Primary study objective is the evaluation of efficacy of autologous hematopoietic cell transplantation (HCT) with core-binding factor (CBF) positive acute myeloid leukemia (AML) in the first CR (CR1) in terms of relapse incidence (cumulative incidence of relapse, CIR) and disease-free survival (DFS).

Secondary study objectives are the engraftment rate / time to engraftment, transplantation-related mortality (TRM) rate, event-free survival (EFS) rate, and Overall survival (OS).

1. BACKGROUND AND RATIONALE 1.1. CLINICAL CHARACTERISTICS OF CBF AML Acute myeloid leukemia (AML) is a disease entity consisting of heterogeneous groups with different clinical features and prognosis. Cytogenetic status of patients with AML is the single most important factor to expect the survival and the treatment responses.

Core binding factor (CBF) AML is characterized by the presence of cytogenetic abnormalities, i.e., the balanced translocation between chromosome 8 and 21 [t(8;21)(q22;q22)] and the pericentric inversion of chromosome 16 [inv(16)(p13q22)] or its less frequent variant, the balanced translocation t(16;16)(p13;q22). Among adults with de novo AML, t(8;21) and inv(16) are found in 7% and 8% of patients, respectively1. All the subtypes of CBF AML share the same chimeric fusion genes that are formed by the disruption of genes encoding different subunits of the core binding factor, a heterodimeric transcription factor complex.

CBF AML has been accepted as a disease entity of favorable prognosis with a high complete remission rate (up to 90%) with conventional induction chemotherapy followed by an intensive consolidation treatment of 3 or 4 cycles of high-dose cytarabine(HDAC). The overall survival of patients in this group rise up to 60 - 70 %, and this encouraging result has supported the opinion that HDAC was a more preferable postremission therapy instead of autologous hematopoietic cell transplantation (HCT) or allogeneic HCT.

1.2. STRATEGY TO REDUCE RELAPSE IN CBF AML It has been thought that patients with CBF AML in first complete remission (CR1) would benefit most from high-dose consolidation chemotherapy and the risk of HCT outweigh the benefit in this group. However, the cumulative incidence of relapse (CIR) has been reported to be up to 54% and 50-60% of patients are cured using contemporary treatment. The survival outcome is unsatisfactory, especially in elderly patients. Prebet et al reported that the 5-year probabilities of overall survival (OS) and leukemia-free survival (LFS) were 31% and 27%, respectively with intensive consolidation or low-dose maintenance chemotherapy among patients with CBF AML who were age 60 years or older. To improve the treatment outcome in this group, alternative strategies of postremission therapy with more efficiency and more tolerability are warranted, especially for patients who are prone to relapse. A number of studies about the stratified intensification of postremission therapy according to the risk of relapse and the appropriate prognostic index for identifying high risk patients have been reported, and some of them are currently under investigation.

1.3. DIFFERENCES BETWEEN AML WITH INV(16) AND WITH T(8;21) Recent studies have reported that these two groups seem to be distinct clinical entities and should be stratified and reported separately. Patients with t(8;21) had shorter OS (hazard ratio [HR] =1.5, p=0.045) and survival after first relapse (HR=1.7, p=0.009) than patients with inv(16). A similar difference was found among patients who had undergone HCT; the 3-year OS of patients with t(8;21) and inv(16) was 50% and 72%, respectively(p=0.002). Based on these results, a discriminative postremission strategy could be applied to patients with CBF AML - patients with t(8;21) should be managed differently from those with inv(16) as to the application of HCT and a prospective trial can be warranted to clarify the significance of HCT over chemotherapy.

1.4. RISK STRATIFICATION IN AML WITH INV(16) Although AML with inv(16) has a relatively good prognosis, a substantial number of these patients (i.e. 40-50%) finally relapse. In this group, timely identification and therapeutic stratification of patients who deemed at high risk for relapse could ultimately result in an improvement of clinical outcomes. The minimal residual disease (MRD) monitoring with real-time quantitative polymerase chain reaction (RQ-PCR) assays for CBFβ/MYH11 fusion transcript has been regarded as a useful surrogate marker for identifying a patient with resistant disease and for predicting relapse early during remission. Lane et al reported that a rise of the same or more than 1 log rise of transcript level relative to the level from a remission bone marrow sample at any time of post-induction follow up correlated with inferior LFS and morphologic relapse (HR 8.6). Bounamici et al suggested that patients whose transcript ratios of bone marrow samples taken during remission were greater than 0.25% finally relapse, and ratio below 0.12% might indicate that patients is in a curable state.

Two conclusions can be deduced from the results above; first, a considerable portion of patients among those with CBF AML finally relapse. Second, post-induction MRD monitoring might be helpful in discriminating patients who are vulnerable to relapse and may have benefit with more intensified consolidation therapy.

1.5. RISK STRATIFICATION IN AML WITH T(8;21) AML with t(8;21) has been accepted as a disease of good prognosis and categorized to favorable cytogenetic risk group along with AML with inv(16). According to recent studies, however, outcomes of AML patients with t(8;21) were disappointing in contrary to those with previous ones. The biologic and prognostic heterogeneity have been recently described for this subgroup (including other subtypes of CBF AML) and a number of promising biologic markers have been suggested.

