Iqbal, Usman

Abstract Abstract 2834 Background: Myelofibrosis (MF), Polycythemia Vera (PV) and Essential Thrombocythemia (ET) are 3 classic BCR ABL negative myeloproliferative neoplasms. MF is comprised of both primary myelofibrosis (PMF) and those with an antecedent PV or ET (Post PV and Post ET MF, respectively). The term secondary MF (SMF) when used sometimes includes those with histologic myelofibrosis without an MPN (i.e. secondary to another myeloid neoplasm or autoimmune conditions) as well as MPN associated MF (PMF and Post PV/ET MF). Limited published epidemiology data is available on these disorders. Literature reports MF incidence in the range of 0.4–1.4 per 100000. The objective of this study was to estimate prevalence and incidence using 2 real world US health insurance claims databases. Method: The US IMPACT® (∼30M enrollees/year) and US MARKETSCAN (∼40M enrollees/year) were used to retrospectively identify unique patients with PMF, SMF, PV and ET between 1/1/08 and 12/31/10. These databases can track patients longitudinally over multiple years, are linked at the patient level by a unique identifier that is consistent across services, health plans, and time and shown to be representative of the US population. ICD9 CM codes were used to identify PMF, SMF, PV and ET. Any MF was defined as having a diagnosis of either PMF or SMF. Incident cohort was constructed by selecting patients who had their first MF (or PV or ET) in 2010 confirmed by absence of diagnosis 3 years prior. Prevalent cohort was constructed by identifying patients who had MF (or PV or ET) in a given year. From the MF cohort, post ET and post PV patients were identified as those who had any ET or PV diagnosis in the calendar year prior. Primary MF was defined as PMF diagnosis alone without prior PV or ET. Annual prevalence rates were calculated for each calendar year from 2008–10. Continuous enrollment was required during the time period being evaluated. Estimates were adjusted for age distribution from census data. Result: About 8.5M enrollees from IMPACT and 17M from MARKETSCAN each year were included in prevalence estimates. Mean age of MF patients was 61 years in IMPACT and 67 in MARKETSCAN. Age adjusted incidence estimates of MF were about 2.4 per 100000 per year using IMPACT database, 1.7 per 100000 per year from MARKETSCAN. There is a slight increase in MF prevalence over time: from 4.9 to 5.6 in IMPACT, 3.6 to 4.2 in MARKETSCAN. Among MF patients, about 35–40% had PMF, 10% were post-PV patients, 15% were post-ET patients and the remaining had MF arising from myelodysplastic syndrome (∼30%), lupus (∼2%) and other heme and solid tumors (∼3%). Mean age of PV patients was 55 and 36% were female in IMPACT database. In MARKETSCAN, mean age was 58 with 35% being female. Using IMPACT, annual age adjusted prevalence for PV was 57 cases per 100000. Estimates using MARKETSCAN varied from 45 to 48 cases per 100000. Age adjusted incidence of PV ranges from 22 to 27 cases per 100000 patients per year. Mean age of ET patients was 53 and 57 years in each database. In both, 67% of patients were female. Using US IMPACT, annual age adjusted prevalence for ET was 54 to 56 cases per 100000. Estimates using MARKETSCAN varied from 39 to 44 cases per 100000. Age adjusted incidence of ET ranges from 22 to 31 cases per 100000 patients per year. Conclusion: This is the first study utilizing two large national US claims databases to estimate prevalence of MPN disorders. In the US, MF prevalence ranges from 3.6–5.7 per 100,000 patients. Incidence estimates of MF range from 1.7–2.4 per 100,000 patients. PV prevalence estimates range from 45–57 cases and ET prevalence ranges from 39–57 cases per 100,000 patients. PV tends to affect more males while ET affects more female. These data provide compelling evidence to suggest that prevalence and burden of MPN associated MF is higher than has been reported in the past. Additional research using other national databases and/or study designs is needed to substantiate these findings. Disclosures: Mesa: Incyte: Research Funding; Lilly: Research Funding; Sanofi: Research Funding; NS Pharma: Research Funding; YM Bioscience: Research Funding. Mehta:Sanofi: Employment. Wang:Sanofi: Employment, Equity Ownership. Iqbal:Sanofi: Employment, Equity Ownership. Neumann:Sanofi-aventis: Employment. Zhang:Sanofi-aventis: Employment.

