Cytogenetics and FLT3-ITD mutation predict clinical outcomes in non transplant patients with acute myeloid leukemia

Background Cytogenetic abnormalities and mutated genes indicate the role of consolidation therapy with hematopoietic stem cell transplantation (HSCT) or chemotherapy in acute myeloid leukemia (AML). In this study, we conducted a retrospective study in adult AML patients with newly diagnosed with de novo AML who did not undergo HSCT, to study long term relapse free survival (RFS) and overall survival (OS) after consolidation chemotherapy. Methods We recruited 141 consecutive AML patients during January 2010–June 2017, the patients received induction chemotherapy with standard dose Ara-C and Idarubicin (7 + 3 or 5 + 2 regimen) followed by intermediate (IDAC) or high dose Ara-c (HiDAC) consolidation therapy. Results Normal karyotype, complex, favorable, intermediate and adverse chromosomal aberrations were found in 59%, 16%, 5%, 14% and 6%, respectively. Mutated NPM1, FLT3-ITD and CEBPA genes in CN-AML were seen in 33%, 18% and 19%, respectively. A 5 year follow up, 5y-RFS was 16% and 5y-OS was 14% in the whole study population. 5y-RFS and 5y-OS in patients completed 4 cycles of consolidation therapy were 25% and 40%, respectively. Adverse cytogenetic risk and mutated FLT3-ITD were significantly associated with poor RFS (9 and 15 months, respectively) and OS (14 and 16 months, respectively), whereas patients with mutant NPM1 had favorable outcomes (RFS/OS = 51/63 months). Patients receiving 4 cycles of consolidation therapy had significantly impacts on median RFS and OS compared with those treated with 1 or 2 cycles; 15 versus 11 months (p = 0.006) and 31 versus 15 months (p < 0.001), respectively. Conclusions Cytogenetic and mutation tests for FLT3-ITD, NPM1 and CEBPA genes were meaningful for predicting outcomes in adult AML patients. Adverse cytogenetic abnormalities and FLT3-ITD mutation showed dismal RFS and OS.


Patients
The study was conducted during 1 January 2010-30 June 2017, we enrolled 141 consecutive patients with newly diagnosed with de novo AML who did not undergo allo-HSCT. All patients were older than 15 years old and had the results of cytogenetic, FLT3-ITD, NPM1 and CEBPA gene mutation analysis. The patients receiving supportive and hypomethylating agent therapies were allowed to enroll into the study. The patients who received prior chemotherapy or previous allogeneic or autologous HSCT were excluded. The patients with acute promyelocytic leukemia, secondary AML or therapy related AML were not eligible for the study.

Induction chemotherapy
The patients younger than 60 years of age received induction chemotherapy with intravenous (i.v.) Ara-C 100 mg/m 2 /day for 7 consecutive days together with i.v. idarubicin 12 mg/m 2 /day for 3 consecutive days (7 + 3 regimen). The patient aged ≥ 60 years or < 60 years with creatinine clearance < 50 mL/min or had septicemia or pulmonary infection at diagnosis of AML were treated with 5 + 2 regimen that consisted of 5 days of i.v. Ara-C 100 mg/m 2 /day (d1-5) combined with 2 days of i.v. idarubicin 12 mg/m 2 /day (d1-2). The bone marrow (BM) study was re-evaluated 28 days after induction therapy. The patient aged ≥ 70 years received azacitidine 100 mg/day subcutaneously for 7 consecutive days, every 4 weeks, and the disease response was re-evaluated after 4-6 cycles. Complete remission (CR) was defined as BM blasts < 5%, absolute neutrophil count ≥ 1000/μL, platelet count ≥ 100,000/μL, absence of circulating blasts and absence of extramedullary disease.

Consolidation chemotherapy
Patients aged < 60 years achieving CR after induction chemotherapy were treated with the first cycle of consolidation chemotherapy with the same regimen as induction therapy followed by 3 cycles of IDAC or HiDAC therapy. IDAC and HiDAC regimen were defined as i.v. Ara-C dose at 1000-2500 mg/m 2 /dose and 2501-3000 mg/ m 2 /dose every 12 h for 3 days (d1, 3, 5), respectively. The dosage of Ara-C in patients receiving IDAC therapy was assigned to 1000 mg/m 2 /dose every 12 h on d1, 3, 5 for patients aged 60 years and older (elderly patients), whereas patients younger than 60 years who had previously experienced severe infection or unfit after the induction or the first cycle of consolidation therapy with 7 + 3 or 5 + 2 were treated with i.v. Ara-C 2000-2500 mg/ m 2 /dose every 12 h on day 1, 3, 5. Conversely, fit patients aged < 60 years with or without t(8;21) received HiDAC regimen. Azacitidine 100 mg subcutaneous route for 7 days every 4 weeks were given as consolidation therapy until the disease relapse in elderly patients who were unfit for receiving IDAC therapy.

