Prognostic significance of inflammatory cytokine polymorphisms in AML: a study of IFNG, TNFA, and IL1B polymorphisms Authors: Tsatsralgerel Munkh-Erdene, Takayuki Saitoh, Miri Kitamura, Asumi Kojima, Takafumi Okawa, Tsukasa Oda, Eiji Miyauchi, Nobuo Sasaki, Ikuko Matsumura, Akira Matsumoto, Hisashi Takei, Nobuhiko Kobayashi, Yuri Miyazawa, Yoshiyuki Ogawa, Hiroshi Handa, Nanami Gotoh Published: 2026 ABSTRACT AbstractBackgroundInflammatory cytokines influence the pathogenesis and progression of acute myeloid leukaemia (AML), not only by shaping the leukemic microenvironment, but also by supporting leukaemia stem cell survival and resistance to therapy. We, therefore, investigated the associations between the cytokine polymorphisms IFNG+874 A/T, TNFA−857 C/T, and IL1B -31 C/T and AML outcomes, as well as their influence on clinical features.ResultsNinety-three patients with AML and 117 healthy controls were analysed. The T allele of TNFA − 857 C/T (i.e., the high-expression allele) was observed at a significantly higher frequency in the patients with AML vs. the controls (AML vs. control = 24.7% vs. 16.2%, p = 0.04). Patients with the high-expression TT genotype had shorter overall survival (TT vs. CC = 46.0 vs. 224.1 months, p < 0.001). Patients with the high-expression non-CC genotype relapsed more frequently than those with the low-expression CC genotype (relapse vs. non-relapse = 55.8% vs. 26.9%, p = 0.01). No significant associations were observed for the IFNG + 874 A/T and IL1B -31 C/T polymorphism.ConclusionTNFA − 857 C/T polymorphisms can influence patient susceptibility to AML, as well as its prognosis. This suggests a link between chronic inflammation and leukemogenesis, as well as the potential value of TNFA genotyping in risk assessments.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12865-026-00837-5.Keywords: Acute myeloid leukaemia, Inflammatory cytokines, Polymorphisms Background Acute myeloid leukaemia (AML) is a clonal haematological malignancy that arises from haematopoietic stem or progenitor cells (HSPCs) in the bone marrow. It is characterised by the accumulation of immature myeloid blasts that proliferate uncontrollably and interfere with normal haematopoiesis, leading to clinical manifestations such as anaemia, recurrent infections, and haemorrhagic complications. AML blasts are organised in a hierarchical structure, with leukaemia stem cells (LSCs) at the apex. Although LSCs represent a minor fraction of the leukemic population—often < 1%—they possess definitive stem cell properties such as self-renewal capacity and long-term survival [1, 2]. In contrast to bulk leukemic blasts, LSCs are inherently resistant to conventional chemotherapy, often allowing them to persist following initial treatments and ultimately driving disease relapse [3]. This persistence is considered a major barrier to curative AML therapy. Emerging evidence highlights the critical role of inflammation in supporting the maintenance and survival of LSCs. The inflamed bone marrow microenvironment creates a permissive niche that favours LSC persistence while impairing normal HSPC function [4, 5]. Proinflammatory cytokines such as interleukin (IL)-1, IL-6, tumour necrosis factor-α (TNF-α), and interferon-γ (IFN-γ) actively reshape the bone marrow niche, enhancing LSC self-renewal, chemoresistance, and maintenance in a quiescent state [4, 6]. The chronic activation of inflammatory signalling pathways, particularly NF-κB and JAK-STAT, further reinforces LSC expansion and contributes to the suppression of normal haematopoiesis [5, 6]. Normal haematopoietic stem cells (HSCs) are quiescent but can enter the cell cycle in response to inflammatory stress. This activation increases proliferation, reactive oxygen species production, and DNA damage, thereby raising the likelihood of acquiring driver mutations. Mutations in genes such as DNMT3A, TET2, ASXL1 create a context in which chronic inflammatory signals confer a selective advantage to mutant HSCs over wild‑type cells, leading to clonal haematopoiesis of indeterminate potential or other