Abstract Abstract 1744 Background: MF, ET and PV are BCR ABL negative myeloproliferative neoplasms (MPN). These are a subset of bone marrow disorders characterized by an overproduction of one or more types of blood or fiber cells in the bone marrow. There is disparate published data on the epidemiological estimates for these diseases in the EU. An evidence based synthesis of MPN epidemiology data can effectively inform population health care planning and economic decision support for these disorders. The aim of our study is to synthesize and provide a comprehensive review of epidemiology estimates from available sources including literature, publicly available registries as well as real world data. Methods: We conducted a search of literature published between 1995–2012 using terms related to MF, PV and ET. Elsevier and Medline databases were searched using Embase platform. Orphan disease registries, specific hematological malignancies registries, country specific databases and registries included RARECARE, Orphanet, Registry of hematological malignancies in France, UK Hematologic Malignancy research network. Finnish cancer registry and Sweden National registry were also evaluated. The RARECARE project is responsible for surveillance of rare cancers in Europe based on data from 88 cancer registries in 22 European countries and includes cancer cases diagnosed in 1995–2002. Orphanet provides an estimate of the prevalence of rare diseases in Europe based on a systematic survey of the literature. Calculation of prevalence in Orphanet is done by multiplying incidence with the mean duration of disease. UK THIN (∼5 million enrollees each year) general practitioner database was used to retrospectively identify patients with primary and secondary MF between 1/1/08 and 12/31/10. READ codes were used to identify MF. This database can track patients longitudinally over multiple years, are linked at the patient level by a unique identifier that is consistent across health related services and time and have been shown to be representative of the UK population. Results: Only six articles looking at the European population were retrieved through the literature. These articles as well as international/national reports were used to summarize epidemiological estimates for Myelofibrosis using several different terms. MF terminology varied among registries. Terms included “Primary myelofibrosis”, “myelosclerosis with myeloid metaplasia” and “myelofibrosis”. Annual prevalence of MF ranged from 0.5–9 cases per 100,000 individuals. For PV, even if terminology and coding is consistent, estimates still varied depending on registry used. RARECARE reported prevalence of 5.5 per 100,000 and Orphanet reported a range of 10–50 cases per 100,000. For ET, prevalence was reported only in RARECARE with 4.4 cases per 100,000 individuals. Prevalence estimates in UK were slightly higher compared to RARECARE; 0.9 cases per 100,000 for MF, 6 cases per 100,000 for PV and ET. Incidence for MF, PV and ET ranged from 0.3–1.9, 0.6–2.8, 0.5–2.2 per 100,000 patients respectively. Conclusions: Few publications and registries report epidemiology data for MF, PV or ET. Especially, prevalence data is very rare. There is wide variation in both prevalence and incidence estimates for MPN across EU data sources that can limit informed application of epidemiological research. The differences stem from lack of a unifying term for MF and coding differences coupled with limitations inherent in registry data including potential sources of bias, missing data, and issues with assessment of statistical significance. Additional research using standardized definitions/coding across appropriate real world databases and enhanced tumor registries are needed to add precision to or substantiate these results. Disclosures: Moulard: Sanofi: Employment, Equity Ownership. Mehta:Sanofi: Employment. Olivares:Sanofi: Employment, Equity Ownership. Iqbal:Sanofi: Employment, Equity Ownership. Mesa:Incyte: Research Funding; Lilly: Research Funding; Sanofi: Research Funding; NS Pharma: Research Funding; YM Bioscience: Research Funding.

