Late excess mortality in essential thrombocythemia: a population-based study in the Netherlands, 2001–2018

Abstract

TO THE EDITOR: Essential thrombocythemia (ET) is a clonal myeloproliferative neoplasm characterized by excessive production of platelets— with occasionally elevated white blood cell counts—resulting in a heightened risk of thrombosis or hemorrhage [1]. Therefore, the main treatment goal is to prevent vascular events using platelet inhibition, cytoreductive therapy, or both [1]. There is an ongoing debate about whether ET patients have excess mortality than equivalent groups from the general population [2, 3]. The few available and outdated population-based studies show that ET patients diagnosed before 2000 have considerable excess mortality [4], whereas studies spanning earlier parts of the first decade of 2000 show lesser excess mortality [5, 6]. Given the scarcity of population-based studies in ET, we performed a nationwide, population-based study to assess trends in primary therapy and relative survival (RS) among ET patients diagnosed in the Netherlands between 2001 and 2018. ET patients diagnosed between 2001 and 2018—with survival follow-up through December 31, 2019—were selected from the Netherlands Cancer Registry (NCR) using the International Classification of Diseases for Oncology morphology code 9962/3. The behavior code was changed from /1 to /3 in 2001 following the 2001 classification of the World Health Organization (WHO) [7]. Case ascertainment in the NCR relies on the Nationwide Histopathology and Cytopathology Data Network and Archive, and the Nationwide Registry of Hospital Discharges—i.e., inpatient (since the start of the registry) and outpatient discharges (since 2015). Data on birth and diagnosis dates, sex, prior malignancies, disease morphology, and first therapy started within one-year post-diagnosis—i.e., no anti-neoplastic therapy (e.g., platelet inhibition only), chemotherapy, and pegylated interferon-α2a—were available in the NCR. Trained registrars of the NCR collect these data through retrospective medical records review. Information on the exact therapeutic regimen was standardly available for patients diagnosed as of 2014. Patients were stratified into three age groups (18–60, 61–70, and >70 years) and two calendar periods (2001–2009 and 2010–2018). The Privacy Review Board of the NCR approved the use of anonymous data for this study. We calculated RS up to 15 years post-diagnosis to estimate disease-specific survival in the absence of information on the cause of death as the ratio of the patients’ overall survival (OS) to the expected survival of equivalent groups in the general population, matched to the patients by age, sex, and calendar year. The expected survival was estimated as per the Ederer II method using national population life tables. RS was measured from diagnosis to death, emigration, or end of follow-up (December 31, 2019), whichever occurred first. A multivariable assessment of RS using Poisson regression was performed to evaluate linear trends in RS across age and the relative excess risk of mortality. A P < 0.05 indicates statistical significance. Full details about the statistical analyses are presented in the Supplementary. Our analysis included 5,880 ET patients (median age, 65 years; and 40% males; Table S1). The number of patients across the three age groups were 2,221 (38%), 1,582 (27%), and 2,077 (35%), respectively (Table S1). The overall age-standardized incidence rate (ASR) increased from 0.8 to 2.0 per 100,000 person-years between 2001 and 2012 (Fig. S1). Thereafter, it remained nearly constant. There was a consistent female predominance over time. The incidence in both sexes increased more profoundly among patients aged ≥60 years (Fig. S2). The application of chemotherapy increased over time, particularly among patients aged 61–70 and >70 years (Fig. 1A and Table S2). Detailed data of 2,554 patients diagnosed during 2014–2018 revealed that patients aged 18–60 years more often received platelet inhibition therapy as the sole modality than patients aged 61–70 and >70 years (45% versus approximately 10%; Fig. 1B and Table S3). The use of hydroxyurea dominated in the latter two age groups (approximately 85% versus 37% in patients aged 18–60 years; Fig. 1B and Table S3). Excess mortality within five years post-diagnosis was low; however, heightened excess mortality emerged thereafter— especially among patients aged 61–70 and >70 years (Fig. 2A). Consequently, age-related survival differences widened over time.