The burden of immune thrombocytopenia in adults: Evaluation of the thrombopoietin receptor agonist romiplostim

Abstract

Background:

Immune thrombocytopenia (ITP) is a chronic, immune-mediated disease characterized by a transient or long-lasting decrease in platelet counts. ITP is associated with numerous serious clinical consequences. Discussed here are clinical aspects of ITP, the humanistic and economic burden of ITP, and current treatment options with a focus on romiplostim, a thrombopoietin (TPO) receptor agonist. The aim of this review is to provide decision-makers with the background information necessary to evaluate the value of romiplostim.

Scope:

PubMed was searched for relevant, English-language papers published from January 2006 through November 2011 relating to the epidemiology and treatment options of chronic ITP, and, focusing on the TPO mimetic romiplostim, patient-reported outcomes (PRO) and economic burden. Recent select conference abstracts were also reviewed.

Findings:

The initial clinical management of ITP (e.g., corticosteroids, immunoglobulins) is often associated with adverse events and recommended for short-term use only. Splenectomy, a potentially curative second-line treatment, is associated with increased risks of bleeding and infection, and patients often require additional long-term drug intervention. ITP and its sequelae are associated with a substantial burden on patients’ health-related quality-of-life (HRQoL) and increased medical costs. Use of TPO receptor agonists in ITP patients may represent a more efficient use of healthcare resources than existing therapies.

Conclusion:

While this literature review is not a systematic review, e.g., it considers only approved therapies and published literature written in English, it provides a comprehensive overview of the clinical, humanistic, and economic factors that should be considered in treating ITP, particularly with new agents such as romiplostim. Among the limited number of safe and effective therapies currently available for chronic ITP, highly effective and well-tolerated medications such as romiplostim may reduce the healthcare resource utilization associated with ITP while improving patients’ HRQoL.

Keywords::

• Review
• Immune thrombocytopenia
• ITP
• Epidemiology
• Patient reported outcomes
• Economics
• Thrombopoietin (TPO)
• TPO receptor agonist
• Romiplostim

Introduction

Immune (idiopathic) thrombocytopenia (ITP), an immune-mediated, acquired disease affecting adults and children, is associated with a transient, persistent, or long-lasting decrease of platelet counts. ITP is considered to be a disease of both increased platelet destruction due to immune mechanisms as well as a relatively decreased rate of platelet production. Depending on the degree of thrombocytopenia, ITP is associated with an increased incidence of spontaneous and induced bleeding, such as bruising, mucosal bleeding, gastrointestinal hemorrhage, and, rarely, intracranial hemorrhage1. Adult ITP is an orphan disease, with an estimated annual incidence ranging from 1.6–3.9 per 100,000 adults per year2. Following a diagnosis of ITP, physicians may simply observe patients whose platelet counts remain sufficiently high, and then generally intervene to increase platelet counts when they have fallen to a level low enough to present a risk of hemorrhage. Management of adult ITP includes medical therapies that modulate the immune system to interfere with platelet destruction and therapies that increase platelet production, such as thrombopoietin (TPO) receptor agonists3–6.

Patients with ITP experience reduced health-related quality-of-life (HRQoL) compared to both the general population and patients with other chronic diseases, such as hypertension, arthritis, and cancer7–9. Adult patients with chronic ITP often require a significant amount of healthcare resources, including frequent, costly hospitalizations and medical therapies to address acute bleeding events and the adverse events resulting from use of those therapies10,11.

Romiplostim is a novel therapy that increases platelet counts by activating the TPO receptor. Because ITP is a rare condition, and the diagnosis is made by exclusion, understanding a new therapeutic option such as romiplostim can be challenging to healthcare providers, payers, and patients. In order to facilitate informed decision-making regarding the value of romiplostim, this review summarizes the current understanding of the clinical, humanistic, and healthcare burden of ITP and describes the economic and patient-related aspects of romiplostim.

Existing literature

A survey of the literature indicates that recent reviews of ITP center primarily on clinical aspects of current treatment options12–18. While some touch briefly on quality-of-life issues, in particular as related to the TPO mimetics13,14,18, there is no comprehensive assessment of quality-of-life effects or the costs or value of these or other treatments. Likewise, recent reviews of TPO receptor agonists as a class or romiplostim in particular center on pharmacodynamics and safety19–23, with only a high-level overview of the beneficial effects these agents have on quality-of-life24–28.

Published data regarding cost-effectiveness of romiplostim is limited to a health technology appraisal (HTA) from the Aberdeen HTA group29 based on the information submitted to the UK’s National Institute for Health and Clinical Excellence (NICE). Health technology assessment documents available at government websites for the UK30, Australia31, and Canada32 provide additional information. These documents are generally limited to a short summary of the efficacy and safety of romiplostim as well as how cost-effectiveness was calculated, arriving at various incremental cost-effectiveness ratios (ICERs), and concluding with under what circumstances romiplostim use should be reimbursed.

HTA documents are country-specific in their conclusions and do not provide a comprehensive view of the costs of ITP and treatments for ITP. Given that ITP is a rare disease, treatment decisions are not routinely made, and decision-makers often are not familiar with associated efficacy and safety outcomes of the various treatment options. Thus, this review was written to provide the context to aid decision-makers in their assessment of romiplostim by describing key concepts regarding direct costs, such as medications and hospitalization, and indirect costs, such as quality-of-life aspects, of ITP and what is known of romiplostim’s effects on these costs.

Methods

A targeted literature search of articles published from January 2006 through November 2011 was conducted to assemble the latest information on the epidemiology, burden of illness, treatment, patient-reported outcomes (PRO), and economics of ITP, with a focus on the TPO receptor agonist, romiplostim, for PRO and economic analyses. Relevant articles prior to that time were only considered if they contained information that was often cited in the literature, particularly as related to traditional treatment and standard of care (SOC). The literature search was conducted in MEDLINE, accessed through PubMed. Search terms were tailored to identify articles related to the topics of interest and are found in the Appendix. The titles and abstracts were reviewed to identify articles for inclusion. Relevant recent abstracts and conference proceedings from relevant conferences (including but not limited to meetings held by the International Society for Pharmacoeconomics and Outcomes Research [ISPOR], the International Society for Pharmacoepidemiology [ISPE], and the American Society of Hematology [ASH]) were also reviewed. A systematic review of the literature for each topic was not performed. Instead, the literature search was limited to articles published in English and included publications with data regarding chronic adult ITP of primary etiology (i.e., not related to chemotherapy, pregnancy, infection, hematologic malignancy, etc.). Observational studies, case series, clinical trials, reviews, and meta-analyses were all included. Case reports, letters to the editor, and non-human studies were excluded.

Results

The following sections provide an overview of the relevant epidemiologic parameters, current treatment options, and treatment guidelines for ITP. We also describe current knowledge regarding the costs of ITP and ITP treatments, as well as PRO for ITP patients in general. Finally, the impact that the use of romiplostim has on these financial and humanistic outcomes is summarized.

Economics of achieving sufficient platelet levels in ITP patients

The economic burden of chronic ITP has been examined in a limited number of recent studies10,33–37. In the US, ITP drug therapy alone is estimated to account annually for hundreds of millions of dollars35, with mean annual per-patient costs estimated at ∼$28,00038. Studies that have examined the costs associated with ITP consistently reveal that adults with chronic ITP incur substantial per-patient medical costs, primarily due to hospitalization required to manage bleeding events, drug costs, and costs of surgical intervention, particularly splenectomy10,34–36. Costs are even higher in refractory patients and those with severe disease. As new treatments are developed, cost analyses for these novel therapies will become increasingly relevant for treatment decision-making.

The burden to payers for any healthcare system is based on both the cost of the disease and the number of patients with the disease. In particular, the important inputs for a budget impact model include the epidemiology of the disease (incidence, prevalence, and mortality), the costs of ITP treatment, and the frequency and costs of clinical events, such as healthcare resource utilization. In addition, the patient’s quality-of-life in relation to both their disease and its treatment is increasingly recognized as important in treatment decision-making39. The sections below describe each of these categories.

Defining ITP

The terminology used to describe ITP in the literature varies widely, often using idiopathic, autoimmune, and immune interchangeably. The clinical criteria used to diagnose ITP and to evaluate disease outcomes are also diverse34. The definition of ITP varies widely among studies, with some studies relying on platelet counts and a clinical history to define ITP40–42 and others relying on the occurrence of particular medical or administrative codes in databases10,43,45. This heterogeneity has made the comparison of results between studies challenging.

In 2009, an International Working Group (IWG) defined standard terminology for primary ITP and criteria for the grading of severity and clinically meaningful outcomes and responses45. The IWG consensus report recommended that a platelet count <100 × 10⁹/L be used as the threshold for thrombocytopenia; however, some studies undertaken prior to 2009 employed a threshold of 150 × 10⁹/L40,43. An ITP diagnosis is established through the exclusion of other diseases45. The IWG defined the following categories to distinguish the different phases of ITP: (1) newly diagnosed (patients within 3 months of diagnosis); (2) persistent (patients who have not achieved spontaneous remission or maintained their response after stopping treatment 3–12 months since diagnosis); (3) chronic ITP (patients with ITP for more than 12 months); (4) severe ITP (patients with bleeding symptoms at presentation that were sufficient to mandate treatment or patients with the occurrence of new bleeding symptoms requiring additional therapeutic intervention). Prior to the IWG review, chronic ITP was defined as persistence of thrombocytopenia for greater than 6 months post-diagnosis45. This definition of chronic ITP was used in most of the published literature cited in this review.

