Romiplostim for Primary Immune Thrombocytopenia in Routine Clinical Practice: Results from a Multicentre Observational Study in Germany

Primary immune thrombocytopenia (ITP) is a rare autoimmune disease characterized by low platelet counts (<100 × 10⁹/L) resulting from increased platelet degradation and impaired platelet production [1-3]. The main symptom of ITP is an increased bleeding tendency, ranging from petechiae and bruising to rare and severe haemorrhagic manifestations such as intracranial haemorrhage [2-5].

Previously, the clinical definition of ITP included an arbitrary distinction between “acute” and “chronic” disease, with the latter described as a duration of ≥6 months [6]. In 2009, the clinical definition of ITP was updated to include three phases, based on the duration of disease: newly diagnosed (<3 months from ITP diagnosis), persistent (3–12 months), and chronic (>12 months) [7]. This update in the clinical definition of ITP has posed a number of challenges in terms of treatment decisions as initial drug indications, and associated clinical trial data were based on the old definition.

Current guideline recommendations include corticosteroids, intravenous immunoglobulin (IVIg), or anti-D as first-line treatment options for newly diagnosed ITP, with romiplostim considered a second-line treatment option [3, 8, 9]. Romiplostim is a thrombopoietin-receptor agonist (TPO-RA) that binds to the TPO receptor, activating the JAK2/STAT5 pathway and ultimately leading to increased platelet production [10]. Romiplostim has demonstrated sustained platelet responses together with low toxicity in patients with chronic ITP [11, 12], with some patients experiencing treatment-free responses [11, 13].

While romiplostim was initially approved for second-line treatment of chronic ITP in patients who are refractory to other treatments, such as corticosteroids, evidence from clinical [14, 15] and real-world studies [16-20] indicated that it could also produce durable platelet responses in patients with newly diagnosed or persistent corticosteroid-refractory ITP [14, 17]. Furthermore, in an integrated analysis of nine studies, treatment-free responses to romiplostim were more likely in patients who had ITP for ≤1 year than in those with ITP for >1 year [13]. Recently, as a result of the evidence, the US Food and Drug Administration and the European Medicines Agency removed the restriction of “chronic” from their romiplostim indications for adult patients with ITP in 2019 and 2021, respectively [21, 22]. Following extension of the regulatory label, romiplostim is now indicated in Europe for the treatment of primary ITP in adult patients who are refractory to other treatments (e.g., corticosteroids and immunoglobulins) [22]. This updated label is aligned with major international and regional guidelines that advocate the second-line use of romiplostim within 1 year of ITP diagnosis. The American Society of Hematology recommends TPO-RAs as treatment options in patients with ITP for ≥3 months who are corticosteroid dependent or have no response to corticosteroids [8]. Similarly, the Joint Working Group of European Haematology Societies in Germany, Austria, and Switzerland recommends that TPO-RAs should be offered as second-line therapy, and that they may be considered for all patients who do not respond to first-line therapy, regardless of disease duration [3].

Decision-making surrounding the optimal treatment pathway for patients with ITP can be challenging in clinical practice. To guide future recommendations and improve outcomes for patients, further real-world data on the use of romiplostim in the newly diagnosed and persistent phases of ITP are warranted. This study describes a population of adults with ITP treated with romiplostim in routine clinical practice in Germany; in particular, the effectiveness and safety of romiplostim were evaluated according to the three phases of ITP.

This was a post hoc analysis of a prospective, multicentre, single-arm, observational study that was conducted in Germany at academic and non-academic clinical practices throughout the country to ensure a representative population of patients. The results from the full study population have been presented previously [23]. Data were collected on the first patient in January 2010, and data of the last patient were collected in July 2016. The original study design is depicted in Figure 1. Patients were followed for 2 years (median [Q1, Q3]: 104.0 [100, 104] weeks) from the day of romiplostim initiation or until death, loss to follow-up, or withdrawal of informed consent (whichever occurred first). Patients were observed for the full period irrespective of continuation of therapy with romiplostim.

The study population included patients aged ≥18 years with a diagnosis of ITP (according to the Joint Working Group of European Haematology Societies in Germany, Austria, and Switzerland, American Society of Hematology, or British Committee for the Standards in Haematology criteria [3, 6, 8]) who had received at least one dose of romiplostim. Patients were treated according to routine clinical practice, and the decision to treat was freely undertaken by the clinician. Exclusion criteria were treatment (or planned treatment) with platelet-related products (e.g., pegylated recombinant human megakaryocyte growth and development factor, recombinant human TPO, other TPO-RAs, or related platelet products), simultaneous participation in other clinical studies, initiation of romiplostim treatment prior to commercial availability in Germany, and hypersensitivity to romiplostim, its excipients, or Escherichia coli-derived proteins. All patients provided informed consent.

