Detection of KIT mutations in core binding factor acute myeloid leukemia.

We have investigated the frequency and the effect of KIT mutations on the outcome of patients with CBF-AML. 69 patients (34 pediatrics and 35 adults) with CBF-AML were enrolled in the study. The frequency of KIT mutations was higher in adults compared to pediatrics (22.9% and 14.7%, p = 0.38) respectively. Leukocytosis ≥ 20 × 109 /L was significantly associated with pediatrics compared to adults. t(8;21)(q22;22) was significantly associated with thrombocytopenia in adults. We conclude that no significant difference is found between KIT mutated and unmutated CBF-AML in adults and pediatrics. Children with CBF-AML present with leukocytosis. t(8;21) is associated with thrombocytopenia.

Introduction

Core binding factor acute myeloid leukemia (CBF-AML) represents 4–12% of all AML, 15% of adults and 25–30% of pediatrics. Patients with CBF-AML are characterized with high complete remission (CR) rates (86–88%), however, 30–50% of patients relapse, and the 5-year survival is only 50% [1].

Mutations in the KIT gene are the most common (15–45%) in CBF-AML. The KIT gene is located on, chromosome band 4q11-12 and encodes a 145-kDa transmembrane glycoprotein that is a member of the type III tyrosine kinase family. Binding of stem cell factor (KIT ligand) to the KIT receptor activates downstream signaling pathways important for cell proliferation, differentiation, and survival [2]. KIT mutations result in ligand- independent activation and most commonly affects the extracellular portion of the receptor (exon 8), and the tyrosine kinase domain (exon 17). Mutations affecting the juxta-membrane domain (exon 10 and 11) are less common KIT mutations have been associated with poor outcome in CBF-AML [3].

The national comprehensive cancer network (NCCN) guidelines have included KIT mutations as a prognostic marker that can change CBF-AML from favorable to intermediate risk group [4]. In contrast, the European Leukemia Net did not add KIT mutations in the routine workup for patients with CBF-AML Unlike the cytogenetic classification, the outcome of CBF-AML is heterogeneous [5].

The aims of this work were to analyze the different clinical and prognostic characteristics of CBF-AML and to investigate the prevalence and prognostic effect of KIT mutations (exon 8 and exon 17) on the outcome of this group of AML patients.

Materials and methods

Patients

Patients were recruited in a period of two years, retrospectively and prospectively from June 2014 to June 2016. 765 patients were diagnosed with AML, 234 pediatrics and 531 adults. A total of 69 patients (34 pediatrics and 35 adults) were diagnosed with CBF AML and included in this study. Sub classification of AML according to French-American-British (FAB) subtypes was based on morphology, cytochemistry (chloroacetate esterase and myeloperoxidase) and immunophenotyping. The diagnosis CBF-AML was confirmed by the detection of inv (16)(p13q22) or t(8,21) (q22;q22)/RUNX1-RUNX1T1 fusion genes using reverse transcriptase–polymerase chain reaction (RT-PCR). All patients gave informed consent and the study was approved by the Institutional Review Board (IRB), (201516028.4) of the National Cancer Institute (NCI), Egypt.

All patients received the 3 and 7 induction chemotherapy at the NCI, induction chemotherapy consisted of Adriamycin 30 mg/m2 for 3 days and ARAC 100 mg/m2 by continuous infusion for 7 days, after complete remission high-dose cytarabine 3 g/m2 IV over 3 h every 12 h on days 1, 3, and 5 for four cycles was given for consolidation (5). Achievement of CR was defined by the detection of less than 5% blasts in normocellular BM. Overall survival (OS) was measured for all living patients from the date of entry to the date of death or last time follow up. Disease free survival (DFS) was calculated from the date of CR to the date of relapse in the first CR.

Detection of C-KIT mutations

DNA extraction: Genomic DNA was extracted from BM samples using GeneJET Whole Blood Genomic DNA Purification Mini Kit(#K0781- Thermo Scientific). DNA quantity and quality were checked using Thermo Scientific NanoDrop™ 1000 Spectrophotometer.