Incidence, prevalence, and mortality

Only five studies have estimated the incidence of ITP using population-based data, with those estimates of total incidence ranging from 1.6–4.7 cases per 100,000 persons per year46 (Table 1). While a considerably higher incidence of ITP is found among women in older studies41,44,47–50, numerous recent studies have not confirmed this finding10,40,51, but suggest that the female preponderance of ITP is restricted to young adulthood41,47,50, with an equalizing female-to-male ratio with increasing age. Recent studies have also found an increasing incidence of ITP with advancing age. A higher incidence was consistently found among those greater than 60 years of age41,42,44,47,48,50.

Table 1.  Incidence of ITP from population-based studies.^a

Reference	Study location	Methodology	Inclusion criteria	Accrual years	Incidence per 100,000 adults/year	Age restrictions (years)
Frederiksen and Schmidt50	County of Funen, Denmark	Retrospective chart review	Patients identified using diagnosis codes from inpatient and outpatient records (ICD-8: 287.10, 287.11, 287.18, 287.19, and 675.09 or ICD-10: D69.3, D69.4, D69.5, D69.6 and O72.3) and included if they had a platelet count <100 × 10^9/L. Patients with solid tumors, other hematologic disorders, hepatic disease or alcoholism, lupus, TTP, drug- or gestation- induced TCP, DIC, or HIV were excluded. Relapsed cases were included if they had normal platelet counts for at least 2 years.	1973–1995	Total = 2.6*;2.3^† M = 2.0*;1.8^† F = 3.3*;2.7^†	≥16
				1985–1995	Total = 3.3*;2.7^† M = 2.4*;2.0^† F = 4.2*;3.3^†
Neylon et al.42	Northern Health Region, UK	Prospective cohort	Patients with a platelet count <50 × 10^9/L and bone marrow confirmation enrolled prospectively. Patients with gestational and therapy-related thrombocytopenia and all those with a proven cause for thrombocytopenia (drug-related) were excluded.	1993–1999	Total = 1.6^†	≥16
Abrahamson et al.47	UK	Retrospective secondary analysis	Patients identified using administrative codes in UK GPRD (42P2-11, D313--12, D313000, D312012, and 2871C). Patients included if they had 1 year of continuous enrollment in GPRD prior to ITP diagnosis code.	1992–2005	Total = 3.9, M = 3.2 F = 4.5	≥18
Schoonen et al.44	UK	Retrospective secondary analysis	Patients identified using administrative codes in UK GPRD (42P2-11, D312-11, D313-12, D313000, D312012, and 2871C).	1990–2005	Total = 3.8, M = 3.4 F = 4.4	All ages
				2000–2005	Total = 4.7, M = 4.4 F = 5.0
Kurata et al.41	Japan	Retrospective secondary analysis	Patients identified from nationwide register with a platelet count <100 × 10^9/L and bone marrow confirmation of ITP. Patients with a secondary immune disease that accounts for the thrombocytopenia were excluded.	2004–2007	Total = 2.2*, M = 1.7* F = 2.7*	≥14

^aOne study was not included in this table because it was not population-based: Franco-Garcia et al.49 reported an incidence rate of 1.15 per 100,000 persons in a hematology department in one hospital in Spain in 1994. The study by Bottiger and Westerholm48 was also not included in this table because the authors evaluated the incidence of thrombocytopenia, as opposed to ITP (8.1 and 4.2 per 100,000 females and males >14 years of age, respectively, per year in the Uppsala region of Sweden 1964–1968). Finally, Kurata et al.41 reported that an incidence rate of 2.3 per 100,000 (platelet count <100) was reported in a study by Takahashi et al. (2004), which is only available in Japanese, and is thus not included in this table.

*Platelet count <100 × 10⁹/L; ^†Platelet count <50 × 10⁹/L.

DIC, disseminated intravascular coagulation; GPRD, General Practice Research Database; HIV, human immunodeficiency virus; ITP, immune/idiopathic thrombocytopenia; IV, intravenous; TCP, thrombocytopenia; TTP, thrombotic thrombocytopenic purpura; M, males; F, females.

The few studies that have estimated the prevalence of ITP report wide variation in the US: 4.5–9.5 cases52, 20.0 cases53, 80 cases10, and 176–189 cases54 per 100,000 persons per year (Table 2). All of the studies except Landgren et al.54 required two ITP-specific ICD-9 codes within 6 months to estimate the prevalence of chronic ITP. The wide variation in these estimates may be due, in large part, to differing source populations, age restrictions, and inconsistent exclusions of secondary causes. The finding from Segal and Powe52, for example, is an under-estimate of the prevalence of chronic ITP because they considered only persons under 65 years of age during one calendar year of data53. Feudjo-Tepie et al.53 and Saleh et al.10 are similarly limited by the use of a large managed care database, which may under-represent elderly persons. Additionally, the prevalence may be over-estimated due to inclusion of cases with thrombocytopenia related to other rare conditions (e.g., Evans syndrome). Most studies excluded patients with disorders associated with secondary thrombocytopenia to improve the specificity of the ICD-9 code 287.3, but it is unclear how effective these exclusions were10,52,53.

Table 2.  Prevalence rates of ITP from population-based studies.

Reference	Location	Methodology	Inclusion criteria	Accrual years	Prevalence rate per 100,000 adults/year	Age restrictions (years)
Landgren et al.54	US	Retrospective secondary analysis of VA database	African American and White males with ITP (ICD codes 287.1 and 287.3) identified from discharge records for inpatient hospitalizations at 142 nationwide VA hospitals.	7/1/1969–9/30/1996	African American M = 189.3* White M = 176.4*	≥18
Segal and Powe52	Maryland, US	Retrospective secondary analysis of private insurance database	Patients with at least two claims, separated by at least 30 or 180 days, for ICD-9 CM 287.3 were identified from a database of privately insured MD residents. Patients with concurrent diagnoses that made ITP unlikely (HIV, malignancy) were excluded.	2002	9.5* (codes separated by 30 days) 4.5*^ (codes separated by 180 days)	≥1 and ≤65
Feudjo-Tepie et al.53	US	Retrospective secondary analysis of managed care database	Patients enrolled in the health plan 2002–2006, with continuous enrollment in 2004 and at least two medical claims with ICD-9 codes 287.3 or 287.5 2000–2004 at least 6 months apart. Patients with HIV, hematological malignancies, or aplastic anemia were excluded.	2004	20.0*^	≥18
Saleh et al.10	US	Retrospective secondary analysis of healthcare database	At least two medical claims with ICD-9 codes for 287.3 or 287.5 2000–2004 at least 6 months apart, with the exclusion of persons with ITP codes indicating other platelet defects, non-thrombocytopenic purpuras, or secondary ITP.	2000–2004	0.08% or 80 per 100,000 persons^a^	≥18 on Jan. 1 2000

*Age-adjusted.

^Estimate of chronic ITP.

^aThe authors state “restricting the definition to chronic primary thrombocytopenia (i.e., ICD-9-CM code 287.3 only) reduced the prevalence estimate to 0.02%.” The reported prevalence estimate of 0.08% was obtained when using chronic idiopathic thrombocytopenia (i.e., ICD-9-CM code 287.3 and 287.5 [unspecified thrombocytopenia]).

HIV, human immunodeficiency virus; ICD, International Classification of Disease; ITP, immune/idiopathic thrombocytopenia; M, males; MD, Maryland VA, veteran’s administration.

While considered a benign hematologic disease, ITP is potentially life-threatening. A study of ITP patients in the UK found a 60% higher mortality rate than age- and sex-matched comparison subjects44. A similar finding was reported in a retrospective series of 152 consecutive ITP patients diagnosed over a 20-year period at a hospital in the Netherlands (relative risk [RR] = 1.3, 95% CI = 0.9–2.0); the mortality risk was similar to the general population among persons with a complete response to therapy, but was substantially greater among persons with no response to therapy (RR = 4.2, 95% CI = 1.7–10.0). Norgaard et al.43 reported that the 5-year risk of death was elevated 2-fold in a cohort of ∼400 chronic ITP patients in Denmark, compared to an age- and gender-matched cohort of the general population (RR = 2.3, 95% CI = 1.8–3.0). A higher relative mortality (RR = 4.21, 95% CI = 3.06–5.79) was reported in a study of 3,121 chronic or persistent ITP patients identified from a medical claims database in the US55, and an even higher risk of mortality was found in a case-control study comparing ITP patients and a matched control group from the General Practice Research Database (odds ratio [OR] = 60.0, 95% CI = 4.47–805.56).

Bleeding and infection are recorded as causes of death in a substantial number of ITP patients56. In a pooled analysis of 17 case-series of ITP patients with persistently low platelet counts, the overall rate of fatal hemorrhage was estimated to be 0.02–0.04 per person-year, with nearly 80% of events occurring in the central nervous system. The rate was substantially higher in older patients, with a rate of 0.004 per person-year for patients less than 40 years of age and 0.13 per person-year for patients greater than 60 years of age57. In a population-based historical cohort study of nearly 1,500 ITP patients, a cause of death related to bleeding was found in 10 ITP patients, and a further 14 patients had infection listed as cause of death (13% and 19% of 75 ITP patients with a recorded cause of death, respectively)44. Neylon et al.42 reported a similar proportion of patients that died due to bleeding, and Portielje et al.58 observed the same percentage of patients that died from an infectious origin in a case series with 11 years of follow-up.

