Jyoti Nangalia, MBBChir of the Wellcome Sanger Institute, Hinxton, United Kingdom, University of Cambridge, Wellcome-MRC Cambridge Stem Cell Institute, Cambridge, United Kingdom, and Cambridge Universities NHS Trust, Cambridge, United Kingdom speaks about the ASH 2020 abstract – LBA-1 Driver Mutation Acquisition in Utero and Childhood Followed By Lifelong Clonal Evolution Underlie Myeloproliferative Neoplasms.
Context:
However the timing of driver mutations and the dynamics of the clonal expansion remain largely uncertain, but repeated mutations in cancer-associated genes drive tumour outgrowth. Philadelphia-negative myeloproliferative neoplasms (MPN) are distinct cancers that trace the earliest stages of tumorigenesis through the evolution of the disease. Most patients have JAK2V617F, which is present as the sole driver mutation or is present in genes such as DNMT3A or TET2 in combination with driver mutations. We aimed at identifying in adult MPN the timing of driver mutations and clonal dynamics.
Method All-genome sequencing of single-cell dependent hematopoietic colonies (n=952) was performed in conjunction with selective resequencing of longitudinal blood samples from 10 patients with MPN presenting with the disease between 20 and 76 years of age. We established 448,553 somatic mutations which were used to recreate phylogenetic trees of hematopoiesis, tracing blood cell lineages back to embryogenesis. We timed the acquisition of driver mutation, characterized the dynamics of tumour evolution, and measured the rates of clonal expansion over a patient lifetime. Bulk blood sample resequencing corroborated clonal trajectories and produced population estimates.
Outcomes:
In all the patients in whom JAK2V617F was the first or only driver mutation, JAK2V617F was acquired in utero or in childhood. The earliest estimates of age were within a few weeks of conception, and, despite broad-ranging MPN presentation ages, the upper estimates of acquisition age were between 4.1 months and 11.4 years. The mean latency was 34 years between the acquisition and clinical presentation of JAK2V617F (range 20-54 years). The subsequent acquisition of driver mutations, including those for JAK2V617F, was separated by decades. The disease latency was still 12-27 years following the acquisition of JAK2V617F as a second driver case. Commonly associated with age-related clonal hematopoiesis (CH), DNMT3A mutations occurred as the first driver occurrence, following mutated-JAK2, and as separate clones representing CH in patients with MPN. In utero or childhood, DNMT3A mutations were also first acquired in 4 patients at the earliest 1.2 weeks after conception and the latest 7.9 weeks of gestation to 7.8 years.
The discovery of identical genetic modifications repeatedly occurring in unrelated clones within the same patient has been a recurring feature of the clonal landscape in MPN. For chr9p loss-of-heterozygosity, chr1q+ and mutations in myeloid cancer genes, such ‘parallel evolution’ are observed, indicating that patient-specific factors flavor selective landscapes in MPN. However, mutant clones, especially those with mutant-JAK2, acquired 1.5-5.5 excess mutations per year and had shorter telomeres, suggesting increased cell divisions during clonal expansion. However, normal hematopoietic stem cells accumulated ~18 somatic mutations per year.
We modelled the clonal expansion rates and found that they differed greatly, both across patients and within people. In one patient, a DNMT3A-mutated clone acquired in utero steadily expanded at <10% /year, taking 30 years to achieve a 1% clonal fraction, while a clone with mutated-JAK2,-DNMT3A and-TET2 expanded at >200% /year, doubling in size every 7 months. JAK2V617F also extended variably across patients as a single driver mutation, demonstrating that other factors that may include germline, cytokine, or stem cell variations between individuals also affect driver mutation selection. Clonal expansion rates associated with JAK2V617F in MPN were greater than those recorded for JAK2-CH. In addition, cohort expansion rates projected time to clinical presentation, more so than the age of acquisition of mutation or tumor burden at diagnosis. This indicates that clonal expansion rates of JAK2-mutant dictate both whether and when there are clinical manifestations. One to four decades before the clinical presentation, driver mutations and rates of clonal expansion may have been observable in blood.
Findings:
MPN comes from the acquisition of driver mutation very early in life, even before birth, with life-long clonal expansion and evolution, creating a new model for the development of blood cancer. Together with the determination of clonal expansion rates, early detection of mutant-JAK2 could provide opportunities for early interventions aimed at reducing thrombotic risk and targeting the mutant clone in individuals at risk.