pre‑leukemic states. These mutant clones frequently adopt a pro‑inflammatory phenotype, establishing a feed‑forward loop between clonal expansion and chronic inflammation that can ultimately facilitate progression to AML [7–9]. Based on these roles, we hypothesised that functional polymorphisms in key inflammatory cytokine genes may influence both susceptibility to AML and its clinical course. In our previous research on haematologic malignancies, including multiple myeloma and myelodysplastic syndromes (MDS), we demonstrated that polymorphisms in genes for cytokines such as TNF-α and IL-10 significantly influence disease susceptibility, clinical features, and prognosis [10–13]. Notably, we found that the high-expression TNF-α polymorphism may serve as a potential poor prognostic marker of MDS, highlighting the broader impact of inflammation in myeloid neoplasms [13]. In this study, we focused on three functional single-nucleotide polymorphisms—IFNG + 874 A/T, TNFA − 857 C/T, and IL1B -31 C/T—located in promoter or transcription factor binding regions known to modulate cytokine expression [14–17]. These polymorphisms have been extensively studied for their roles in immune regulation; however, their effects on AML, particularly in Japanese populations, remain largely unexplored. Therefore, we investigated the associations between these cytokine polymorphisms and AML susceptibility, as well as their potential influence on clinical features, in a Japanese patient cohort. Understanding these genetic factors may offer insights into the inflammatory mechanisms underlying AML pathogenesis and progression. Materials and methods In this retrospective cohort study performed at two hospitals, we investigated the associations between IFNG + 874 A/T, TNFA − 857 C/T, and IL1B -31 C/T and AML susceptibility, as well as the potential influence of these cytokine polymorphisms on clinical features. Study population This study was conducted on a Japanese cohort of 93 patients with AML and 117 healthy controls. The observation period was between 1991 and 2013. All patients were classified according to the French-American-British (FAB) classification and the UK Medical Research Council (MRC) classification (2010) at Gunma University Hospital or Saiseikai Maebashi Hospital. Any cases in which polymerase chain reaction (PCR) testing failed were excluded from the analysis. Among the study population, the most commonly administered treatment regimen was daunorubicin (DNR) plus cytarabine (AraC) (31 patients, 33.3%), followed by all-trans retinoic acid (23 patients, 24.7%), idarubicin plus AraC (16 patients, 17.2%), and enocitabine plus DNR (15 patients, 16.1%). Two patients (2.1%) did not receive any treatment (Additional file 1). Written informed consent was obtained from all participants, and the study received approval from the Institutional Review Board of Gunma University (approval no.: #770). Genotyping Genomic DNA was extracted from peripheral blood samples obtained from each participant using a QIAamp DNA Blood Kit (Qiagen Sciences, Germantown, MD, USA). Three single-nucleotide polymorphisms were analysed: TNFA − 857 C/T, IL1B -31 C/T, and IFNG + 874 A/T. Genotyping of TNFA − 857 C/T and IL1B -31 C/T was executed using the PCR-restriction fragment length polymorphism method. Each PCR reaction was conducted in a total volume of 20 µL, containing 1 µL of genomic DNA, 0.2 mM of dNTPs, 0.5 U of TaKaRa Ex Taq HS DNA polymerase (TaKaRa Bio, Japan), and 0.5 µM of each primer (Additional file 2). The resulting PCR products were digested with the appropriate restriction enzymes (New England BioLabs, Tokyo, Japan), as indicated in Additional file 2. Genotyping of IFNG + 874 A/T was conducted using allele-specific PCR. The primers used for amplification are listed in Additional file 2. The PCR reaction was performed in a total volume of 20 µL, comprising 1 µL of genomic DNA, 0.2 mM of dNTPs, 0.5 U of TaKaRa Ex Taq HS DNA polymerase (TaKaRa Bio), and 0.5 µM of each primer. Statistical analysis All statistical analyses were performed using SPSS version 