Eltrombopag versus romiplostim in treatment of adult patients with immune thrombocytopenia: A systematic review incorporating an indirect-comparison meta-analysis

Purpose In absence of direct comparison randomized controlled trials (RCTs), indirect comparison was conducted to evaluate the efficacy and safety of thrombopoietin-receptor agonists (TPO-RAs) in treatment of adult immune thrombocytopenia (ITP). Methods We searched PubMed, Embase and Cochrane Library, Clinical Trials.gov, China National Knowledge Infrastructure, and Chinese Biomedical Literature Database from their earliest records to May 2017. RCTs comparing the TPO-RAs with placebo in adult ITP were included. Primary outcomes were the overall response rate. Secondary outcomes included safety, durable response, overall or clinically significant bleeding, and the proportion of patients receiving rescue medication. Results Nine randomized placebo-controlled trials (786 participants) were included in this systematic review. Overall response [Risk Ratio(RR) = 0.59, 95%Confidence Interval(CI): 0.24–1.45], the incidence of adverse events (RR = 0.98, 95%CI: 0.79–1.21), durable response (RR = 0.47, 95%CI: 0.08–2.81), the incidence of overall bleeding (RR = 1.15, 95%CI: 0.52–2.57) and clinically significant bleeding (RR = 1.09, 95%CI: 0.37–3.24), and the proportion of patients receiving rescue treatment (RR = 0.95, 95%CI: 0.47–1.90) were similar between eltrombopag and romiplostim. Conclusions Eltrombopag and romiplostim might be equivalent in efficacy and safety for adult ITP, however, physicians should still take into account drug cost and comorbidities of the specific patient while making decisions on the treatment of ITP with TPO-RAs. Registration PROSPERO International Prospective Register of Systematic Review (PROSPERO 2017: CRD42017068661).

Introduction Immune thrombocytopenia (ITP) is an immune-mediated disease characterized by transient or persistent decrease in the platelet count and increased risk of bleeding [1]. ITP in adults is a clinically distinct condition from that in children, with a lower likelihood of spontaneous remission, a higher incidence of underlying diseases and comorbidities, and often a higher risk of bleeding. The incidences of adult ITP reported in recent studies varied among countries: 2.20 per 10 5 in Japan [2], 2.94 per 10 5 in France [3], and 3.70 per 10 5 in Korea [4], respectively. ITP is often a chronic disease in adults, and the prevalence exceeds the incidence [5].
The first-line treatments of adult ITP include longer courses of corticosteroids, intravenous immunoglobulin (IVIg) and anti-D immunoglobulin which has been approved for ITP only in a few countries (North America) [6]. But for those refractory to the first-line treatments, subsequent treatment may include splenectomy, rituximab, thrombopoietin-receptor agonists (TPO-RAs), or more potent immunosuppression [6]. TPO-RAs stimulating the TPO-receptor to increase the production of platelets are recommended for adults at risk of bleeding who relapse after splenectomy, or who have a contraindication to splenectomy and who have failed at least one other therapy [6]. Two TPO-RAs, eltrombopag (ELT) and romiplostim (ROM), have been approved for the treatment of adults with ITP in the United States. Recent evidence showed that TPO-RAs were effective and safe second-line options for primary ITP patients [7].
Nevertheless, ROM and ELT have different mechanisms of action and routes of administration: ROM is a subcutaneously administered peptide mimetic binding to the extracellular TPO-receptor, while ELT is an oral non-peptide binding to a transmembrane site of the TPOreceptor [8,9]. Unfortunately, there are no head-to-head randomized controlled trials (RCTs) comparing ROM with ELT in treatment of adult ITP. Hence indirect comparisons, which preserves within-trial randomization by comparison treatment effects(RR) relative to a common comparator (placebo) from each trial [10], are recommended in the UK National Institute for Health and Clinical Excellence (NICE) methods guide [11]. An indirect comparison between ROM and ELT in treatment of adult patients with ITP was previously conducted [12], but the conclusions became controversial with the publication of later RCTs. Therefore, this study aims to evaluate the efficacy and safety of ELT versus ROM for adult patients with ITP using an indirect-comparison meta-analysis.

