Mutation-Enhanced International Prognostic Systems for Essential Thrombocythemia and Polycythemia Vera

Polycythemia vera (PV) and essential thrombocythemia (ET) are chronic myeloproliferative neoplasms (MPNs) driven by overactive Janus kinase (JAK) signal transducer and activator of transcription (STAT) signaling. Patients with these conditions typically enjoy long-term survival (survival in ET is felt to not differ significantly from that of an age- and sex-matched healthy individual, although some studies have arrived at different conclusions), and the major goal in the management of these conditions is prevention of thrombohemorrhagic complications. That being said, a minority of these patients do develop progression to myelofibrosis (MF) and a smaller proportion undergo transformation to acute myeloid leukemia (AML), a devastating complication. Unfortunately, current therapies do not impact these adverse outcomes.

Classic adverse prognostic features in PV include advanced age, leukocytosis, history of venous thrombosis, and abnormal karyotype, and in ET include advanced age, prior thrombosis, and leukocytosis. The widespread availability of next-generation sequencing has enabled the incorporation of this powerful technique into diagnostic algorithms and prognostic models across hematologic malignancies. Indeed, in primary MF, work from investigators at the University of Florence in Italy and the Mayo Clinic in Rochester, MN, has identified several “high molecular risk” mutations; that is, the spliceosome genes SRSF2 and U2AF1^Q157, and the epigenetic regulators ASXL1, EZH2, IDH1, and IDH2. In this study published in the British Journal of Haematology, these groups again join forces to study 906 patients with PV and ET in an effort to see if molecular information can further refine the existing risk-stratification systems for these diseases.

The factors independently predicting inferior survival that emerged from this effort were: in PV, age >67 years, leukocyte count ≥15 x 10⁹/L, SRSF2 mutation and prior thrombosis; and, in ET, age ≥60 years, male sex, leukocyte count ≥11 x 10⁹/L and mutations in SF3B1 or SRSF2. Additionally, in ET, TP53 mutations predicted for leukemic transformation, whereas mutations in the RNA splicing genes SF3B1 and U2AF1 were associated with worse MF-free survival. Using these factors (age, leukocyte count, “adverse” mutations, gender [in ET], and thrombosis history [in PV]), weighted based on their individual hazard ratios for survival, the authors were able to construct three-tiered prognostic models for both PV and ET, termed Mutation-enhanced International Prognostic Scoring System (MIPSS)-PV and MIPSS-ET, respectively, with clear separation of the survival curves among the low-, intermediate-, and high-risk groups. In view of the disparate results between the Mayo and Florence cohorts with respect to the influence of abnormal karyotype on outcomes in PV (only half of the Florence patients had karyotypic information available), and the fact that patients with PV/ET often do not have karyotyping done in routine clinical practice, karyotype was intentionally left out of the multivariable analysis performed on the entire cohort that was used to generate the models. Of interest, “adverse” mutations occurred in 10% of ET patients but in only 2% of PV patients, contrary to what might have been expected, given that all major clinical outcome measures in ET are more favorable than those in PV.

Ultimately, while an important academic endeavor that provides some additional insights into the underlying biology of PV and ET, the study has limited clinical utility at the present time. Indeed, formal prognostication of patients with PV or ET is rarely performed in everyday practice. In MF, the main clinical utility of the many prognostic models described in the literature is to select patients for potentially curative allogeneic stem cell transplantation, a procedure considered too risky for the much more indolent PV and ET. This, coupled with the lack of disease-modifying drugs for patients with PV/ET, makes it hard to envision the role of calculating the MIPSS-PV/MIPSS-ET in the current scenario. Hopefully, as novel agents, such as ropeginterferon alfa-2b, murine double minute 2 (MDM2) and histone deacetylase (HDAC) inhibitors, are developed for patients with PV/ET, these models will be better able to inform therapeutic decisions.