• Morgan received her Ph.D. in Pharmaceutical Sciences from the University of Montréal, after which she was a postdoc in the Department of Organismic and Evolutionary Biology at Harvard University. Her research focuses on the development of predictive, mechanistic models in a variety of disease contexts to identify pathophysiological mechanisms and tailor therapeutic regimens according to patient-specific characteristics.

Current Work

• Pathological hematopoietic stem cell dynamics: In healthy individuals, approximately a trillion blood cells are produced each day through a tightly controlled process in which hematopoietic stem cells (HSCs) produce lineage restricted progenitor cells that in turn generate fully differentiated, mature blood cells. When disrupted, immature myeloid cells can accumulate in the blood and bone marrow due to their uncontrolled proliferation. This accumulation is a hallmark of leukemias, including acute myeloid leukemia (AML).
  
  Our work combines stochastic and deterministic approaches to hematopoiesis, with PK/PD drug models to understand how clones are generated during (pre-)leukemia. Together with collaborators, we explore these dynamics in ex vivo xenotransplants to provide a comprehensive picture of the changing clonal HSC landscape.

• Cancer heterogeneity and treatment resistance: Heterogeneity (in solid tumours and within the hematopoietic system) is a barrier to anti-cancer therapy success, and complicates clinical care strategies. Together with experimental and clinical collaborators, we work to understand the mechanisms of drug tolerance within heterogenous tumour and/or blood cell populations. In turn, the processes we uncover are used to modify current therapeutic designs to improve patient outcomes.

• Reconstructing immune networks: Disordered hematopoietic conditions, like cyclic neutropenia and cyclic thrombocytopenia, give us a window into the multitude of control networks that regulate the production of blood cells. Using data from individuals with perturbed hematopoiesis and applying dynamical systems and statistical techniques like convergent cross mapping and periodogram analysis, we reconstruct immunological networks of cytokines and blood cells. Clustering and threshold measures allow us to zoom in on the "hubs" that control hematopoiesis to give us a clearer picture of how hematopoiesis is regulated at homeostasis.

• Identifying pathophysiological mechanisms and personalizing therapies through in silico clinical trials: Modern treatments frequently combines multiple drugs. For example, combination chemotherapy can target different mechanisms of action against cancerous cells, and HAART integrates different classes of antiretrovirals to best control viral loads and disease symptoms. Unfortunately, combination therapy can carry a high therapy burden and may increase overall toxicity. Running clinical trials to test different combination therapies is a long and expensive process. Overall attrition along the drug development pipeline is high for a variety of reasons, including trial failures.

We have developed an in silico clinical trial platform to efficiently test different drug combinations and treatment regimens before clinical trials are run. Our approach puts together our various mechanistic disease and quantitative systems pharmacology models in a rational, quantitative approach to therapy scheduling and optimization that allows us to tailor or personalize regimens to patient cohorts or individuals.

• COVID-19: We have developed a range of quantitative tools with which we can interrogate SARS-CoV-2 infection and COVID-19 clinical manifestation. We focus particularly on within-host immunological models to identify pathophysiological mechanisms leading to severe COVID-19, including differential responses to viral variants. By improving our understanding of SARS-CoV-2 infection and immune responses to novel coronavirus, our results rationalize the search for candidate therapies and the preclinical decision making process, contributing to reducing clinical trial failures and bettering patient outcomes.

• Studying viral respiratory diseases using virtual patient cohorts: We use mathematical and computational models to describe the within-host immunological responses to viral respiratory diseases like flu and SARS-CoV-2. Our approach relies on virtual patient cohorts that are generated using experimental and clinical data, with each virtual patient selected based on realistic disease trajectories. Since virtual patient cohorts are identical, it is possible to establish causal relationships between changes in treatment strategy and disease status and to find mechanisms driving severe outcomes.

Tags: Vaccine, Vaccines, Vaccination