Literature search
Pubmed, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) published in Cochran Library were searched using the search strategies detailed in S2 Table, from their earliest records to May 2017. Clinical Trials.gov was searched using the terms "immune thrombocytopenia", "adult", "eltrombopag", and "romiplostim". The China National Knowledge Infrastructure(CNKI) and Chinese Biomedical Literature Database(CBLD) were also searched in Chinese.

Eligibility criteria
All included studies met the following criteria: (1) Randomized controlled studies; (2) Participants were adult (! 18 years) with ITP; (3) the intervention was ELT or ROM irrespective of dosage and schedule; (4) the comparison was placebo; (5) studies included at least one of the following outcomes: overall platelet response(primary outcome), defined as achieving at least once platelet response (! 50×10 9 /L) during treatment; incidence of overall and serious adverse events (SAEs); durable platelet response, defined as maintaining platelet counts ! 50×10 9 /L for at least 60% of the duration of TPO-RAs treatment or for six or more weeks during the final eight weeks of TPO-RAs treatment; incidence of clinically significant bleeding (WHO Grade 2-4 or rated as severe, life threatening, or fatal); all bleeding events; and proportion of patients who received rescue treatment [e.g. receiving any unscheduled or new treatment (including new drugs, increase dose of a concomitant drug from baseline, platelet transfusion or splenectomy) for immediate risk or treatment failure]; (6) publications written in English or Chinese. We excluded studies on patients with secondary ITP and those including both children and adults when data of adults could not be extracted separately.

Study selection and data extraction
Two authors independently screened the titles and abstracts of all studies identified by the search strategies, and assessed the studies using predetermined inclusion criteria. The full texts of all potentially relevant articles were retrieved for detailed review. We resolved any disagreements by discussion until consensus was achieved. We used a pre-designed data collection form to extract data from each eligible study. The following data were extracted: (1) authors; (2) year of publication; (3) country or region where the study conducted; (4) study design and use of control; (5) number of participants randomized into each group; (6) gender, age, and disease duration of participants; (7) baseline platelet count, previous ITP medication, and splenectomy status; (8) dose and schedule of TPO-RAs; (9) outcomes of each study and their definitions; (10) numerical data for assessment of included outcomes; (11) sources of funding.

Quality assessment
Two authors independently assessed the risk of bias of each included study using the checklist developed by Cochrane Collaboration [13]. The items included random sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other bias. We categorized the judgments as low, high or unclear risk of bias and created plots of risk of bias assessment in Review Manager Software (RevMan 5.3).

Statistical synthesis
We calculated a kappa statistic for measuring the agreement level between two authors making decisions on study selection. The value of kappa (K) between 0.40 and 0.59 was considered as fair agreement, between 0.60 and 0.74 as good and 0.75 or more as excellent.
If more than one study reported the same outcome, the pairwise meta-analysis was conducted to calculate the pooled estimate of the risk ratio (RR) of different TPO-RAs versus placebo by RevMan 5.3. Statistical heterogeneity among studies was examined by the Chi-square test and quantified by the I 2 statistic [14]. We used a fixed-effect model to synthesize data when heterogeneity was not significant (P>0.1 and I 2 <50%). When heterogeneity was significant (P 0.1 and I 2 !50%) and could not be explained by subgroup analyses or in terms of clinical or methodological features of the trials, the random-effect model was used. If both the ELT-placebo and ROM-placebo trials reported the same outcome, the relative treatment effect (RR) for ELT versus ROM was estimated using indirect procedure of Stata12.0 software [15], with the formula as follow: For the overall platelet response, sensitivity analysis was conducted by comparing the results of intention-to-treat (ITT) analysis with per-protocol (PP) analysis to improve the robustness of the results. The subgroup analysis was also conducted according to the different types of adverse events(AEs).

