Clinical Characteristics and Outcomes of Acute Myeloid Leukemia Patients Harboring NPM1/FLT3-ITD/DNMT3A Triple Mutations and the Potential Prognostic Value of GNG4

Abstract

Introduction

Nucleophosmin 1 (NPM1), FMS-like tyrosine kinase 3-internal tandem duplication (FLT3-ITD), and de novo methyl transferase 3 A (DNMT3A) triple-mutated acute myeloid leukemia (AML) represents a distinct entity with poor outcomes.

Methods

We explored the gene mutation spectrum and clinical characteristics of 165 AML patients retrospectively, particularly comparing patients with NPM1/FLT3-ITD/DNMT3A triple-mutations and those without.

Results

Our results demonstrated significantly elevated white blood cell counts (P < 0.001), bone marrow blast percentages (P = 0.037), and platelet counts (P = 0.007) in the triple-mutated cohort (6.7%) compared to the non-triple-mutated patients. Furthermore, all triple-mutated cases were classified as the M4/M5 subtype of the French-American-British classification (P = 0.017). Although no significant difference in complete remission rates was observed between the groups after initial treatment, the median overall survival for triple-mutated AML patients was only 4 months. Using the Gene Expression Omnibus (GEO) database and bioinformatics, we compared AML^{NPM1mutFLT3-ITDmutDNMT3Amut} and AML^{NPM1mutFLT3-ITDmutDNMT3Awt}. A total of 246 AML patients from the GEO dataset were included to evaluate the expression profiles of differentially expressed genes. The guanine nucleotide-binding protein subunit γ 4 (GNG4) was differentially expressed between AML^{NPM1mutFLT3-ITDmutDNMT3Amut} and AML^{NPM1mutFLT3-ITDmutDNMT3Awt}, which had the most adjacent nodes among hub genes. The prognostic value of GNG4 was further validated in AML patient samples through qRT-PCR.

Conclusion

Clinical validation indicated a substantial downregulation of GNG4 in AML^{NPM1mutFLT3-ITDmutDNMT3Amut} compared to AML^{NPM1mutFLT3-ITDmutDNMT3Awt} patients. Thus, GNG4 may play a role in the low survival rate of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients, offering novel insights into the prognosis, therapeutic targets, and prognostic evaluation of AML.

Plain Language Summary

Acute myeloid leukemia (AML) is a type of blood cancer that can be influenced by various genetic mutations. In this study, we focused on a specific group of AML patients who have mutations in three genes: Nucleophosmin 1 (NPM1), FMS-like tyrosine kinase 3-internal tandem duplication (FLT3-ITD), and de novo methyl transferase 3 A (DNMT3A). These patients tend to have a poor prognosis. We compared 165 AML patients, analyzing those with these three mutations (referred to as the triple-mutated group) and those without. Our findings showed that patients with the triple mutations had higher white blood cell counts, more blast cells in the bone marrow, and higher platelet counts compared to patients without the triple mutations. All of the triple-mutated patients were also classified under a particular subtype of AML, the M4/M5 subtype. However, even though the two groups had similar remission rates after initial treatment, patients with the triple mutations had a much shorter overall survival rate, with a median survival of just 4 months. We also used a large dataset to examine gene expression patterns in these patients. One gene, called GNG4, was found to be significantly downregulated in patients with the triple mutations. This gene may help explain the poor survival outcomes for these patients. Our results suggest that GNG4 could serve as a potential target for future therapies and may help doctors predict the prognosis of AML patients with this triple mutation.

Graphical Abstract

Keywords

acute myeloid leukemia (AML)prognosis

Introduction

Acute myeloid leukemia (AML) is a heterogeneous hematologic malignancy with a median overall survival (OS) of less than 1 year.¹ Recent advancements in high-throughput sequencing have emphasized the critical role of gene mutations in the diagnosis, prognostic risk stratification, and therapeutic decision-making of AML.² The most frequently observed mutations in AML include those in nucleophosmin 1 (NPM1) and FMS-like tyrosine kinase 3-internal tandem duplication (FLT3-ITD).³ NPM1 mutations are present in approximately 25-35% of adult AML cases and about 8% of pediatric cases^4,5 and are associated with higher complete remission (CR) rates, longer event-free survival, and extended OS.^6-8 FLT3-ITD mutations, which occur in around 25% of AML patients,⁹ are linked to increased leukemic cell counts, shorter OS, and a higher cumulative incidence of relapse. NPM1 and FLT3-ITD mutations frequently co-occur in adult AML patients.¹⁰ Patients with NPM1 and FLT3-ITD mutations are classified as low-risk or intermediate-risk according to the ELN-2017 guidelines.¹¹ However, these patients with coexisting mutations of NPM1 and FLT3-ITD have variable clinical courses.¹² A sizable proportion of patients have a poor prognosis and short survival,¹² particularly when co-occurring with de novo methyl transferase 3 A (DNMT3A) mutations.¹³

The DNMT3A gene encodes a 130-kDa protein that is involved in DNA methylation modification.¹⁴ Aberrant DNA methylation patterns are a hallmark of AML.^15,16 DNMT3A mutations are found in approximately 20% of newly diagnosed AML patients.¹⁷ Notably, the co-occurrence of DNMT3A, NPM1, and FLT3-ITD mutations is relatively common, with approximately 7% of AML patients harboring these triple gene mutations.^3,18 Co-occurrence of these three mutations is more frequent than can be explained by chance⁸ and is associated with typical clinical features, such as significantly higher white blood cell counts.¹³ Recent studies revealed genetic interaction of these three mutations in mice¹⁹ and humans,³ further suggesting that triple-mutated AML represents a distinct entity with very poor outcome.¹³ However, there are relatively few studies investigating NPM1/FLT3-ITD/DNMT3A triple-mutated AML,²⁰ and little is known about its underlying molecular mechanisms. Therefore, further research is required to investigate the interactions between DNMT3A mutations and co-mutations in NPM1 and FLT3-ITD in triple-mutated AML patients.

GNG4 encodes a G-protein γ subunit involved in transmembrane signaling, and recent studies suggest it acts as a context-dependent regulator of leukemogenic pathways. Notably, GNG4 is frequently epigenetically silenced (eg, via promoter hypermethylation) in various cancers, including hematologic malignancies, and this loss is associated with more aggressive disease and poorer outcomes.²¹ Mechanistically, GNG4 normally modulates GPCR signaling – for example, restoring GNG4 expression can dampen SDF1α/CXCR4-driven migration signaling in tumor cells and help maintain MHC class I expression via NF-κB signaling, an effect that bolsters anti-tumor immune surveillance.²² Thus, downregulation of GNG4 may promote leukemogenesis and immune evasion, while high GNG4 levels have been linked to improved therapy responses in certain contexts.^21,22 GNG4’s prognostic impact is under active investigation as a potential biomarker in leukemia, given its influence on signaling pathways and correlation with patient outcomes.²¹ One recent example is Kawase et al, 2023, who identified GNG4 as a crucial NF-κB–dependent mediator of immune checkpoint efficacy.²²

In acute myeloid leukemia (AML), recurrent driver mutations commonly affect genes regulating epigenetics, signaling, or transcription, notably NPM1, FLT3-ITD, DNMT3A, CEBPA double, ASXL1, RUNX1, TET2, IDH1/2, KIT, and TP53.²³ These mutations often co-occur in non-random patterns that define molecular subgroups. For instance, NPM1-mutated AML frequently harbors concurrent FLT3-ITD and DNMT3A mutations (often alongside IDH1/2 or TET2), reflecting a common genotype in normal-karyotype AML.²⁴ By contrast, mutations associated with secondary or adverse-risk disease (such as ASXL1 and RUNX1) – as well as TP53 mutations – tend to cluster together and rarely overlap with NPM1 or CEBPA mutations.²⁵ Likewise, c-KIT mutations are typically restricted to core-binding factor AML subsets (eg, cases with t(8;21) or inv(16) translocations) and are known to co-occur in those leukemias.²³ Collectively, the presence or combination of these mutations has strong prognostic and biologic significance.