30.0.0.0 (IBM Corp., Armonk, NY, USA). Results are reported as means ± standard deviations for continuous variables, and as percentages for categorical variables. Clinical features were compared using Student’s t and the Chi-squared tests. Kaplan–Meier curves were used to show survival time, and the log-rank test was used to compare survival between patient groups. The Cox proportional hazard model was used to calculate hazard ratios (HRs). Statistical significance was defined as a two-tailed p value of < 0.05. Results Study population This study’s cohort included 93 Japanese patients with AML (51 men and 42 women aged between 15 and 86 years, median age: 58 years; Table 1) and 117 healthy Japanese controls. Table 1Clinical characteristics of patients in this studyNumber of patients106Male/Female51/42Age, median (range)58.0 (15–86)WBC (×109/L), median (range)7.0 (0.3–317.4)LDH (IU/L), median (range)354 (134–3770)Karyotype, n (%) t(8;21)7 (7.5%) t(15;17)23 (24.7%) inv(16)5 (5.4%) Normal41 (44.1%) Complex5 (5.4%) Others12 (12.9%)FAB classification, n (%) M06 (6.5%) M114 (15.1%) M232 (34.4%) M323 (24.7%) M410 (10.8%) M55 (5.4%) M62 (2.2%) M71 (1.1%)MRC classification, n (%) Favourable35 (37.6%) Intermediate52 (55.9%) Adverse6 (6.5%) Allo-SCT, n (%)12 (12.9%)WBC white blood cell, LDH lactate dehydrogenase, FAB classification French-American-British classification, MRC classification UK Medical Research Council classification, Allo-SCT allogeneic stem cell transplantation Clinical characteristics of patients in this study WBC white blood cell, LDH lactate dehydrogenase, FAB classification French-American-British classification, MRC classification UK Medical Research Council classification, Allo-SCT allogeneic stem cell transplantation Comparison of genotype and allele frequencies Table 2 summarises the genotype distributions of TNFA, IL1B, and IFNG polymorphisms in the patients with AML vs. the healthy controls. We compared the genotype frequencies of each polymorphism using the co-dominant, dominant, and recessive models. We also compared allele frequencies. Table 2Genotype and allele frequency of inflammatory cytokine gene polymorphisms in AML and controlsPolymorphismModelGenotypeAMLn = 93Controln = 117p-valueOR95% CITNFA − 857 C/TCC56(60.2%)83(72.8%)refco-dominantCT28(30.1%)25(21.9%)0.141.650.83–3.30TT9(9.7%)6(5.3%)0.172.210.66–8.00dominantTT + CT vs. CC0.071.760.95–3.31recessiveTT vs. CT + CC0.281.920.58–6.84C allele140(75.2%)191(83.8%)0.040.590.35–0.98T allele46(24.7%)37(16.2%)IL1B -31 C/TCC24(25.8%)37(31.9%)co-dominantCT41(44.1%)54(46.6%)0.741.170.58–2.38TT28(30.1%)25(21.6%)0.191.720.77–3.89dominantTT + CT vs. CC0.361.340.70–2.60recessiveTT vs. CT + CC0.201.560.80–3.08C allele89(47.8%)128(55.1%)0.140.750.50–1.12T allele97(52.2%)104(44.8%)IFNG + 874 A/TAA78(83.9%)87(74.4%)co-dominantAT15(16.1%)28(23.9%)0.170.600.28–1.26TT0(0.0%)2(1.7%)0.500.000.00–6.07dominantTT + TA vs. AA0.170.600.28–1.26recessiveTT vs. TA + AA0.500.000.00–6.07A allele171(91.9%)202(86.3%)0.091.800.91–3.71T allele15(8.1%)32(13.7%)AML acute myeloid leukaemia, OR odds ratio, CI confidence intervalP value < 0.05 represents statistical significance (chi-squared test) Genotype and allele frequency of inflammatory cytokine gene polymorphisms in AML and controls AML acute myeloid leukaemia, OR odds ratio, CI confidence interval P value < 0.05 represents statistical significance (chi-squared test) The T allele of TNFA − 857 C/T (i.e., the high-expression allele) was observed at a significantly higher frequency among the patients with AML than among the controls (AML vs. control = 24.7% vs. 16.2%, p = 0.04). This polymorphism also showed a trend toward being associated with AML risk under the dominant model (p = 0.07). These findings suggest a potential association between the high-expression allele of TNFA − 857 C/T and the risk of developing AML. No significant associations were observed for the other two polymorphisms. Comparison of clinical features Table 3 presents a comparison of clinical characteristics according to TNFA, IL1B, and IFNG polymorphisms. The patients