Aging of preleukemic thymocytes drives CpG island hypermethylation in T-cell acute lymphoblastic leukemia

Cancer cells display DNA hypermethylation at specific CpG islands in comparison to their normal healthy counterparts, but the mechanism that drives this so-called CpG island methylator phenotype (CIMP) remains poorly understood. Here, we show that CpG island methylation in human T-cell acute lymphoblastic leukemia (T-ALL) mainly occurs at promoters of Polycomb Repressor Complex 2 (PRC2) target genes that are not expressed in normal or malignant T-cells and which display a reciprocal association with H3K27me3 binding. In addition, we revealed that this aberrant methylation profile reflects the epigenetic history of T-ALL and is established already in pre-leukemic, self-renewing thymocytes that precede T-ALL development. Finally, we unexpectedly uncover that this age-related CpG island hypermethylation signature in T-ALL is completely resistant to the FDA-approved hypomethylating agent Decitabine. Altogether, we here provide conceptual evidence for the involvement of a pre-leukemic phase characterized by self-renewing thymocytes in the pathogenesis of human T-ALL.

Chromatin accessibility landscape of pediatric T-lymphoblastic leukemia and human T-cell precursors

We aimed at identifying the developmental stage at which leukemic cells of pediatric T-ALLs are arrested and at defining leukemogenic mechanisms based on ATAC-Seq. Chromatin accessibility maps of seven developmental stages of human healthy T cells revealed progressive chromatin condensation during T-cell maturation. Developmental stages were distinguished by 2,823 signature chromatin regions with 95% accuracy. Open chromatin surrounding SAE1 was identified to best distinguish thymic developmental stages suggesting a potential role of SUMOylation in T-cell development. Deconvolution using signature regions revealed that T-ALLs, including those with mature immunophenotypes, resemble the most immature populations, which was confirmed by TF-binding motif profiles. We integrated ATAC-Seq and RNA-Seq and found DAB1, a gene not related to leukemia previously, to be overexpressed, abnormally spliced and hyper-accessible in T-ALLs. DAB1-negative patients formed a distinct subgroup with particularly immature chromatin profiles and hyper-accessible binding sites for SPI1 (PU.1), a TF crucial for normal T-cell maturation. In conclusion, our analyses of chromatin accessibility and TF-binding motifs showed that pediatric T-ALL cells are most similar to immature thymic precursors, indicating an early developmental arrest.

Single-cell analysis of structural variations and complex rearrangements with tri-channel processing

Structural variation (SV), involving deletions, duplications, inversions and translocations of DNA segments, is a major source of genetic variability in somatic cells and can dysregulate cancer-related pathways. However, discovering somatic SVs in single cells has been challenging, with copy-number-neutral and complex variants typically escaping detection. Here we describe single-cell tri-channel processing (scTRIP), a computational framework that integrates read depth, template strand and haplotype phase to comprehensively discover SVs in individual cells. We surveyed SV landscapes of 565 single cells, including transformed epithelial cells and patient-derived leukemic samples, to discover abundant SV classes, including inversions, translocations and complex DNA rearrangements. Analysis of the leukemic samples revealed four times more somatic SVs than cytogenetic karyotyping, submicroscopic copy-number alterations, oncogenic copy-neutral rearrangements and a subclonal chromothripsis event. Advancing current methods, single-cell tri-channel processing can directly measure SV mutational processes in individual cells, such as breakage-fusion-bridge cycles, facilitating studies of clonal evolution, genetic mosaicism and SV formation mechanisms, which could improve disease classification for precision medicine.

PDX models recapitulate the genetic and epigenetic landscape of pediatric T-cell leukemia

We compared 24 primary pediatric T‐cell acute lymphoblastic leukemias (T‐ALL) collected at the time of initial diagnosis and relapse from 12 patients and 24 matched patient‐derived xenografts (PDXs). DNA methylation profile was preserved in PDX mice in 97.5% of the promoters (ρ = 0.99). Similarly, the genome‐wide chromatin accessibility (ATAC‐Seq) was preserved remarkably well (ρ = 0.96). Interestingly, both the ATAC regions, which showed a significant decrease in accessibility in PDXs and the regions hypermethylated in PDXs, were associated with immune response, which might reflect the immune deficiency of the mice and potentially the incomplete interaction between murine cytokines and human receptors. The longitudinal approach of this study allowed an observation that samples collected from patients who developed a type 1 relapse (clonal mutations maintained at relapse) preserved their genomic composition; whereas in patients who developed a type 2 relapse (subset of clonal mutations lost at relapse), the preservation of the leukemia's composition was more variable. In sum, this study underlines the remarkable genomic stability, and for the first time documents the preservation of the epigenomic landscape in T‐ALL‐derived PDX models.

Identification of a genetically defined ultra-high-risk group in relapsed pediatric T-lymphoblastic leukemia

In the search for genes that define critical steps of relapse in pediatric T-cell acute lymphoblastic leukemia (T-ALL) and can serve as prognostic markers, we performed targeted sequencing of 313 leukemia-related genes in 214 patients: 67 samples collected at the time of relapse and 147 at initial diagnosis. As relapse-specific genetic events, we identified activating mutations in NT5C2 (P=0.0001, Fisher's exact test), inactivation of TP53 (P=0.0007, Fisher's exact test) and duplication of chr17:q11.2-24.3 (P=0.0068, Fisher's exact test) in 32/67 of T-ALL relapse samples. Alterations of TP53 were frequently homozygous events, which significantly correlated with higher rates of copy number alterations in other genes compared with wild-type TP53 (P=0.0004, Mann–Whitney's test). We subsequently focused on mutations with prognostic impact and identified genes governing DNA integrity (TP53, n=8; USP7, n=4; MSH6, n=4), having key roles in the RAS signaling pathway (KRAS, NRAS, n=8), as well as IL7R (n=4) and CNOT3 (n=4) to be exclusively mutated in fatal relapses. These markers recognize 24/49 patients with a second event. In 17 of these patients with mostly refractory relapse and dire need for efficient treatment, we identified candidate targets for personalized therapy with p53 reactivating compounds, MEK inhibitors or JAK/STAT-inhibitors that may be incorporated in future treatment strategies.