HRM analysis for KIT exon 8 mutations

HRM analysis was used for the analysis of KIT exon 8 mutations. All samples were tested in triplicates on 7500 fast real-time pcr-Applied Biosystems. Positive and negative controls were included in each run. Twenty nanograms of DNA were amplified in 20 µl reaction volume containing 0.5 µM forward and reverse primers designed by [6] and 10 µl of MeltDoctor™ HRM (Applied Biosystems) master Mix with its thermal profile for 45 cycles with a ramp of 0.02 °C/ S. The expected PCR product was 219 bp.

Upon completion of the run, data were analyzed as fluorescence versus temperature graphs (temperature shifting, difference plots, and derivative melting curves) using High Resolution Melting (HRM) Software version 2.0 (#4,397,808).

PCR amplification and cycle sequencing for KIT exon 8 mutations

Samples positive for exon 8 mutations with same primers were confirmed by sequencing. PCR products were purified using QIAquick PCR Purification Kit (#28,104). Cycle sequencing was performed using BigDye™ Terminator v3.1 Cycle Sequencing Kit and the sequencing product was purified using Centri-Sep™ Spin Columns (#401,762) according to manufacturer instructions. Sequencing products were then resuspended with 10 µl of Hi-Di™ Formamide (#4,311,320), incubated at 95C for 5 minutes and then chilled on ice for 5 min. Bidirectional sequencing was performed on the Applied Biosystems™ 3500 Genetic Analyzer. Sequencing traces were analyzed by Applied Biosystems SeqScape Software v2.5. The analysis of data was done according to Gene Bank accession number (U63834.1).

Fragment analysis for KIT exon 8 mutations

Fragment analysis was used to help in the analysis of indel mutations of KIT exon 8. PCR reaction was performed as previously described by [7] using a Fluorescently labeled forward primer (FAM). Next, these PCR products were added to 7 μL of Hi-Di™ Formamide (#4,311,320) and 2 μL of GeneScan™ 500 LIZ™ dye Size Standard (#4,322,682).The mixture was then injected to Applied Biosystems™ 3500 Genetic Analyzer, analysis and verification of fragments size done using GeneMapper® Software Version 4.1 Microsatellite Analysis.

PCR-RFLP for KIT exon 17 mutations (D816)

KIT exon 17 was amplified as previously described by [8]. PCR products were digested by AatII (10 U/µL) (#ER0991) at 37 °C overnight. Heterozygous samples create 106, 85 and 21 bp fragments, while wild samples create 85 and 21-bp fragments.

Statistical analysis

Descriptive statistics were calculated for all variables. Patient follow-up was updated on April 1, 2017. Disease-free survival (DFS) and relapse-free survival (RFS) were estimated from the date of complete remission. Differences in proportions were assessed using the v2 or Fisher exact statistic. Survival was plotted with Kaplan–Meier curves and the data for the various groups were compared with independent T- test, [9]. All survival estimates were reported 1 standard error (SE). All P values were 2-sided, P value less than 0.05 was considered statistically significant. Statistical analysis was performed using SPSS version 24.0 software statistical package (SPSS, Chicago, IL, USA).

Results

Comparison of initial clinical and laboratory characteristics of CBF-AML between pediatrics and adults patient's groups