MRD monitoring with RQ-PCR or flow cytometry is also thought as a useful method for the stratification of patients with t(8;21), as well as for those with inv(16). C-kit mutation has been generally accepted as a discriminating marker of CBF AML which increase the relapse risk. A difference in race has been considered as an important predictor for t(8;21) AML, in that nonwhites failed induction more often and had shorter OS than white. Other biomarkers which has been being considered are leukocyte count or white blood cell, CD56 positivity, loss of sex chromosome and secondary cytogenetic abnormalities6, submicroscopic deletion during chromosome rearrangement, loss of MRP gene during translocation / inversion and presence of RAS/FLT3 mutation. Gene-expression profile is suggested as a relevant way of molecular characterization to discriminate substantial biologic and clinical heterogeneity within CBF AML.

1.6. COMPARISON OF POSTREMISSION THERAPY FOR CBF AML The optimal postremission therapy of CBF AML remains to be determined. Despite being considered as patients in more favorable risk group in AML, only approximately half of the patients are cured with current strategy, significant portion of patients finally relapse and the overall survival is unsatisfactory. Heterogeneity of the treatment outcome in this group also suggests that a tailored approach might be preferred to a unique predefined strategy to treat. Current state of CBF AML indicates the need for improved therapeutic approaches, including more intensive consolidation to obtain improved LFS.

There were a few prospective studies that support the role of HCT. A prospective trial on the impact of cytogenetics and the kind of consolidation therapy performed by Visani et al showed that patients in the favorable group had significantly longer disease-free survival(DFS) when treated with an intensive induction and allogeneic HCT as an intensive consolidation therapy. According to the result of MRC AML 10 randomized controlled trial comparing the addition of autologous HCT with intensive chemotherapy alone for AML in CR1, addition of autologous HCT to four course of intensive chemotherapy reduced the risk of relapse, increased disease-free survival significantly, and improved the overall survival, although there were more death in remission in the autologous HCT group. However, a number of studies support the classic concept that CBF AML in CR1 would benefit most from HDAC and the risk of HCT outweigh the benefit. Delaunay et al reported that outcome of patients with inv(16) in CR1 was similar among patients allocated to receive allogeneic HCT vs HDAC8. According to the meta-analysis performed on 392 adults with CBF AML in prospective German AML treatment trial, type of postremission therapy revealed no difference between intensive chemotherapy and autologous HCT in the t(8;21) group and between chemotherapy, autologous HCT, and allogeneic HCT in the inv(16) group. Recent results suggest the possibility of improving overall survival with HCT. Subgroup analysis of EORTC-LG/GIMEMA AML-10 trial in which patients younger than 46 years were assigned to allogeneic HCT or autologous HCT according to the availability of HLA-identical sibling donor, DFS rate were similar in patients with good risk cytogenetics. Kuwatsuka et al reported a retrospective analysis on the results of HCT performed on CBF AML. OS was not different between patients in CR1 who received allogeneic HCT and those who received autologous HCT for both t(8;21) AML (84% vs 77%; p=0.49) and inv(16) AML (74% vs 59%, p=0.86).

1.7. OPTIMAL POSTREMISSION THERAPY FOR CBF AML In summary, HDAC chemotherapy has been recommended as a relevant postremission therapy for patients with CBF AML in CR1 on the basis that risk of HCT outweigh the clinical benefit of reducing the incidence of relapse and prolonging LFS. With the advances in hematopoietic cell transplantation technique and biomarkers for risk stratification, HCT is now considered for a promising method to improve the overall outcome of patients with CBF AML. Until now, previous results support the introduction of autologous HCT rather than allogeneic HCT for intensified postremission therapy in CBF AML in that the benefit of HCT was not yet proven and the risk of allogeneic HCT might outweigh the benefit in this group. We hereby intend to evaluate the efficacy of autologous HCT in patients with CBF AML in CR1.

Inclusion Criteria:

• Patients with CBF positive AML in CR1. CBF AML includes t(8;21)(q22;q22) [AML1(RUNX1)/ETO(CBFα2T1)], inv(16)(q13q22) (CBFβ/MYH11), t(16;16)(p13;q22) (CBFβ/MYH11) Using RT-PCR, FISH, or standard karyotype analysis technique.
• Patients who plan to receive the second cycle of HDAC consolidation chemotherapy.
• 15 years old or older and 65 years or younger
• Adequate performance status (Karnofsky score of 70 or more).
• Adequate hepatic and renal function (AST, ALT, and bilirubin < 3.0 x upper normal limit, and creatinine < 2.0 mg/dL).
• Adequate cardiac function (left ventricular ejection fraction over 40% on heart scan or echocardiography)
• Signed and dated informed consent must be obtained from patient.

Exclusion Criteria:

• Presence of significant active infection
• Presence of uncontrolled bleeding
• Any coexisting major illness or organ failure
• Patients with psychiatric disorder or mental deficiency severe as to make compliance with the treatment unlike, and making informed consent impossible
• Nursing women, pregnant women, women of childbearing potential who do not want adequate contraception
• Patients with a diagnosis of prior malignancy unless disease-free for at least 5 years following therapy with curative intent (except curatively treated nonmelanoma skin cancer, in situ carcinoma, or cervical intraepithelial neoplasia)