Poster Session II: Acute myeloid leukemia - Biology & translational research
Intra-tumor heterogeneity is a common feature of many blood cancers and is correlated with poor clinical outcomes. Single-cell DNA sequencing can unambiguously resolve the genetic heterogeneity of tumors missed with traditional bulk next generation sequencing (NGS). Despite the advantages of single-cell DNA sequencing, previous studies have relied upon laborious, expensive and low-throughput approaches that are not scalable for routine analysis of tumors. Additionally, to provide maximum insight into disease progression, clonality of mutations within a tumor would also be correlated with cell lineage.
We sought to demonstrate how single-cell analysis of both cell surface markers as well as variant calling across hundreds of disease relevant loci can provide unprecedented insight into tumor architecture.
To enable high-resolution profiling of intra-tumor heterogeneity, we previously developed a unique microfluidic droplet platform that can accurately identify genome variants at hundreds of disease relevant loci across 10,000 individual cells per sample run. This platform has a rapid, robust and easy to use workflow that, for the first time, enables routine analysis of genetic heterogeneity for hematologic malignancies.
Here, we report our continued work developing and applying targeted single-cell sequencing panels for AML and other hematologic malignancies such as CLL. The high-throughput single-cell approach we developed has consistently revealed rare tumor cells at diagnosis and remission (MRD) that are critically important with respect to therapy response and disease progression. Additionally, for the first time, we also demonstrate integrated multi-omic capability on our platform by enabling single-cell measurement of cell surface protein expression levels together with genomic variant identification in a single NGS readout. In our modified workflow, cell lines, PBMCs or bone marrow aspirates are stained with a pool of monoclonal antibodies (e.g., CD34, CD33, CD19, CD45, CD30 and CD3) that are conjugated with oligonucleotides containing unique protein identifying tags. The sequencing data is analyzed with a bioinformatic pipeline that quantifies antibody levels associated genotyping data from a myeloid specific targeted sequencing panel on a cell-by-cell basis. With this approach the need for upstream flow sorting of cells can often be eliminated and the well-characterized catalog of hematology cell surface markers employed to more directly correlate phenotype and cell-state information of single cells to mutational burden.
The multi-omic capability we present, together with other improvements to the platform, make our technology ideally suited to fully resolve clonal architecture of hematologic malignancies. Correlation of mutation status with phenotype using high-throughput single-cell multi-omics is a powerful approach with the potential to better predict disease progression and inform therapy selection.