Mean age was (7.1 ± 4.7 vs 32.1 ± 10.2) years for pediatrics and adults respectively, and median was (6.0 (1.0–16) vs 30.0 (19–59)) years for pediatrics and adults respectively. The prevalence of CBF-AML was 69/765 (9%), (14.9%, 6.36%) for pediatrics and adults respectively. t(8;21)(q22;22) was detected in (64.7% vs 62.9%, p = 0.87) for pediatrics and adults respectively. Inv(16) (p13q22) was detected in (35.3% vs 37.1%, p = 0.87) for pediatrics and adults respectively. Leukocytosis ≥ 20 × 109 /L was significantly associated with pediatrics compared to adults, (64.7% vs 40%, p = 0.04). The frequency of FLT3-ITD and FLT-3 TKD mutations was (2.9% vs 5.7%, p = 0.57, 2.9% vs 8.6%, p = 0.3) for pediatrics and adults respectively. 26 pediatrics and 26 adults were followed till the end of induction chemotherapy, 61.5% of pediatrics and 69.2% of adults achieved complete remission with no significant difference, p = 0.771. The total death rate was 31(44.9%). Four (12.9%) patients died before starting chemotherapy from disease progression and 12 (38.7%) patients died during induction chemotherapy. Death was contributed mainly to infections and febrile neutropenia. OS at 6 months was significantly higher in adults compared to pediatrics (63.3% vs 35.7%, p = 0.043).

Comparison of initial clinical and laboratory characteristics between inv (16) and t(8;21) in CBF AML

Inv (16) (p13q22) was significantly associated with FAB subtypes M4 + M5 and t(8;21))(q22;22) was significantly associated with M1 + M2 in pediatrics and adults (p < 0.001, p = 0.004) respectively.

In pediatrics, inv (16) (p13q22) CBF-AML was associated with hepatomegaly, splenomegaly and lymphadenopathy compared to t (8;21)(q22;22) (41.7% vs 22.7%, p = 0.27; 25% vs 13.6%, p = 0.64; 8.3%vs 4.5%, p = 1) respectively. inv (16) (p13q22) positive CBF-AML was also associated with anemia (HGB ≤ 8 g/dl) and leukocytosis ≥ 20 × 109/L compared to t (8;21)(q22;22) (91.7% vs 59.1%, p = 0.06; 83.8% vs 54.5%, p = 0.14) respectively.

In adults, t(8;21)(q22;22) was significantly associated with thrombocytopenia ≤ 20 × 109/L compared to inv (16) (50% vs 7.7%, p = 0.01). However, t(8;21)(q22;22) was significantly associated anemia (HGB ≤ 8 g/dl) with compared to inv (16) (p13q22) (72.7% vs 30.8%, p = 0.03) respectively. No significant difference was found in response to induction chemotherapy was found between in inv 16 compared to t(8;21) in pediatrics and adults (p = 0.18, p = 0.19 respectively, Table 1.

Table 1

Initial patients characteristics for t (8;21) and inv(16) CBF AML patients.

CBF AML	Pediatric CBF-AML N = 34(%)		p-value	Adult CBF-AML N = 35(%)		p value
	inv 16 12 (35.4)	t(8;21) 22(64.7)		Inv 16 13(37)	t(8;21) 22(62.8)
Age
Mean ± SD	5.4 ± 5.3	8.01 ± 4.3	0.132	34.9 ± 12.7	30.5 ± 8.3	0.221
Gender
Males	9(42.9)	12(57.1)	0.292	9(47.4)	10(52.6)
females	3(23.1)	10(76.9)		4(25)	12(75)	0.29
Hepatomegaly	5(41.7)	5 (22.7)	0.27	4(30.8)	7(31.8)	1
Splenomegaly	3 (25)	3 (13.6)	0.641	4(30.8)	5(22.7)	0.698
Lymphadenopathy	1(8.3)	1(4.5)	1	2(15.4)	1(4.5)	0.541
FLT3-ITD positive	0	1(4.5)	1	0	2(9.1)	0.519
FLT3-TKD positive	1(100)	0	0.35	1(7.7))	2(9.1)	1
KIT mutations
Wild	10(83.3)	19(86.4)		11(84.6)	16(72.7)
mutants	2(16.7)	3(13.6)	1	2(15.4)	6(27.3)	0.68
FAB subtypes
M1 + M2	2(16.7)	19(86.4)		4(30.8)	18(81.8)
M4 + M5	10(83.3)	3(13.6)	<0.001	9(69.2)	4(18.2)	0.004
TLC
< 20 × 109 /L	2(16.7)	10(54.5)		6(46.2)	15(68.2)
≥ 20 × 109/L	10(83.8)	12(54.5)	0.14	7(53.8)	7(31.8)	0.288
Platelets
≤ 20 × 109 /L	5(41.7)	5(22.7)		1(7.7)	11(50)
> 20 × 109/L	7(58.3)	17(77.3)	0.271	12(92.3)	11(50)	0.01
HGB
≤ 8 g/dl	11(91.7))	13(59.1)		4(30.8)	16(72.7)
> 8 g/dl	1(8.3)	9(40.9)	0.06	9(69.2)	6(27.3)	0.03
BM blasts
Mean ± SD	44.7 ± 13.9	54.9 ± 22.7	0.117	59.3 ± 25.8	49.6 ± 22.4	0.276
PB blasts
Mean ± SD	32.9 ± 16.3	37.3 ± 19.4	0.490	47.3 ± 30.2	31.6 ± 17.2	0.103
Reponses to induction chemotherapy
CR	3(37.5)	13(72.2)		4(50)	14(77.8)
RD	5(62.5)	5(27.8)	0.18	4(50)	4(22.2)	0.197
OS
6 months (%)	39.3	31.9		37	75
median	0.8	2.5	0.772	-	-	0.91