In this study, we explored the gene mutation spectrum of 165 AML patients, focusing on distinct co-mutation patterns, particularly comparing AML^{NPM1mutFLT3-ITDmutDNMT3Amut} and AML^{NPM1mutFLT3-ITDmutDNMT3Awt}. We compared the clinical characteristics of these two subsets and used bioinformatics to validate their prognoses. And we have focused on DNMT3A in order to determine whether the presence of DNMT3A mutation, alongside NPM1 and FLT3-ITD, defines a particularly high-risk subset. Furthermore, we investigated the underlying mechanisms and found that guanine nucleotide-binding protein subunit γ 4 (GNG4) may be the key to AML^{NPM1mutFLT3-ITDmutDNMT3Amut}.

Materials and Methods

Patient Samples

This retrospective study included 165 patients newly diagnosed with AML between January 2018 and August 2021 at the First Hospital of Lanzhou University. Patients diagnosed with acute promyelocytic leukemia were excluded from the study. All patients were diagnosed based on a comprehensive evaluation including peripheral blood cell count, bone marrow blast percentage, bone marrow cell morphology, cytochemical staining, flow cytometry (FCM) immunophenotyping, chromosomal karyotype analysis, and molecular biological assessment. All diagnoses were made in accordance with the World Health Organization diagnostic criteria.²⁶ Clinical and survival data were collected through electronic medical records and telephone follow-ups. This study adhered to the principles of the Declaration of Helsinki and was approved by the Ethics Committee of the First Hospital of Lanzhou University (Approval No. LDYYLL-2024-803; Date of approval: December 30, 2024). All patients or their family members provided written informed consent for bone marrow aspiration, biopsy, and chemotherapy-related procedures. For this retrospective analysis of fully de-identified data, the requirement for obtaining informed consent was formally waived by the above Ethics Committee (waiver date: 30 Dec 2024). We did not perform a formal prospective power calculation due to the exploratory, retrospective nature. Because AML patients are relatively uncommon, our sample size was determined by the availability of patients meeting these criteria rather than a prospective power calculation.

Cytogenetic Analysis

The karyotypes were described according to the International System for Human Cytogenetic Nomenclature 2009.¹²

Mutation Detection

Mutations were detected using targeted next-generation sequencing and Sanger sequencing. Targeted sequencing of the following 11 genes was performed for all samples: NPM1, FLT3-ITD, DNMT3A, CEBPA double, ASXL1, RUNX1, TET2, IDH1, IDH2, c-KIT, and TP53.Genomic DNA was extracted from bone marrow aspirate samples and used for targeted NGS on an Illumina MiSeq platform to screen for mutations in 11 genes. A custom panel covering 11 recurrently mutated genes in acute myeloid leukemia (AML) – including DNMT3A, FLT3 (internal tandem duplication, FLT3-ITD), NPM1, CEBPA (double mutations), ASXL1, RUNX1, TET2, IDH1, IDH2, c-KIT, and TP53 – was sequenced for all samples. Library preparation and target enrichment were performed using standard protocols (eg, multiplex PCR-based amplification of target regions), and paired-end sequencing was carried out on the Illumina MiSeq platform. Sequence reads were aligned to the human reference genome (GRCh37/hg19), and variants in the target genes were identified using standard bioinformatics pipelines. Key mutations detected by NGS were validated by Sanger sequencing. Specifically, 11 gene mutations identified through NGS were confirmed by polymerase chain reaction (PCR) amplification and direct Sanger sequencing of the relevant gene regions. Gene-specific primers flanking each mutation site were used to amplify the target regions from genomic DNA, and the PCR products were purified for sequencing. Sanger sequencing was performed in both forward and reverse directions using BigDye® Terminator chemistry on an ABI capillary DNA sequencer. The resulting chromatograms were analyzed and compared to reference gene sequences, confirming the presence of the mutations initially observed in the NGS panel.

Response to Treatment

All patients were classified according to the ELN 2017 criteria and received cytotoxic chemotherapy, except those who received supportive therapy only. Chemotherapy responses were assessed using the International Working Group criteria. Overall responses included CR, partial remission (PR), and non-remission (NR). CR was defined as less than 5% bone marrow blasts with no Auer rods or evidence of persistent extramedullary leukemia. PR was defined as a reduction in bone marrow blasts to below 20% after treatment, with a decrease of more than 50% from pretreatment levels. NR was defined as the number of bone marrow blasts remaining >20% after the treatment. OS was defined as the time from initial diagnosis to the date of death or last follow-up.

Bioinformatics Analysis

The data used for bioinformatics analysis in this study were extracted from the Gene Expression Omnibus (GEO) database (GSE146173²⁷), focusing on AML patients harboring NPM1/FLT3-ITD co-mutations. We downloaded the RNA sequencing data from 246 AML samples and found that 27 samples had NPM1/FLT3-ITD co-mutations. Data processing and presentation were performed using R packages (Version 4.0.3). Pan-cancer RNA-seq data in TPM format from the TCGA and GTEx databases, uniformly processed through the Toil pipeline, were downloaded via UCSC Xena (https://xenabrowser.net/datapages/). Patients with missing OS data were excluded. We included patients with complete clinical data and normalized gene expression data (Figure 1). AML patients were divided into subgroups based on the mutation status of NPM1, FLT3-ITD, and DNMT3A. Survival differences among these subgroups were analyzed and compared using Kaplan–Meier survival curves. AML^{NPM1mutFLT3-ITDmut} patients were further stratified into two groups based on their DNMT3A mutation status: the control group AML^{NPM1mutFLT3-ITDmutDNMT3Awt} and the experimental group AML^{NPM1mutFLT3-ITDmutDNMT3Amut}. Differential expression analysis of RNA-seq data between these two groups was performed using the limma package in R. Genes were considered differentially expressed if they met the threshold criteria of |log2(FC)| > 1 and P-value <0.05. Differentially expressed genes (DEGs) were subjected to Gene Ontology (GO) annotation, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis, and protein-protein interaction (PPI) network analysis. Hub genes were defined based on node degree (the number of adjacent nodes in the network), as this provided a straightforward indicator of each gene’s connectivity in the PPI analysis.

Figure 1.

Workflow Diagram

RNA Isolation, cDNA Preparation, and Quantitative PCR (qPCR)

RNA was isolated using the miRNeasy Kit (QIAGEN, Hilden, Germany), and reverse transcription for cDNA was performed utilizing the SuperScript® III First-Strand Synthesis System (#18080-051, Invitrogen, Carlsbad, CA). SYBR Green qPCR (#4309155; Applied Biosystems, Foster City, CA) was used for all genes. The sequences of the primer sets were as follows: GNG4 forward primer, 5′-GAGGGCATGTCTAATAACAGCAC-3′; GNG4 reverse primer, 5′-AGACCTTGACCCTGTCCATAC-3′; GAPDH forward primer, 5′-GTCTCCTCTGACTTCAACAGCG-3′; GAPDH reverse primer, 5′-ACCACCCTGTTGCTGTAGCCAA-3′. GNG4 expression was determined through the comparative Ct method using GAPDH levels for normalization.

