhaus A, Goh YT, Masszi T, Pasquini R, Maloisel F et al. Dasatinib in imatinib-resistant or imatinib-intolerant chronic myeloid leukemia in blast phase after 2 years of follow-up in a phase 3 study: efficacy and tolerability of 140 milligrams once daily and 70 milligrams twice daily. Cancer 2010; 116: 3852–3861.

  8. 8

    Giles FJ, Kantarjian HM, le Coutre PD, Baccarani M, Mahon FX, Blakesley RE et al. Nilotinib is effective in imatinib-resistant or -intolerant patients with chronic myeloid leukemia in blastic phase. Leukemia 2012; 26: 959–962.

  9. 9

    Gratwohl A, Brand R, Apperley J, Crawley C, Ruutu T, Corradini P et al. Allogeneic hematopoietic stem cell transplantation for chronic myeloid leukemia in Europe 2006: transplant activity, long-term data and current results. An analysis by the Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation (EBMT). Haematologica 2006; 91: 513–521.

  10. 10

    Luo Y, Zhao Y, Tan Y, Shi J, Han X, Zheng Y et al. Imatinib combined with myeloablative allogeneic hematopoietic stem cell transplantation for advanced phases of chronic myeloid leukemia. Leuk Res 2011; 35: 1307–1311.

  11. 11

    Saussele S, Lauseker M, Gratwohl A, Beelen DW, Bunjes D, Schwerdtfeger R et al. Allogeneic hematopoietic stem cell transplantation (allo SCT) for chronic myeloid leukemia in the imatinib era: evaluation of its impact within a subgroup of the randomized German CML Study IV. Blood 2010; 115: 1880–1885.

  12. 12

    Menzel H, von Bubnoff N, Hochhaus A, Haferlach C, Peschel C, Duyster J . Successful allogeneic stem cell transplantation in second chronic-phase CML induced by the tyrosine kinase inhibitor nilotinib (AMN107) after blast crisis under imatinib. Bone Marrow Transplant 2007; 40: 83–84.

  13. 13

    Baccarani M, Cortes J, Pane F, Niederwieser D, Saglio G, Apperley J et al. Chronic myeloid leukemia: an update of concepts and management recommendations of European LeukemiaNet. J Clin Oncol 2009; 27: 6041–6051.

  14. 14

    Baccarani M, Deininger MW, Rosti G, Hochhaus A, Soverini S, Apperley JF et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia. Blood 2013; 122: 872–884.

  15. 15

    Vardiman JW, Harris NL, Brunning RD . The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002; 100: 2292–2302.

  16. 16

    Huang XJ, Liu DH, Liu KY, Xu LP, Chen H, Han W et al. Haploidentical hematopoietic stem cell transplantation without in vitro T-cell depletion for the treatment of hematological malignancies. Bone Marrow Transplant 2006; 38: 291–297.

  17. 17

    Huang XJ, Liu DH, Liu KY, Xu LP, Chen H, Han W et al. Treatment of acute leukemia with unmanipulated HLA-mismatched/haploidentical blood and bone marrow transplantation. Biol Blood Marrow Transplant 2009; 15: 257–265.

  18. 18

    Xiao-Jun H, Lan-Ping X, Kai-Yan L, Dai-Hong L, Yu W, Huan C et al. Partially matched related donor transplantation can achieve outcomes comparable with unrelated donor transplantation for patients with hematologic malignancies. Clin Cancer Res 2009; 15: 4777–4783.

  19. 19

    Lu DP, Dong L, Wu T, Huang XJ, Zhang MJ, Han W et al. Conditioning including antithymocyte globulin followed by unmanipulated HLA-mismatched/haploidentical blood and marrow transplantation can achieve comparable outcomes with HLA-identical sibling transplantation. Blood 2006; 107: 3065–3073.