Treatment guidelines

Due to a paucity of randomized, controlled trials in patients with ITP, treatment guidelines are primarily based on expert opinion. The guidelines provide consensus opinion on the data supporting different treatment options and emphasize the need to individualize patient treatment. Guidelines from an expert panel selected by the American Society of Hematology (ASH) were initially published in 199657. These guidelines centered on use of corticosteroids and intravenous immunoglobulin (IVIg) as first-line therapies, followed by splenectomy as necessary; for patients not responding to those interventions, recommendations included high-dose IVIg, high-dose corticosteroids, or anti-D (i.e., immunoglobulin [Ig] against the rhesus antigen). Seven years later, in 2003, the British Committee for Standards in Haematology General Haematology Task Force made similar recommendations59. More recent guidance, from the International Consensus Report (ICR)60 and updated guidelines from ASH61, also included the TPO receptor agonists, romiplostim and eltrombopag, which increase platelet counts.

Medical treatment options

Corticosteroids, received by most ITP patients, result in variable initial response rates (39–100%), with responses typically of short duration1,62,63. As toxicity is dose- and duration-dependent, most patients receiving long-term corticosteroids experience adverse events or complications including diabetes, hypertension, anxiety, insomnia, infections, fractures, obesity, and psychosis56,64–72.

IVIg therapy is typically used to rapidly increase platelet counts, often as emergency therapy1,60,73. While up to 80% of patients respond to IVIg, most relapse after a few weeks1,59. Anti-D also results in a rapid response in most patients74. Serious adverse events described in a black box warning in the prescribing information include potentially fatal intravascular hemolysis and acute renal failure60. Like IVIg, anti-D is a pooled blood product and carries risks of infection75.

Rituximab, an anti-CD20 monoclonal antibody, is frequently used off-label for ITP76. In a comprehensive review of response rates of rituximab, it was reported that the highest response rates (>60%) were from studies with small sample sizes (20 patients or less)77–79. Long-term response rates are relatively low, reported to be 20% at 5 years. Infusion-related adverse events have also been reported80.

Splenectomy

Since chronic ITP is now defined as starting at 1 year post-diagnosis and there is a chance of spontaneous remission, current guidelines recommend delaying splenectomy until patients have had ITP for 1 year and have failed systemic therapy, had troublesome adverse events, or both45,60,81,82. As splenectomy is associated with an increased risk of infection83, it is recommended that patients be vaccinated for pneumococcal meningococcal C conjugate and Haemophilus influenzae b (Hib) at least 4 weeks prior to elective splenectomy or 2 weeks after splenectomy60. Complications associated with splenectomy include bleeding, infection, thrombosis, prolonged and repeated hospitalizations, and requirement for additional intervention58,60,84, with mortality rates ranging from 0.2–3.6%65,85.

TPO receptor agonists

Recent research has revealed the crucial role of a relative deficiency of TPO, and hence inadequate platelet production, in the pathogenesis of ITP86. By stimulating TPO receptor signaling, the TPO receptor agonists class of agents, romiplostim and eltrombopag, increase platelet counts in patients with ITP.

Eltrombopag

Eltrombopag, a once-daily oral treatment, was assessed in a randomized, double-blind, 6-week Phase 3 placebo-controlled study of previously treated adult chronic ITP patients5. Patients on eltrombopag were more likely to achieve platelet counts ≥50 × 10⁹/L compared to those on placebo (59% vs 16%, p < 0.001). In a 6-month, randomized, Phase 3 study (eltrombopag n = 135, placebo n = 62), most patients (79%) receiving eltrombopag responded to treatment at least once compared to 28% of placebo-treated patients5. Response was defined as a platelet count of 50–400 × 10⁹/L at assessment (weekly during the first 6 weeks and at least once every 4 weeks thereafter)5.

Safety concerns with eltrombopag include hepatic toxicity, bone marrow fibrosis, worsened thrombocytopenia, and hemorrhage risk after cessation of eltrombopag, thrombotic and thromboembolic events, malignancy progression, cataracts, phototoxicity, and renal tubular toxicity88. To avoid significant reduction in eltrombopag absorption due to chelation, there must be an interval of at least 4 hours between eltrombopag administration and intake of medications, foods, or supplements containing polyvalent cations (e.g., iron, calcium, aluminum, magnesium, selenium, and zinc)88. In the eltrombopag label, there is a black box warning of the increased risk of hepatotoxicity, with regular liver function monitoring required. Thromboembolic complications may result from excessive increases in platelet counts due to excessive doses of eltrombopag5.

Romiplostim

In multi-center, randomized (2:1 romiplostim:placebo), 24-week Phase 3 trials of romiplostim in splenectomized and non-splenectomized adult chronic ITP patients (n = 125), platelet counts were above 50 × 10⁹/L by Week 4 for most patients receiving romiplostim6. Nearly half of romiplostim-treated patients achieved the primary end-point of durable response vs 2% of placebo-treated patients. Durable platelet response was defined as a weekly platelet count ≥50 × 10⁹/L for 6 or more of the last 8 weeks of treatment6. Further, in an open-label 52-week trial in 234 non-splenectomized ITP patients that compared romiplostim with medical standard of care (SOC) (2:1, romiplostim:SOC)89, splenectomy occurred less frequently with romiplostim (9% vs 36%, p < 0.001).

Bleeding events occurring in chronic ITP patients were also examined in a combined analysis of the Phase 3 trials of romiplostim and during subsequent treatment in an open-label extension study (n = 125). In the Phase 3 trials, 34% of patients treated with placebo had bleeding events of moderate or greater severity (at least grade 2) compared to 15% of romiplostim-treated patients (p = 0.018). As medications such as corticosteroids, platelet transfusions, and IVIg may be either introduced or increased in dose to “rescue” patients from having platelet counts fall to levels at which serious or fatal bleeding events are more likely to occur, accurate measurement of bleeding can be confounded. Thus, a composite end-point of bleeding-related episodes (BRE) was developed91,92. A BRE is defined as an actual bleeding event and/or the use of rescue medication, with events (bleeding and/or the use of rescue medication) that occurred concurrently or within 3 days of each other collapsed into a single BRE. The duration-adjusted rate of BREs was reduced by 55% with romiplostim vs placebo (95% CI = 41–65%)91,92, with an accompanying 88% reduction in the use of IVIg rescue medication92. Similar results were seen when romiplostim was compared to SOC in non-splenectomized ITP patients; the BRE rate was reduced by 67% (95% CI = 60–73%).

A recent analysis of pooled data from the controlled studies described above found that mortality rates in romiplostim-treated patients were ∼5-fold less than those in placebo and SOC-treated patients (hazard ratio [HR] = 0.187; 95% confidence interval [CI] = 0.048–0.931; p = 0.04), indicating a statistically significant survival benefit for romiplostim. The survival benefit in romiplostim-treated patients was maintained in the extension study (HR = 0.120; 95% CI = 0.035–0.410; p = 0.0007)94.

Across all trials, most adverse events were mild-to-moderate in severity; the most commonly reported adverse events were headache, nasopharyngitis, and fatigue95. Adverse event frequency, including Grade 2/3 bleeding events and thrombotic events, did not increase with time95. Bone marrow biopsies, performed at the investigators’ discretion on a small proportion of patients in the open-label extension study, revealed reticulin in 11 patients95. Neutralizing antibodies to romiplostim (but not to endogenous TPO) developed in two patients during that study; these antibodies were absent on re-testing after drug withdrawal. Other safety concerns, as mentioned in the prescribing information for romiplostim, include: thrombotic and thromboembolic events, bone marrow reticulin deposition, and rebound thrombocytopenia. Also, patients with myelodysplastic syndrome (for which romiplostim is not indicated) treated with romiplostim may experience increased circulating blast cell counts and progression to acute myelogenous leukemia.

Economic considerations for ITP

Cost of illness studies

The determination of ITP-attributable direct medical costs requires an in-depth analysis of existing literature, as non-ITP related costs are generally included in direct-cost estimates. Long-term management and hence increased resource utilization is required for ITP due to the burden of disease and possibility of repeated bleeding events. Cost-related events might include splenectomy-related hospitalizations, hospitalizations for low platelet counts and bleeding events, costs of prescription drugs and ancillary products, and diagnostic tests.

Table 3 describes the data sources, cost types, and findings in the cost-of-illness studies for ITP; Table 4 describes the studies reporting costs associated with interventions in ITP. While many of the cost of illness studies were based in the US (n = 3; 50% (Table 3)), studies were also conducted in Italy and France. Studies that examined costs associated with ITP interventions were mostly conducted outside of the US (n = 7; 64% (Table 4)). A variety of study designs (e.g., cross-sectional, retrospective, prospective) and data sources (e.g., commercial claims data, national survey data, single or multi-center data) were used.

Table 3.  Costs of illness studies for ITP.

Reference	Country (Currency)	Types of costs included (source)	Findings
Aledort et al.38	US (unknown $)	Inpatient, outpatient, laboratory, pharmacy (PharMetrics, 2000–2003)	Mean annual cost of care per patient was $23,420, 43% attributed to inpatient hospitalizations.
Danese et al.33	US (2008$)	Hospital costs (Nationwide Inpatient Sample, 2003–2006)	Mean cost per hospital discharge was $16,476 for ITP patients, compared to $10,039 for non-ITP patients; highest mean costs per discharge were associated with splenectomy ($21,754).
Khellaf et al.36	France (2005 €)	Hospital costs (French health care tariffs)	Mean annual costs were €7,293 for all patients, €15,334 for patients with at least one hospitalization, and €26,581 for patients with severe disease.
Lucioni et al.37	Italy (2008 €)	Inpatient, outpatient, laboratory, pharmacy (Italian healthcare tariffs)	Cost of illness to the NHS was estimated at €4 million per year, assuming 1200 adult patients.
Saleh et al.10	U.S. (unknown $)	Inpatient, outpatient, laboratory, pharmacy (HealthCore Integrated Research Network database, January 2000–February 2004)	Average ITP-attributable hospitalization cost per year was $4986; other ITP-attributable costs include $913 for diagnostic and treatment procedures, $367 for other outpatient visits, $299 for pharmacy, $260 for emergency department visits, $78 for office visits, and $27 for laboratory tests.
Tarantino et al.35	NA	Number of physician visits, work loss, work productivity (self-report)	Compared to controls, ITP patients were significantly more likely to have a primary care visit and a specialist care visit in the past month, more likely to have taken sick leave, and had difficulties with six productivity-related items. Visits, work loss, and productivity were not consistently related to years since diagnosis.