Study selection
A total of 3,499 citations were obtained from the literature search and the selection process was shown in Fig 1. Nine randomized, placebo-controlled studies (786 participants) [16][17][18][19][20][21][22][23] were included in this systematic review, and two of them were published in one article [22]. Agreement between two reviewers for study selection was excellent (K = 0.85). As shown in Table 1, all studies were multicenter, double-blind, RCTs from different countries in North and South America, Asia, Europe, Africa, and Oceania.

Description of patients
All patients were aged !18 years old, with disease duration more than 3 months and baseline platelet count less than 30×10 9 /L. Five studies (606 patients) evaluated the efficacy and safety of ELT in comparison to placebo [16][17][18][19][20] (Table 1). The initial dose of ELT was ranged from 12.5 to 75mg, and the following dose was adjusted according to individual platelet count, with a target of 50-200×10 9 /L [16]. Four studies (180 patients) evaluated the efficacy and safety of ROM [21][22][23]. It was administrated at an initial dose of 1 or 3 μg/kg and was also adjusted according to platelet counts ( Table 1). The characteristics of patients were shown in Table 2.

Overall platelet response
The overall platelet response was reported in all studies (five for ELT [16][17][18][19][20] and four for ROM [21][22][23], respectively) including 785 patients (ITT). The heterogeneity was not statistically significant (I 2 = 32%, P = 0.21 and I 2 = 4%, P = 0.37, respectively). The pooled results with a fixed-effect model (Table 3) showed that proportion of patients achieving overall response was significantly higher in the TPO-RAs group than in the placebo group (RR = 4.07, 95%CI: 2.91-5.70 for ELT and RR = 8.81, 95%CI: 4.01-19.35 for ROM, respectively). However, the result of indirect comparison (Fig 3) indicated that the overall response between ELT and ROM was not significantly different (RR = 0.59, 95%CI: 0.24-1.45). And sensitivity analysis showed that the results of PP analysis were consistent with the ITT analysis (Table 3 and Fig 3).

Safety
Eight studies (764 participants) [16][17][18][19][20][22][23] reported the overall incidence of any AEs reported in patients receiving TPO-RAs or placebo. The pooled analysis showed that the incidence was not significantly different between two groups (RR = 1.05, 95%CI: 0.84-1.32 for ELT and RR = 1.05, 95%CI: 0.97-1.14 for ROM) ( Table 3). And the result of indirect Patients!18 years old, with a diagnosis of ITP (duration of !6 months), had received at least one previous treatment for ITP, had a platelet counts<30×10 9 /L, and had a splenectomy for the treatment of ITP greater than or equal to 24 weeks prior to study entry.

Romiplostim vs Placebo
Starting dose of 1ug/kg subcutaneously weekly for 24 weeks; dose was adjusted to achieve target platelet counts of 50 to 200×10 9 /L.
Patients!18 years old, with a diagnosis of ITP (duration of !6 months), had received at least one previous treatment for ITP, had a platelet counts<30×10 9 /L, and had nonsplenectomized status.

Romiplostim vs Placebo
Starting dose of 1ug/kg subcutaneously weekly for 24 weeks; dose was adjusted to achieve target platelet counts of 50 to 200×10 9 /L.
Patients!20 years old, with a diagnosis of ITP (duration of !6 months), had received at least one previous treatment for ITP, and had a platelet counts<30×10 9 /L. comparison (Fig 4) also showed that the overall incidence of any AEs in ELT group was similar to that in ROM group (RR = 0.98, 95%CI: 0.79-1.21). SAEs were reported in all studies [16][17][18][19][20][21][22][23], and the results of both direct and indirect comparison (Table 3  The common AEs in TPO-RAs or placebo group were headache, fatigue, thrombosis, arthralgia, nausea, nasopharyngitis, diarrhea, peripheral edema, epistaxis, pain in extremity, dizziness, contusion, upper abdominal pain, upper respiratory tract infection, cough, myalgia, anxiety and back pain. However, both the direct and indirect comparison of the incidences (Table 3 and Fig 4) demonstrated no significant difference among ELT, ROM and placebo, except that the incidence of peripheral edema was significantly lower in ELT than in placebo (RR = 0.20, 95%CI: 0.06-0.66). As liver abnormalities and cataract were only reported in ELTplacebo trials, indirect comparison was not conducted. Head to head meta-analysis results indicated that incidences of increased level of ALT or AST and cataract were not significantly different between ELT and placebo (ALT: RR = 1.   ROM treatment, but this level of reticulin subsequently returned to baseline after termination of ROM [22]. Bussel 2006 reported that two patients receiving ROM exhibited reversible increases in bone marrow reticulin levels during the subsequent extension study [21].  Eltrombopag versus romiplostim in treatment of adult patients with immune thrombocytopenia

