Five Decades of Experience With MPNs: Characterizing Survival and Outcomes

A report based on data from more than 3,000 patients with myeloproliferative neoplasms (MPNs) evaluated at the Mayo Clinic in Rochester, Minnesota, over 50 years found that even patients with low-risk disease have inferior survival compared with the general population – despite the perception of some MPNs as “indolent” diseases. The results, which represent “the most mature survival and outcomes data in MPNs to date,” according to lead author Natasha Szuber, MD, and colleagues, were published in Mayo Clinic Proceedings.

“Although they share common oncogenic pathways, the distinct MPN subgroups (polycythemia vera [PV], essential thrombocythemia [ET], and primary myelofibrosis [PMF]) have been documented to differ in their survival and complication rates,” Dr. Szuber and co-authors wrote. “[Our data] highlight MPN subgroup risk categorization as key in appraising disease natural history.”

To establish risk-stratified survival and complication rates in a large population of adult patients with MPN, the researchers reviewed medical records from 3,023 patients who were diagnosed with PV, ET, or PMF (according to World Health Organization 2016 criteria) between 1967 and 2017. All patients were followed from diagnosis until death or date of last follow-up.

Patients had the following diagnoses: 665 with PV; 1,076 with ET; and 1,282 with PMF.

The authors confirmed that “MPNs were validated as diseases more common in middle- or advanced-age adults,” with a median age at diagnosis of 62 years (range = 18-96 years). Patients with PMF presented at an older age, compared with those with either ET or PV (median age = 65 years vs. 58 years and 62 years, respectively; p<0.001), and most patients (n=1,682; 56%) were older than 60 years at diagnosis.

The researchers performed conventional risk stratification on 2,925 evaluable patients using clinical and pathological variables, since many patients were seen before the discovery of JAK2 V617F mutations in 2005 and subsequent MPN-associated mutations. They found that nearly half of the patients with PV had “high-risk” disease (n=283; 45%), while patients with ET were more evenly distributed among risk categories and most patients with PMF had intermediate-risk disease (n=1,075; 78%).

A review of patients’ clinical and laboratory data revealed that PV was associated with excess thrombotic risk and higher risk of fibrotic progression, compared with PMF and ET (TABLE).

“Not surprisingly, patients with PMF, compared with their PV and ET counterparts, displayed lower hemoglobin and platelet values and more frequent cytogenetic abnormalities, transfusion dependence, leukocyte count of at least 25×109/L, and blast counts of at least 1 percent, supporting their documented prognostically detrimental phenotype,” the authors wrote. PMF also was associated with the highest rates of blast transformation and death compared with the other subtypes.

However, they added, “The present data ascribed a less ‘innocuous’ phenotype to ET than perhaps historically perceived.” This included similar incidence of blast transformation in PV and ET (4% for each), “reflecting potentially more aggressive disease behavior in ET regarding leukemic potential.”

Median overall survival (OS) among the disease subtypes was 18 years for ET, 15 years for PV, and 4.4 years for PMF (p<0.05 for all). Survival in ET was time-dependent, with inferior survival observed for patients diagnosed more recently (21 years for pre-1990 vs. 17 years for post-1990; p<0.01). Survival in PV and PMF did not differ by year of diagnosis.

“This not only exposes the absence of any breakthrough survival-modifying therapies in the past 30 years,” the researchers commented, “but also raises the following question: What factors are driving seemingly less favorable survival patterns over time in ET?”

Life expectancy differed among MPN subtypes and risk levels: The authors found “nearly indistinguishable patterns” of median OS between patients with low-risk PV and ET, at 28.1 and 26.7 years, respectively (p=0.89). For each subtype, however, survival was inferior to age- and sex-matched controls in the general population (28.1 and 26.7 years vs. 39.2 and 37.5 years).

Survival curves for low-risk PMF and intermediate-risk PV also overlapped (median = 14.3 years and 18 years; p<0.06). “Together, these data suggest that, from a survival standpoint, disease nomenclature in MPNs may be superseded by disease risk level,” the authors concluded. “The fact that despite their favorable prognostic features … patients [with low-risk PV and ET] still display excess mortality relative to matched controls is somewhat sobering and reiterates the need for effective disease-modifying agents in patients with MPNs.”

The study’s implications are limited by its retrospective design, and the authors noted that “further studies explicitly examining risk-stratified data in MPN are warranted to validate these findings.”

The authors report relationships with Novartis and Bristol-Myers Squibb.


Szuber N, Mudireddy M, Nicolosi M, et al. 3023 Mayo Clinic patients with myeloproliferative neoplasms: risk-stratified comparison of survival and outcomes data among disease subgroups. Mayo Clin Proc. 2019;94:599-610.