ITP, immune/idiopathic thrombocytopenic; NHS, National Health Service; NA, not available.

Table 4.  Costs associated with interventions in ITP.

Reference	Country (Currency)	Types of costs included (source)	Findings
Adams et al.95	US (2001 $)	Pharmacy (average wholesale price)	Annual cost of IVIg therapy in the US is $115–$693 million; assuming similar patient response, annual cost of anti-D is $49–$292 million if replacing IVIg; costs could be cut by 40% in RhD-positive patients.
Aledort et al.38	US (unknown $)	Inpatient, outpatient, laboratory, pharmacy (PharMetrics Integrated Medical and Pharmaceutical Database, 2000–2003)	Majority (82%) of patients had at least one claim for corticosteroid therapy during 12-month follow-up; 12% underwent splenectomy during follow-up; in splenectomized patients, 90% received corticosteroids and anti-D was administered more frequently than pre-splenectomy. Mean annual cost of care for splenectomized patients was $48,424; in non-splenectomized patients, costs were $20,110. Inpatient hospitalizations were the primary cost driver for splenectomized patients (39.6% of costs) and non-splenectomized patients (43.8% of costs).
Arocho et al.126	Mexico (2010 MXP$)	Cost of therapy and rescue medication over a 24-week period (Public Mexican Healthcare Perspective)	Cost per response^a was MXP$727,317 for romiplostim and infinite for placebo in splenectomized patients and MXP$456,912 for romiplostim and MXP$1,290,655 for placebo in non-splenectomized patients. Mean treatment costs in splenectomized patients were MXP$574,580 for romiplostim and MXP$301,218 for placebo and in non-splenectomized patients were MXP$402,083 for romiplostim and MXP$180,692 for placebo.
Brosa et al.97	Spain (2009 €)	Cost of therapy, rescue medication, and bleeding events over a 24-week period (national reimbursement rate)	Treatment costs over 24 weeks were €15,781 for all ITP patients treated with romiplostim, compared to €8,111 for placebo group (splenectomized patients €15,436 vs €10,263; non-splenectomized patients €16,125 vs €5958, respectively). Cost per response^b was €19,013 and €115,871, respectively, for romiplostim compared to placebo. ICER was not calculated.
Estrada-Gomèz et al.99	Spain (unknown US $)	Costs of treatment (20-year single institution study)	Costs of various treatments for a 60 kg adult, in US dollars, were $8,400–$30,000, $3,200 for anti-D, and $170 for anti-D opsonized autologous erythrocytes.
Lucioni et al.37	Italy (2008 €)	Inpatient, outpatient, laboratory, pharmacy (Italian health care tariffs)	Mean annual total medical costs for ITP were €3293 per patient. Of these, 56% were due to medications and 38% were due to hospitalizations. IVIg/anti-D treatments accounted for 97% of pharmacy costs.
Nelson et al.129	US (unknown $)	Hospitalizations and drug costs (Nationwide Inpatient Sample, 2003–2004)	Mean cost per splenectomy hospitalization was $16,557 (average LOS = 6.02 days); for hospitalizations without splenectomy, costs were $13,211 (LOS = 5.55 days).
Northridge et al.124	US (2010 $)	Medicare-covered costs of treatment, administration, on-going patient management, adverse events, and bleeding-related events (Medicare Fee Schedule)	Model estimates the 6-month cost per patient with durable response in splenectomized patients were $110,589 and $325,320 for those treated with romiplostim compared to watch-and-rescue, respectively; in non-splenectomized patients, 6-month costs per durable response were $58,598 and $89,486, respectively.
Ruggeri et al.34	Italy (2005 €)	Hospital costs (Italian Health Service, 1997–2007)	Total cost was €10,780 per patient; major expenses were associated with splenectomy interventions and recurrent admissions for high-cost rescue therapies.
Vainchtock et al.106	France (2007 €)	Hospital costs (French healthcare tariffs)	Mean charge per ITP-related hospitalization were €3,653 for public hospitals, €3,177 for private hospitals, and €6,367 for ITP-related splenectomy. Patients had an average of 2.8 (public hospital) to 3.4 (private hospital) stays per year.
Xie et al.100	Canada (2007 $CAD)	Modeled lifetime medical costs	Incremental cost-effectiveness of IVIg compared to prednisone was $1.13 million per QALY.

^a≥4 weekly platelet responses ≥50 × 10⁹/L from Weeks 2 to 25.

^bPlatelet response ≥50 × 10⁹/L.

CAD, Canadian dollars; ICER, incremental cost-effectiveness ratio; ITP, immune thrombocytopenic; IV, intravenous; IVIg, intravenous immunoglobulin; LOS, length of stay; MXP, Mexican peso; QALY, quality-adjusted life-year.

Studies evaluating only ITP-attributable direct medical costs reveal lower, although still substantial, costs. For example, in the US, a retrospective analysis of 2004 medical claims for adult chronic ITP patients found that medical diagnostic and treatment procedures and laboratory visits accounted for almost 14% of ITP-attributable costs, with mean annual per-patient ITP-related costs of almost $7,00010. In Italy, a multi-center retrospective cohort evaluation estimated annual ITP-attributable medical costs based on 2008 diagnosis-related group (DRG) tariffs, 56% of which were attributed to medications and 38% to hospitalizations; laboratory tests comprised the remaining 6% of costs37. It must be noted that most of these studies were either cross-sectional or had limited follow-up (less than 3 years); thus, changes over time could not be evaluated. Further, patients who had higher expenses prior to the study period are likely to incur similarly high expenses during the study period. None of the reviewed studies considered this potential bias. Regardless, these studies represent the most up-to-date information regarding economic issues associated with ITP and its treatments.

Cost of medical treatment

Although there is no current consensus, primarily due to the need for personalized as-needed therapy, first-line treatment for refractory ITP patients usually consists of IVIg and corticosteroids directed at preventing platelet destruction, with IVIg therapy typically given to chronic ITP patients96,97. Drug-related costs, including acquisition costs, drug administration, and costs related to managing treatment-related adverse events all contribute to the direct medical costs associated with ITP. A number of costing studies examined drug-related costs associated with ITP (Table 5). Study designs varied, including a US retrospective multi-center cohort, an Italian prospective cohort, a Canadian HTA submission, a comprehensive review, clinical trial data, and inputs to various costing models. Most of these studies had short time horizons, with only one capturing costing information for a long period of time (10 years). In the US, pharmacy costs were ∼20% of total direct medical costs10. Pharmacy costs were also the most significant cost driver of ITP-attributable direct medical costs, accounting for ∼50–60%37 of these costs in Italy.

Table 5.  Treatments and costs for ITP therapies*.

Reference	Country (currency)	Treatment	Reported costs (dosage)
Adams et al.95	US (2001 $)	Anti-D	$3,780 (low dose)–$5,680 (high dose) (50 µg/kg × 1 day–75 µg/kg × 1 day) With infusion: $3,824 (low dose)–$5,724 (high dose)
Adams et al.95	US (2001 $)	IVIg	$4,850 (low dose)–$13,300 (high dose) (0.8 g/kg × 1 day–1.0 g/kg × 2 days) With infusion: $4,986 (low dose)–$13,436 (high dose)
Blackhouse et al.96	Canada (2007 $)	IVIg	$550–$1,000 per infusion (0.5–1.0 g/kg for 20 kg child) ,$4,000 per infusion (1 g/kg for 70 kg adult)
Brosa et al.97	Spain (2009 €)	IVIg	Unit cost €5,376
Chiche et al.110	France (unknown $)	IVIg	€6,462 per dose (1 g/kg × 2 days)
Pettigrew et al.125	Canada (unknown $)	IVIg	Annual cost per patient was $107,041 (1 g/kg × 4 weeks, average administration of 3.5 hours)
Ruggeri et al.34	Italy (2005 €)	IVIg	Total cost per patient €1639
Xie et al.100	Canada (2007 $CAD)	IVIg	$59.19 per gram (range, $51–64)
Adams et al.95	U.S. (2001 $)	Prednisone	$25 per course (4 mg/kg taper × 17 days)
Blackhouse et al.96	Canada (2007$)	Prednisone	Dose: $0.09 per 50 mg tablet Course: $0.73 per 20 kg child
Chiche et al.110	France (unknown $)	Prednisone	€6 per dose (€0.045/mg unit dose) (2 mg/kg daily × 2 weeks)
Xie et al.100	Canada (2007 $CAD)	Prednisone	$0.09 per 50 mg tablet
Chiche et al.110	France (unknown $)	Rituximab	€2,647 per dose (375 mg/m^2, 4 weekly infusions)
Chiche et al.110	France (unknown $)	Romiplostim	€688 per dose (2.5 mcg/kg once weekly)
Pettigrew et al.125	Canada (unknown $)	Romiplostim	Average annual cost of $50,950 based on 3 mcg/kg weekly
Perreault and Burzynski28	US (2008 $)	Romiplostim	Average whole price of 250 mcg vial is $1,062.50 and $2,125.00 for 50 mcg vial. Annual drug cost per patient is ∼ $55,250.
Adams et al.95	US (2001 $)	Splenectomy	$16,000**
Blackhouse et al.96	Canada (2007 $CAD)	Splenectomy	Mean = $11,709 (95% CI: $3,251–$85,452)
Xie et al.100	Canada (2007 $CAD)	Splenectomy	$11,709 (range, $3,251–$85,452)

CAD, Canadian dollars; CI, confidence interval; IVIg, intravenous immunoglobulin.