Durable platelet response
Five studies [18][19][20]22] reported the durable platelet response, including 499 patients. The direct and indirect comparison analysis (Table 3 and Fig 3) showed that proportion of patients achieving durable platelet response in TPO-RAs was significantly higher than placebo (RR = 6.

Discussion
This systematic review incorporating an indirect-comparison meta-analysis summarized the efficacy and safety of TPO-RAs in adults with ITP. Our study suggests that the use of TPO-RAs may improve the durable and overall platelet response, and reduce the use of rescue medication, without increasing the incidence of AEs, compared to placebo. While ELT might resemble ROM in the overall and durable platelet response, the incidence of AEs (including SAEs), the incidence of overall and clinically significant bleeding, and the proportion of patients receiving rescue treatment.
A published indirect comparison demonstrated that ROM significantly improved overall platelet response compared with ELT for adult patients with ITP, while the durable platelet response of the two TPO-RAs was similar [12,24]. Our research drew a different conclusion that ELT and ROM were similar in both overall and durable platelet response for adults with ITP, as additional studies led to negative results in the indirect comparison analysis.
Nonetheless, an updated systematic review and meta-analysis of RCTs suggested that TPO-RAs were associated with a higher risk of thromboembolic events compared with placebo or standard care [28]. And another meta-analysis of three large, population-based observational studies concluded that the risk of arterial and venous thromboembolism should be considered when evaluating the risk of thromboembolism attributed to ITP treatments (e.g. TPO-RAs) [29]. Moreover, a disproportionality analysis in the World Health Organization global individual case safety report (ICSR) database (VigiBase) suggested the presence of a signal for an increased risk of thrombosis with ELT compared to ROM (adjusted reporting odds ratio = 1.72, 95%CI 1.47-2.02) [30]. However, our study did not find there was a significant difference, probably due to the small sample size. The physicians should still be cautious when administering TPO-RAs to patients with higher risk of thromboembolism, especially for ELT.
Long-term observational studies (2 for ROM [31][32] and 4 for ELT [33][34][35][36], respectively) indicated that a minority of patients treated with ROM or ELT developped bone marrow fibrosis and those adverse events (3 for ROM [37][38][39] and 1 for ELT [40], respectively) were usually reversible and dose dependent. Probably due to short-term follow-up and small sample size of the RCTs included in this review, the reticulin fibrosis in bone marrow was only reported in the ROM trials. But careful monitoring is still needed during use of ROM or ELT.
There are several limitations in this study. We only included RCTs in this review, the results might not have good generalizability for strict inclusion criteria and small sample size in those studies. These studies were not sensitive to find rare AEs (e.g. reticulin fibrosis in bone marrow) related to the drug as the sample size was relatively small. In addition, the results of indirect comparisons between ELT and ROM should be interpreted with caution due to lower power of test and heterogeneity caused by the study designs, patient populations, different dosage and course of treatment, and response definitions.

Conclusions
In summary, this meta-analysis suggests that ELT and ROM might be similar in efficacy and safety for adult ITP. However, physicians should still take into account drug cost and comorbidities of the specific patient while making decisions on the treatment of ITP with TPO-RAs.
Supporting information S1