Transposable Element Expression Predicts Prognosis in Acute Myeloid Leukemia

Over half of the human genome is comprised of transposable elements (TE). TE have been implicated in cancer pathogenesis. Despite large-scale studies of the transcriptome in cancer, a comprehensive look at TE expression investigating its relationship to various mutations and its role in predicting prognosis has not been performed. We characterized TE expression in 178 adult acute myeloid leukemia (AML) patients using transcriptome data from The Cancer Genome Atlas (TCGA). We identified significant dysregulation of TE, with distinct patterns of TE expression correlated to specific mutations and distinct coding gene networks. TP53 mutated AML had a unique TE expression signature and was associated with significantly suppressed expression of TE and various classes of non-coding RNA. We identified 17 candidate prognostic TE transcripts that can classify AML subtypes as either high or low risk. These 17 TE were able to further sub-stratify low risk AML (based on mutational profile and coding gene expression) into favorable and unfavorable prognostic categories. The expression signature of the 17 TE was able to predict prognosis in an independent cohort of 284 pediatric AML patients, and was also able to predict time to relapse in an independent dataset of relapsed adult cases. This first comprehensive study of TE expression in AML demonstrates that TE expression can be used as a biomarker for predicting prognosis in AML and also provides novel insights into the biology of TP53 mutated AML. Studies characterizing its role in other cancers are warranted.