Researchers Identify Genetic Prognostic Predictors in Myeloproliferative Neoplasms

A team of European and American researchers have identified and characterized associations between specific genetic mutations and prognosis in patients with myeloproliferative neoplasms (MPNs), findings that suggest a tailored prediction model based on genomic associations could predict outcomes and improve care in this population.

Lead author Jacob Grinfield, MBChB, and co-authors published their findings in the New England Journal of Medicine. The development of a genomics-guided prediction model could overcome some of the challenges in identifying patients who would benefit from targeted therapies, the authors wrote. “A genomic classification has the virtue of identifying patients with shared causative biologic factors, is stable over time, and does not rely on blood-count thresholds for assigning particular disease labels.”

The investigators collected tumor DNA samples derived from bone marrow mononuclear cells, blood granulocytes, or whole blood from all participants.

The final analysis included 2,035 patients; targeted sequencing for the full coding sequence of 69 genes and genome-wide copy-number information was performed in 1,887 patients, while 148 patients underwent whole-exome sequencing. Patients presented with the following diagnoses:

  • essential thrombocythemia (n=1,321)
  • polycythemia vera (n=356)
  • myelofibrosis (n=309)
  • other diagnoses of MPNs (n=49)

Thirty-three genes had driver mutations in at least five patients. Mutations in JAK2, CALR, or MPL were the sole abnormality in 45 percent of the patients. A total of 1,075 driver mutations were identified across other genes. Age and advanced disease were associated with an increase in the number of driver mutations.

MPNs are currently classified by clinical and laboratory criteria such as those included in the World Health Organization diagnostic classification, but, the authors noted, “the biologic factors underlying these distinctions are incompletely understood.” In this analysis, the number of driver mutations differed according to MPN subtype.

Germline predispositions to MPN subtypes also differed. For example, patients with the JAK2 46/1 haplotype were more likely to have polycythemia vera (odds ratio [OR] = 2.3; 95% CI, 1.7-3.3; p=0.004).

The authors were able to define eight genomic subgroups that showed distinct clinical phenotypes, including blood counts, risk of leukemic transformation, and event-free survival. One of the most striking prognostic factors was TP53 mutation status. Compared with patients with TP53 wild-type disease, those with TP53 mutations were more likely to transform to acute myeloid leukemia (AML; hazard ratio [HR] = 15.5; 95% CI, 7.5-31.4; p<0.001) and die early (HR=2.4; 95% CI 1.6-3.6; p<0.001).

Patients with MPL-mutated myelofibrosis were more likely than patients with JAK2-heterozygosity to progress to AML (HR=8.6; 95% CI 1.4-49.1; p=0.02). Similarly, a subgroup of patients with NFE2 mutations and patients with polycythemia vera had higher likelihood of myelofibrosis transformation (HR=3.0; 95% CI 1.3-6.6; p=0.007).

Patients with the following characteristics also had higher risks of transformation to myelofibrosis (HR=5.4; 95% CI 2.7-11.0; p<0.001) and shorter event-free survival (HR for disease progression or death = 2.6; 95% CI 2.1-3.2; p<0.001):

  • ≥1 mutation in 16 myeloid cancer genes (e.g., chromatin and spliceosome regulators)
  • loss of heterozygosity at chromosome 4q
  • abnormalities in chromosomes 7 and 7q

“Our model accurately identified a minority of patients with chronic-phase MPNs who were at substantial risk for disease progression,” the researchers wrote. “Such patients could be considered for clinical trials of new therapeutic agents, since they are the most likely to benefit and the trials would be more efficient if higher-risk patients are preferentially enrolled.”

A subgroup that included patients without identified drivers “had particularly benign outcomes,” the authors reported. “Only one patient (0.5%) had myelofibrosis transformation and two (1%) had AML transformation during a median follow-up of eight years.” For these patients, “experimental therapy would be unnecessary, and a conservative treatment strategy that is based on cytoreduction and reduction of vascular risk will suffice to give long-term, event-free survival,” the investigators recommended.

According to the researchers, the observed and predictive outcomes in this study were associated with good correlation in an internal cross-validation of a training cohort of 515 patients with MPNs, as well as in an independent external cohort of 270 patients with MPNs.

“Comprehensive gene sequencing of patients with blood cancers is becoming increasingly accessible and routine,” the investigators concluded, allowing for the integration of clinical data with genomic-profiling data. “Regarding patients with MPNs, such information will empower the clinician and support complex decisions around the choice and intensity of therapy, recruitment into clinical trials, and long-term clinical outlook.”

The authors of the paper reported financial relationships with the Wellcome Trust, Wellcome-MRC Stem Cell Institute, National Institute for Health Research Cambridge Biomedical Research Centre, Cancer Research UK, Bloodwise, the Kay Kendall Leukemia Fund, the Leukemia & Lymphoma Society, the European Hematology Association, the Li Ka Shing Foundation, and the Medical Research Council.


Grinfeld J, Nangalia J, Baxter EJ, et al. Classification and Personalized Prognosis in Myeloproliferative Neoplasms. N Engl J Med. 2018;379:1416-30.