Longitudinal transcriptomic study of relapsing T-cell acute lymphoblastic leukaemia

T-cell Acute Lymphoblastic Leukemia (T-ALL) is an hematological malignancy caused by transformation of immature T cells. Despite current treatment regimes allowing stable remission of the disease in most of the cases, some patients experience relapse and have a poorer prognosis. In epidemiological terms, T-ALL occurs in 15% of pediatric patients and 25% of adult patients with Acute Lymphoblastic Leukemia. T-ALL is a clinically and biologically-molecularly heterogeneous tumor caused by high clinical and molecular heterogeneity, based on genetic alterations, epigenetics modifications and gene expression regulation in order to interfere with thymic T cells maturation. In this thesis, I've studied a 19 patients’ cohort with T-ALL, 7 adults and 12 children, who had a relapse after therapy and for whom we obtained samples adequate both at diagnosis (TALL-D) and at relapse (TALL-R). The samples were profiled with high-depth Illumina RNA-seq, starting from ribodepleted total RNA, to conduct transcriptome analysis. RNA-seq data of purified populations of thymocytes at different maturation stages, from more stem to the so-called “double positive” stage, CD4+ and CD8+ were used as normal counterparts (THYM). RNA-seq data were analyzed using CircComPara2 pipeline for gene identification, annotation and quantification. Downstream analyses were performed in R/Bioconductor, with data import and normalization (TPM) using txi. Differential expression analysis was conducted with DESeq2. Functional and pathway enrichment analyses were carried out using GSEA (GO for ontology, signaling pathway database KEGG, and MSigDB, molecular signatures) and SPIA. Preliminary exploratory analyses included Relative Log Expression (RLE) plots and Principal Component Analysis (PCA). Setting corrected p-value < 0.01, in TALL-D vs THYM comparison, we found 5,053 up-regulated and 2,994 down-regulated genes, and in TALL-R vs THYM 5,224 up-regulated and 2,954 down-regulated genes. As regards the direct comparison between TALL-R and TALL-D, the overall number of differentially expressed genes is lower than in the two previous comparisons, 334 up-regulated genes and 581 down-regulated genes. This result was expected because relapse disease derives from evolution of previous maturation steps. In both T-ALL disease versus control, clear gene transcription reorganisation, characterised by chromatin modulation, chromosome separation reorganisation, DNA replication, and homologous recombination aberrancies emerged. Suppressed pathways were associated with the body's inflammatory response, tumor drug resistance, alterations in cellular metabolism, and signalling pathway activation. At diagnosis, accelerated biosynthesis of proline-associated tRNA linked to improved survival and proliferation of T-ALL tumour cells, at relapse, ribosomal biosynthesis is more activated, suggesting efficient bioactivity of chemo-resistant leukaemia cells. I also performed analyses using SPIA, based on gene networks whose topology is derived from pathways. At diagnosis, there is a notable alteration in the NOTCH pathway involved in cell replication and signalling, and at relapse, in the MAPK pathway, a key genetic cascade involved in cell signalling and the uncontrolled proliferation of tumour cells. In conclusion, we highlighted distinct alterations in cellular and molecular processes,in the two T-ALL stages studied. Further insights can be obtained by experimental studies and using the same RNA-seq data for the analysis of circular RNAs, emerging molecules for the development and maintenance of cancer.