  20. 20

    Xiao-Jun H, Lan-Ping X, Kai-Yan L, Dai-Hong L, Huan C, Wei H et al. HLA-mismatched/haploidentical hematopoietic stem cell transplantation without in vitro T cell depletion for chronic myeloid leukemia: improved outcomes in patients in accelerated phase and blast crisis phase. Ann Med 2008; 40: 444–455.

  21. 21

    Huang XJ, Xu LP, Liu KY, Liu DH, Chen H, Liu YR et al. Individualized intervention guided by BCR-ABL transcript levels after HLA-identical sibling donor transplantation improves HSCT outcomes for patients with chronic myeloid leukemia. Biol Blood Marrow Transplant 2011; 17: 649–656.

  22. 22

    Huang XJ, Liu DH, Liu KY, Xu LP, Chen H, Han W . Donor lymphocyte infusion for the treatment of leukemia relapse after HLA-mismatched/haploidentical T-cell-replete hematopoietic stem cell transplantation. Haematologica 2007; 92: 414–417.

  23. 23

    Huang XJ, Wang Y, Liu DH, Xu LP, Chen H, Chen YH et al. Modified donor lymphocyte infusion (DLI) for the prophylaxis of leukemia relapse after hematopoietic stem cell transplantation in patients with advanced leukemia—feasibility and safety study. J Clin Immunol 2008; 28: 390–397.

  24. 24

    Beillard E, Pallisgaard N, van der Velden VH, Bi W, Dee R, van der Schoot E et al. Evaluation of candidate control genes for diagnosis and residual disease detection in leukemic patients using 'real-time' quantitative reverse-transcriptase polymerase chain reaction (RQ-PCR)—a Europe against cancer program. Leukemia 2003; 17: 2474–2486.

  25. 25

    Qin YZ, Liu YR, Zhu HH, Li JL, Ruan GR, Zhang Y et al. Different kinetic patterns of BCR-ABL transcript levels in imatinib-treated chronic myeloid leukemia patients after achieving complete cytogenetic response. Int J Lab Hematol 2008; 30: 317–323.

  26. 26

    Qin YZ, Jiang Q, Jiang H, Li JL, Li LD, Zhu HH et al. Which method better evaluates the molecular response in newly diagnosed chronic phase chronic myeloid leukemia patients with imatinib treatment, BCR-ABL(IS) or log reduction from the baseline level? Leuk Res 2013; 37: 1035–1040.

  27. 27

    Jiang Q, Xu LP, Liu DH, Liu KY, Chen SS, Jiang B et al. Imatinib mesylate versus allogeneic hematopoietic stem cell transplantation for patients with chronic myelogenous leukemia in the accelerated phase. Blood 2011; 117: 3032–3040.

  28. 28

    Cross NC, White HE, Muller MC, Saglio G, Hochhaus A . Standardized definitions of molecular response in chronic myeloid leukemia. Leukemia 2012; 26: 2172–2175.

  29. 29

    Sawyers CL, Hochhaus A, Feldman E, Goldman JM, Miller CB, Ottmann OG et al. Imatinib induces hematologic and cytogenetic responses in patients with chronic myelogenous leukemia in myeloid blast crisis: results of a phase II study. Blood 2002; 99: 3530–3539.

  30. 30

    Hehlmann R . How I treat CML blast crisis. Blood 2012; 120: 737–747.

  31. 31

    Jabbour E, Kantarjian HM, O'Brien S, Shan J, Quintas-Cardama A, Garcia-Manero G et al. Front-line therapy with second-generation tyrosine kinase inhibitors in patients with early chronic phase chronic myeloid leukemia: what is the optimal response? J Clin Oncol 2011; 29: 4260–4265.

Download references


We would like to thank the contributions of all the people who have participated in this research. This includes our laboratory staff, our colleagues and our nurses. The study was supported by the National Outstanding Young Scientists’ Foundation of China (grant no. 30725038) and by the Program for Innovative Research Team at the University of China (grant no. IRT0702). The study was also supported by the Collaborative Innovation Center of Hematology, China.

Author information

Correspondence to X-J Huang.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Further reading