Single-Cell Multi-Omic Analysis of KMT2A-Rearranged Pediatric Acute Leukemia Clonal Evolution

Advancements in the genetic understanding of pediatric leukemias have significantly enhanced risk stratification and treatment. Specific genetic subtypes correlate with poorer prognoses in both acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). Unlike adult leukemias, pediatric leukemias frequently exhibit oncogenic fusions as key molecular drivers. Notably, rearrangements in the KMT2A gene are prevalent in a subset of ALL and AML, most commonly in infant leukemia, and are consistently linked to poor outcomes across all disease subtypes. KMT2A fusions are typically detected through cytogenetic analysis and reverse transcriptase PCR (RT-PCR) targeting common fusion partners. Methods to identify fusions, however, are not integrated with bulk sequencing platforms used to uncover additional genetic mutations in leukemic cells, leading to a fragmented understanding of the clonal architecture in KMT2A-rearranged (KMT2Ar) leukemias.

Large-scale DNA sequencing studies have identified substantial divergences in the mutational landscape of pediatric and adult AML, suggesting distinct clonal evolution and mutational synergies in adult and pediatric leukemia. Previous work has elucidated the clonal architecture of adult myeloid malignancies through single-cell multi-omic analysis of patient samples (Miles, L.A. et al, 2020; Morita, K. et al, 2020). While studies utilizing single-cell multi-omic sequencing have uncovered critical transcriptional alterations in pediatric leukemia, none have utilized the combination of simultaneous molecular profiling and gene expression at the single cell level. This technical gap has precluded an understanding of the clonal architecture of pediatric leukemia and the impact of specific co-mutations on transcriptional regulation.

To investigate the mutational landscape of pediatric KMT2Ar leukemia and the impact of specific co-mutations, we designed and optimized a novel, custom multi-omic amplicon panel that simultaneously detects KMT2A-fusion products, somatic co-mutations, and targeted gene expression. The custom DNA panel provides coverage of 197 amplicons spanning 22 genes represented in KMT2Ar leukemias in our biorepository. This panel is combined with an RNA panel targeting 5 KMT2A fusion partners in addition to 6 transcripts previously shown to be important in KMT2Ar leukemias. We first optimized the DNA + RNA multi-omic workflow using this custom multi-omic panel on a 1:1:1 mixture of 3 established leukemia cell lines (2 KMT2Ar, 1 KMT2A WT). We found that our panel could successfully delineate amongst all 3 cell lines through the identification of KMT2A-fusion specific transcripts as well as known co-mutations. Moreover, we observed increased expression of specific KMT2A targets in our KMT2Ar cell lines, including HOXA9, HOXA10 and MEIS1 through the targeted single cell gene expression analysis, which correlated with trends we observed by bulk qPCR. Surprisingly, we found that the expression levels of these targets varied significantly within cells from the same cell line, highlighting the potential heterogeneity that is uncovered through single cell analysis. We also determined the limit of detection for the KMT2A-fusions through sequencing of serial dilution mixtures of the KMT2Ar cell lines in the WT cell line. We next performed single cell DNA + RNA sequencing on primary patient samples from pediatric patients with KMT2A-rearranged ALL (n = 3), AML (n = 6), and mixed phenotype acute leukemia (MPAL; n = 2). These samples consist of single timepoint samples as well as longitudinal samples while patients were undergoing therapy. Single-cell analysis revealed unique fusion and genotype specific alterations in clonal architecture and target gene expression. In conclusion, our study optimizes and utilizes simultaneous single cell molecular profiling, fusion identification, and targeted transcriptomic sequencing to delineate the clonal architecture and uncover relationships between genetic fusions/alterations and transcriptional regulation in KMT2A-rearranged pediatric leukemias.