Cytogenetic and molecular techniques
Short-term culture of BM cells, metaphase spread preparation and G-banding were performed. Karyotypes were described according to ISCN 2016 [7]. FLT3-ITD, NPM1, and CEBPA gene mutations were studied using genomic DNA isolated from BM or peripheral blood samples with QIAamp DNA Blood Mini Kit (Qiagen, Germany). FLT3 exon 14, 15 and 20 was amplified by PCR using specific primers and subsequently digested with EcoRV as previously described [8,9]. The PCR products were analyzed by 2% agarose gel electrophoresis. NPM1 gene mutations were performed by fluorescent PCR [10], PCR products were separated by 3130 Genetic Analyzer (Applied Biosystem, USA), and the results were analyzed using Gene Mapper software version 4 (Applied Biosystems, USA). CEBPA gene mutations were investigated by amplification of entire coding region and bidirectional sequencing was performed using Bigdye Terminator Version 1.1 Cycle Sequencing Kit (Qiagen, Germany). Patient's sequences were compared to CEBPA reference gene to evaluate the mutation status. The cytogenetic and molecular stratification risks were classified by 2017 ELN recommendations [2].

Outcome assessment
The objectives of this study were to evaluate RFS and OS of the entire study population and subgroup analysis that we divided the study population into four subgroups according to the factors that would affect to RFS and OS, including AML patients receiving best supportive care, AML patients completed 4 cycles of consolidation therapy, cytogenetically normal AML patients with or without completed consolidation therapy. RFS was defined as the length of time from the date of CR to relapse. OS was defined as the interval between the dates of diagnosis and death.

Statistical analysis
The parameters which included age, white blood cell (WBC) count, cytogenetics, molecular data and treatment regimens were compared between patients with and without CR by using Chi-square. OS and RFS were calculated using the Kaplan-Meier method, difference between groups were calculated using the log-rank test for univariate analysis. Cox's Regression model was used for multivariate survival analysis. All calculations were performed using the statistical package of social sciences software, SPSS statistics version 17 (Chicago: SPSS Inc; 2008), p < 0.05 was considered significant. In this study, three gene mutation tests (NPM1, FLT3-ITD and CEBPA) were performed in all AML patients, 88 patients (62%) had no mutation and 53 patients (38%) had at least one gene mutation. Single gene mutation was found in 43 patients (30%); mutation of NPM1 (18 patients; 13%), FLT3-ITD (11 patients; 8%) and CEPBA genes (14 patients; 10%). Combination of NPM1 and FLT3-ITD gene mutations was found in 7 out of 141 patients (5%), the coexisting mutation of CEBPA/FLT3-ITD/NPM genes was detected in 1 patient (0.7%) and the remaining 2 patients (1.3%) had either mutant FLT3-ITD or NPM1 gene with CEBPA gene mutations.
A 5 year follow up, median RFS and OS were 12 and 9 months, respectively, and 5y-RFS was 16% and 5y-OS was 14% in the whole study population. Median OS in treated and untreated AML patients were 13 and 2 months, respectively. The median RFS and OS were significantly longer in patients treated with 4 cycles of consolidation chemotherapy (41 patients) than those treated with 1 or 2 cycles of consolidation therapy (17 patients); RFS were 15 and 11 months, respectively (p = 0.006) and OS were 31 and 15 months, respectively (p < 0.001). 5y-RFS and 5y-OS in group of patients completed consolidation therapy were 25 and 40%, respectively, which were significantly greater than those in the whole study population, p < 0.001 (Fig. 1). A total of 141 AML patients, WBC < 100,000/μL (p = 0.004) and wild-type FLT3-ITD (p = 0.047) were associated with significantly longer RFS than those in groups of WBC ≥ 100,000/μL and FLT3-ITD positive AML in the univariate analysis, however, they were not found to be independently significant in multivariate analysis. In univariate analysis of OS, five individual factors significantly affected OS in the overall study population in order of increasing significance, these included patients aged < 60 years, WBC < 100,000/ μL, non adverse cytogenetic risk, wild type FLT3-ITD and CR patients, nevertheless, the multivariate analysis indicated only non adverse cytogenetic risk (p = 0.035) and CR patients (p < 0.001) were significantly favorable prognostic factors for OS. The status of high WBC count and CR affected the survival in 141 AML patients are shown in Fig. 2. Among 41 AML patients who completed consolidation therapy, the results of univariate analysis showed that the patients with favorable and intermediate risk cytogenetics had prolonged RFS (p = 0.044) and OS (p = 0.007) than those with adverse karyotype. In CN-AML, consolidation with IDAC had longer median RFS than HiDAC, 22 versus 12 months. Nevertheless, consolidation with HiDAC had 5y-RFS than IDAC, 40% versus < 15% (Figs. 3,  4), and patients with WBC < 100,000/μL at diagnosis was also associated with prolonged OS (p = 0.013), but the difference was not significant in multivariate analysis.
A total of 83 patients with CN-AML, patients with WBC < 100,000/μL (p = 0.003) and wild-type FLT3-ITD (p < 0.001) were found to be significantly associated with increased RFS, while patients with WBC < 100,000/μL (p = 0.008) and patients aged < 60 years (p = 0.003) were associated with longer OS according to the univariate analysis. In multivariate analysis, shorter RFS was found in positive FLT3-ITD AML patients (p = 0.025) and shorter OS were observed in group of WBC ≥ 100,000/μL (p = 0.026) and non-CR patients (p = 0.027) ( Table 2).
In patients with CN-AML who completed consolidation therapy, only patients with WBC < 100,000/μL had longer RFS (p = 0.029) and OS (p = 0.017) than those with WBC ≥ 100,000/μL. Mutant NPM or CEBPA gene illustrated longer RFS and OS than those in wild type NPM or CEBPA, but the difference was not significant. In group of untreated patients (30 patient), CEBPA mutation was also a significant factor for prolonged OS (p = 0.016). Higher RFS and OS were found in biallelic CEBPA mutation compared to single CEBPA mutation in the entire study population and every subgroups of study but no statistically significant differences. The factors associated with survival of 141 AML patients are shown in Table 2. The impact of IDAC consolidation on survival in AML patients with and without gene mutations were analyzed and found that AML patients harboring FLT3-ITD and both FLT3-ITD and NPM1 mutations had poor RFS and OS compared with those in AML patients with mutant NPM1 or no mutated gene (Fig. 5). HiDAC was treated in 11 patients without mutation and only 1 patient with mutated CEBPA, therefore, we had no result on survival issue after HiDAC therapy in mutant NPM1 and FLT3-ITD AML (Table 3).

