Abstract

Chronic myelomonocytic leukemia (CMML) is an aggressive clonal hematopoietic malignancy hallmarked with monocytosis, cytopenia, and marrow dysplasia. CMML biology remains poorly understood, with a 30% chance of progressing to acute myeloid leukemia, and an overall unfavorable prognosis with a median survival of 34 months, metrics which have not appreciably changed over the last three decades. Population-level variability, here termed intraleukemic heterogeneity (ILH), suggests that there are diverent evolutionary trajectories that shape disease progression. Currently, the only curative treatment option is an allogeneic stem cell transplant, which many patients are not eligible for due to co-morbidities and advanced age. CMML patients over a unique opportunity to better identify molecular and cell-phenotypic predictors of evolution and therapeutic consequences of ILH in hematologic malignancies, because most patients can be longitudinally followed in a treatment naive state before they progress. To develop a toolkit to quantify this evolutionary process, we focused on individual leukemia patients' bone marrow mononuclear cell samples, in comparison to healthy subjects, to statistically describe ILH. We use a generalized diversity measure to quantify ILH, and ffind that it can characterize disease stage at the level of sub-population structures derived from single-cell RNA sequencing. Furthermore, using single-cell, high parameter flow cytometry data, we can create nearest-neighbor graphs in a high-dimensional parameter space of patient cytokine receptor proffiles. Then, we use community detection to determine phenotypically similar cells that reflect biologically meaningful phenotypes. These statistical and computational inferences lay the foundation for novel mathematical modeling techniques that seek to understand potential mechanisms underlying CMML progression and timing of clonal expansions.