Statistical Analyses

This study was conducted and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.²⁸ All patient information was anonymized and de-identified prior to analysis to ensure confidentiality. We included all consecutive patients diagnosed with AML at our institution between January 2018 and August 2021 at the First Hospital of Lanzhou University who met the inclusion criteria. General clinical data, such as peripheral blood count, bone marrow blasts, and age, are expressed as median values and compared using the non-parametric Mann–Whitney U-test. The frequency of gene mutations, proportions of French-American-British (FAB) categories, and remission rates are expressed as percentages and compared using the Pearson chi-square or Fisher’s exact test. Statistical analyses were performed using the statistical package SPSS 25. GO annotation and KEGG pathway analysis were performed employing the clusterProfiler package in R. PPI network analysis was conducted using the STRING online database. The network was further analyzed, and hub genes were selected using the Cytoscape software. R package (version 4.0.3) and Microsoft Office Excel were used to draw the figures.

Results

Patients Characteristics

We retrospectively analyzed the clinical and survival data of 165 patients (Figure 1) with newly diagnosed AML and found that 141 (85.5%) had de novo AML, while 24 patients (14.5%) had secondary AML, primarily owing to myelodysplastic syndrome and chronic myelomonocytic leukemia.

The baseline patient characteristics are listed in Table 1. The demographic characteristics of the 165 AML patients are as follows: 86 (52.1%) were male and 79 (47.9%) were female, with a median age of 53 years (range: 3-84 years). In the first peripheral blood count taken upon hospital admission, the median hemoglobin level was 73.0 g/L (minimum 27 g/L), the median white blood cell (WBC) count was 12.1 × 10⁹/L (maximum 392.5 ×10⁹/L), and the median platelet count was 28.0 × 10⁹/L (minimum 1× 10⁹/L). Bone marrow blasts ranged from 20.0% to-98.0%, with a median percentage of 62.0%. The FAB category had a high percentage of patients in M5, M4, and M2 and only one patient each in M0, M1, and M6. The karyotype cytogenetic profile of 107 (64.9%) AML patients suggested an intermediate prognosis, of whom 71 had normal karyotype AML. Combining karyotype cytogenetics and gene mutations with the latest National Comprehensive Cancer Network guideline risk stratification by genetics, we identified 53 favorable patients (32.2%), 56 intermediate patients (33.9%), and 56 poor patients (33.9%), noting that patients with an unknown karyotype and no detected gene mutations were categorized into the intermediate risk stratification group.

Table 1.

Patient Characteristics

Characteristics	Value
Age, years
Median	53
Range	3-84
Sex, n (%)
Male	86 (52.1)
Female	79 (47.9)
Hemoglobin, g/L
Median	73.0
Range	27-153
Platelet count, ×10⁹/L
Median	28.0
Range	1-283
WBC count, ×10⁹/L
Median	12.1
Range	0.22-392.5
Bone marrow blasts, %
Median	62.0
Range	20.0-98.0
Origin of disease, n (%)
de Novo AML	141 (85.5)
s-AML	24 (14.5)
FAB category, n (%)
M0	1 (0.6)
M1	1 (0.6)
M2	49 (29.7)
M4	51 (30.9)
M5	60 (36.4)
M6	1 (0.6)
Undetermined	2 (1.2)
Karyotype, n (%)
Favorable	15 (9.1)
Intermediate	107 (64.9)
Poor Unknown	20 (12.1) 23 (13.9)
Risk stratification, n (%)
Favorable	53 (32.2)
Intermediate	56 (33.9)
Poor	56 (33.9)
Patients receiving different therapy, n (%)
Conventional chemotherapy alone	86 (52.1)
HMAs alone	11 (6.7)
Conventional chemo & HMAs	26 (15.8)
Venetoclax & HMAs	16 (9.7)
Other	3 (1.8)
No chemotherapy	23 (13.9)
Response to chemotherapy (n = 115) , n (%)
1 cycle CR	73 (63.5)
1 cycle PR	16 (13.9)
1 cycle NR	26 (22.6)
MRD (1 cycle CR)(n = 73) , n (%)
negative	33 (45.2)
positive	12 (16.4)
unknown	28 (38.4)

Abbreviations: WBC white blood cell, s-AML secondary AML, FAB French-American British, HMAs hypomethylating agents, CR complete remission, PR partial remission, NR non-remission, MRD minimal residual disease.

The 165 patients received various therapies. Of these, 23 (13.9%) were discharged with supportive therapy only because of factors such as poor performance status, advanced age, and/or limited family economic resources. Among the remaining 142 patients, 86 (52.1%) received conventional chemotherapy alone; 11 (6.7%), hypomethylating agents (HMAs) alone; 26 (15.8%), conventional chemo- and HMAs; 16 (9.7%), venetoclax and HMAs; and three (1.8%), other therapies. After the first cycle of treatment (n = 115), 73 (63.5%) patients achieved CR, with 45.2% of them being minimal residual disease-negative. Additionally, 16 (13.9%) patients achieved PR, and 26 (12.6%) patients were evaluated as having NR.

Mutational Landscape: NPM1, FLT3-ITD, and DNMT3A Were the Most Frequent Mutated Genes

The mutational landscape of 165 patients with newly diagnosed AML is shown in Figure 2. The number of coexisting gene mutations in each patient ranged from 0 to 4. Mutations in NPM1, FLT3-ITD, and DNMT3A were the most frequent. NPM1 mutations were observed in 37 (22%) patients; FLT3-ITD mutations, in 31 (19%); and DNMT3A mutations, 31 (19%) patients. Other gene mutations were identified in TET2 (19%), IDH2 (16%), ASXL1 (16%), CEBPA double (15%), RUNX1 (13%), IDH1 (9%), c-KIT (8%), TP53 (8%), CEBPA other (7%), and FLT3 other (7%).

Figure 2.

Diagram of Gene Mutations Landscape of 165 Newly Diagnosed Non-APL AML Patients. Age, Sex, FAB Category, Risk Stratification, Gene Mutation Frequency/Number, Gene Mutation Status, Coexisting Gene Mutations, and Co-mutation Numbers in AML Patients at the Initial Diagnosis. Abbreviations: CEBPA Double: Biallelic CEBPA Mutation; CEBPA Other: No Biallelic CEBPA Mutations; FLT3 Other: No ITD FLT3 Mutations

The NPM1-W288 hotspot mutation region was detected in 89% (33/37) of NPM1-mutated AML cases. We explored the FLT3 allelic ratio (AR) value in 31 FLT3-ITD-mutated AML cases and found that 29% (9/31) of patients had high AR (≥0.5), while 45% (14/31) had low AR (<0.5); AR was unknown in 26% (8/31) of patients. DNMT3A-R882 mutations were detected in 55% of DNMT3A-mutated AML cases (17/31), most of which were identified as R882H (82%, 14/17), followed by R882 C (18%, 3/17). The co-mutation patterns of NPM1, FLT3-ITD, and DNMT3A in 165 AML patients are shown in Table 2. The frequencies of AML^{NPM1mutFLT3-ITDmut} was 11.5% (19/165); AML^{NPM1mutDNMT3Amut}, 12% (20/165); and AML^{FLT3-ITDmutDNMT3Amut}, 7% (12/165). The frequency of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} was 6.7% (11/165), which was unexpectedly high (Table 2). We confirmed the co-mutation patterns of the 11 genes using Spearman’s correlation. The correlation between coexisting gene mutations is shown in Figure 3. Mutations in NPM1, FLT3-ITD, and DNMT3A were positively correlated.

Table 2.

Co-Existing Pattern of Gene Mutations of NPM1, FLT3-ITD and DNMT3A

Group	n (%)
NPM1+	37/165 (22 %)
FLT3-ITD+	31/165 (19%)
DNMT3A+	31/165 (19%)
NPM1+/FLT3-ITD+	19/165 (11.5%)
NPM1+/DNMT3A+	20/165 (12%)
FLT3-ITD+/DNMT3A+	12/165 (7%)
NPM1+/FLT3-ITD+/DNMT3A+	11/165 (6.7%)

Figure 3.

