Chromatin-Spliceosome Mutations in Acute Myeloid Leukemia

Precision Medicine Approaches in Acute Myeloid Leukemia with Adverse Genetics

The treatment of acute myeloid leukemia (AML) with adverse genetics remains unsatisfactory, with very low response rates to standard chemotherapy and shorter durations of remission commonly observed in these patients. The complex biology of AML with adverse genetics is continuously evolving. Herein, we discuss recent advances in the field focusing on the contribution of molecular drivers of leukemia biogenesis and evolution and on the alterations of the immune system that can be exploited with immune-based therapeutic strategies. We focus on the biological rationales for combining targeted therapy and immunotherapy, which are currently being investigated in ongoing trials, and could hopefully ameliorate the poor outcomes of patients affected by AML with adverse genetics.

Comprehensive genomic profiling reveals molecular subsets of ASXL1-mutated myeloid neoplasms

A large-scale genomic analysis of patients with ASXL1-mutated myeloid disease has not been performed to date. We reviewed comprehensive genomic profiling results from 6043 adults to characterize clinicopathologic features and co-mutation patterns by ASXL1 mutation status. ASXL1 mutations occurred in 1414 patients (23%). Mutation co-occurrence testing revealed strong co-occurrence (p < 0.01) between mutations in ASXL1 and nine genes (SRSF2, U2AF1, RUNX1, SETBP1, EZH2, STAG2, CUX1, CSF3R, CBL). Further analysis of patients with these co-mutations yielded several novel findings. Co-mutation patterns supported that ASXL1/SF3B1 co-mutation may be biologically distinct from ASXL1/non-SF3B1 spliceosome co-mutation. In AML, ASXL1/SRSF2 co-mutated patients frequently harbored STAG2 mutations (42%), which were dependent on the presence of both ASXL1 and SRSF2 mutation (p < 0.05). STAG2 and SETBP1 mutations were also exclusive in ASXL1/SRSF2 co-mutated patients and associated with divergent chronic myeloid phenotypes. Our findings support that certain multi-mutant genotypes may be biologically relevant in ASXL1-mutated myeloid disease.

Spliceosome mutations are associated with clinical response in a phase 1b/2 study of the PLK1 inhibitor onvansertib in combination with decitabine in relapsed or refractory acute myeloid leukemia

PLK1 is overexpressed in acute myeloid leukemia (AML). A phase 1b trial of the PLK1 inhibitor onvansertib (ONV) combined with decitabine (DAC) demonstrated initial safety and efficacy in patients with relapsed/refractory (R/R) AML. The current study aimed to identify molecular predictors of response to ONV + DAC in R/R AML patients. A total of 44 R/R AML patients were treated with ONV + DAC and considered evaluable for efficacy. Bone marrow (BM) samples were collected at baseline for genomic and transcriptomic analysis (n = 32). A 10-gene expression signature, predictive of response to ONV + DAC, was derived from the leading-edge genes of gene set enrichment analyses (GSEA). The gene signature was evaluated in independent datasets and used to identify associated mutated genes. Twenty percent of the patients achieved complete remission, with or without hematologic count recovery (CR/CRi), and 32% exhibited a ≥50% reduction in bone marrow blasts. Patients who responded to treatment had elevated mitochondrial function and OXPHOS. The gene signature was not associated with response to DAC alone in an independent dataset. By applying the signature to the BeatAML cohort (n = 399), we identified a positive association between predicted ONV + DAC response and mutations in splicing factors (SF). In the phase 1b/2 trial, patients with SF mutations (SRSF2, SF3B1) had a higher CR/CRi rate (50%) compared to those without SF mutations (9%). PLK1 inhibition with ONV in combination with DAC could be a potential therapy in R/R AML patients, particularly those with high OXPHOS gene expression and SF mutations.

Clinical and prognostic profile of SRSF2 and related spliceosome mutations in patients with acute myeloid leukemia

BACKGROUND

Mutations in splicing factor (SF) genes are frequently detected in myelodysplastic syndrome, but their clinical and prognostic relevance in acute myeloid leukemia (AML) have rarely been reported.