Data are presented as n (%). FAB: French-American-British classification, FLT3 ITD: Fms-like Tyrosine Kinase Internal Tandem Duplication, FLT3 TKD: Fms-like Tyrosine Kinase Tyrosine Kinase Domain, HGB: Hemoglobin, PLT: Platelets count, TLC: Total Leukocyte Count, PB Blast: Peripheral Blood Blast, BM Blast:Bone Marrow Blast. CR: Complete Remission, RD: Resistant Disease, OS: Overall survival.

Effect of the initial clinical and laboratory characteristics on OS of CBF-AML

When we investigated the effect of different pretreatment clinical and laboratory parameters on the OS at 12 month period for adults and 6 months for pediatric CBF-AML, because the number of pediatric cases within the strata was too small at one year to be presented in the results. We have found that thrombocytopenia ≤ 20 × 109 /L had a significantly adverse effect on OS in adults (p = 0.04) CBF-AML, Table 2.

Table 2

The association between different prognostic factors and OS at 6 and 12 months in CBF-AML.

Prognostic factors	Pediatrics N = 34				Adults N = 34*
	N	6 months	Median	N	6 months		12 months	Median
t(8;21) (q22.q22)
Negative	12	39.3	0.8	13	37.0		37.0	1.6
Positive	22	31.9	2.5	21	75.0		62.9	-
p value		0.844			0.17
FAB
M1 = M2	21	28.7	2.5	21	76.5		62.6	−
M4 + M4	13	49.2	0.8	13	43.3		43.3	5
p value		0.77			0.09
Inv 16
Negative	22	31.9	2.5	21	75.0		62.9	–
Positive	12	39.3	0.8	13	37.0		37.0
p value		0.77			0.09
Platelets
≤ 20 × 109 /L	10	54.0	6	12	81.8		81.8	−
> 20 × 109/L	24	30.0	2.3	22	51.8		39.3	8.7
p value		0.7			0.047
TLC
< 20 × 109 /L	12	45.0	3	21	68.4		61.9	−
≥ 20 × 109/L	22	32.0	2.3	13	54.9		44.0	8.7
p value		0.415			0.211
HGB
≤ 8 g/dl	24	41.2	3	19	70.1		55.2	-
> 8 g/dl	10	25.9	2.3	15	55.3		55.3	-
p value		0.235			0.758
Hepatomegaly
Negative	24	24.1	1.9	23	69.1		69.8	–
Positive	10	66.7	6.0	11	51.1		40.9
p value		0.10			0.262
Splenomegaly
Negative	28	29.1	2.3	25	69.3		53.0	-
Positive	6	60	-	9	59.3		59.3	-
p value		0.183			0.832

Data are presented as n (%). FAB: French-American-British classification,FLT3 ITD: Fms-like Tyrosine Kinase Internal Tandem Duplication, FLT3 TKD: Fms-like Tyrosine Kinase Tyrosine Kinase Domain, HGB: Hemoglobin, PLT: Platelets count, TLC: Total Leukocyte Count, PB Blast: Peripheral Blood Blast, BM Bone marrow blasts. *Data are presented for 34 adult patients only.