Abstract Abstract 972 Background: Myelofibrosis (MF) is a Philadelphia chromosome–negative myeloproliferative neoplasm. Common disease manifestations include bone marrow failure, enlarged spleen due to extramedulary hematopoiesis (splenomegaly), debilitating symptoms including fatigue, night sweats, pruritus, early satiety, abdominal pain and discomfort, and marked decrement in patient's quality of life (QoL). Decreased survival is a hallmark of the disease as a result of infections, bleeding and leukemic transformations. There are no recent literature estimates looking at clinical and economic burden of illness of this rare disease. The objective of this study was to describe patient demographics, prevalence, comorbidities, utilization and costs for MF using real-world data. Methods: The US IMPACT® insurance claims database was used to retrospectively identify unique patients with any MF (including all primary and secondary MF cases) between 1/1/08 and 12/31/10. The IMPACT database is a fully de-identified, HIPAA compliant national database that captures the complete medical history for over 100 million managed-care individuals, including patient demographics, disease description, laboratory results, details of medical, pharmacy, outpatient, and inpatient claims. These databases can track patients longitudinally over multiple years, are linked at the patient level by a unique identifier that is consistent across services, health plans, and time and are representative of the US population. ICD-9-CM codes were used to identify MF. The Charlson Comorbidity Index (CCI) was employed to assess overall comorbid disease status. Enrollment was restricted to those with full year of medical and pharmacy benefit. Control group was age and gender matched but without any diagnosis of myelofibrosis. Medical costs include inpatient, outpatient and emergency room cost. Results: In 2010, 433 patients with MF (mean age 60, 50% female) were identified from about 12 million enrollees. This corresponds to an age-adjusted prevalence of 5.4 per 100,000 patients. Compared with age-gender matched control patients, MF patients had higher overall comorbidities (mean CCI of 2.1 vs. 0.9), were hospitalized more often (34% vs. 11%), had higher number of average hospital days (7 vs. 1 day), and had more outpatient office visits (58 vs. 22) in 2010. Accordingly, MF patients incurred much higher average annual cost ($54,168 vs $10,203) driven by both medical ($45,646 vs $7,987) and pharmacy ($8,523 vs $2,216) cost. Over a period of 3 years, annual cost of MF ranged from $54,000-$68,000. The utilization rate and total cost in each of conditions were significantly higher than those of their matched patients in each of the three years. Conclusions: Myelofibrosis is associated with a significant burden of illness. Patient with MF incurred about five times healthcare expenditure than those in the control group. Our study indicates that MF-associated medical resource utilization and the corresponding expenditures for those services are substantive. Continued efforts in the development of more efficacious treatments for myelofibrosis are needed in order to reduce the burden of illness associated with this disease. Disclosures: Wang: Sanofi: Employment, Equity Ownership. Mehta:Sanofi: Employment. Iqbal:Sanofi: Employment, Equity Ownership. Mesa:Incyte: Research Funding; Lilly: Research Funding; Sanofi: Research Funding; NS Pharma: Research Funding; YM Bioscience: Research Funding.

Abstract Abstract 2071 Background: PV and ET are clonal stem cell diseases belonging to the 3 BCR ABL negative myelopoliferative neoplasms. Patients with PV and ET suffer from splenomegaly and disease associated symptoms such as pruritus, night sweats, fatigue, and bone pain. Both diseases, if progressive despite standard therapies, are associated with an increased risk of thrombosis, bleeding, and progression to MF or even acute myeloid leukemia. Literature estimates looking at burden of illness are lacking for these disorders. Hence, real world analyses evaluating economic burden are imperative. The objective of this study was to evaluate the clinical and economic burden of these disorders by describing the patient demographics, prevalence, comorbidities, utilization and costs using large scale databases. Method: The US IMPACT® claims database was used to retrospectively identify unique patients with PV and ET between 1/1/08 and 12/31/10. This database is a fully de-identified, HIPAA compliant national database that captures the complete medical history for over 100M managed care individuals, including patient demographics, disease description, laboratory results, and details of medical, pharmacy, outpatient, and inpatient claims. ICD9 CM codes were used to identify PV and ET. Charlson Comorbidity Index (CCI) was used to assess overall comorbid