*Studies reporting costs solely based on national health tariffs were not included in this table. Some studies relied on national health tariffs (e.g., Chiche et al.110), but included additional administration and other costs to determine total costs; these studies were included for comparative purposes.

**Based on literature reviews.

While the low cost of corticosteroid therapy appears to be a cost-effective treatment option, the resources required to address short-term complications and drug-induced comorbidities may add significantly to the total costs of this therapy58. For example, in a single-center retrospective case series in the Netherlands, patients who continued to use corticosteroids and immunosuppressive drugs 2 years post-diagnosis required 5-times as many ITP-related hospital admissions as patients who did not use corticosteroids58. Platelet counts were similar for all patients, leading the authors to hypothesize that corticosteroid adverse events, rather than ITP disease severity, were the cause of increased hospitalization among corticosteroid users58.

Regardless of geography or study design, several studies report that costs related to rescue drugs, and specifically IVIg, are the primary cost drivers in ITP patient care34,36,98–103, with IVIg accounting for up to 40% of total direct costs, depending on severity of disease, in a recent French study36. IVIg-related costs are difficult to compare due to IVIg being less expensive in France compared to the US and other European countries36. In Italy, Lucioni et al.37 reported that 97% of the cost for medications was attributed to IVIg or anti-D treatments using 2008 DRG tariffs; however, costs based on reimbursement tariffs likely under-estimate the actual healthcare costs, again confounding cross-study and cross-country comparisons34,36. In the US, the annual cost of IVIg therapy for ITP has been estimated at between $115–693 million, depending on dosage99. In Canada, a cost-utility analysis conducted as part of a health technology assessment on behalf of the Canadian Agency for Drugs and Technologies in Health indicated that IVIg was unlikely to be a cost-effective treatment in adult chronic ITP unless decision-makers were willing to pay in excess of $500,000 per quality-adjusted life-year (QALY) when compared to oral prednisone99,103. Costs of managing treatment-related adverse events, such as infusion reactions, and ancillary infusion supply costs further contribute to the cost of IVIg therapy104. In one study, ancillary product use accounted for 76% of total costs in ITP patients who received short-term (6 month) IVIg therapy104.

Studies in the US and Canada suggest that anti-D, another rescue drug, may offer a lower cost alternative for treatment of non-splenectomized adults with ITP96,98,99. Although a viable alternative first-line treatment to IVIg, anti-D is most effective in Rh D-positive patients102, with a cost reduction of ∼40% compared to IVIg in Rh-positive ITP patients as opposed to only 10% in other ITP patients in the US98. The annual US estimated cost of ITP therapy if IVIg was replaced with anti-D is estimated at $49 million to $292 million98, with costs approximately one-third higher for patients treated with anti-D rescue therapy than those who were not38. Treatment efficiency with ex vivo opsonized autologous anti-D coated RBCs were 50–150-times cheaper than treatment with IVIg, and had efficacy comparable to other Fc-receptor blockers, in a 20-year single-institution Spanish study of chronic, refractory, Rh D-positive ITP patients102. The lower cost, as well as the shorter infusion time, of anti-D compared to IVIg has also been reported96,98,105. However, because of safety concerns, anti-D is no longer recommended for ITP patients in the European Union (EU).

Alternate treatment is required as patients become refractory to first-line therapy96,97. This might include unapproved medications such as rituximab, or approved agents including TPO -receptor agonists, such as romiplostim and eltrombopag, as second-line therapy97. In addition to high per-unit costs, in a multi-center study of community oncology sites, rituximab was also found to require significant staff resources (31–80% more time), higher human resource costs (increase of $17–65), and patient time to administer and care for infusion reactions106 compared to healthcare utilization patients without infusion reactions.

Costs of splenectomy

In recent years, the rate of splenectomy has decreased, primarily due to complications associated with the procedure and reluctance of patients to undergo surgery97; however, between 70–80% of adults relapse following primary treatment and require splenectomy to maintain acceptable platelet counts107. Most relapses from splenectomy occur within the first 2 years post-surgery, with a failure rate of 28% 5-years post-splenectomy108. While the cost-effectiveness of splenectomy vs no splenectomy in ITP patients is unknown, in general, the mean charge for splenectomy-related hospital admissions is higher than that for other ITP-related inpatient admissions38,109 in both a US managed-care setting and a French national database that records public and private hospitalizations for billing purposes. In addition, mean annual direct medical costs of treatment for splenectomized ITP patients are 2–3-fold higher than those for non-splenectomized patients, with inpatient hospitalization costs being the primary cost driver for those undergoing splenectomy38,109. Post-operative complications and medical management for non-responders also contribute to downstream medical resource utilization38. Further, data from the 2003–2006 US Nationwide Inpatient Sample found the average cost for splenectomy was ∼$21,000, a figure that excludes any management outside the hospitalization for splenectomy (e.g., vaccinations, preventive procedures to reduce surgical bleeding risk prior to hospitalization, separately billed physician services, and any costs incurred after discharge)33.

Healthcare resource utilization

A retrospective database analysis revealed that ITP patients use a number of healthcare resources, including inpatient and outpatient visits, pharmacy and prescription drugs, and hospital visits, including emergency room (ER) use10. A recent Internet-based survey of ITP patients and age- and gender-matched comparisons found that ITP patients used statistically significantly (p ≤ 0.001) more resources in terms of both specialist (28% vs 11%) and primary care visits (20% vs 11%) when compared to matched comparisons35. ITP patients were more likely to visit a primary care physician for reasons other than their ITP110. ITP patients were significantly more likely to have one or more specialist visits per month than the comparison group (43% vs 26%, p < 0.05) and were more likely to see a specialist, rather than a primary care physician, for treatment (41% vs 20%, respectively)110. Given the higher costs typically associated with specialist care, as well as the higher number of visits per patient, the increased resource utilization has clear impacts for payers and healthcare providers.

Hospitalizations, outpatient visits, and emergency room visits

Data from a large US-managed care system and retrospective database analysis indicate that hospitalization accounts for the most expensive resource use by ITP patients10,38, with a mean cost of $16,476 compared to $10,039 for non-ITP discharges33. Based on the per-patient mean annual cost of care, these studies reported that up to 40–50% of the cost of care for an ITP patient was attributed to inpatient hospitalizations10,38; similar results (38%) were reported in Italy37 (Table 4). On average, adult ITP patients in the US have 0.7 annual inpatient hospitalizations10,10; similar data are reported for Italy37. In France, higher hospitalization rates were reported, with the highest rates found in the private sector109. Non-responders, defined as those without a hematologic response on maintenance therapy (oral prednisone or immunosuppressive therapy), required more than 3-times as many ITP-related hospital admissions (183 per 1,000 patient-years) than partial (27 per 1,000 patient-years) or complete responders (53 per 1,000 patient-years), and patients with severe thrombocytopenia required more than 3-times as many hospital admissions than patients with moderate thrombocytopenia in a single-center French study58. In addition to higher costs, the mean hospital length of stay in both a US and EU setting is also longer for ITP patients than for the general population, averaging 6 or 7 days33,37,111,112 compared to 5 days33. Hospitalization is typically required to manage bleeding events, which are particularly costly for ITP patients33,35,113, surgical complications, or infections33,34. Two US studies in a managed care setting reported that the most common reasons (38%) for hospitalizations for ITP patients were purpura and other hemorrhagic conditions112,114, while a retrospective single-center study in France of adult ITP patients found that ∼57% of hospitalizations were for splenectomy and associated complications, and that 24% of visits involved medical management of thrombocytopenia or therapy-related adverse events58.

ITP accounts for a significant proportion of per-patient mean-annualized outpatient medical resource utilization and costs in the US. Common reasons for outpatient visits include purpura and hemorrhagic conditions (29%), although respiratory symptoms and pelvis/abdomen symptoms were also reported114. A phase IV clinical trial revealed that, on average, ITP patients visit an outpatient care facility at least 3 times per year10. While 23% of ITP patients visit their primary care physician at least once per month (compared to 34% of matched controls), ITP patients were more likely to visit a specialist at least once per month (43% compared to 26% of matched controls)110.

On average, ITP patients in France visit the ER a little less than once per year10. However, in the US, because of the low prevalence of ITP, only 0.03% of all US ER visits in 2007 included an ITP diagnosis115. Data from a phase IV clinical trial reported that, while ER visits were infrequent, they accounted for nearly half of ITP-attributable costs10 and, in the US, were associated with worse clinical outcomes and greater resource utilization than for non-ITP patients115. While not every ER visit requires admission to the hospital, a large US retrospective database analysis showed that over three-quarters of ER visits with an ITP diagnosis resulted in inpatient hospitalizations, compared to 15% of other ER visits115. Costs included those associated with splenectomy, bone marrow biopsy, and emergency platelet transfusion98. As expected, total charges for ER visits by ITP patients are more costly than other visits, totaling $1,653 per visit compared to $1,490 per visit for non-ITP patients based on data from the 2007 Nationwide Emergency Department Sample115. Mean total charges for combined hospitalization and ER services were $47,300 in ITP patients, compared to $29,300 in non-ITP patients115.

