In 1892, the French cardiologist Louis Henri Vaquez (1860-1936) made a pioneering observation, detailing the case of a 40-year-old man with strikingly high red blood cell count, enlarged spleen, and a distinctive ruddy complexion - "a special form of persistent polycythaemia, associated neither with the cyanosis of high altitude nor of congenital heart disease". This case report marked the first clinical description of what we now recognize as a myeloproliferative neoplasm (MPN). Since Vaquez's pioneering work, our understanding of MPNs has dramatically evolved, leading to the integration of phenotypic assessment and molecular profiling in diagnostics.

So, what is MPN?
Myeloproliferative neoplasms (MPNs) are a group of hematologic malignancies characterized by the clonal proliferation of one or more myeloid cell lineages in the bone marrow, leading to an increase in red blood cells, white blood cells, or platelets in the peripheral blood. These disorders share a common pathophysiological basis rooted in the dysregulation of signaling pathways that control hematopoiesis. The latest WHO classification delineates these disorders into distinct categories, including chronic myeloid leukemia (CML) and the BCR::ABL1-negative entities: polycythemia vera (PV), primary myelofibrosis (PMF), and essential thrombocythemia (ET).

How common is MPN?
The incidence of MPNs is estimated to be around 1 to 2.5 cases per 100,000 people annually. This includes all MPN subtypes, with PV being the most common, followed by ET and PMF. Incidence rates across North America, Europe, and Australia show remarkable consistency. However, reports from Asian countries indicate lower incidence rates of MPNs compared to those in Western regions. Data on MPN incidence in Africa and South America are scarce, likely due to challenges in diagnosis and case reporting.    

What is the role of genetics in MPN classification?
The discovery of the JAK2 V617F mutation in 2005 marked a pivotal moment in the understanding of MPNs, providing the first molecular signature common to several MPN subtypes and highlighting the role of aberrant JAK-STAT signaling in their pathogenesis. Subsequent research has identified additional driver mutations, such as those in the CALR and MPL genes. The most commonly identified co-mutations are ASXL1, TET2, SRSF2, and EZH2, further elucidating the genetic landscape of these diseases. The newest WHO guidelines now integrate genetic information alongside traditional clinical and hematological criteria. The integration of genetics into the classification of MPNs paves the way for more personalized medicine. It not only facilitates a more nuanced understanding of these diseases but also directly impacts patient care by enabling more accurate diagnoses, refined prognostic assessments, and personalized therapeutic approaches. 

Why did we work on a genetic-based classification approach?
Despite recent advances, overlaps, borderline findings or transitions between MPN subtypes occur and incomplete clinical data often complicates diagnosis. Moreover, the mechanisms contributing to MPN initiation, progression, and transformation into more aggressive forms, such as acute myeloid leukemia (AML), are still not completely understood and remain areas of active investigation. The aim of this study was to overcome diagnostic ambiguity by using genetic markers for the stratification of MPN entities and, hence, to also provide prognostic information based on a patients’ genetic profile. We further aimed to thoroughly characterize MPN cases with progression to blast phase (BP) to provide guidance for a better upfront genetic risk stratification and therapeutic management.

Cohorts & Methods:

The study entailed comprehensive analyses across various cohorts, starting with a genome-wide characterization of MPN cases by whole genome transcriptome analysis. This analysis paved the way for the subsequent generation and validation of a machine learning model based on genetic features. This model was further distilled into a streamlined decision tree for routine clinical application (refer to Figure 1A, left). Additionally, the research included an examination of whole exome data for MPN cases that progressed to the blast phase, enriching our understanding of disease evolution (see Figure 1A, right). Figure 1B provides an overview and succinct description of the analyzed cohorts on the left, while a Venn diagram on the right illustrates the overlap of patients among the different cohorts.

The key findings of our work:

• Expression levels of 18 subgroup-specific genes effectively stratify MPN patients into diagnostic groups.
• A set of 12 genetic markers accurately classifies chronic phase MPN patients as per WHO guidelines.
• Molecular markers enable genetic classification of MPN-NOS patients and identification of high-risk individuals without traditional diagnostic criteria.
• Paired exome analysis shows most genetic variants in MPN-CP and MPN-BP are patient-specific, likely germline, and not disease progression markers.
• During progression to blast phase, one-third of MPN-CP patients lose their driver-gene mutation, while splicing and chromatin modifying gene mutations remain stable, suggesting a common origin of chronic and blast phases with distinct progression paths.
• Mutations in SRSF2, TET2, and RUNX1 signal the transition to blast phase, with TP53 mutations also common during this shift.
• The study reveals MPN's clonal hierarchy is diverse, with independent clones potentially driving leukemic transformation alongside the primary MPN clone.

Limitations of our work:

• Limited availability of histopathology data for PMF diagnosis.
• Incomplete OS data for the different cohorts used.

Take home message: 
The genetic landscape of MPNs is heterogenous and expanding genetic analysis beyond JAK2, CALR, and MPL at diagnosis is crucial for accurate MPN classification, early high-risk patient identification, and timely intervention.