Alternatively spliced cell surface proteins as drivers of leukemogenesis and targets for immunotherapy

Project Summary While impressive progress has been made in tumor immunology and clinical application of immunomodulatory agents in solid tumors, we don’t fully understand how to effectively incorporate immunotherapeutic strategies in Acute Myeloid Leukemia (AML) (Perna F et al., Cancer Treat Res 2022). Advances in genomic and epigenomic

characterization of AML have fostered better understanding of leukemogenesis. The average AML genome may have only 13-16 coding mutations, of which around five are recurrent mutations in suspected driver genes. Despite approval of novel targeted therapies, the clinical outcome of AML patients remains dismal, with a 5-year

overall survival of approximately 27%, prompting the search for additional and synergistic therapeutic rationales. Developing immune-based therapies for AML has been challenged by the lack of surface targets. I previously developed a target discovery strategy to identify Chimeric Antigen Receptor (CAR) targets in AML (Perna F. et

al., Cancer Cell 2017). As we could not identify single targets with favorable expression profiles, we showed that combinatorial pairings hold promise for CAR T-cell therapy of AML. Over the past two years, my lab investigated whether cancer-specific mechanisms such as alternative mRNA splicing driven by founder mutations (i.e. genetic mutations affecting splicing factors) may shape the

cancer surface proteome, providing targets for promoting disease initiation and progression and use of immunotherapy (Perna F. Molecular Therapy 2021 and Dong et al., Oncogene 2021). We have now developed a novel pipeline that we called Spliced-ImmuneTargetFinder based on 5 steps to identify splice variants of cell

surface molecules: 1) RNA-seq analysis of normal cord blood CD34+ HSPCs virally expressing splicing factor mutations such as SRSF2P95H mutation 2) custom transcriptome analysis for isoform quantification 3) isoform switch analysis and prediction of functional consequences based on isoform sequences such as coding potential

and domain gain 4) cell surface molecule annotation based on an integrated dataset developed in our laboratory, and 5) protein candidates validation in primary AML patient samples beyond SRSF2 mutant patients. Given that, we identified six top candidate targets that are a) involved in at least one switch isoform b) upregulated in SRSF2

mutant cells compared to controls and c) derive from genes coding for cell surface proteins. We validated their aberrant protein expression in several AML cell lines and primary AML patient samples by developing custom antibodies specifically recognizing isoform unique domains derived from splicing events.

Thus, we propose to investigate the functional roles of these promising variants and the path for targeting the extracellular isoform-specific domains with immune-therapeutic interventions. This method for unconventional target antigens discovery is generalizable to other cancers as well and will provide useful

information to prepare for a phase-I clinical trial in AML patients. Identification of these attractive antigens may also open new avenues for understanding the role of splice isoforms in leukemogenesis.