METHODS

A total of 368 newly diagnosed non-M3 AML patients were included in this study. Next generation sequencing including four SF genes was performed on the genomicDNA. The clinical features and survival were analyzed using statistical analysis.

RESULTS

We found that 64 of 368 patients harbored SF mutations. The SF mutations were much more frequently found in older or male patients. SRSF2 mutations were shown obviously co-existed with IDH2 mutation. The level of measurable residual disease after first chemotherapy was higher in SF-mutated patients compared to that in SF-wild patients, while the complete remission rate was significantly decreased. And the overall survival of SF-mutated patients was shorter than that of SF-wild patients. Moreover, our multivariable analysis suggests that the index of male, Kit mutation or ZRSR2 mutation was the independent risk factor for overall survival. SRSF2mut was associated with older age, higher proportion of peripheral blasts or abnormal cell proportion by flow cytometry.

CONCLUSION

SF mutation is a distinct subgroup of AML frequently associated with clinic-biological features and poor outcome. SRSF2mut could be potential targets for novel treatment in AML.

Biomolecular Condensates in Myeloid Leukemia: What Do They Tell Us?

Recent studies have suggested that several oncogenic and tumor-suppressive proteins carry out their functions in the context of specific membrane-less cellular compartments. As these compartments, generally referred to as onco-condensates, are specific to tumor cells and are tightly linked to disease development, the mechanisms of their formation and maintenance have been intensively studied. Here we review the proposed leukemogenic and tumor-suppressive activities of nuclear biomolecular condensates in acute myeloid leukemia (AML). We focus on condensates formed by oncogenic fusion proteins including nucleoporin 98 (NUP98), mixed-lineage leukemia 1 (MLL1, also known as KMT2A), mutated nucleophosmin (NPM1c) and others. We also discuss how altered condensate formation contributes to malignant transformation of hematopoietic cells, as described for promyelocytic leukemia protein (PML) in PML::RARA-driven acute promyelocytic leukemia (APL) and other myeloid malignancies. Finally, we discuss potential strategies for interfering with the molecular mechanisms related to AML-associated biomolecular condensates, as well as current limitations of the field.

Genetic landscape of chronic myeloid leukemia

Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm caused by the BCR::ABL1 fusion gene, which aberrantly activates ABL1 kinase and promotes the overproduction of leukemic cells. CML typically develops in the chronic phase (CP) and progresses to a blast crisis (BC) after years without effective treatment. Although prognosis has substantially improved after the development of tyrosine kinase inhibitors (TKIs) targeting the BCR::ABL1 oncoprotein, some patients still experience TKI resistance and poor prognosis. One of the mechanisms of TKI resistance is ABL1 kinase domain mutations, which are found in approximately half of the cases, newly acquired during treatment. Moreover, genetic studies have revealed that CML patients carry additional mutations that are also observed in other myeloid neoplasms. ASXL1 mutations are often found in both CP and BC, whereas other mutations, such as those in RUNX1, IKZF1, and TP53, are preferentially found in BC. The presence of additional mutations, such as ASXL1 mutations, is a potential biomarker for predicting therapeutic efficacy. The mechanisms by which these additional mutations affect disease subtypes, drug resistance, and prognosis need to be elucidated. In this review, we have summarized and discussed the landscape and clinical impact of genetic abnormalities in CML.