Co-Mutation Patterns of 165 Newly Diagnosed Non-APL AML Patients. Red and Blue Indicate Positive and Negative Correlations, Respectively, and the Depth Signifies the Correlation Strength

AML^{NPM1mutFLT3-ITDmutDNMT3Amut} Patients had Reduced OS and Higher WBC Count and Were all in the M4/M5 FAB Category

Baseline characteristics, gene mutations, and karyotypes of the 11 AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients are summarized in Table 3. Among these patients, 72.7% (8/11) were female and 27.3% (3/11) were male, with the median age was 59 years. All 11 AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients were classified as M4 (18%, 2/11) or M5 (82%, 9/11), according to the FAB classification. The median WBC count was 67.45 × 10⁹/L, and the WBC count of 55% (6/11) of these AML cases was over 50 × 10⁹/L. The 11 AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients’ treatment history and efficacy assessment are summarized in Figure 4. The follow-up cutoff date was late August 2021. Only one patient (No. 11) received supportive therapy. The median OS of these patients was only 4 months, and only Patient No. 5 had a survival time of over 6 months. This suggested that AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients generally exhibit a reduced OS.

Table 3.

The NPM1/FLT3-ITD/DNMT3A Triple-Mutated AML Patient Characteristics

Case No./Sex/Age	FAB	WBC	HB	PLT	BM	Karyotype	Gene mutation
1/Female/60	M5	76.03	57	47	82	46,XX	NPM1(W288),DNMT3A(R882H),FLT3-ITD(High),JAK2
2/Male/59	M4	37.14	55	22	75.5	46,XY	NPM1(W288),DNMT3A(R882 C),FLT3-ITD(Low),FLT3-TKD
3/Female /39	M5	86.12	55	120	74.5	46,XX	NPM1(W288),DNMT3A(R882H),FLT3-ITD(High)
4/Female/35	M5	6.34	70	35	62.5	46,XX	NPM1(W288),DNMT3A(R882H),FLT3-ITD(High)
5/Female/49	M5	36.03	114	122	54.5	Unknown	NPM1(W288),DNMT3A(R882H),FLT3-ITD(Low),NRAS
6/Female/70	M4	129.11	94	23	80	46,X,-X,+mar	NPM1(W288),DNMT3A(non-R882),FLT3-ITD(High),TET2
7/Male/45	M5	254.86	88	103	98	46,XY	NPM1(W288),DNMT3A(non-R882),FLT3-ITD(High),CEBPA
8/Female/72	M5	40.14	121	55	73	46,XX	NPM1(W288),DNMT3A(R882H),FLT3-ITD(High),ATM,TET2
9/Female/73	M5	67.45	84	96	65.5	46,XX	NPM1(W288),DNMT3A(R882H),FLT3-ITD(Low),LYST,TET2
10/Male/41	M5	34.98	71	93	60	46,XY	NPM1(W288),DNMT3A(R882H),FLT3-ITD(Low),TET2
11/Female/77	M5	134.5	84	44	80	Unknown	NPM1(W288),DNMT3A(R882H),FLT3-ITD(Unknown)

Abbreviations: FAB French-American British, WBC white blood cell count ×10⁹/L, HB hemoglobin g/L, PLT platelet count ×10⁹/L, BM bone marrow blasts %, High: AR ≥0.5, Low:<0.5.

Figure 4.

Swimmer Plots of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} Patients. Abbreviations: HMAs: Hypomethylating agents, CR: complete remission, PR: partial remission, NR: non-Remission, Ambiguous: Bone Marrow Morphology is Hypoplastic

Compared to non-triple-mutated patients, AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients had a higher WBC count (67.45 vs 10.70 × 10⁹/L; P < 0.001), higher bone marrow blasts (74.5% vs 60.5%; P = 0.037), and higher platelet count (55 vs 27 × 10⁹/L; P = 0.007) and were all in the M4/M5 FAB category (100.0% vs 64.9%; P = 0.017). Although it was not statistically significant between two groups, we observed a trend for higher proportions of advanced age (≥70 years), female sex, and normal karyotype in AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients (P = 0.199, P = 0.088, and P = 0.062). There were no significant differences between the two groups in terms of disease origin, hemoglobin level, and CR rate after the first cycle of treatment. (Table 4).

Table 4.

Patient Characteristics Comparison According to the NPM1/FLT3-ITD/DNMT3A Mutations

Characteristics	Triple-mutated n (%)	Non-triple mutated n (%)	P Value
Age, years
≥70	4/11 (36.4%)	25/154 (16.2 %)	0.199
<70	7/11 (63.6%)	129/154 (83.8%)
Median(Range)	59.0 (35-77)	53.0 (3-84)	0.511
Gender
Male	3/11 (27.3%)	83/154 (53.9%)	0.088
Female	8/11 (72.7%)	71/154 (46.1%)
Origin of disease
de Novo AML	11/11 (100%)	130/154 (84.4%)	0.369
s-AML	0/11 (0%)	24/154 (15.6%)
Hemoglobin, g/L
Median(Range)	84 (55-121)	73 (27-153)	0.425
Platelet count, ×10⁹/L
Median(Range)	55 (22-122)	27 (1-283)	0.007*
WBC count, ×10⁹/L
≥20	10/11 (90.9%)	52/154 (33.8%)	<0.001*
<20	1/11 (9.1%)	102/154 (66.2%)
Median(Range)	67.45 (6.34-254.86)	10.70 (0.22-392.5)	<0.001*
Bone marrow blasts, %
Median(Range)	74.5 (54.5-98.0)	60.5 (20-96)	0.037*
FAB category
M4/M5	11/11 (100%)	100/154 (64.9%)	0.017*
Karyotype
Normal	8/11 (72.7%)	65/154 (42.2%)	0.062
NPM1+	-	26/154 (16.9%)	-
FLT3-ITD+	-	20/154 (13.0%)	-
DNMT3A+	-	20/154 (13.0%)	-
1 cycle CR (n = 115)	4/7 (57.1%)	69/108 (63.9%)	0.705

*P < 0.05. p-values have been presented in bold to indicate statistical significance (p < 0.05).

The Hub Gene of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} –– GNG4

As illustrated in the workflow (Figure 1), in addition to analyzing the AML clinical data obtained from our hospital, our study extended the examination of the prognostic outcomes of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients identified in the GEO database.

To clarify the prognostic impact of DNMT3A on AML^{NPM1mutFLT3-ITDmut} patients, we divided the 27 AML^{NPM1mutFLT3-ITDmut} patients from GSE146173 into two groups based on DNMT3A status: wild-type or mutated. A comparison of survival analysis between the two groups is shown in Figure 5C. We found that AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients had a significantly lower OS than AML^{NPM1mutFLT3-ITDmutDNMT3Awt} patients (P < 0.05). DEGs between AML^{NPM1mutFLT3-ITDmutDNMT3Amut} and AML^{NPM1mutFLT3-ITDmutDNMT3Awt} patients were visualized using a heatmap (Figure 6A). In total, 480 DEGs were identified, including 196 upregulated and 284 downregulated genes. The results of the biofunctional analyses, which included GO and KEGG pathway enrichment analyses, are shown in Figure 5D–I. We found that the DEGs played important roles in (1) biological processes: positive regulation of cell adhesion, cell-cell junction organization, granulocyte migration, and granulocyte chemotaxis; (2) cell composition: extrinsic components of the membrane, collagen-containing extracellular matrix, and extrinsic components of the plasma membrane; and (3) molecular function: receptor ligand activity, signaling receptor activator activity, cytokine activity, and cytokine receptor binding. The main signaling pathways involved were cytokine-cytokine receptor interactions, osteoclast differentiation, and human T-cell leukemia virus 1 infection. The PPI network was obtained from the STRING database using the Cytoscape software (version 3.8.0) (Figure 6B). The number of adjacent nodes is shown in Figure 6C. Notably, we identified hub genes using a degree-centrality approach; by this measure, GNG4 showed the highest connectivity with 13 adjacent nodes, marking it as the top hub gene, although we acknowledge that alternative centrality metrics (eg, maximal clique centrality) could also be applied for hub selection.