with the high-expression TNFA − 857 non-CC genotype were significantly older than those with the low-expression CC variant (non-CC vs. CC = 59.7 ± 16.3 vs. 51.8 ± 16.4 years, p = 0.03). Table 3Comparison of clinical features by inflammatory cytokine gene polymorphisms in patients with AMLTNFA − 857 C/TCC (n = 56)low expressionNon-CC (n = 37)high expressionp-valueOR95% CIMale/female29/2722/150.471.370.59–3.16Age (years), mean ± SD51.8 ± 16.459.7 ± 16.30.03WBC count (×109/L), mean ± SD36.2 ± 59.122.1 ± 32.60.45LDH level (IU/L), mean ± SD590 ± 612538 ± 4570.66Karyotype, n (%)t(8;21)6(10.7%)1(2.7%)0.154.320.50–37.5t(15;17)14(25.0%)9(24.3%)0.941.040.40–2.72inv(16)4(7.1%)1(2.7%)0.060.150.02–1.40IL1B -31 C/TCC (n = 24)low expressionNon-CC (n = 69)high expressionp-valueOR95% CIMale/female14/1037/320.690.830.32–2.11Age (years), mean ± SD56.6 ± 16.354.4 ± 17.00.58WBC count (×109/L), mean ± SD35.0 ± 65.629.1 ± 44.60.63LDH level (IU/L), mean ± SD624 ± 759550 ± 4680.58Karyotype, n (%)t(8;21)3(12.5%)4(5.8%)0.282.320.48–11.2t(15;17)3(12.5%)20(29.0%)0.110.350.09–1.31inv(16)2(8.3%)3(4.3%)0.462.000.31–12.8IFNG + 874 A/TAA (n = 78)low expressionNon-AA (n = 15)high expressionp-valueOR95% CIMale/female40/3811/40.122.610.77–8.92Age (years), mean ± SD54.1 ± 17.159.4 ± 14.40.27WBC count (×109/L), mean ± SD26.9 ± 45.749.7 ± 69.70.11LDH level (IU/L), mean ± SD587 ± 585482 ± 3510.51Karyotype, n (%)t(8;21)7(9.0%)0(0.0%)0.230.830.75–0.91t(15;17)22(28.2%)1(6.7%)0.085.500.68–44.4inv(16)4(5.1%)1(6.7%)0.810.760.08–7.29AML acute myeloid leukaemia, OR odds ratio, CI confidence interval, WBC white blood cell, LDH lactate dehydrogenase, MRC classification UK Medical Research Council classificationP value < 0.05 represents statistical significance (Student’s t-test or chi-squared test) Comparison of clinical features by inflammatory cytokine gene polymorphisms in patients with AML AML acute myeloid leukaemia, OR odds ratio, CI confidence interval, WBC white blood cell, LDH lactate dehydrogenase, MRC classification UK Medical Research Council classification P value < 0.05 represents statistical significance (Student’s t-test or chi-squared test) No significant associations were observed for the IL1B -31 C/T or IFNG + 874 A/T polymorphism. Comparison of overall and event-free survivals To investigate the impact of TNFA, IL1B, and IFNG polymorphisms on survival, univariate analyses of overall survival (OS) and event-free survival (EFS) were performed using log-rank tests (Fig. 1). The TNFA − 857 C/T polymorphism was found to be significantly associated with OS and EFS. Patients with the high-expression TT genotype had shorter OS (TT vs. CC = 46.0 vs. 224.1 months, p < 0.001) and EFS (TT vs. CC = 22.1 vs. 178.9 months, p < 0.001) than those carrying the low-expression CC variant. This polymorphism was also found to be significantly associated with OS and EFS in patients with non-M3 AML (Additional file 3). Patients with the high-expression TT genotype had shorter OS (TT vs. CC = 20.2 months vs. not reached, p < 0.001) and shorter EFS (TT vs. CC = 14.7 vs. not reached months, p = 0.003). In addition, to reduce the effect of allogenic stem cell transplantation (allo-SCT), OS analysis was conducted using a model in which allo-SCT cases were treated as censored cases. Among patients with non-M3 AML, those carrying the TNFA − 857 C/T TT genotype (high-expression type) had shorter OS than those with CT or CC genotypes (TT vs. CC = 17.2 months vs. not reached, p = 0.001) (Additional File 4). We then performed a multivariate analysis to control for the effect of age-based differences, which identified the TT genotype as an independent adverse prognostic factor of OS (HR = 3.10, 95% confidence interval = 1.10–8.71, p = 0.03, Table 4). Comparison of OS and EFS in TNFA, IL1B, and IFNG polymorphisms. Survival was compared using the log-rank test, and a p-value of < 0.05 was considered statistically significant. OS, overall survival; EFS, event-free survival Table 4Multivariate analysis to evaluate the impact of TNFA − 857 C/T on OSVariables n p valueHR95% CIAge < Median461.00 > Median470.0099.321.75–49.7Sex Male511.00 Female420.1200.400.13–1.27MRC classification Favourable351.00 Adverse and intermediate580.00916.01.97–128.9WBC [×109/L] < 20601.00 > 20330.0200.200.05–0.75LDH [IU/L] < 230241.00 > 230690.3301.740.57–5.36Allo-SCT No811.00 Yes120.5402.100.19–22.7TNFA − 857 C/T CC and TT841.00 TT90.0303.101.10–8.71The data were analysed using Cox proportional hazards regression analysis. P < 0.05 was considered to represent statistical significanceOS overall survival, HR hazard ratio, CI confidence interval, MRC classification UK Medical Research Council classification, WBC white blood cell, LDH lactate dehydrogenase, Allo-SCT allogeneic stem cell transplantation; ref, reference Multivariate analysis to evaluate the impact of TNFA − 857 C/T on OS The data were analysed using Cox proportional hazards regression analysis. P < 0.05 was considered to represent statistical significance OS overall survival, HR hazard ratio, CI confidence interval, MRC classification UK Medical Research Council classification, WBC white blood cell, LDH lactate dehydrogenase, Allo-SCT allogeneic stem cell transplantation; ref, reference No significant associations were observed for the IL1B -31 C/T or IFNG + 874 A/T polymorphisms. Comparison of relapse-related clinical features To further assess clinical features, we investigated our cohort’s relapse rate and relapse-free survival (RFS) time in relation to TNFA, IL1B, and IFNG polymorphisms, among the patients who achieved complete remission (CR). Prior to these assessments, we compared CR rates among the polymorphisms and found no significant differences associated with any of the TNFA, IL1B, or IFNG polymorphisms (Additional file 5). The patients with the high-expression TNFA − 857 non-CC genotype relapsed more frequently than those with the low-expression CC alternative (relapse vs. non-relapse = 55.8% vs. 26.9%, p = 0.01, Table 5). In our analysis of RFS stratified by the three genotypes for this polymorphism (CC, CT, and TT), an overall log-rank test demonstrated a significant difference among the groups (p = 0.02, Fig. 2). However, no statistically significant differences were observed during pairwise comparisons, although a trend toward shorter RFS was noted in the patients with the high-expression TT genotype vs. those with the low-expression CC genotype (TT vs. CC = 29.1 vs. 185.4 months, p = 0.06). We also analysed RFS in patients with non-M3 AML; however, no significant association was observed for this polymorphism (Additional file 3). Table 5Comparison of relapse rate by inflammatory cytokine gene polymorphisms in patients with AMLRelapsen = 34non-relapsen = 52p valueOR95% CITNFA − 857 C/T CC15 (44.1%)38 (73.1%)ref non-CC19 (55.8%)14 (26.9%)0.013.381.26–9.47IL1B -31 C/T CC8 (23.5%)12 (23.1%)ref non-CC26 (76.5%)40 (76.9%)1.000.980.32–3.15IFNG + 874 A/T AA29 (85.3%)45 (86.5%)ref non-AA5 (14.7%)7 (13.5%)1.001.110.25–4.50AML acute myeloid leukaemia, OR odds ratio, CI confidence intervalP value < 0.05 represents statistical significance (chi-squared test) Comparison of relapse rate by inflammatory cytokine gene polymorphisms in patients with AML AML acute myeloid leukaemia, OR odds ratio, CI confidence interval P value < 0.05 represents statistical significance (chi-squared test) Comparison of RFS in the TNFA polymorphism. Survival was compared using the log-rank test, and a p-value of < 0.05 was considered statistically significant. RFS, relapse-free survival Discussion This study examined the clinical impact of functional polymorphisms in inflammatory cytokine genes among patients with AML. Our findings suggest that the high-expression genotype of the TNFA − 857 C/T polymorphism may be associated with increased susceptibility and poorer clinical outcomes in patients with AML. In particular, patients harbouring the high-expression genotype exhibited shorter OS, EFS, and RFS than those carrying the low-expression variant. Notably, this trend for OS and EFS persisted even after excluding patients with M3 (acute promyelocytic leukaemia). The TNFA − 857 C/T polymorphism