Indirect medical costs

There is limited recent research on the impact of chronic ITP on indirect medical costs35. Health utility measurements and assessments are an important consideration in economic evaluations and in healthcare decision-making, with some studies indicating a loss of value in chronic ITP patients116,117. In addition, work productivity may be reduced. In an internet-based study35, ITP patients were surveyed and compared to the general population on various aspects of work loss and work productivity. In general, ITP was consistently associated with poor work and productivity outcomes as evidenced by six work productivity items included in the survey35. Specifically, ITP patients were statistically significantly (p ≤ 0.003) more likely to have taken sick leave (56% vs 30%) or missed chores (18% vs 13%) compared to non-ITP patients35. Similar results regarding extended sick leave (>1 week) were reported by Canadian investigators (53% vs 28%), who also noted that 38% of ITP patients had difficulty concentrating at work, compared to 29% of controls, with 25% reporting that they could not complete their normal work activities, compared to 18% of controls (p < 0.05 for all comparisons)110. Time since diagnosis was also important, with more ITP patients diagnosed in the past year visiting specialists and missing chores than those diagnosed more recently, and with long-term ITP sufferers having more frequent loss of productivity35. A large percentage of chronic ITP patients (40%) are between the ages of 25–64; therefore, issues related to work productivity particularly impact employers.

Patient reported outcomes in ITP

ITP has a well-documented effect on patient functioning and well-being due to the symptoms of ITP and adverse events of treatment7–9,118. Symptoms can include spontaneous bruising, menorrhagia, mucosal bleeding, fatigue, and prolonged bleeding with injury. Treatment of ITP can also involve adverse events including mood swings, headaches, and weight gain56,119–121. Based on a review of the literature and input from focus groups of patients from three US sites (New York, California, and Oklahoma), Mathias et al.120 constructed a conceptual model of the impact of ITP on HRQoL. This conceptual model specified that platelet count affects symptoms and treatment (and, therefore, treatment adverse events), which in turn impact the HRQoL domains of emotional health, functional health, social and leisure activities, reproductive health, and work.

HRQoL measures

There are several questionnaires that have been used to assess the impact of ITP and its treatment on HRQoL. The Short Form-36 (SF-36)122 and the EuroQol (EQ-5D)123 are generic instruments that have been used in patients with ITP. In addition to providing information about HRQoL changes associated with ITP and its treatment, generic measures are also relevant to use with both the general population and other patient populations. These measures allow for comparisons to be made with published data. An ITP-specific questionnaire, the ITP-Patient Assessment Questionnaire (ITP-PAQ), has recently been developed to further characterize the HRQoL impact of ITP and its treatment124. This measure includes 44 items consisting of 10 scales: symptoms, bother-physical health, fatigue/sleep, activity, fear, psychological health, work, social activity, women’s reproductive health (fertility and menstrual symptoms), and overall quality-of-life (QoL). Since this questionnaire was developed specifically for use in patients with ITP, findings cannot be used to compare to other non-ITP populations. Instead, it has the advantage of assessing the issues most highly impacted by ITP, and thus is likely more sensitive to changes in disease status and more responsive to change over time. The ITP-PAQ is based on the conceptual model developed by Mathias et al.120, and has sound psychometric properties124,125 and established data on minimally important differences (MID) to aid in interpretation of scores126.

Humanistic burden of ITP

Patients with ITP have been shown to experience decrements in HRQoL when compared to other populations. A set of analyses describing the ITP-related HRQoL was undertaken from a study of 1002 ITP patients identified through the Platelet Disorder Support Association (PDSA) and 1031 age- and gender-matched consumer panel comparisons9,35. Participants in this study completed online questions about work loss, work productivity, and physician visits; the SF-36 and the EQ-5D; ITP patients also completed the ITP-PAQ. Snyder et al.9 reported that ITP patients scored statistically significantly worse than age- and gender-matched comparisons on every SF-36 domain except bodily pain, and on the SF-36 physical component summary and mental component summary scores (p < 0.05). Scores for the US general population aged 45–54 years were higher (better QoL) than those of the comparisons in this study, indicating that the decrement in HRQoL reported by ITP patients is even greater when compared to the US general population9. Decrements were also found in the group with ITP vs comparisons on the EQ-5D Visual Acuity Scale (VAS) score (65.5 vs 82.3, p = 0.002), but not on the EQ-5D index score. More ITP patients than comparisons reported “some problems” on dimensions measuring “usual activities” and “anxiety/depression”, while more comparisons reported “no problems” on these EQ-5D dimensions.

HRQoL has been shown to differ by ITP clinical characteristics. Based on data at baseline from two clinical trials of romiplostim vs placebo7 (in which rescue medication use was allowed in both treatment groups), splenectomized patients reported significantly worse ITP-PAQ scores on symptoms, bother, fear, psychological health, work, social activity, and overall QoL (p < 0.05). In the above-mentioned cross-sectional study, Snyder et al.9 found that splenectomy status was not associated with significant differences in SF-36 or EQ-5D, but splenectomized patients reported worse scores than non-splenectomized patients on the ITP-PAQ scales of bother, psychological, fear, social activity, work, and fertility (p < 0.05). Snyder et al.9 also found that time since diagnosis was associated with bother (p = 0.042) and overall QoL (p = 0.019) from the ITP-PAQ, for which more recently diagnosed patients reported worse functioning. Platelet count was not significantly associated with SF-36 scores or the EQ-5D index score in this study, but was significantly associated with EQ-5D VAS score (p = 0.026), and with all ITP-PAQ scales except overall QoL and fertility.

Tarantino et al.35 also reported on the impact of ITP using data from the PDSA study. Scores on the ITP-PAQ symptoms, bother, activities, and fatigue scales were significantly associated with the number of primary care and specialty visits, with work-related outcomes, and with productivity outcomes such that those with worse ITP-PAQ scores had higher utilization, more work-related problems, and more productivity problems.

Evaluation of romiplostim

As described above, the clinical, economic, and humanistic burden of ITP has a significant impact on patients’ lives and health costs. Before 2009, available treatment options for ITP were of variable efficacy and associated with significant adverse events which patients often did not find acceptable. A new class of drugs, the TPO receptor agonists, romiplostim and eltrombopag, were approved for use in ITP in many countries throughout the world. While eltrombopag data were described briefly above, we chose to comprehensively review health outcomes for romiplostim, the first-in-class TPO mimetic, for which a larger body of literature exists. The goal is that this description of romiplostim, in the background context of ITP provided above, will assist decision-makers in making informed decisions regarding the use of romiplostim for patients with ITP. Specifically, as detailed below, prolonged use of TPO receptor agonists in ITP patients appears to be a viable therapeutic option and may be effective even in patients considered to have refractory disease97.

Economic analyses for romiplostim

When compared to a watch-and-rescue approach over a 6-month period in a trial-based simulation, the cost per non-splenectomized patient with a durable response was ∼50% lower for romiplostim; in splenectomized patients, the cost difference was almost 3-fold127. In general, a net cost impact model showed that romiplostim is less expensive to prepare and administer than IVIg and is associated with lower indirect costs from a Canadian Healthcare perspective128. Romiplostim is also associated with lower cost per response over 6 months when compared to rituximab, regardless of splenectomy status (from the French National Health System perspective)113 and when compared to placebo100. Over a 24-week period, a clinical trial conducted in Spain indicated that romiplostim was cost-effective in both splenectomized and non-splenectomized ITP patients, with a significantly lower cost per treatment response for romiplostim than for placebo100. These lower costs were primarily driven by resource utilization charges for non-responders who had higher rates of bleeding and required rescue medication, including IVIg. Similar findings were reported in Mexico, where, from a Public Mexican Healthcare perspective, lower costs per responder for romiplostim compared to placebo were attributed to reduced frequency of using IVIg as a rescue medication129.

Patient reported outcomes associated with romiplostim treatment

HRQoL in ITP patients has been shown to improve with romiplostim treatment. In one trial of splenectomized patients, romiplostim treatment was associated with significantly greater improvement than placebo on ITP-PAQ symptoms, bother, social activity, and women’s reproductive health (p < 0.05), while in another trial in non-splenectomized patients, romiplostim treatment was associated with significantly greater improvement than placebo in the ITP-PAQ activity scale (p = 0.0458)7. When data were pooled from both studies, romiplostim treatment was associated with significantly greater improvements than placebo in the ITP-PAQ scales of symptoms, bother, activity, fear, psychological, social activity, and women’s reproductive health. In addition, patients who achieved a durable response in platelet count had significantly greater improvement than non-responders in ITP-PAQ bother and psychological health (<0.05)7.

In a 52-week open-label study of romiplostim vs SOC in non-splenectomized ITP patients (157:77 romiplostim:SOC), Rummel et al.¹³⁰ found that patients in both treatment groups reported significant improvement that exceeded the MID in all 10 ITP-PAQ scale scores except fatigue in both arms and activity in the SOC arm. The improvement in score was significantly greater for the romiplostim group than the SOC group on symptoms, bother, activity, psychological health, fear, overall QoL, and social QoL (p ≤ 0.05)¹³⁰. When compared to the SOC group, the romiplostim group showed statistically significant improvement from baseline on all scales except fatigue; none of the scores exceeded the MID. The investigators compared responders vs non-responders in both groups based on the criteria of having a platelet count > 20 × 10⁹/L for 4 consecutive weeks at the highest recommended dose and schedule, not having a major bleeding event, not receiving blood transfusions or rescue medications, and not having a change in therapy due to intolerable adverse events or bleeding symptoms. Although responders in both groups combined showed statistically and clinically significant improvement in nearly all ITP-PAQ scales, responders in the romiplostim group had a significantly greater improvement than responders in the SOC group on bother, fear, overall QoL, social QoL, and work QoL scales (p ≤ 0.05)¹³⁰.

