Predicting progression in myeloproliferative neoplasm patients by reconstructing the history of disease in each patient
Sahand Hormoz
PhDDana-Farber Cancer Institute
Project Term: October 1, 2023 - September 30, 2026
Blood cancers called myeloproliferative neoplasms occur when one of the blood stem cells picks up a mutation. Some patients stay in the chronic phase of the disease for years whereas others rapidly progress with poor outcome. We recently measured when the cancer mutation first occurs and the rate of expansion of the cancer cells in individual patients. We will develop a method that uses the history of disease in each patient to identify those that are at risk of progression.
Blood is maintained by a population of hematopoietic stem cells (HSCs) that reside in the bone marrow. Certain types of blood cancers, called myeloproliferative neoplasms (or MPNs), occur when one of the HSCs picks up a mutation and thus gains a fitness advantage over the other HSCs. The progeny of the mutated HSC expands exponentially, eventually taking over the stem cell population. Intriguingly, the same mutation can result in drastically different disease dynamics in different patients - some remaining in the chronic phase of MPN for years and being treated with a daily aspirin, while others rapidly progress to more advanced stages of the disease such as acute myeloid leukemia and have poorer outcomes. To understand this heterogeneity, we recently reconstructed the history of expansion of mutated HSCs in individual MPN patients. To do so, we sequenced the genome of single HSCs and reconstructed the phylogenetic tree of the HSCs from the pattern of genetic alterations that they had accrued. Astonishingly, we observed that the JAK2 mutation occurred decades before diagnosis and that the fitness of the mutated HSCs varied significantly across different patients. The objective of this proposal is to carry out this measurement on a large cohort of MPN patients with known outcomes. Whole genome sequencing of hundreds of single cells from the bone marrow is invasive and expensive, however, so we have recently developed a method that accurately infers the age of onset and fitness of mutated HSCs from bulk sequencing of blood samples. This also means we can obtain the disease dynamics regardless of when the measurement is made - be it at the time of diagnosis or when the patient has progressed - effectively carrying out a retrospective study of MPN patients with known outcomes without the need for access to patient samples banked at the time of diagnosis. We will first validate our approach on MPN patients with banked samples (Aim 1). We will then study samples from 200 MPN patients with known outcomes across the US and use the data to construct an accurate prognostic model that relates disease history in a patient to outcome (Aim 2). Clinical intervention after a patient has progressed is simply too late; existing therapies may help stop progression if we can identify which patients are at a high risk of progression and should therefore be treated, despite the risk of side effects. Our approach can also be applied to other types of blood cancers.