is located in the promoter region of the TNF-α gene, where the T allele increases TNF-α transcription compared with the C allele. Functionally, this makes the − 857T allele a high-producer allele, consistent with data from reporter assays [17]. These results suggest that enhanced TNF-α expression associated with this functional polymorphism may contribute to poorer clinical outcomes. Our analysis demonstrated an association between the TNFA − 857 C/T polymorphism and poor prognosis in patients with AML. We hypothesise that the high-expression genotype of this polymorphism increases the frequency of AML relapse, thereby contributing to shorter OS. Relapse in AML is primarily driven by the persistence of LSCs in the bone marrow after therapy. Inflammatory signals within the bone marrow microenvironment create a protective niche that supports LSC survival and stemness at the expense of normal HSPCs [4]. Proinflammatory cytokines such as IL-1, IL-6, TNF-α, and IFN-γ remodel the marrow architecture, directly promoting LSC self-renewal, quiescence, and chemoresistance [6]. Notably, LSCs have been shown to express transmembrane TNF-α (tmTNF-α), which confers chemoresistance and anti-apoptotic properties. The knockdown or pharmacological inhibition of tmTNF-α sensitises LSCs to chemotherapeutic agents, indicating its direct role in drug resistance and LSC maintenance. Consistent with these findings, high tmTNF-α expression has been shown to correlate with poor clinical outcomes [18]. Collectively, these findings suggest that TNF-α-mediated inflammatory signalling may contribute to relapse through its protective effects on LSCs. This mechanism may underlie the association we observed between the high-expression TNFA genotype and poor patient prognoses. Our findings also revealed an association between the TNFA − 857 C/T polymorphism and the risk of developing AML. Verma et al. reported that serum TNF-α levels were significantly higher in patients with AML than in healthy controls and that these levels decreased following chemotherapy [19]. This finding suggests that a TNF-α-rich inflammatory environment supports the survival and persistence of AML clones. Furthermore, chronic inflammation has been shown to induce DNA damage and stress responses in normal HSPCs, leading to their depletion, while inflammation-adapted mutant clones—for example, those harbouring mutations in genes such as DNMT3A, TET2, ASXL1, or TP53—gain a selective growth advantage and multiply preferentially as LSCs [20]. Further supporting this notion, in vivo studies have demonstrated that chronic inflammatory stimulation with poly(I: C) or lipopolysaccharide suppresses the proliferation of TP53-wild-type HSPCs while selectively promoting the expansion of TP53-mutant HSPCs [21]. These findings indicate an association between TNF-α-driven chronic inflammation and AML development, possibly reflecting the influence of an inflammatory microenvironment on pre-leukemic clone dynamics. To our knowledge, limited data are available regarding the association between the TNFA − 857 C/T polymorphism and AML. Meta-analyses in solid tumours have suggested that this polymorphism may influence cancer risk, although the direction and magnitude of this effect appear to vary depending on cancer type and ethnicity, indicating a context-dependent role [22]. In MDS, we previously reported that this polymorphism was associated with adverse risk classification and a higher frequency of leukemic transformation [13]. Other TNFA polymorphisms, such as -308 G/A, have been linked to lower haemoglobin levels, severe cytopenias, and increased transfusion dependency, suggesting a role in disease severity [23–25]. Collectively, although our findings may provide insights into the potential clinical relevance of TNFA polymorphism in AML, they should be interpreted with caution in light of several limitations. First, the sample size was relatively modest, and the retrospective design may have introduced some inherent bias. Therefore, larger prospective studies are warranted to validate our