The above group of 234 non-splenectomized patients from this 52-week open-label study was also discussed by Kuter et al.89 At baseline, no significant differences were observed in scores between the two groups on any of the 10 ITP-PAQ scales evaluated. At 52 weeks, the romiplostim group had significantly greater improvements in scores on seven of the 10 scales than the SOC group (p = 0.01 for symptoms, p = 0.008 for bother, p = 0.02 for activity, p = 0.049 for psychological, p < 0.001 for fear, p = 0.002 for social QoL, and p = 0.02 for overall QoL)89. Scores for women’s reproductive health and work QoL could not be analyzed due to “inadequacies in the statistical model.”

In the 2 phase 3 registrational trials, 125 subjects (on romiplostim n = 83, and placebo n = 42), the use of romiplostim in chronic ITP patients was associated with improvement in health utility as measured by the EQ-5D¹³¹. Mean change scores, using multiple linear regression models adjusting for age, gender, splenectomy status, and baseline score, showed that improvement in EQ-5D index score over a 24-week study was greater for romiplostim vs placebo (0.05 vs −0.03, p = 0.015) and greater for the EQ-5D VAS score (6.42 vs 0.48, p = 0.066). EQ-5D change scores did not differ significantly for those with a durable response vs non-responders, but were borderline significantly greater for the EQ-5D index score for those who did not experience a bleeding event compared to those who did (p = 0.066).

Discussion

This targeted review establishes that adult ITP is a chronic disease with an unpredictable course with potentially serious clinical, humanistic, and economic consequences, including increased risk of severe bleeding, decreased QoL, and substantial per-patient medical costs due to hospitalizations, management of bleeding events, and drug costs. Traditional ITP treatment focuses on preventing the destruction of platelets to reduce bleeding events. These traditional therapies often are adequate for short-term use only due to treatment-related toxicities, transient response, and high costs.

In treating ITP patients, clinical, humanistic, and economic factors should be considered when evaluating potential treatment options. All of the costs associated with ITP must be understood when evaluating new treatments. The available data suggest that the major cost drivers that should be considered include hospitalization, drug costs, and costs associated with surgery. There are currently a limited number of safe and cost-effective therapies for chronic ITP. The introduction of highly effective and well-tolerated medications may reduce the healthcare resource burden of ITP, specifically those treatments that reduce the need for costly hospitalizations, transfusions, IVIg infusions, and splenectomy. Romiplostim presents a lower cost treatment alternative to existing therapies, allowing for more efficient use of healthcare resources, and has been demonstrated in recent clinical trials to lead to a durable response in patients with limited adverse events and toxicity and improved QoL on several areas.

While this review is not comprehensive in nature, and focused on English-language literature published since 2006, it does provide a framework which decision-makers can use to evaluate therapies of ITP, particularly new agents such as the TPO receptor agonist, romiplostim. As clinical research continues in the field of ITP, knowledge regarding the cost-effectiveness of romiplostim and other various treatment options will grow and provide answers to remaining questions regarding this field of health economics research.

Transparency

Declaration of funding

This review was sponsored and funded by Amgen Inc.

Declaration of financial/other relationships

RD and MS are Amgen Inc. employees and stockholders. MD, SM, and JF have consulted for Amgen Inc. JF was previously an employee of Exponent.

The peer reviewers on this manuscript have disclosed that they have no relevant financial relationships.

Acknowledgments

Medical writing support was provided by Exponent (www.exponent.com), previous employer of JF, and Amgen Inc. (Susanna Mac). Exponent received funding from Amgen Inc. for their medical writing support.

Appendix

Medical subject headings were identified to create search strings for each component of the topics of interest. The search strings were then combined to create the best combination of terms to return the most relevant articles from the MEDLINE database. The following queries were used to identify relevant literature:

Epidemiology search

(“immune thrombocytopenic purpura” OR “immune thrombocytopaenic purpura” OR “idiopathic thrombocytopenic purpura” OR “idiopathic thrombocytopaenic purpura” OR “autoimmune thrombocytopenic purpura” OR “autoimmune thrombocytopaenic purpura” OR “immune thrombocytopenia” OR “immune thrombocytopaenia” OR “autoimmune thrombocytopenia” OR “autoimmune thrombocytopaenia” OR “immune thrombocytopenia purpura” OR “immune thrombocytopaenia purpura” OR “idiopathic thrombocytopenia purpura” OR “idiopathic thrombocytopaenia purpura” OR “autoimmune thrombocytopenia purpura” OR “autoimmune thrombocytopaenia purpura” OR “idiopathic thrombocytopenia” OR “idiopathic thrombocytopaenia”) AND (epidemiology OR prevalence OR incidence OR descriptive OR “cancer patterns” OR “cancer trends” OR burden OR treatment outcome)

• Limits: English, Publication Date from 2006/01/01.

• Results: 306 articles, 73 articles identified for further full text review based on pre-determined inclusion criteria.

Treatment search

(“immune thrombocytopenic purpura” OR “immune thrombocytopaenic purpura” OR “idiopathic thrombocytopenic purpura” OR “idiopathic thrombocytopaenic purpura” OR “autoimmune thrombocytopenic purpura” OR “autoimmune thrombocytopaenic purpura” OR “immune thrombocytopenia” OR “immune thrombocytopaenia” OR “autoimmune thrombocytopenia” OR “autoimmune thrombocytopaenia” OR “immune thrombocytopenia purpura” OR “immune thrombocytopaenia purpura” OR “idiopathic thrombocytopenia purpura” OR “idiopathic thrombocytopaenia purpura” OR “autoimmune thrombocytopenia purpura” OR “autoimmune thrombocytopaenia purpura” OR “idiopathic thrombocytopenia” OR “idiopathic thrombocytopaenia”) AND (Treatment and (frequency or outcome or proportion or pattern or management))

• Limits: English, Publication Date from 2006/01/01.

• Results: 322 articles, 10 articles identified for further full text review based on pre-determined inclusion criteria.