A distinct subpopulation of leukemia initiating cells in acute precursor B lymphoblastic leukemia: quiescent phenotype and unique transcriptomic profile

In leukemia, a distinct subpopulation of cancer-initiating cells called leukemia stem cells (LSCs) is believed to drive population expansion and tumor growth. Failing to eliminate LSCs may result in disease relapse regardless of the amount of non-LSCs destroyed. The first step in targeting and eliminating LSCs is to identify and characterize them. Acute precursor B lymphoblastic leukemia (B-ALL) cells derived from patients were incubated with fluorescent glucose analog 2-(N-(7-Nitrobenz-2-oxa-1, 3-diazol-4-yl) Amino)-2-Deoxyglucose (NBDG) and sorted based on NBDG uptake. Cell subpopulations defined by glucose uptake were then serially transplanted into mice and evaluated for leukemia initiating capacity. Gene expression profiles of these cells were characterized using RNA-Sequencing (RNA-Seq). A distinct population of NBDG-low cells was identified in patient B-ALL samples. These cells are a small population (1.92% of the entire leukemia population), have lower HLA expression, and are smaller in size (4.0 to 7.0 μm) than the rest of the leukemia population. All mice transplanted with NBDG-low cells developed leukemia between 5 and 14 weeks, while those transplanted with NBDG-high cells did not develop leukemia (p ≤ 0.0001-0.002). Serial transplantation of the NBDG-low mouse model resulted in successful leukemia development. NBDG-medium (NBDG-med) populations also developed leukemia. Interestingly, comprehensive molecular characterization of NBDG-low and NBDG-med cells from patient-derived xenograft (PDX) models using RNA-Seq revealed a distinct profile of 2,162 differentially-expressed transcripts (DETs) (p<0.05) with 70.6% down-regulated in NBDG-low cells. Hierarchical clustering of DETs showed distinct segregation of NBDG-low from NBDG-med and NBDG-high groups with marked transcription expression alterations in the NBDG-low group consistent with cancer survival. In conclusion, A unique subpopulation of cells with low glucose uptake (NBDG-low) in B-ALL was discovered. These cells, despite their quiescence characteristics, once transplanted in mice, showed potent leukemia initiating capacity. Although NBDG-med cells also initiated leukemia, gene expression profiling revealed a distinct signature that clearly distinguishes NBDG-low cells from NBDG-med and the rest of the leukemia populations. These results suggest that NBDG-low cells may represent quiescent LSCs. These cells can be activated in the appropriate environment in vivo, showing leukemia initiating capacity. Our study provides insight into the biologic mechanisms of B-ALL initiation and survival.

RNA-Binding Proteins CLK1 and POP7 as Biomarkers for Diagnosis and Prognosis of Esophageal Squamous Cell Carcinoma

The abnormality of RNA-binding proteins (RBPs) is closely related to the tumorigenesis and development of esophageal squamous cell carcinoma (ESCC), and has been an area of interest for research recently. In this study, 162 tumors and 11 normal samples are obtained from The Cancer Genome Atlas database, among which 218 differentially expressed RBPs are screened. Finally, a prognostic model including seven RBPs (CLK1, DDX39A, EEF2, ELAC1, NKRF, POP7, and SMN1) is established. Further analysis reveals that the overall survival (OS) rate of the high-risk group is lower than that of the low-risk group. The area under the receiver operating characteristic (ROC) curve (AUC) of the training group and testing group is significant (AUCs of 3 years are 0.815 and 0.694, respectively, AUCs of 5 years are 0.737 and 0.725, respectively). In addition, a comprehensive analysis of seven identified RBPs shows that most RBPs are related to OS in patients with ESCC, among which EEF2 and ELCA1 are differentially expressed at the protein level of ESCC and control tissues. CLK1 and POP7 expressions in esophageal cancer tumor samples are undertaken using the tissue microarray, and show that CLK1 mRNA levels are relatively lower, and POP7 mRNA levels are higher compared with non-cancerous esophageal tissues. Survival analysis reveals that a higher expression of CLK1 predicts a significant worse prognosis, and a lower expression of POP7 predicts a worse prognosis in esophageal cancer. These results suggest that CLK1 may promote tumor progression, and POP7 may hinder the development of esophageal cancer. In addition, gene set enrichment analysis reveals that abnormal biological processes related to ribosomes and abnormalities in classic tumor signaling pathways such as TGF-β are important driving forces for the occurrence and development of ESCC. Our results provide new insights into the pathogenesis of ESCC, and seven RBPs have potential application value in the clinical prognosis prediction of ESCC.