Prevalence and types of KIT mutations

The frequency of KIT mutations was higher in adults compared to pediatrics (22.9% and 14.7%, p = 0.38) respectively. KIT exon 8 mutations were positive (11.8% and 8.6%, p = 0.66) for pediatrics and adults respectively. Melting curve analysis revealed a single peak with a mean Tm at (80ͦC) for wild cases, and a clear difference between wild and mutant cases for KIT exon 8 mutations. Fragment analysis and sequencing confirmed the insertion/deletion mutations. Sequencing confirmed positive cases, Table 3. KIT exon 17 was positive in (2.9% and 14.3%, p = 0.095) pediatrics and adults respectively.

Table 3

Sequencing analysis of KIT exon8 mutations.

Patient ID	Blast percentage	Cytogenetics	Fragment analysis	Nucleotide change	Amino acid change
65	75	t(8:21)	Deletion 6 bases	Del GACTTACGA ins TGC	L416_D419
2	67	Inv (16,16)	Deletion 6 bases	1346_1351delACTTAC	T417_Y418
53	20	t(8:21)	Insertion 5 bp	1352delinsTTCCT	R419Ffs*5
40	50	Inv (16,16)	Deletion 4 bp	1350_1353delACGA	Y418Sfs *4
23	79	Inv (16,16)	Deletion 6 bp	Del ACGACAGGCTCG insGCGC	Y418_V422
19	30	t(8:21)	Deletion 3 bp	1350_1352delACG	D419
31	62	t(8:21)	Insertion 5 bp	1351_1356delinsTTCCT	D419Sfs*5

Fs: frame shift mutation.

Clinical and laboratory characteristics for KIT mutations

We have compared the initial laboratory and clinical characteristics for patients positive and negative for KIT mutations and found no significant difference was found between KIT mutated and unmutated CBF-AML in adults and pediatrics, Table 4.

Table 4

Association of KIT mutations with prognostic factors in CBF-AML.