Figure 5.

Bioinformatics Analysis of GEO data: GSE146173 Dataset. a-c. Kaplan–Meier Survival Curves for the Patients From Different Groups. d-f. Results of the GO Analyses—Bar Plot, Bubble, and Circus. g-i. Results of the KEGG Analyses—Bar Plot, Bubble, and Circus

Figure 6.

Bioinformatics Analysis of GEO data: GSE146173 Dataset a. Heat Map of the Differentially Expressed Genes (DEGs) Between the NPM1 + FLT3-ITD + DNMT3A + AML (Red) Patients and the NPM1 + FLT3-ITD + DNMT3A-(Blue) Patients. (B) Protein-Protein Interaction of DEGs Using STRING. (C) Number of Adjacent Nodes of the DEGs. (D) Kaplan–Meier Survival Curves for the LAML Dataset Between High GNG4 Expression Group and Low GNG4 Expression Group. (E) Kaplan–Meier Survival Curves for the AML^FLT3-ITDmut Patients Between High GNG4 Expression Group and Low GNG4 Expression Group. (F) Validation of the GNG4 Expression Through qPCR Using Human Bone Marrow Samples From AML^{NPM1mutFLT3-ITDmutDNMT3Amut} and AML^{NPM1mutFLT3-ITDmutDNMT3Awt} Patients. Data are Presented as Mean ± Standard Deviation (SD), n = 10. **P < 0.01, vs the AML^{NPM1mutFLT3-ITDmutDNMT3Amut} Group

We found that GNG4, a hub gene, had 13 adjacent nodes. Based on the expression level of GNG4 (with the top 50% expression serving as the threshold), AML patients were categorized into two groups: a high GNG4 expression group and a low GNG4 expression group. Using the TCGA database LAML dataset, we extracted OS data to conduct survival analysis for these two groups. Across the entire AML cohort, patients in the high GNG4 expression group exhibited longer OS (P = 0.027) (Figure 6D). Further stratification of AML patients revealed that among AML^FLT3-ITDmut patients, the high GNG4 expression group showed slightly improved OS compared to the low expression group, although the difference was not statistically significant (P = 0.3) (Figure 6E). Using the TCGA, TIMER and TIDE database, we analyzed the GNG4 expression in pan-cancer and immune infiltration. We explored the differential expression levels of GNG4 in AML and analyzed the correlation between GNG4 expression levels and clinical features (Supplemental Figure S1).

Low GNG4 Expression may Contribute to the Low Survival Rate of AML ^{NPM1mutFLT3-ITDmutDNMT3Amut} Patients

To investigate the role and clinical significance of GNG4, identified as a potential distinguishing factor between AML^{NPM1mutFLT3-ITDmutDNMT3Amut} and AML^{NPM1mutFLT3-ITDmutDNMT3Awt} patients in our previous bioinformatics analysis, we conducted qPCR experiments to further validate the expression of GNG4 using the bone marrow samples from AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients (n = 4) or AML^{NPM1mutFLT3-ITDmutDNMT3Awt} patients (n = 6). All samples were obtained from patients who were newly diagnosed with AML at the First Hospital of Lanzhou University. RNA was extracted from these samples, and PCR was performed to quantify the transcript levels of GNG4.

Our results indicated that the transcript level of GNG4 in AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients was significantly lower than that in AML^{NPM1mutFLT3-ITDmutDNMT3Awt} patients (P = 0.0095) (Figure 6F). This finding is consistent with our bioinformatic predictions, suggesting that GNG4 expression may serve as a critical molecular marker for differentiating between these two patient groups. The lower expression of GNG4 in AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients underscores its potential role in the pathogenesis and progression of AML, warranting further investigation into its clinical implications.

Discussion

This study is among the first to investigate the hub genes of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients. We identified 11 AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients from a cohort of 165 newly diagnosed AML patients, yielding a mutation frequency of 6.7%, which was consistent with previous reports.^18,20 These AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients exhibited distinct clinical features, including elevated WBC counts, higher platelet counts, increased BM blasts, and predominantly monocytic leukemia (an increased number of monocytes in the bone marrow and peripheral blood),²⁹ particularly within the M4/M5 subtypes of the FAB classification system.^9,20,30 These findings align with earlier studies and provide further insight into the clinical presentation of this subtype of AML.

Additionally, there is a difference in age and sex distribution between triple-mutated and non-triple-mutated AML.^9,20,31 We found a notable trend toward a higher incidence of triple mutations in patients aged >70 years and women, although these differences were not statistically significant. This study identified a novel characteristic—the platelet count in patients with triple mutations at the time of initial diagnosis was significantly higher than that in non-triple mutated AML patients. This suggests that triple mutated patients may not present with bleeding symptoms caused by thrombocytopenia in the early stages of diagnosis. Based on the results of the chromosomal karyotype analysis, the majority of the 11 triple-mutated patients included in the study exhibited a normal karyotype. Only one patient presented with a chromosomal deletion, while two patients showed no metaphase cells available for analysis. A previous study showed that NPM1 mutations are strongly associated with normal karyotype AML and are commonly accompanied by DNMT3A and FLT3-ITD mutations.³²

From a cytomorphological perspective, nearly half of the triple-mutated patients (Patient No. 5, 8, 9, 10, and 11) displayed bone marrow characteristics consistent with predominantly monocytic leukemia. The monocyte lineage showed abnormal proliferation, comprising mature monocytes as well as primitive and immature monocytes. The proportion of abnormal cell populations ranged from 65% to 85%. Although AML^NPM1mut has a wide range of morphologies that are not directly related to FAB classification, numerous studies have demonstrated a strong association between NPM1 mutations and both acute myelomonocytic and acute monocytic leukemia.^33,34 FCM results have led to the diagnosis of acute monocytic leukemia (AML-M5). For example, in Patient No. 5, immune cell phenotyping indicated that abnormal cells comprised approximately 80% of the bone marrow nucleated cells. These cells predominantly expressed markers, such as HLA-DR, CD4, CD11b, CD13, CD14, CD15, CD33, CD38, CD56, CD58, CD64, and CD123. Among these, about 24.5% were CD64^bri+CD14⁺ mature monocytes, while approximately 54.5% were CD64^bri+CD14^- naïve monocytes. Similar immune cell profiles were observed in Patient Nos. 8, 9, 10, and 11. These findings suggested that certain cases of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} exhibit distinct monocyte characteristics.

All AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients showed rapid disease progression, a high tendency for recurrence, and a poor overall prognosis. Previous studies have indicated that AML patients harboring both NPM1 and FLT3-ITD mutations have a poor prognosis, especially in cases with a high FLT3-ITD AR. Additional research has confirmed the negative prognostic impact of DNMT3A mutations on AML^{NPM1mutFLT3-ITDmut} patients, suggesting that DNMT3A further exacerbates poor outcomes in this subset of AML. According to previous research findings, AML patients with NPM1/FLT3-ITD/DNMT3A triple mutations have shorter median OS and disease-free survival (DFS) than those with only FLT3-ITD mutation.⁹ Further analysis showed no significant difference in the CR rate between patients with triple mutations and those without. However, the relapse rate was significantly higher in the triple mutated group, which may explain the poorer prognosis observed in these patients.²⁰ Allo-HSCT was found to improve OS and DFS in both groups (triple-mutated and non-triple-mutated AML patients). However, in triple-mutated AML, the benefit was less pronounced, with the 1-year OS rate remaining below 30%.³⁰

To understand the reasons behind the poor prognosis of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} patients and to explore the underlying mechanisms, we used the GEO dataset GSE146173 to identify specific genes associated with poor outcomes. Our analysis identified the GNG4 gene as playing a significant role in the prognosis of triple-mutated AML patients. GPCR, comprising α, β, and γ subunits, responds to various extracellular stimuli, such as hormones, growth factors, and sensory stimulating signals. GNG4 is one of 14 γ subunit proteins of GPCR.³⁵ GNG4 is important in MHC-I expression and immunotherapy resistance. Previous studies have shown that GNG4 is involved in PI3K/protein kinase B (Akt) signaling pathway and NF-κB signaling pathways.^36-40 In addition, NF-κB signaling pathways are reportedly related to MHC-I expression.^41-45 Kawase et al. found that MHC-I expression was reduced by GNG4 knockdown via the NF-κB signaling pathway, and GNG4 knockdown induced resistance to PD-1 blockade.⁴⁶

In addition, Pal et al reported that the promoter region of GNG4 was significantly hypermethylated and that its transcript level was significantly downregulated in glioblastoma, renal cell carcinoma, and thymic carcinoma.^36,47 It functions as a tumor suppressor gene. Relapse-free survival was significantly worse in tumors with higher GNG4 DNA methylation levels than in those with lower DNA methylation levels in all thymic epithelia tumors.⁴⁷ Indeed, it has been proposed that malignancies can hijack central neuroendocrine homeostatic mechanisms to reset systemic physiology in a manner that favors tumor growth at the host’s expense.⁴⁸

A systems biology- and machine learning-based study revealed the potential benefits of midostaurin in FLT3-mutated AML. GNG4 has been identified as a protein that is differentially modulated in response to midostaurin treatment in FLT3-mutated AML, specifically through midostaurin’s non-FLT3 target effects. The upregulated GNG4 protein was related to a beneficial effect from the treatment, suggesting that it may play a role in enhancing the drug’s therapeutic efficacy by impacting signaling pathways beyond FLT3 inhibition.⁴⁹ Thus, GNG4 expression is consistently downregulated across various solid tumors and in AML. Its promoter is often highly methylated in these tumors, contributing to reduced expression levels. Moreover, hypermethylation of GNG4 frequently compromises relapse-free survival in cancer patients.⁴⁷

Consistent with this, our findings confirmed that GNG4 was downregulated in AML^{NPM1mutFLT3-ITDmutDNMT3Amut} compared to AML^{NPM1mutFLT3-ITDmutDNMT3Awt} and survival analysis showed that the low-expression GNG4 group had lower OS in AML^FLT3-ITDmut, a finding that aligns with the survival analysis results of Reina et al.⁴⁷ Our PCR experiments confirmed that GNG4 is a significant prognostic factor distinguishing AML^{NPM1mutFLT3-ITDmutDNMT3Amut} from AML^{NPM1mutFLT3-ITDmutDNMT3Awt}.

Emerging evidence suggests that expression patterns of certain hub genes in leukemia can reflect key clinical aspects, including a propensity for central nervous system involvement,⁵⁰ differential responses to therapy,⁵¹ and underlying developmental lineage biases.⁵² In this context, GNG4 expression may hold promise as a diagnostic and prognostic biomarker for AML patients harboring NPM1/FLT3-ITD/DNMT3A triple mutations. However, we emphasize that this potential utility of GNG4 remains to be validated, and further studies are needed to determine its clinical significance.

Patients with triple mutations showed significantly lower transcript levels of GNG4, which is consistent with previous studies and our bioinformatics predictions. This suggests that lower expression of GNG4 may contribute to the pathogenesis and progression of AML. These findings highlight the clinical value of GNG4 expression for diagnosing and potentially guiding the treatment of AML, warranting further investigation into its therapeutic implications.

This study must be interpreted in light of several limitations. First, it was a retrospective observational analysis with a relatively small sample size – especially in the triple‐mutated subgroup (only 11 patients) – which reduces statistical power and limits the generalizability of our findings. Second, our identification of hub genes in the PPI network was based on node degree (adjacent node count) alone, a relatively simplistic criterion; using more robust centrality metrics (eg, betweenness or eigenvector centrality) might have revealed a different or more nuanced set of hub genes. Third, due to the limited size of our patient cohort and the low number of events, we assessed the prognostic value of GNG4 using ROC analysis in an external dataset rather than in our own patient data. Another limitation of our study is the use of qRT-PCR for gene expression analysis, which cannot capture the full complexity of transcriptional regulation. Fifth, due to the low number of NPM1/FLT3-ITD/DNMT3A triple-mutated cases in our cohort, we relied on external cohorts to assess GNG4 expression and its prognostic impact. Therefore, any conclusions regarding GNG4’s specificity as a marker for this subgroup should be interpreted with caution, and further validation in larger patient cohorts is warranted. Relying on external data provided an independent perspective but also introduces potential biases and differences in patient characteristics, which may limit the direct applicability of those results to our population. Taken together, these limitations mean that our conclusions are tentative and should not be overstated. Further validation in larger, prospective cohorts and additional functional studies are warranted to confirm and expand upon our findings.

Conclusion

In conclusion, our study confirmed a high frequency of NPM1/FLT3-ITD/DNMT3A triple mutations (approximately 6.7%). Patients with triple-mutated AML consistently had higher WBC counts, platelet counts, and bone marrow blast percentages, and all fall within the M4/M5 FAB category. Most importantly, they had a high relapse rate and shorter OS, although there was no significant difference in the CR rate. Using bioinformatics analysis, we found that poor prognosis may be associated with low expression of GNG4, and further research may explore the mechanisms linking GNG4 to the prognostic significance of AML. However, the main limitation of this study was the small sample size of triple-mutated AML patients. Further studies with a larger case series are warranted.

Supplemental Material

Supplemental Material - Clinical Characteristics and Outcomes of Acute Myeloid Leukemia Patients Harboring NPM1/FLT3-ITD/DNMT3A Triple Mutations and the Potential Prognostic Value of GNG4

Supplemental Material for Clinical Characteristics and Outcomes of Acute Myeloid Leukemia Patients Harboring NPM1/FLT3-ITD/DNMT3A Triple Mutations and the Potential Prognostic Value of GNG4 by Yujie Niu, MD, Xingchun Luo, MD, Xiaoxiao Yang, MD, Yuancheng Guo, Xiao Tang, Long Zhao, Jinli Jian, Bei Liu in Cancer Control.

Footnotes

Author Note

The authors declare that this study was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.

Acknowledgments

We thank the patients and investigators who submitted data to the GEO database for providing the high-quality data analyzed in this study. We also express our gratitude to the patients who provided clinical information and to the medical testing institutions for their support.

The graphical abstract and were created using BioRender.com.

ORCID iD

Yujie Niu

Ethical Considerations

The study protocol was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Lanzhou University (Approval No. LDYYLL-2024-803) and was conducted in accordance with the Declaration of Helsinki.

Author Contributions

Yujie Niu: Conceptualization, Investigation, Data curation, Visualization, Writing - original draft, Writing - review & editing. Xingchun Luo: Investigation, Methodology, Formal analysis, Validation, Writing - original draft. Xiaoxiao Yang: Resources, Investigation, Formal analysis, Visualization, Validation, Writing - original draft. Yuancheng Guo: Investigation, Methodology, Validation, Project administration, Writing - review & editing. Xiao Tang: Formal analysis, Methodology, Software, Writing - review & editing. Long Zhao: Resources, Visualization, Writing - review & editing. Jinli Jian: Data curation, Writing - review & editing. Bei Liu: Conceptualization, Investigation, Funding acquisition, Supervision, Writing - original draft, Writing - review & editing.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by The First Hospital of Lanzhou University under Grant number ldyyyn2023-12; The First Hospital of Lanzhou University under Grant number ldyyyn2018-64; The Natural Science Foundation of Gansu province under Grant number 18JR3RA356.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data supporting the bioinformatic analysis findings of this study are openly available in the Gene Expression Omnibus database ().