disease status. Enrollment was restricted to those with a full year of medical and pharmacy benefit. Control group was age and gender matched but without any diagnosis of PV or ET. Medical costs include inpatient, outpatient and ER cost. Result: In 2010, we identified 5752 PV patients from ∼12M enrollees. This corresponds to an age adjusted prevalence of 56.5 cases/100000 patients. Compared with age gender matched control patients, PV patients had higher overall comorbidities (mean CCI 1.2 vs 0.7), were hospitalized more often (16% vs 8%), had higher average number of hospital days spent (1.7 vs 0.8), and had more outpatient visits (31 vs 18). PV patients incurred much higher average annual cost ($14,903 vs $7,913) than age gender matched controls driven by both medical ($12,006 vs $6,188) and pharmacy ($2,897 vs $1,724) cost. In 2010, we identified 5483 ET patients from ∼12M enrollees. This corresponds to an age adjusted prevalence of 56.1 cases/100000 patients. Compared with age gender matched control patients, ET patients had higher overall comorbidities (mean CCI 1.4 vs 0.7), were hospitalized more often (30% vs 9%), had higher average number of hospital days spent (5 vs 0.9 days), and had more outpatient visits (37 vs 19). ET patients incurred much higher average annual cost ($29,553 vs $8,026) than age gender matched controls driven by both medical ($26,287 vs $6,394) and pharmacy ($3,267 vs $1,631) cost. Similar trend was observed in 2008 and 2009. Over a period of 3 years, annual cost of PV ranged from $14,000-$16,000 and those of ET ranged from $29,000-$31,000. The utilization rate and total cost in each of conditions were significantly higher than those of their matched patients in each of the 3 years. Conclusion: PV and ET are associated with significant burden of illness. Healthcare expenditure was twice as higher in PV patients compared with the control group. ET patients incurred more than 3 times the healthcare expenditure than the control group. Our study indicates that PV and ET associated medical resource utilization and the corresponding expenditures for those services are substantive. In order to reduce the burden of illness associated with these diseases, continued efforts in the development of more efficacious treatments for these disorders are needed. Disclosures: Mehta: Sanofi: Employment. Wang:Sanofi: Employment, Equity Ownership. Iqbal:Sanofi: Employment, Equity Ownership. Mesa:Incyte: Research Funding; Lilly: Research Funding; Sanofi: Research Funding; NS Pharma: Research Funding; YM Bioscience: Research Funding.

Abstract Background Primary myelofibrosis ( PMF ), essential thrombocythemia ( ET ), and polycythemia vera ( PV ) are BCR ABL ‐negative myeloproliferative neoplasms ( MPN ). Published epidemiology data are scarce, and multiple sources are needed to assess the disease burden. Methods We assembled the most recent information available on the incidence and prevalence of myelofibrosis ( MF ), ET , and PV by conducting a structured and exhaustive literature review of the published peer‐reviewed literature in EMBASE and by reviewing online documentation from disease registries and relevant health registries in European countries. The search was restricted to human studies written in English or French and published between January 1, 2000, and December 6, 2012. Results Eleven articles identified from EMBASE , three online hematology or oncology registries, and two Web‐based databases or reports were used to summarize epidemiological estimates for MF , PV , and ET . The incidence rate of MF ranged from 0.1 per 100 000 per year to 1 per 100 000 per year. Among the various registries, the incidence of PV ranged from 0.4 per 100 000 per year to 2.8 per 100 000 per year, while the literature estimated the range of PV incidence to be 0.68 per 100 000 to 2.6 per 100 000 per year. The estimated incidence of ET was between 0.38 per 100 000 per year and 1.7 per 100 000 per year. While a few studies reported on the MPN s' prevalences, it is difficult to compare them as various types of prevalence were calculated (point prevalence vs. period prevalence) and standardization was made according to different populations (e.g., the world population and the European population). Conclusion There is a wide variation in both prevalence and incidence estimates observed across European data sources. Carefully designed studies, with standardized definitions of MPN s and complete ascertainment of patients including both primary and secondary MF s, should be conducted so that estimates of the population aimed to receive novel treatments for these neoplasms are better understood assist public health planning and provide valuable information about the burden of illness to policy makers, funding agencies, resource planners, healthcare insurers, and pharmaceutical manufacturers.