Changes in the Incidence and Overall Survival of Patients with Myeloproliferative Neoplasms between 2002 and 2016 in the United States

Background

Patients (pts) with myeloproliferative neoplasms (MPNs), including polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF), have reduced overall survival (OS) compared with the general population. Previous reports have shown MPN incidence rates of 0.9–1.1, 0.7–1.0, and 0.3 per 100,000 person-years for PV, ET, and primary MF (PMF), respectively (Srour et al. Br J Haematol. 2016;174[3]:382; Deadmond et al. J Cancer Res Clin Oncol. 2015;141[12]:2131), with evidence of an increase in MPN incidence over time (Mesa et al. Blood. 2012;120[21]:2834). The median OS for pts with PV, ET, and PMF has been previously reported in the literature as 13.5, 19.8, and 5.2–5.9 years, respectively (Tefferi et al. Blood. 2014;124[16]:2507; Cervantes et al. Blood. 2009;113[13]:2895; Cervantes et al. J Clin Oncol. 2012;30[24]:2981). Contemporary analyses of MPN incidence and pt survival are needed, as the most recent available real-world analyses were conducted on time periods up to 2012. The objective of this analysis was to describe trends in incidence and OS among pts with MPNs in the United States using data through 2016 from the Surveillance, Epidemiology, and End Results (SEER) database.

Study Design and Methods

Adult pts with PV, ET, or PMF were identified from the SEER 18 registry (2002–2016) using the primary site of bone marrow (C421) and International Classification of Diseases for Oncology (ICD-O) histology codes (ICD-O-3: 9950 [PV], 9962 [ET], and 9961 [PMF]). Pts were excluded if they were aged <18 y on the index date (date of diagnosis) or were missing demographic or survival information. Age-adjusted incidence rates (per 100,000 person-years) were standardized to the 2000 US population by dividing the incidence rates among adult pts per year by the adult US population in the year 2000. Confidence intervals for rates and rate ratios were calculated using the Tiwari method (Tiwari et al. Stat Methods Med Res. 2006;15[6]:547). Kaplan-Meier methodology was used to compare mortality rates by diagnosis year, categorized into 3 groups: 2002–2006, 2007–2011, and 2012–2016. Pts were censored at the end of each 5-year analysis timeframe.

Results

Data for 34,031 pts (mean age, 65 y; female, 50.5%) were included in this analysis, including 15,141 pts diagnosed with PV, 14,676 with ET, and 4214 with PMF. Over the entire study period, incidence rates (95% CI) for PV, ET, and PMF were 1.57 (1.55–1.60), 1.55 (1.52–1.57), and 0.44 (0.43–0.45) per 100,000 person-years, respectively. The annual incidence rates of PV and PMF were higher for male vs female pts (1.94 vs 1.23 and 0.59 vs 0.33, respectively); however, the ET incidence rate was higher for female pts (1.73 vs 1.34; Table 1). ET and PMF incidence increased across the 3 time periods, whereas PV incidence remained relatively stable.

Mortality rates at 1, 2, and 5 years for each MPN subtype are presented by time period in Table 2. Over the 3 time periods, mortality rates decreased for PV and PMF, but not for ET. Improved OS was observed in pts with PMF over time (median [95% CI]: 2002–2006, 3.3 [2.4–3.6] y; 2007–2011, 3.6 [3.3–4.3] y; 2012–2016, 3.8 [3.5–4.2] y). The median (95% CI) OS for the entire time period investigated was 12.0 (11.7–12.4) years for pts with PV, 12.0 (11.7–12.3) years for those with ET, and 3.6 (3.4–3.8) years for pts with PMF.

Conclusions

In this nationally representative real-world study, incidence of ET and PMF appeared to increase over time from 2002–2016. Median OS of patients with PV, ET, and PMF was shorter than previous reports. A trend of improved survival over time was observed in pts with PV and PMF, which was not observed in pts with ET. Further investigation into the varying OS rates between MPN subtypes is needed, as these data may suggest that ET is lacking improvements in supportive care strategies or therapies that are currently available in PV and MF.