Supplemental Material

Supplemental material for this article is available online.

References

Burd

Levine

Ruppert

, et al. Precision medicine treatment in acute myeloid leukemia using prospective genomic profiling: feasibility and preliminary efficacy of the Beat AML Master Trial. Nat Med. 2020;26(12):1852-1858. doi:10.1038/s41591-020-1089-8

Tyner

Tognon

Bottomly

, et al. Functional genomic landscape of acute myeloid leukaemia. Nature. 2018;562(7728):526-531. doi:10.1038/s41586-018-0623-z

Papaemmanuil

Gerstung

Bullinger

, et al. Genomic classification and prognosis in acute myeloid leukemia. N Engl J Med. 2016;374(23):2209-2221. doi:10.1056/NEJMoa1516192

Eisfeld

Kohlschmidt

Mims

, et al. Additional gene mutations may refine the 2017 European LeukemiaNet classification in adult patients with de novo acute myeloid leukemia aged <60 years. Leukemia. 2020;34(12):3215-3227. doi:10.1038/s41375-020-0872-3

Sasaki

Kanagal-Shamanna

Montalban-Bravo

, et al. Impact of the variant allele frequency of ASXL1, DNMT3A, JAK2, TET2, TP53, and NPM1 on the outcomes of patients with newly diagnosed acute myeloid leukemia. Cancer. 2020;126(4):765-774. doi:10.1002/cncr.32566

Shafik

Darwish

Allam

Elsayed

. FLT3-ITD allele frequency is an independent prognostic factor for poor outcome in FLT3-ITD-positive AML patients. Clin Lymphoma Myeloma Leuk. 2021;21(10):676-685. doi:10.1016/j.clml.2021.05.005

Herold

Rothenberg-Thurley

Grunwald

, et al. Validation and refinement of the revised 2017 European LeukemiaNet genetic risk stratification of acute myeloid leukemia. Leukemia. 2020;34(12):3161-3172. doi:10.1038/s41375-020-0806-0

Ley

Miller

Ding

, et al. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med. 2013;368(22):2059-2074. doi:10.1056/NEJMoa1301689

Loghavi

Zuo

Ravandi

, et al. Clinical features of de novo acute myeloid leukemia with concurrent DNMT3A, FLT3 and NPM1 mutations. J Hematol Oncol. 2014;7:74. doi:10.1186/s13045-014-0074-4

10.

Patel

Kuo

Gibson

, et al. High NPM1-mutant allele burden at diagnosis predicts unfavorable outcomes in de novo AML. Blood. 2018;131(25):2816-2825. doi:10.1182/blood-2018-01-828467

11.

Döhner

Estey

Grimwade

, et al. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood. 2017;129(4):424-447. doi:10.1182/blood-2016-08-733196

12.

Chen

Zeng

, et al. Clinical features and prognostic significance of DNMT3A, FLT3, and NPM1 mutations in de novo acute myeloid leukemia patients. Int J Lab Hematol. 2023;45(6):899-907. doi:10.1111/ijlh.14133

13.

Garg

Reyes-Palomares

, et al. Hepatic leukemia factor is a novel leukemic stem cell regulator in DNMT3A, NPM1, and FLT3-ITD triple-mutated AML. Blood. 2019;134(3):263-276. doi:10.1182/blood.2018862383

14.

Ali

Salih

. Molecular and clinical significance of FLT3, NPM1, DNMT3A and TP53 mutations in acute myeloid leukemia patients. Mol Biol Rep. 2023;50(10):8035-8048. doi:10.1007/s11033-023-08680-2

15.

Tang

Shen

Mao

, et al. FLT3-ITD with DNMT3A R882 double mutation is a poor prognostic factor in Chinese patients with acute myeloid leukemia after chemotherapy or allogeneic hematopoietic stem cell transplantation. Int J Hematol. 2017;106(4):552-561. doi:10.1007/s12185-017-2256-7

16.

DiNardo

Lachowiez

. Acute myeloid leukemia: from mutation profiling to treatment decisions. Curr Hematol Malig Rep. 2019;14(5):386-394. doi:10.1007/s11899-019-00535-7

17.

Martinez-Verbo

Estrada

Cabezón

, et al. Mutational profile and relative telomere length in Chronic Myelomonocytic Leukemia subgroups according to the 2016 World Health Organization classification. Leuk Res. 2021;111:106726. doi:10.1016/j.leukres.2021.106726

18.

. The impact of DNMT3A/FLT3-ITD/NPM1 on patients with acute myeloid leukemia after allogeneic hematopoietic stem cell transplantation. Turk J Haematol. 2019;36(1):64-66. doi:10.4274/tjh.galenos.2018.2018.0274

19.

Guryanova

Shank

Spitzer

, et al. DNMT3A mutations promote anthracycline resistance in acute myeloid leukemia via impaired nucleosome remodeling. Nat Med. 2016;22(12):1488-1495. doi:10.1038/nm.4210

20.

Bezerra

Lima

Piqué-Borràs

, et al. Co-occurrence of DNMT3A, NPM1, FLT3 mutations identifies a subset of acute myeloid leukemia with adverse prognosis. Blood. 2020;135(11):870-875. doi:10.1182/blood.2019003339

21.

Song

Yang

Lin

Cai

Zheng

Chen

. Molecular heterogeneity of guanine nucleotide binding-protein γ subunit 4 in left- and right-sided colon cancer. Oncol Lett. 2020;20(6):334. doi:10.3892/ol.2020.12197

22.

Ebrahimi

Abdulwahid

A-HRR

Mansouri

, et al. Targeting the NF-κB pathway as a potential regulator of immune checkpoints in cancer immunotherapy. Cell Mol Life Sci. 2024;81(1):106. doi:10.1007/s00018-023-05098-8

23.

Peker

. Navigating through mutations in acute myeloid leukemia. What Do We Know and What Do We Do with It? Erciyes Tıp Dergisi/Erciyes Medical Journal. 2018;40:183-187. doi:10.5152/etd.2018.18136

24.

Kayser

Levis

. The clinical impact of the molecular landscape of acute myeloid leukemia. Haematologica. 2023;108(2):308-320. doi:10.3324/haematol.2022.280801

25.

Lachowiez

Long

Saultz

, et al. Comparison and validation of the 2022 European LeukemiaNet guidelines in acute myeloid leukemia. Blood Adv. 2023;7(9):1899-1909. doi:10.1182/bloodadvances.2022009010

26.

Peng

Zuo

, et al. Chronic myelomonocytic leukemia with nucleophosmin (NPM1) mutation. Eur J Haematol. 2016;96(1):65-71. doi:10.1111/ejh.12549

27.

Bamopoulos

Batcha

AMN

Jurinovic

, et al. Clinical presentation and differential splicing of SRSF2, U2AF1 and SF3B1 mutations in patients with acute myeloid leukemia. Leukemia. 2020;34(10):2621-2634. doi:10.1038/s41375-020-0839-4. https://www.ncbi.nlm.nih.gov/geo/

28.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147(8):573-577. doi:10.7326/0003-4819-147-8-200710160-00010

29.

Matarraz

Almeida

Flores-Montero

, et al. Introduction to the diagnosis and classification of monocytic-lineage leukemias by flow cytometry. Cytometry B Clin Cytom. 2017;92(3):218-227. doi:10.1002/cyto.b.21219

30.

Huang

, et al. Acute myeloid leukemia patient with FLT3-ITD and NPM1 double mutation should undergo allogeneic hematopoietic stem cell transplantation in CR1 for better prognosis. Cancer Manag Res. 2019;11:4129-4142. doi:10.2147/cmar.S194523

31.

