Genetic basis and molecular profiling in myeloproliferative neoplasms

Abstract

BCR::ABL1-negative myeloproliferative neoplasms (MPNs) are clonal diseases originating from a single hematopoietic stem cell that cause excessive production of mature blood cells. The 3 subtypes, that is, polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF), are diagnosed according to the World Health Organization (WHO) and international consensus classification (ICC) criteria. Acquired gain-of-function mutations in 1 of 3 disease driver genes (JAK2, CALR, and MPL) are the causative events that can alone initiate and promote MPN disease without requiring additional cooperating mutations. JAK2-p.V617F is present in >95% of PV patients, and also in about half of the patients with ET or PMF. ET and PMF are also caused by mutations in CALR or MPL. In ∼10% of MPN patients, those referred to as being "triple negative," none of the known driver gene mutations can be detected. The common theme between the 3 driver gene mutations and triple-negative MPN is that the Janus kinase-signal transducer and activator of transcription (JAK/STAT) signaling pathway is constitutively activated. We review the recent advances in our understanding of the early events after the acquisition of a driver gene mutation. The limiting factor that determines the frequency at which MPN disease develops with a long latency is not the acquisition of driver gene mutations, but rather the expansion of the clone. Factors that control the conversion from clonal hematopoiesis to MPN disease include inherited predisposition, presence of additional mutations, and inflammation. The full extent of knowledge of the mutational landscape in individual MPN patients is now increasingly being used to predict outcome and chose the optimal therapy.

Graphical abstract

Figure 1.

Reconstructing the timeline of clonal evolution in myeloproliferative neoplasms. Schematic drawing of the timeline (not to scale) and the different hematopoietic compartments during the clonal evolution of MPNs caused by JAK2-p.V617F. Starting in embryogenesis with the first division of an ancestral HSC, the daughter cells acquire a number of mutations in their genomes that can be used as markers to distinguish them from all other cells that underwent cell division. New mutations are added during each of the next cell divisions, and by comparing the sequence similarities and differences of individual HSCs, a phylogenetic tree can be reconstructed. This analysis relies on taking bone marrow cells after MPN has been diagnosed and depositing single HSCs or progenitor cells into wells where they can be expanded in liquid culture to obtain single cell–derived colonies. DNA from each of these colonies is then analyzed by WGS. By comparing the sequence similarities and differences of individual HSCs, a phylogenetic tree can be reconstructed that originates in a common ancestor HSC that first divided at the time of gastrulation. The estimate of the time when JAK2-p.V617F mutation was acquired is calculated by assuming a constant mutational rate of 19 mutations per HSC per year. This estimate is not confounded by the increased cell division rate of JAK2-p.V617F mutant HSCs, because only sequence alterations that occurred before JAK2-p.V617F was acquired were used for deriving the estimate.

Figure 2.

Evolution of MPNs originating from a single HSC that acquired a disease driver mutation. Model summarizing the events from the acquisition of JAK2-p.V617F until the development of MPN disease. The early events occur inside the bone marrow. A single HSC with JAK2-p.V617F can divide to yield 2 HSC daughter cells that carry the mutation, which leads to persistence, and later limited expansion of the mutated HSCs. Alternatively, the mutated HSC can differentiate into committed progenitors that produce a wave of mutant hematopoietic cells, but eventually are exhausted, due to loss of stemness. The mutant HSCs can also be eliminated at this early stage by cells of the immune system. During this phase, cells carrying JAK2-p.V617F are not yet detectable in peripheral blood (“pre-CHIP phase”). After expansion of the mutated HSCs and with a latency of years or decades, mature hematopoietic cells carrying JAK2-p.V617F are produced that become detectable in peripheral blood as “clonal hematopoiesis of undetermined potential (CHIP).” In only a minority of cases, the JAK2-p.V617F mutant HSC clone expands and produces committed progenitors that become dominant in bone marrow and can be diagnosed as MPN with elevated blood counts in peripheral blood. The factors favoring this conversion from CHIP to MPN are listed under the red arrow. The size of the JAK2-p.V617F mutant HSC clone can be reduced by interferon-α (IFNα), which acts by pushing the mutant HSCs into the cell cycle and thereby exhausting them,

Figure 3.

Frequencies of germline variants and the associated likelihood for developing a phenotype. Graph depicting the frequencies of germline gene mutations in the general population (x-axis) and the likelihood that they will promote the manifestation of MPN or MPN-like disease (y-axis). The highest penetrance (close to 100%) is observed in rare families with erythrocytosis or thrombocytosis caused by mutations in a single gene that are most frequently inherited as a Mendelian trait with autosomal dominant transmission (EPOR, EPO, THPO, and MPL, etc), or autosomal recessive transmission (Chuvash polycythemia due to mutations in VHL). Rare familial cases with inherited predisposition to MPN require the acquisition of somatic disease driver mutations for disease manifestation, most frequently JAK2-p.V617F, or mutations in CALR or MPL. The predisposing germline mutations are inherited as autosomal dominant traits with reduced penetrance, in most cases ∼20% to 40%. Inherited predisposition to MPN due duplication of chr.14q32 is an exception, reaching higher penetrance around 80%. Finally, common polymorphisms found at high frequencies in the general populations located in, for example, the JAK2, MECOM, and TERT genes, are associated with a much weaker predisposing effect and also require the acquisition of somatic disease driver mutations for MPN.

Figure 4.

Correlation plot showing the likelihood of co-occurrence of gene mutations in the same patient. The association between gene mutations is based on the compilation of data from 5 published cohorts comprising a total of 3002 MPN patients, including 1497 ET, 535 PV, and 970 PMF or secondary MF patients.^,,, The color code and numbers on the y-axis represent the Pearson R-coefficient, which indicates the strength of positive or negative associations between 2 gene mutations. Only the significant associations are depicted (P-value < .01).