Variable	Pediatrics			Adults
	KIT negative N = 29	KIT positive N = 5	p value	KIT negative N = 27	KIT positive N = 8	p- value
Sex
Male: n (%)	16(55.2)	5(100)		14(51.9)	5(62.5)
Female: n (%)	13(44.8)		0.132	13(41.8)	3(37.5)	0.700
FAB subtypes
M1 + M2	18(62.1)	3(60)		16(59.3)	6(75)
M4 + M5	11(37.9)	2(40)	1.000	11(40.7)	2(25)	0.683
Hepatomegaly
negative: n (%)	21(72.4)	3(60)		20(74.1)	4(50)
positive: n (%)	8(27.6)	2(40)	0.678	7(25.9)	4(50)	0.226
Splenomegaly
negative: n (%)	24(82.8)	4(80)		22(81.5)	4(50)
positive: n (%)	5(17.2)	1(20)	1.000	5(18.5)	4(50)	0.162
Lymphadenopathy
negative: n (%)	27(93.1)	5(100)		2(7.4)	1(12.5)
positive: n (%)	2(6.9)		1.000	2(7.4)	1(12.5)	0.553
t(8:21)
t(8:21) negative	10(34.5)	2(40)		11(40.7)	2(25)
t(8:21) positive	19(65.5)	3(60)	1.000	16(59.3)	6(75)	0.680
Inv (16)
Inv16 wild	19(65.5)	3(60)		16(59.3)	6(75)
Inv 16 positive	10(34.5)	2(40)	1.000	11(40.7)	2(25)	0.680
FLT3- ITD
FLT3 ITD wild	28(69.6)	5(100)		25(92.6)	8(100)
FLT3 ITD mutant	1(3.4)	0	1.000	2(7.4)	0	1.000
FT3- TKD
FLT3 TKD wild	28(69.6)	5(100)		25(92.6)	7(87.5)
FLT3 TKD mutant	1(3.4)	0	1.000	2(7.4)	1(12.5)	0.553
HGB (g/dL)
≤ 8 g/dl	21(72.4)	3(60)		16(59.3)	4(50)
> 8 g/dl	10(34.5)	2(40)	0.618	11(40.7)	4(50)	0.700
PLT (x109/L)
< = 20 × 109/L	8(27.6)	2(40)		8(29.6)	4(50)
> 20 × 109/L	8(27.6)	2(40)	0.618	8(29.6)	4(50)	0.402
TLC (x109/L)
< 20,000	10(34.5)	2(40)		16(59.3)	5(62.5)
> 20 × 109/L	19(65.5)	3(60)	1.000	11(40.7)	3(37.5)	1.000
Pb Blast (%)
Mean ± SD	34.483 ± 18.13	43.2 ± 19.52	0.392	38.47 ± 25.5	34.12 ± 17.36	0.558
BM Blast (%)
Mean ± SD	50.65 ± 20.03	55.2 ± 24.87	0.715	51.074 ± 24.21	60.62 ± 22.66	0.323
Response to induction chemotherapy
CR	15(56.2)	1(33.3)		13(68.4)	2(28.6)
RD	8(34.8)	2(66.7)	1.000	6(31.6)	5(71.4)	1.000

Data are presented as n (%). FAB: French-American-British classification, FLT3 ITD: Fms-like Tyrosine Kinase Internal Tandem Duplication, FLT3 TKD: Fms-like Tyrosine Kinase Tyrosine Kinase Domain, HGB: Hemoglobin, PLT: Platelets count, TLC: Total Leukocyte Count, PB Blast: Peripheral Blood Blast, BM Bone marrow blasts. CR: Complete Remission, RD: Resistant Disease, OS: Overall survival.

Discussion

In the present study, we have primarily investigated CBF-AML as one group in (34 pediatrics and 35 adults) patients for different clinical, laboratory, and secondary molecular aberrations (c-KIT and FLT-3) that might affect the patient's outcome. Then we have separated each CBF-AML into two subtypes t(8;21) and inv (16) searching for the different role of each subtypes on the clinical outcome of CBF-AML. Based on the current controversy regarding the effect of KITmutations on the prognosis of CBF-AML, we have analyzed KIT mutations status on CBF-AML as a one group because of the small number of cases.

When we considered all CBF-AML patients and compared pediatrics with adult CBF-AML, leukocytosis was significantly associated with pediatrics compared to adults (64.7% vs 40%, p = 0.04) respectively, similar results were previously reported [10], [11].

FLT-3-ITD mutation is frequent in cytogenetically normal AML with adverse prognostic effect, however, it is relatively uncommon in CBF-AML with uncertain prognostic significance. In this study, the frequency of FLT-ITD mutations was low (2.9% and 5.7%) in pediatrics and adults respectively. These results were consistent with previous studies [7], [12]. In contrast, FLT-3 TKD D835 mutation has been associated with favorable prognostic effect in inv (16) positive CBF-AML [13], [14]. In this study, the frequency of FLT-3 TKD D835 mutation was (2.9% and 8.6%) for pediatrics and adults respectively and it was comparable to previous studies (6–24%) [15].

The frequency of c-KIT mutations was 14.2% and 23.5% for pediatrics and adults respectively, which was in the range reported by previous studies in pediatrics (11–41.5%) and adults (17–46%) [5], [7], [11].