Shen

Zhu

Fan

, et al. Gene mutation patterns and their prognostic impact in a cohort of 1185 patients with acute myeloid leukemia. Blood. 2011;118(20):5593-5603. doi:10.1182/blood-2011-03-343988

32.

Xing

Wang

Gao

, et al. Cytogenetics and associated mutation profile in patients with acute monocytic leukemia. Int J Lab Hematol. 2019;41(4):485-492. doi:10.1111/ijlh.13030

33.

Falini

Mecucci

Tiacci

, et al. Cytoplasmic nucleophosmin in acute myelogenous leukemia with a normal karyotype. N Engl J Med. 2005;352(3):254-266. doi:10.1056/NEJMoa041974

34.

Falini

Nicoletti

Martelli

Mecucci

. Acute myeloid leukemia carrying cytoplasmic/mutated nucleophosmin (NPMc+ AML): biologic and clinical features. Blood. 2007;109(3):874-885. doi:10.1182/blood-2006-07-012252

35.

Clapham

Neer

. G protein beta gamma subunits. Annu Rev Pharmacol Toxicol. 1997;37:167-203. doi:10.1146/annurev.pharmtox.37.1.167

36.

Pal

Patil

Mondal

, et al. Epigenetically silenced GNG4 inhibits SDF1α/CXCR4 signaling in mesenchymal glioblastoma. Genes Cancer. 2016;7(3-4):136-147. doi:10.18632/genesandcancer.105

37.

Tanaka

Kanda

Miwa

, et al. G-protein subunit gamma-4 expression has potential for detection, prediction and therapeutic targeting in liver metastasis of gastric cancer. Br J Cancer. 2021;125(2):220-228. doi:10.1038/s41416-021-01366-1

38.

Bai

Ueno

Vogt

. Akt-mediated regulation of NFkappaB and the essentialness of NFkappaB for the oncogenicity of PI3K and Akt. Int J Cancer. 2009;125(12):2863-2870. doi:10.1002/ijc.24748

39.

Ozes

Mayo

Gustin

Pfeffer

Donner

. NF-kappaB activation by tumour necrosis factor requires the Akt serine-threonine kinase. Nature. 1999;401(6748):82-85. doi:10.1038/43466

40.

Romashkova

Makarov

. NF-kappaB is a target of AKT in anti-apoptotic PDGF signalling. Nature. 1999;401(6748):86-90. doi:10.1038/43474

41.

Forloni

Albini

Limongi

, et al. NF-kappaB, and not MYCN, regulates MHC class I and endoplasmic reticulum aminopeptidases in human neuroblastoma cells. Cancer Res. 2010;70(3):916-924. doi:10.1158/0008-5472.Can-09-2582

42.

Kalbasi

Tariveranmoshabad

Hakimi

, et al. Uncoupling interferon signaling and antigen presentation to overcome immunotherapy resistance due to JAK1 loss in melanoma. Sci Transl Med. 2020;12(565):eabb0152. doi:10.1126/scitranslmed.abb0152

43.

Dejardin

Deregowski

Greimers

, et al. Regulation of major histocompatibility complex class I expression by NF-kappaB-related proteins in breast cancer cells. Oncogene. 1998;16(25):3299-3307. doi:10.1038/sj.onc.1201879

44.

Zhang

Wang

, et al. Therapeutically increasing MHC-I expression potentiates immune checkpoint blockade. Cancer Discov. 2021;11(6):1524-1541. doi:10.1158/2159-8290.Cd-20-0812

45.

Dersh

Phelan

Gumina

, et al. Genome-wide screens identify lineage- and tumor-specific genes modulating MHC-I- and MHC-II-restricted immunosurveillance of human lymphomas. Immunity. 2021;54(1):116-131.e110. doi:10.1016/j.immuni.2020.11.002

46.

Kawase

Kawashima

Nagasaki

, et al. High expression of MHC class I overcomes cancer immunotherapy resistance due to IFNγ signaling pathway defects. Cancer Immunol Res. 2023;11(7):895-908. doi:10.1158/2326-6066.Cir-22-0815

47.

Kishibuchi

Kondo

Soejima

, et al. DNA methylation of GHSR, GNG4, HOXD9 and SALL3 is a common epigenetic alteration in thymic carcinoma. Int J Oncol. 2020;56(1):315-326. doi:10.3892/ijo.2019.4915

48.

Slominski

Raman

Chen

Slominski

. How cancer hijacks the body's homeostasis through the neuroendocrine system. Trends Neurosci. 2023;46(4):263-275. doi:10.1016/j.tins.2023.01.003

49.

Díaz-Beyá

García-Fortes

Valls

, et al. A systems biology- and machine learning-based study to unravel potential therapeutic mechanisms of midostaurin as a multitarget therapy on FLT3-mutated AML. BioMedInformatics. 2022;2(3):375-397. doi:10.3390/biomedinformatics2030024

50.

Guo

Xiao

Chen

Liu

Zhou

. Simulating neuronal development: exploring potential mechanisms for central nervous system metastasis in acute lymphoblastic leukemia. Front Oncol. 2023;13:1331802. doi:10.3389/fonc.2023.1331802

51.

Zhong

Yao

Zhang

Liu

Wang

. Identification and validation of hub genes and molecular classifications associated with chronic myeloid leukemia. Front Immunol. 2024;14:2023. doi:10.3389/fimmu.2023.1297886

52.

Mukherjee

Kar

Paul

Dey

Biswas

Barik

. In silico integration of transcriptome and interactome predicts an ETP-ALL-specific transcriptional footprint that decodes its developmental propensity. Front Cell Dev Biol. 2022;10:2022. doi:10.3389/fcell.2022.899752

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.52 MB

Clinical Characteristics and Outcomes of Acute Myeloid Leukemia Patients Harboring NPM1/FLT3-ITD/DNMT3A Triple Mutations and the Potential Prognostic Value of GNG4

Abstract

Introduction

Methods

Results

Conclusion

Plain Language Summary

Keywords

Introduction

Materials and Methods

Patient Samples

Cytogenetic Analysis

Mutation Detection

Response to Treatment

Bioinformatics Analysis

RNA Isolation, cDNA Preparation, and Quantitative PCR (qPCR)

Statistical Analyses

Results

Patients Characteristics

Mutational Landscape: NPM1, FLT3-ITD, and DNMT3A Were the Most Frequent Mutated Genes

AML NPM1mutFLT3-ITDmutDNMT3Amut Patients had Reduced OS and Higher WBC Count and Were all in the M4/M5 FAB Category

The Hub Gene of AML NPM1mutFLT3-ITDmutDNMT3Amut –– GNG4

Low GNG4 Expression may Contribute to the Low Survival Rate of AML NPM1mutFLT3-ITDmutDNMT3Amut Patients

Discussion

Conclusion

Supplemental Material

Supplemental Material - Clinical Characteristics and Outcomes of Acute Myeloid Leukemia Patients Harboring NPM1/FLT3-ITD/DNMT3A Triple Mutations and the Potential Prognostic Value of GNG4

Footnotes

Author Note

Acknowledgments

ORCID iD

Ethical Considerations

Author Contributions

Funding

Declaration of Conflicting Interests

Data Availability Statement

Supplemental Material

References

Supplementary Material

AML^{NPM1mutFLT3-ITDmutDNMT3Amut} Patients had Reduced OS and Higher WBC Count and Were all in the M4/M5 FAB Category

The Hub Gene of AML^{NPM1mutFLT3-ITDmutDNMT3Amut} –– GNG4

Low GNG4 Expression may Contribute to the Low Survival Rate of AML ^{NPM1mutFLT3-ITDmutDNMT3Amut} Patients