Prognostic Value of Response to First-Line Hydroxyurea According to Ipset Stratification in Essential Thrombocythemia

INTRODUCTION

Essential thrombocythemia (ET) is the most frequent chronic myeloproliferative neoplasm and the one with the best prognosis. ET patients are classically stratified into two thrombotic risk categories based on age over 60 years and history of thrombosis. Recently, the increased thrombotic risk associated with JAK2 mutation has been incorporated into the revised IPSET-thrombosis scoring system. Hydroxyurea (HU) is the first-line treatment for the majority of patients with ET, but criteria for treatment change are not well-established. The prognostic value of the response according to the revised-IPSET thrombosis risk score has not been studied.

OBJECTIVES

Our aim was to analyze the prognostic impact of achieving complete hematological response (CHR) and resistance/intolerance (R/I) to HU, as defined by the European LeukemiaNet, in a contemporary cohort of 1080 patients from the Spanish registry of ET. Subgroup analyses were also conducted based on both classical and revised IPSET-thrombosis risk stratification.

METHODS

The main outcomes of the study were survival, the occurrence of thrombotic events, bleeding, and disease progression. The period at risk was defined as the time elapsed from HU start to the first event.

RESULTS

Patients were classified at the time of cytoreduction according to the classical risk scale (low- n=144, high-risk n=936) and revised IPSET-Thrombosis stratification (very low- n= 61, low- n=83, intermediate- n= 261, and high-risk n=675). With a median treatment duration of 5 years (IQR: 2.4-8.2), 720 (67%) patients achieved CHR and 219 (20%) developed R/I to HU. The median time to CHR was 352 days, with probabilities of CHR at 12 and 24 months being 51% and 61%, respectively. The probabilities of R/I to HU at 1, 3, and 5 years were 4%, 9%, and 13%, respectively. Patients achieving CHR had significantly longer survival compared to non-responders, both in the intermediate-risk category (median survival: 17 and 11 years, respectively, p<0.001) and high-risk category (15.5 and 12.5 years, respectively, p=0.02). No differences in survival were observed according to R/I. High-risk patients achieving CHR also had a lower risk of arterial thrombosis (HR: 0.35, 95% CI: 0.2-0.6, p=0.001). There was not significant association between the development of R/I to HU and the probability of thrombosis. CHR and R/I rates were not associated with an increased risk of bleeding. CHR was associated with a lower probability of progression to myelofibrosis (MF). The absence of CHR was associated with a higher probability of MF in intermediate-risk (10-year probability: 5% vs 23% in responders and non-responders, respectively, p=0.01) and high-risk patients (10-year probability: 6% vs 15% in responders and non-responders, respectively, p=0.03), according to revised IPSET-Thrombosis. Subgroup analysis showed a higher probability of acute myeloid leukemia (AML) only in non-responders intermediate-risk patients. Fulfilling any of the criteria of R/I was associated with a higher probability of MF, but not of AML. When analyzed separately, only cytopenia has a higher risk of disease progression (10-year probability of MF: 5,7% vs 17,4% in responders and no-responders, respectively, p<0.001; 10-year probability of AML: 3% vs 11% in responders and no-responders, respectively, p=0.01).

CONCLUSION

In conclusion, CHR with first-line HU is a relevant prognostic factor that correlates with significant better survival and a lower rate of disease progression in patients categorized as high- and intermediate-risk according to revised IPSET-thrombosis stratification. Moreover, high risk patients achieving CHR face a lower risk of arterial thrombosis. The results of the present study support that failure to achieve CHR could be regarded as a valuable criterion for selecting candidates for second-line therapy.