Previous reports have implicated, older age, initial TLC, percentage of BM blasts (both are linked), and platelets count in influencing the CBF-AML outcome. [1], [16], [17], [18]. In this study, multivariant analysis revealed thrombocytopenia ≤ 20 × 109 /L as the only significant prognostic factor affecting OS in adults (81.8% vs 54.6%, p = 0.04). In accordance with Cancer and Leukemia Group B (CALGB), thrombocytopenia was significantly associated with t(8;21) compared to inv (16) in adults (50% vs 7.7%, p = 0.01). Lower OS was reported by the (CALGB) for CBF-AML t(8;21) positive patients presenting with thrombocytopenia [19]. Appelbaum et al. [18], found t(8;21) by itself, had a poorer outcome compared to inv (16) on OS, after adjusting for age and BM blasts. In this study, neither leukocytosis or the percentage of BM

We confirm previous reports showing the significant association between, t(8;21) (q22;q22) with FAB M1 + M2 and inv (16) with M4 + M5 subtypes in both adults and pediatrics (p < 0.001, p = 0.004) [1], [15]. In addition, extramedullary manifestations (hepatosplenomegaly and lymphadenopathy) were more common with inv (16) compared to t(8;21) (q22;q22) (41.7% vs 22.7%, p = 0.64; 25% vs 13.6%, p = 0.64; 8.3%vs 4.5%, p = 1). In addition, leukocytosis was frequent with inv(16) compared to t(8;21), (83.8% vs 54.5%, p = 0.14) these finding were in accordance with previous studies [1], [15], [16], [20]

In this study, KIT mutation status was not associated with gender, age, initial WBC count, platelet count, and percentage of BM blasts. Similar results were obtained by Riera et al. [10]. In contrast, [11], [13], [21] reported that KIT exon 17 was associated with leukocytosis in t(8;21) compared to inv(16). Paschka et al. [5] observed higher WBC and BM blasts in patients positive for KIT and FLT3 mutations. Previous studies have also found an association between exon 8 mutations and exon 17 D816 with inv(16) and t(8;21) respectively [3], [15], [21]. Unfortunately, we could not analyze this association because of the small number of positive cases.

Few studies have addressed the prognostic effect of KIT mutations in pediatrics. In accordance with previous studies [3], [22], [23], [24], we found no significant effect for KIT mutations in pediatric CBF-AML patient's outcome. On the other hand, Shimada et al. [21] and Manara et al. [25] found an adverse effect of KIT mutation on t(8;21) subtype of CBF-AML. (both were letters to the editors). The lack of prognostic effect for KIT mutations in pediatrics compared to adults could be related to the difference in treatment protocols or to the maturation stage of leukemic progenitors at which the mutation achieve clonal dominance [3].

In contrast to pediatrics, the effect ofKIT mutations on the prognosis of adult CBF-AML was studied extensively. It was found that KIT mutations have no effect on the response to induction chemotherapy [5], [11], [26] but it has a different prognostic effects in t(8;21) and inv (16) CBF-AML subtypes. Pashka et al. [5]; Care et al. [13] found an adverse effect for KITr mutations in adults inv(16) AML. Other studies [11], [12], [2] found a prognostic effect for KIT exon 17 mutation in t(8;2) AML. Contrary to previous results, we and Riera et al. [10] found no difference between mutated and unmutated CBF-AML in adults.

Conclusion

The difference in the clinical and laboratory characteristics between inv(16) and compared t(8;21) positive AML, suggests dealing with these cytogenetic abnormalities as two separate entities and not as one group. t(8;21) is associated with thrombocytopenia, and it has an adverse effect on the OS of adult CBF-AML .inv (16) is associated with leukocytosis and extramedullary manifestations. KIT mutations are frequent in CBF-AML. FLT3 mutations are rare in CBF-AML. The prognostic effect of KIT mutations requires studying larger number of samples.

Funding

This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit-sectors.

Disclosure

Passant Badr, Ghada M Elsayed, Dalia Negm Aldin, Bahia Y Riad, Nayera Hamdy declare that they have no conflict of interest.