Discussion
This study demonstrated long-term outcomes of consolidation chemotherapy in adult AML patients who didn't undergo transplantation, because of lack of an HLA matching donor, financial problem, unfit or older patients. The prognostic impact of cytogenetic abnormalities on survival have significant implication for AML even in the era of molecular risk stratification in AML [11][12][13]. Numerous somatic gene mutations have been reported as a potential tool to predict survival outcomes in cytogenetically normal AML [14][15][16][17][18][19][20][21][22][23]. Our results illustrated that normal cytogenetic was found in 59% of all de novo AML patients and it was observed more in patients aged ≥ 40 years. In group of patients aged < 40 years, 41% had cytogenetic abnormalities and found adverse cytogenetic abnormalities (31%) than intermediate (15.5%) and favorable risk (13%), despite almost all of them didn't have AML with myelodysplasia related changes or prior history of myelodysplastic syndrome. The prevalence of CN-AML in this series was slightly higher than that in the other previous studies (42-48%) from MRC AML10 [11], CALGB 8461 [12,   consolidation with chemotherapy, but there was no statistically significant difference because of short follow up duration of azacitidine group, small number of patients receiving azacitidine and patients completed consolidation chemotherapy. Elderly patients had longer OS than younger patients, this may be because most elderly patients had normal cytogenetic (75%) and had low incidence of adverse cytogenetics (11%). Besides, they received consolidation with IDAC or azacitidine, which were low intensity, less myelosuppressive and severe infection complications. Thus azacitidine might be suitable for consolidation therapy in older AML patients. WBC ≥ 100,000/μL was also significantly related to shorter RFS and OS in the whole study population and all subgroup analysis, however, the number of WBC did not affect the CR rate and patients achieving CR had longer OS compared with that in patients who failed to achieve CR.
The limitation of this study were a retrospective study, small number of patients who completed 4 cycles of consolidation therapy and short follow up duration in AML patients with CEBPA mutation, however, the consolidation therapy regimen in this study was chosen follow the patients' status during AML treatment without selection bias which representing the real results in the clinical practice under limited resource.

Conclusions
Cytogenetic and mutation test for FLT3-ITD, NPM1 and CEBPA genes were useful for identify prognostic outcomes in adult AML. Adverse cytogenetic abnormalities and FLT3-ITD mutation exhibited dismal RFS and OS. Authors' contributions PN designed the study, collected the cytogenetic, molecular and clinical data and data analysis. NL, SP and BR performed molecular study. SP, PC, SP, KB, TP, PA, AU, BR and SC collected the sample. All authors read and approved the final manuscript.