observations. Confirmation in cohorts with more even distributions of disease subtypes may also strengthen the robustness of the results. Second, our genetic analysis was confined to three cytokine polymorphisms (TNFA − 857 C/T, IL1B -31 C/T and IFNG + 874 A/T) and, thus, inherently excluded other potentially functional variants. Future studies should consider exploring additional polymorphisms that may influence cytokine expression or disease pathogenesis. Third, the study population consisted exclusively of Japanese patients, which may limit the generalizability of our findings to other ethnic groups. Thus, multicentre studies involving diverse populations are warranted to assess the global applicability of our results. Taken together, these limitations underscore the need for cautious interpretation of the present findings. Conclusion This study’s findings indicate that inflammatory cytokine polymorphisms, particularly TNFA − 857 C/T, are associated with the susceptibility to and prognosis of AML. The high-expression genotype of this polymorphism appears to increase the risk of disease recurrence, which may explain the shorter survival times we observed in the patients who harboured it. These observations provide further insights into the link between chronic inflammation and AML pathogenesis and highlight the potential utility of TNFA polymorphism analysis as a future risk stratification and prognostic marker for AML. Supplementary Information Additional file 1: Treatment regimens of the study population. Additional file 2: This is the primer sequences for genotyping. Additional file 3: Survival analysis of the TNFA −857 C/T polymorphism in non-M3 patients. Statistical significance was defined as p < 0.05. Additional file 4: Survival analysis of the TNFA −857 C/T polymorphism in non-M3 patients. In this analysis, patients who underwent allo-SCT were treated as censored at the time of transplantation. Statistical significance was defined as p < 0.05. Additional file 5: This is the table data for complete remission rate of our population. Acknowledgements Not applicable. Abbreviations Acute Myeloid Leukaemia Complete Remission Event-free Survival French-American-British Hazard Ratio Haematopoietic Stem or Progenitor Cells Interferon-γ Interleukin Leukaemia Stem Cell Myelodysplastic Syndromes UK Medical Research Council Overall Survival Polymerase Chain Reaction Relapse-free Survival Transmembrane TNF-α Tumour necrosis factor-α Authors’ contributions TME and NG performed this research; MK, AK, TO, TO, EM, and NS discussed the research data; TS, IM, AM, HT, NK, YM, YO, and HH collected the study population; TME and NG analysed the data; TME and NG wrote the manuscript; TS supported writing the manuscript; TS designed this study; HH supported designing the study. All authors read and approved the final manuscript. Funding This work was supported by JST Grant Number JPMJPF2301. Data availability The data that support the findings of this study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of Gunma University (approval no.: #770). Written informed consent was obtained from all participants. Although some participants were minors at the time of diagnosis, this study is a retrospective observational study, and all participants had reached adulthood at the time the study was initiated. Therefore, informed consent was obtained directly from each participant. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1: Treatment regimens of the study population. Additional file 2: This is the primer sequences for genotyping. Additional file 3: Survival analysis of the TNFA −857 C/T polymorphism in non-M3 patients. Statistical significance was defined as p < 0.05. Additional file 4: Survival analysis of the TNFA −857 C/T polymorphism in non-M3 patients. In this analysis, patients who underwent allo-SCT were treated as censored at the time of transplantation. Statistical significance was defined as p < 0.05. Additional file 5: This is the table data for complete remission rate of our population. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.