Costs/economics search

((“immune thrombocytopenic purpura”[All Fields] OR “immune thrombocytopaenic purpura”[All Fields] OR “idiopathic thrombocytopenic purpura”[All Fields] OR “idiopathic thrombocytopaenic purpura”[All Fields] OR “autoimmune thrombocytopenic purpura”[All Fields] OR “autoimmune thrombocytopaenic purpura”[All Fields] OR “immune thrombocytopenia”[All Fields] OR “immune thrombocytopaenia”[All Fields] OR “autoimmune thrombocytopenia”[All Fields] OR “autoimmune thrombocytopaenia”[All Fields] OR “immune thrombocytopenia purpura”[All Fields] OR (immune[All Fields] AND (“purpura, thrombocytopenic”[MeSH Terms] OR (“purpura”[All Fields] AND “thrombocytopenic”[All Fields]) OR “thrombocytopenic purpura”[All Fields] OR (“thrombocytopaenia”[All Fields] AND “purpura”[All Fields]) OR “thrombocytopaenia purpura”[All Fields])) OR “idiopathic thrombocytopenia purpura”[All Fields] OR “idiopathic thrombocytopaenia purpura”[All Fields] OR “autoimmune thrombocytopenia purpura”[All Fields] OR ((“autoimmune thrombocytopaenia”[All Fields] OR “purpura, thrombocytopenic, idiopathic”[MeSH Terms] OR (“purpura”[All Fields] AND “thrombocytopenic”[All Fields] AND “idiopathic”[All Fields]) OR “idiopathic thrombocytopenic purpura”[All Fields] OR (“autoimmune”[All Fields] AND “thrombocytopenia”[All Fields]) OR “autoimmune thrombocytopenia”[All Fields]) AND (“purpura”[MeSH Terms] OR “purpura”[All Fields])) OR “idiopathic thrombocytopenia”[All Fields] OR “idiopathic thrombocytopaenia”[All Fields]) AND ((“economics”[Subheading] OR “economics”[All Fields] OR “cost”[All Fields] OR “costs and cost analysis”[MeSH Terms] OR (“costs”[All Fields] AND “cost”[All Fields] AND “analysis”[All Fields]) OR “costs and cost analysis”[All Fields]) OR (“healthcare utilization”[Text Word] OR “hospitalization”[Text Word] OR “insurance claim”[Text Word] OR “home health”[Text Word] OR “doctor visit”[Text Word] OR “physician visit”[Text Word] OR “emergency room visit”[Text Word] OR “emergency room”[Text Word] OR “urgent care visit”[Text Word] OR “insurance”[MeSH Terms] OR “insurance”[All Fields]) OR (“medicare”[MeSH Terms] OR “medicare”[All Fields]) OR (“Medicaid”[MeSH Terms] OR “Medicaid”[All Fields]) OR “out of pocket”[All Fields] OR reimburse[All Fields] OR payor[All Fields] OR payer[All Fields] OR (“quality-adjusted life years”[MeSH Terms] OR (“quality-adjusted”[All Fields] AND “life”[All Fields] AND “years”[All Fields]) OR “quality-adjusted life years”[All Fields] OR “qaly”[All Fields]) OR “quality-adjusted life-year”[All Fields] OR “quality adjusted life year”[All Fields] OR “quality-adjusted life year”[All Fields] OR “quality adjusted life-year”[All Fields] OR “life years gained”[All Fields] OR “life year”[All Fields] OR “incremental cost-effectiveness ratio”[All Fields] OR “incremental cost effectiveness ratio”[All Fields] OR “disability-adjusted life-year”[All Fields] OR “disability adjusted life-year”[All Fields] OR “disability adjusted life year”[All Fields] OR “disability-adjusted life year”[All Fields] OR “cost of illness”[MeSH Terms] OR “models, economic”[MeSH Terms] OR “quality-adjusted life-years”[MeSH Terms] OR “cost comparison”[All Fields] OR “direct cost”[All Fields] OR “resource utilization”[All Fields] OR “cost-benefit analysis”[All Fields] OR “health resource”[All Fields] OR DALY[All Fields] OR ICER[All Fields] OR “budget impact”[All Fields] OR “BIM”[All Fields] OR “Cost utility”[All Fields] OR “health care costs”[MeSH Terms] OR “direct service costs”[MeSH Terms] OR “cost of illness”[MeSH Terms] OR “healthcare expenditure”[All Fields] OR “cost minimization”[All Fields] OR “cost effectiveness”[All Fields])) OR (((“immune thrombocytopenic purpura”[All Fields] OR “immune thrombocytopaenic purpura”[All Fields] OR “idiopathic thrombocytopenic purpura”[All Fields] OR “idiopathic thrombocytopaenic purpura”[All Fields] OR “autoimmune thrombocytopenic purpura”[All Fields] OR “autoimmune thrombocytopaenic purpura”[All Fields] OR “immune thrombocytopenia”[All Fields] OR “immune thrombocytopaenia”[All Fields] OR “autoimmune thrombocytopenia”[All Fields] OR “autoimmune thrombocytopaenia”[All Fields] OR “immune thrombocytopenia purpura”[All Fields] OR (immune[All Fields] AND (“purpura, thrombocytopenic”[MeSH Terms] OR (“purpura”[All Fields] AND “thrombocytopenic”[All Fields]) OR “thrombocytopenic purpura”[All Fields] OR (“thrombocytopaenia”[All Fields] AND “purpura”[All Fields]) OR “thrombocytopaenia purpura”[All Fields])) OR “idiopathic thrombocytopenia purpura”[All Fields] OR “idiopathic thrombocytopaenia purpura”[All Fields] OR “autoimmune thrombocytopenia purpura”[All Fields] OR ((“autoimmune thrombocytopaenia”[All Fields] OR “purpura, thrombocytopenic, idiopathic”[MeSH Terms] OR (“purpura”[All Fields] AND “thrombocytopenic”[All Fields] AND “idiopathic”[All Fields]) OR “idiopathic thrombocytopenic purpura”[All Fields] OR (“autoimmune”[All Fields] AND “thrombocytopenia”[All Fields]) OR “autoimmune thrombocytopenia”[All Fields]) AND (“purpura”[MeSH Terms] OR “purpura”[All Fields])) OR “idiopathic thrombocytopenia”[All Fields] OR “idiopathic thrombocytopaenia”[All Fields]) AND (bleeding-related[All Fields] AND episode[All Fields]) OR ((“hemorrhage”[MeSH Terms] OR “hemorrhage”[All Fields] OR “bleeding”[All Fields]) AND related[All Fields] AND episode[All Fields]) OR BRE[All Fields] OR (“splenectomy”[MeSH Terms] OR “splenectomy”[All Fields])) AND ((“economics”[Subheading] OR “economics”[All Fields] OR “cost”[All Fields] OR “costs and cost analysis”[MeSH Terms] OR (“costs”[All Fields] AND “cost”[All Fields] AND “analysis”[All Fields]) OR “costs and cost analysis”[All Fields]) OR (“healthcare utilization”[Text Word] OR “hospitalization”[Text Word] OR “insurance claim”[Text Word] OR “home health”[Text Word] OR “doctor visit”[Text Word] OR “physician visit”[Text Word] OR “emergency room visit”[Text Word] OR “emergency room”[Text Word] OR “urgent care visit”[Text Word] OR “insurance”[MeSH Terms] OR “insurance”[All Fields]) OR (“medicare”[MeSH Terms] OR “medicare”[All Fields]) OR (“medicaid”[MeSH Terms] OR “medicaid”[All Fields]) OR “out of pocket”[All Fields] OR reimburse[All Fields] OR payor[All Fields] OR payer[All Fields] OR (“quality-adjusted life years”[MeSH Terms] OR (“quality-adjusted”[All Fields] AND “life”[All Fields] AND “years”[All Fields]) OR “quality-adjusted life years”[All Fields] OR “qaly”[All Fields]) OR “quality-adjusted life-year”[All Fields] OR “quality adjusted life year”[All Fields] OR “quality-adjusted life year”[All Fields] OR “quality adjusted life-year”[All Fields] OR “life years gained”[All Fields] OR “life year”[All Fields] OR “incremental cost-effectiveness ratio”[All Fields] OR “incremental cost effectiveness ratio”[All Fields] OR “disability-adjusted life-year”[All Fields] OR “disability adjusted life-year”[All Fields] OR “disability adjusted life year”[All Fields] OR “disability-adjusted life year”[All Fields] OR “cost of illness”[MeSH Terms] OR “models, economic”[MeSH Terms] OR “quality-adjusted life-years”[MeSH Terms] OR “cost comparison”[All Fields] OR “direct cost”[All Fields] OR “resource utilization”[All Fields] OR “cost-benefit analysis”[All Fields] OR “health resource”[All Fields] OR DALY[All Fields] OR ICER[All Fields] OR “budget impact”[All Fields] OR “BIM”[All Fields] OR “Cost utility”[All Fields] OR “health care costs”[MeSH Terms] OR “direct service costs”[MeSH Terms] OR “cost of illness”[MeSH Terms] OR “healthcare expenditure”[All Fields] OR “cost minimization”[All Fields] OR 2cost effectiveness”[All Fields])) AND (English AND “2006/01/01”[PDAT]: “2011/11/16”[PDAT]) AND (“humans”[MeSH Terms] AND English)

• Limits: English, Humans, Publication Date from 2006/01/01.

• Results: 154 articles, 19 identified for full text review based on pre-determined inclusion criteria.

Patient reported outcomes

((“immune thrombocytopenic purpura”[All Fields] OR “immune thrombocytopaenic purpura”[All Fields] OR “idiopathic thrombocytopenic purpura”[All Fields] OR “idiopathic thrombocytopaenic purpura”[All Fields] OR “autoimmune thrombocytopenic purpura”[All Fields] OR “autoimmune thrombocytopaenic purpura”[All Fields] OR “immune thrombocytopenia”[All Fields] OR “immune thrombocytopaenia”[All Fields] OR “autoimmune thrombocytopenia”[All Fields] OR “autoimmune thrombocytopaenia”[All Fields] OR “immune thrombocytopenia purpura”[All Fields] OR (immune[All Fields] AND (“purpura, thrombocytopenic”[MeSH Terms] OR (“purpura”[All Fields] AND “thrombocytopenic”[All Fields]) OR “thrombocytopenic purpura”[All Fields] OR (“thrombocytopaenia”[All Fields] AND “purpura”[All Fields]) OR “thrombocytopaenia purpura”[All Fields])) OR “idiopathic thrombocytopenia purpura”[All Fields] OR “idiopathic thrombocytopaenia purpura”[All Fields] OR “autoimmune thrombocytopenia purpura”[All Fields] OR ((“autoimmune thrombocytopaenia”[All Fields] OR “purpura, thrombocytopenic, idiopathic”[MeSH Terms] OR (“purpura”[All Fields] AND “thrombocytopenic”[All Fields] AND “idiopathic”[All Fields]) OR “idiopathic thrombocytopenic purpura”[All Fields] OR (“autoimmune”[All Fields] AND “thrombocytopenia”[All Fields]) OR “autoimmune thrombocytopenia”[All Fields]) AND (“purpura”[MeSH Terms] OR “purpura”[All Fields])) OR “idiopathic thrombocytopenia”[All Fields] OR “idiopathic thrombocytopaenia”[All Fields]) AND (PRO[All Fields] OR “patient reported outcome”[All Fields] OR “quality of life”[MeSH Terms] OR “quality-of-life”[All Fields] OR “utilities”[All Fields] OR “satisfaction”[All Fields] OR HRQL[All Fields] OR HRQoL[All Fields] OR “time trade off”[All Fields] OR “rating scale”[All Fields] OR “standard gamble”[All Fields] OR “QTwist”[All Fields] OR Q-Twist[All Fields] OR QOL[All Fields] OR EORTC[All Fields] OR FACT[All Fields] OR FACIT[All Fields] OR “questionnaires”[MeSH Terms] OR “Health status”[MeSH Terms] OR “Health status”[All Fields] OR “symptom score”[All Fields] OR “symptom scale”[All Fields] OR “I-HRQL”[All Fields] OR “EORTC”[All Fields] OR “FACT-General”[All Fields] OR “FACT-Anemia”[All Fields] OR “FACT-Fatigue”[All Fields] OR “brief pain inventory”[All Fields] OR BPI [All Fields] OR “health utility”[All Fields] OR “patient report”[All Fields] OR “health preference”[All Fields] OR “reported symptoms”[All Fields] OR “EQ-5D”[All Fields] OR “SF-36”[All Fields] OR “preference”[All Fields] OR “health related quality of life”[All Fields] OR “quality of life”[All Fields] OR “health utility”[All Fields] OR humanistic) AND (English AND “2006/01/01”[PDAT]: “2011/11/16”[PDAT]) AND (“humans”[MeSH Terms] AND English)

• Limits: English, Humans, Publication Date from 2006/01/01.

• Results: 70 articles, 10 identified for full text review based on pre-determined inclusion criteria.