Genomic Profiling of T-Cell Acute Lymphoblastic Leukemia/Lymphoma (T-ALL) for Identification of Driver Aberrations Using Combined Next Generation Sequencing and Optical Genome Mapping in Clinical Diagnostic Laboratory

Background. T-cell acute lymphoblastic leukemia (T-ALL) is caused by the accumulation of genomic alterations that disrupt the development and proliferation of T-cells. These include somatic mutations in multiple genes, with NOTCH1 seen in about 50% of patients. In addition to mutations, driver chromosomal rearrangements, such as those affecting BCL11B (14q32) form the basis of sub-classification per latest WHO (5^th ed) and ICC classifications and treatment decisions. In addition, ICC has proposed multiple provisional entities based on the underlying driver genomic rearrangements, often affecting T-cell receptors (TRB/TRG) with genes such as TAL1/3, TLX1/3, LMO1/2, KMT2A rearrangements, PICALM::MLLT10, SET::NUP214 among others. Consequently, clinical genomic profiling and work-up of T-ALL cases requires the use of advanced, high-throughput genome-wide techniques for detecting both somatic mutations and chromosomal structural variants. NGS offers comprehensive mutational profiling, detecting point mutations, insertions, deletions, and copy number variations at a nucleotide level. Optical genome mapping (OGM) is a DNA-based technique, that enables genome-wide structural variant analysis at a high-resolution, including those alterations often missed by conventional cytogenetics. Integrating OGM and NGS in the diagnostic workflow allows for a more detailed and accurate genomic characterization of T-ALL, facilitating better classification and diagnosis. In this study, we present the results of targeted NGS and OGM data from our clinical lab in a large cohort of T-ALL patients.

Methods. All patients diagnosed with T-ALL and who underwent bone marrow (BM) examination were identified from the medical records. Somatic mutation analysis was performed on BM aspirate samples using a clinical-grade 81-gene panel through next-generation sequencing (NGS). All patients underwent chromosome banding analysis (CBA). A subset of cases underwent OGM using ultra-high molecular weight DNA to identify driver genomic aberrations.

Results. Our cohort included a total of 125 patients diagnosed with T-ALL [63 (50.4%) treatment-naïve; 62 (49.6%) previously treated]; 31% also met the criteria for early T-cell precursor acute lymphoblastic leukemia (ETP-ALL). The median age of the patients was 36 years, with 94 men (75.2%) and 31 women (24.8%).

By NGS, 98 (78.4%) patients had at least 1 mutation. The number of mutations per case ranged between 1 and 14. The most frequent mutations observed were in NOTCH1 (50.4%), PHF6 (24%), JAK3 (19%), DNMT3A (18%), NRAS (17%), and TP53 (17%). The majority of these mutations were missense, while frameshift and nonsense mutations occured less frequently. The highest median variant allele frequency (VAF) was seen in PHF6 (64%), followed by WT1, FBXW7, JAK3 and DNMT3A, all of which showed median >30%. The median NOTCH1 VAF was 22%. NOTCH1 mutations aggregated in the C-terminal PEST domain (nonsense or frameshift), and NOD/NODP domains (missense).

CBA showed normal karyotype in 50 (40%) and abnormal karyotypes in 75 (60%) patients; complex karyotype was seen in 14 (11.2%) cases. Due to the limitations of CBA in T-ALL, primarily due to the poor growth of T-ALL cells under in vitro culture and the cryptic nature of most abnormalities, we performed OGM on 18 cases. Seventeen (94%) patients showed Tier 1/2 cytogenetic abnormalities. In 14 (78%) patient, a putative gene fusion implicated in T-ALL pathogenesis could be identified. These included known driver rearrangements involving BCL11B [TLX3::BCL11B (n=3), t(14;16) and t(2;14;10)], ETV6 [ETV6::MN1, CCDN2::ETV6, HOX transcription factors [TRB::HOXA13; fus(7;7), PICALM::MLL10, KMT2A::MLLT1, SPTAN1::NUP214], as well as novel fusions, such as NUP98::YY1, CBL [t(1;11)]. Among copy number changes, CDKN2A/B deletions were most frequent, seen in 5 patients. Four patients showed chromoanagenesis indicating a genomic catastrophe resulting in multiple chromosomal rearrangements and copy number changes.

Conclusions. Combining data from both NGS and OGM, enabled accurate sub-classification of T-ALL cases, including those into the provisional entities defined by the ICC. This study demonstrates the feasibility and effectiveness of using genomic analysis (NGS and OGM) to fully understand the major regulators of leukemia cell growth and metabolism in treating T-ALL.