Skip to main content

Coronavirus situation at the University of Eastern Finland

Pipettes.

Identifying targetable genes across hematological malignancies

  • Health and well-being

Advances in sequencing technologies and genome-wide transcriptomic profiling have revolutionized our understanding of genetics and transcriptomics of hematological malignancies. Hematological malignancies have previously been classified by morphology and translocation detection. In the last decade, the detection of somatic mutations has yielded better disease classifications and the identification of patients with adverse outcomes. Various tumorigenic processes have been identified, such as aberrant regulation of hematopoietic or lineage determining transcription factors (TFs) and signaling pathways, activation of proliferative signaling, immune evasion and resistance to apoptosis. Targeting these vulnerabilities is warranted, but heterogeneity in the treatment responses poses a significant challenge. A detailed understanding of the molecular processes is required to identify patients who benefit from the treatment.

Vast amount of genomic data is being generated, but the current limitation is to interpret the genetic and genomic information and translate it into clinically actionable strategies. In the doctoral study by Petri Pölönen, MSc, the molecular subtypes of hematological malignancies were identified and characterized by analyzing nearly 10000 transcriptomes. These diseases and subtypes were analyzed in the context of cell differentiation, pathway activities and immunological profiles to find potential drug targets and to expose vulnerabilities. Furthermore, a comparative analysis was performed between the diseases and normal hematological cell types, to better understand if the identified process is related to cell-of-origin, lineage, certain differentiation state, or is abnormally expressed in disease. Moreover, thousands of samples were analyzed with multi-omic data to link driver genetic, epigenetic and immunological properties to discover the potential mechanism driving the tumorigenic processes.

In study I, the aim was to study to study drugs and their target gene expression across hematological malignancies. A statistical framework was developed to associate drug targets to disease molecular subtypes and compared the gene expression to normal cells. The approach was validated using ex vivo drug screening data and known vulnerabilities. Furthermore, the approach was applied to identify novel surface targets and DPEP1 was found to be specifically expressed in Acute lymphoblastic leukemia (ALL) and is a suitable immunotherapy candidate. Also, it was found that CDK6 could be targeted using Palbociclib in acute myeloid leukemia (AML).

In study II, the researchers comprehensively characterized AML subtypes using multi-omics data sets and integrated these subtypes with ex vivo drug response profiles. They found that AML contains seven molecular subtypes that are characterized by fusion genes and distinct mutational and epigenetic landscape and have different survival. The analysis further dissected normal karyotype samples to four different groups. First, it was discovered that CEBPA mutated samples form a distinct molecular subtype that was characterized by global hypermethylation. Secondly, NPM1 and FLT3 mutated samples form two molecular subtypes that have the progenitor-like and monocyte-like phenotypes. Fourth, myelodysplastic syndrome (MDS) mutations, such as TP53 and RUNX1 and MDS-derived gene expression signature are specific for AML subtype associated with poor prognosis and complex cytogenetics. Furthermore, the identified subtypes were found to have a major impact on drug responses.

In study IV, BCL2-family genes and venetoclax responses were investigated. The analysis revealed that venetoclax is particularly sensitive against AML progenitor-like cells, whereas resistant to monocyte-like cells. It was also discovered that BCL2-family gene expression predicts venetoclax sensitivity.

In study III, the researchers characterized the immunogenomic landscape of hematological malignancies. They discovered that the Infiltration of cytotoxic immune cells is associated with distinct microenvironmental responses and driver alterations in different cancers, such as TP53 in acute myeloid leukemia and DTX1 in diffuse large B cell lymphoma (DLBCL). In AML, MDS-like subtype was associated with high cytolytic infiltration. Also, CIITA epigenetic silencing in NPM1 and IDH1, IDH2 or TET2 mutated FAB M1/M2 AML patients resulted in diminished HLA II expression. In addition, VISTA co-inhibitory receptor is expressed in monocyte-like AML cells. Furthermore, they found that the lymphoma microenvironment cytolytic cell infiltration is correlated to myeloid-cell co-infiltration, chemokine, complement, and IFNγ responsive gene expression. In addition, cancer germline antigens were frequently expressed in multiple myeloma (MM) and were especially common in a subtype that is associated with high proliferation and poor prognosis. Finally, immunologic features were associated with different outcomes and improve the existing prognostic classification in DLBCL and MM.

In conclusion, the cell-of-origin and the differentiation state of the disease strongly influence the disease characteristics and vulnerabilities to certain treatments. Furthermore, our results highlight the importance of in-depth characterization of disease subtypes. Moreover, the study showed that the disease microenvironment and immunogenic properties should be considered when designing novel immunotherapies.

The doctoral dissertation of Petri Pölönen, Master of Science, entitled Identifying targetable genes across hematological malignancies, will be examined at the Faculty of Health Sciences on 21 February 2020. The Opponent in the public examination will be Associate Professor Jüri Reimand of the University of Toronto, and the Custos will be Associate Professor Merja Heinäniemi of the University of Eastern Finland.

Photo available for download at https://mediabank.uef.fi/A/UEF+Media+Bank/35783?encoding=UTF-8