Abstract 3701

Introduction:

ITP is an autoimmune disease characterized by increased platelet destruction and suboptimal platelet production. ITP pts are at increased risk of bleeding, and those with persistently low platelet counts are at increased risk of fatal bleeding (Cohen, Arch Intern Med, 2000). Romiplostim increases platelet production and offers an alternative treatment option to chronic immune suppression in pts with ITP. We analyzed mortality rates in ITP pts who participated in randomized, controlled romiplostim clinical studies (in which pts received either romiplostim or placebo/standard of care [SOC]) and their subsequent time in an open-label extension study (in which all pts received romiplostim). Methods: Data were pooled from two 24-week placebo-controlled phase 3 studies and a 52-week SOC-controlled study, which included an off-treatment safety follow-up period for those pts who did not enter the extension study. Additional analyses included data from pts' subsequent time in an extension study. The Cox regression model was used to estimate mortality rates. Results: In total, 354 pts (238 romiplostim, 41 placebo, 75 SOC) from the controlled studies were included in the analysis. Of these, 238 enrolled into the extension study (187 romiplostim, 33 placebo, 18 SOC). At the time of enrolment into the controlled study, the median age was 55 years (Q1 –Q3, 42 – 68), and 22% were splenectomized; their median time since diagnosis was 3.5 years (range, 0.02 – 45); pts had received a median of 3 prior ITP therapies (range, 1 – 9), and 24% were receiving baseline concurrent ITP treatment. Pts were followed for up to 87 weeks in the controlled studies; 0.8% (2/238) of pts in the romiplostim arm died and 6.9% (8/116) of pts in the placebo/SOC arm died. Pts who entered the extension study were followed for up to 215 weeks; 5.9% (14/238) of pts died during this time. Pts died from various causes (see Table). The median age of the pts who died was 73 years (Q1 – Q3, 58 – 78). When data from the controlled studies were pooled, mortality rates in the romiplostim arm were approximately 5-fold less than those in the placebo/SOC arm (HR, 0.187; 95% CI, 0.048–0.931; p=0.04). The survival benefit in the romiplostim arm was maintained when time spent in the extension study was added to the analysis (HR, 0.120; 95% CI, 0.035–0.410; p=0.0007) [see Figure]. Factors predicting increased mortality risk in a multiple Cox regression model were: treatment with placebo/SOC vs romiplostim (p<0.0001), age ≥ 65 years (p=0.002), greater number of prior ITP treatments (p=0.064), and concurrent baseline ITP treatment (p=0.016). Furthermore, there was a positive correlation between the grade of worst bleeding event and risk of death (p<0.0001), and there was an association between mortality and experiencing a bleeding event with a worst grade of ≥2 (p=0.0154), ≥3 (p=0.0002), and ≥4 (p=0.0002). Conclusions: In this large retrospective analysis of an ITP pt study population, the mortality rate was significantly lower in pts receiving romiplostim than in those receiving placebo or other ITP therapies. Associations were found between mortality risk and age, prior treatment with other ITP therapies, and on-study bleeding; however, there was no clear pattern of disease-related or treatment-related cause of death, and determining the mechanism for increased survival in the romiplostim arm requires further study.

Table:

List of cause of death

Study TreatmentaAge (years)Time since ITP diagnosis (years)Cause of death
Controlled studies Placebo 70 6 Pneumonia primary atypical 
 Placebo 51 8 Pulmonary embolism 
 Placebo 52 21 Cerebral hemorrhage 
 Romiplostim 80 1 Hemorrhage intracranial 
 SOC 49 0.1 Hepatic neoplasm malignant 
 SOC 71 0.01 Cardio-respiratory arrest 
 SOC 73 5 Hepatic failure 
 SOC 77 7 Lung cancer metastatic 
 Romiplostim 73 2 Pneumonia 
 SOC 83 10 Left ventricular failure 
Extension study 48 9 Hepatic neoplasm malignant 
88 12 Cardiac failure congestive 
82 3 Meningitis listeria 
35 1 Pneumonia streptococcal 
47 1 Pneumococcal sepsis 
69 0.2 Cardiac arrest 
73 14 Myocardial infarction 
79 0.04 Myocardial infarction 
78 1 Myocardial infarction 
62 6 Angina unstable 
60 6 Narcotic overdose 
76 15 Cardiac tamponade 
78 0.4 Cardiac failure congestive 
76 5 Lung neoplasm malignant 
Study TreatmentaAge (years)Time since ITP diagnosis (years)Cause of death
Controlled studies Placebo 70 6 Pneumonia primary atypical 
 Placebo 51 8 Pulmonary embolism 
 Placebo 52 21 Cerebral hemorrhage 
 Romiplostim 80 1 Hemorrhage intracranial 
 SOC 49 0.1 Hepatic neoplasm malignant 
 SOC 71 0.01 Cardio-respiratory arrest 
 SOC 73 5 Hepatic failure 
 SOC 77 7 Lung cancer metastatic 
 Romiplostim 73 2 Pneumonia 
 SOC 83 10 Left ventricular failure 
Extension study 48 9 Hepatic neoplasm malignant 
88 12 Cardiac failure congestive 
82 3 Meningitis listeria 
35 1 Pneumonia streptococcal 
47 1 Pneumococcal sepsis 
69 0.2 Cardiac arrest 
73 14 Myocardial infarction 
79 0.04 Myocardial infarction 
78 1 Myocardial infarction 
62 6 Angina unstable 
60 6 Narcotic overdose 
76 15 Cardiac tamponade 
78 0.4 Cardiac failure congestive 
76 5 Lung neoplasm malignant 
a

Treatment received at time of death; all pts in the extension study were receiving romiplostim.

View Large
Figure:
Figure:. Cox regression model of time to death, including both romplostim and placebo/SOC exposure in controlled and extension studies.
View largeDownload slide

Cox regression model of time to death, including both romplostim and placebo/SOC exposure in controlled and extension studies.

Figure:
Figure:. Cox regression model of time to death, including both romplostim and placebo/SOC exposure in controlled and extension studies.
View largeDownload slide

Cox regression model of time to death, including both romplostim and placebo/SOC exposure in controlled and extension studies.

Close modal
Disclosures:

Gernsheimer:Amgen Inc.: Consultancy; Cangene: Consultancy; Baxter: Honoraria; Shionogi Corp.: Research Funding. Kuter:Amgen Inc.: Consultancy, Research Funding; GlaxoSmithKline: Consultancy, Research Funding; ONO: Consultancy, Research Funding; Eisai: Consultancy, Research Funding. Cines:Amgen Inc.: Consultancy; GlaxoSmithKline: Consultancy. Stasi:Amgen Inc.: Honoraria, Speakers Bureau; GlaxoSmithKline: Honoraria, Speakers Bureau. Godeau:Amgen France: Consultancy, Research Funding; Roche France: Consultancy, Research Funding; LFB France: Consultancy. Guo:Amgen Inc.: Employment, Equity Ownership. Franklin:Amgen Inc: Employment, Equity Ownership.

Topics:
brachial plexus neuritis, inosine triphosphate, purpura, thrombocytopenic, idiopathic, romiplostim, thrombocytopenia due to immune destruction, hemorrhage, myocardial infarction, congestive heart failure, liver cancer, pneumonia

Author notes

*

Asterisk with author names denotes non-ASH members.

© 2010 by The American Society of Hematology
2010
Sign in via your Institution