Following the first dose of romiplostim, follow-up data were collected at 3-monthly intervals for up to 2 years using a standardized electronic clinical response form (eCRF) at all sites. Baseline demographics, patient characteristics, and ITP-specific medical history were abstracted from medical records. Data were also collected for prior ITP-related hospitalizations, defined as those that occurred within a 2-year period before initiation of romiplostim treatment. Reasons for hospitalization included red blood cell transfusion, platelet transfusion, ITP treatment administration, bleeding event, infection, splenectomy, or others. Data on bleeding events up to 6 months prior to initiation of romiplostim were abstracted from medical records. Bleeding events were classified according to the World Health Organization scale [24]. Data collected after initiation of romiplostim included romiplostim dose, number of doses, duration of treatment, adverse drug reactions (ADRs), platelet count over time, bleeding events, ITP-related hospitalization, and discontinuations.

Outcomes included the proportion of patients with an overall platelet response (≥1 platelet count measurement ≥50 × 10⁹/L during 2–24 weeks after romiplostim initiation); the proportion of patients with a durable platelet response (≥75% of all platelet count measurements ≥50 × 10⁹/L during 14–24 weeks after romiplostim initiation); time to first platelet response (≥50 × 10⁹/L) during 2–24 weeks after romiplostim initiation; median platelet count during 2–24 weeks after romiplostim initiation; and the proportions of patients with ITP therapies, bleeding events, transfusion, or ITP-related hospitalization. All ADRs (including thrombotic and bone marrow ADRs) that occurred from the day after initiation of romiplostim therapy through 2 years were evaluated, irrespective of causality to treatment (coded according to Common Terminology Criteria for Adverse Events [CTCAE]).

This was a post hoc analysis of an observational study with no formal hypotheses and hence no power calculations; the planned sample size of approximately 150 patients to enrol into the study was based on the cumulative exposure to romiplostim. Analyses were conducted in patients who met the study eligibility criteria and with known dates for ITP diagnosis and romiplostim initiation. Baseline was defined as the time of romiplostim initiation. All outcomes were analysed by ITP phase at romiplostim initiation: newly diagnosed, persistent, or chronic ITP was defined as time from ITP diagnosis to romiplostim initiation of <3 months, 3–12 months, and >12 months, respectively (i.e., as defined by the International Working Group in 2009 [7]).

Analyses were descriptive in nature. In each ITP phase, categorical outcomes were summarized using frequencies and percentages (with 95% confidence intervals [CIs] where appropriate), and continuous data were summarized using mean and standard deviation (or 95% CI where appropriate) or median and interquartile range (IQR). Additionally, unweighted mean dose and median platelet count were calculated for each patient over the follow-up period. Missing data were not imputed, except for start/end dates for ADRs and medication. Where a start date for an ADR/medication was partial or missing, the following imputation rules were applied: if the day portion was missing, it was set to day 1; if the month portion was missing, it was set to January 1; if the date was completely missing or if the imputation date was prior to the first dose date, the date was set to the first dose date. Where an end date for an ADR/medication was partial or missing for ADRs that were resolved (end date was not imputed for not resolved ADRs), the following imputation rules were applied: if the day portion was missing, the date was set to the last day of the month; if the month portion was missing, the date was set to December 31; if the date was completely missing or if the imputation date was after the end of the patient’s observational period, the date was set to the date of discontinuation.

A total of 159 patients were enrolled at 38 sites in Germany (online suppl. Table S1; for all online suppl. material, see www.karger.com/doi/10.1159/000521689) and were included in the study; 22 patients were excluded from the ITP phase analysis set due to protocol violations (exclusion criteria met [n = 12], inclusion criteria not met [n = 3], documentation incomplete/not valid [n = 4], and screening failure [n = 3]). The ITP phase for 41 patients could not be determined due to a missing date of ITP diagnosis, and these patients were also excluded from the analyses (shown in Fig. 2). Of the remaining 96 patients, 18 patients (18.8%) had newly diagnosed ITP (<3 months since diagnosis), 25 (26.0%) had persistent ITP (3–12 months), and 53 (55.2%) had chronic ITP (>12 months).

Baseline demographics and characteristics are presented in Table 1. Across ITP phases, median age was 64–70 years, and median baseline platelet counts were similar, ranging from 28.0 to 31.5 × 10⁹/L; for most patients, at least one ITP therapy prior to initiation of romiplostim was documented. The most common ITP therapies received by patients were corticosteroids and IVIg. The proportion of patients who had received a platelet transfusion was 11.1–18.9% across ITP phases. A bone marrow biopsy had been performed prior to initiation of romiplostim in 54.7–72.0% of patients across ITP phases. At baseline, no patients with newly diagnosed ITP were splenectomized, whereas 12.0% and 11.3% of patients with persistent or chronic ITP, respectively, were splenectomized.

Most patients received a romiplostim starting dose of approximately 1 µg/kg (Table 2). The median weekly doses of romiplostim during the 2-year follow-up period ranged from 1.5 µg/kg to 2.4 µg/kg in patients with newly diagnosed, persistent, and chronic ITP.

Over the 2-year follow-up, patients with persistent ITP received a lower median number of romiplostim administrations (37 administrations) across a lower median number of weeks (87 weeks) compared with patients with newly diagnosed or chronic ITP (67 and 72 administrations across a median of 102 and 104 weeks, respectively; Table 2). Mean romiplostim dose over time is presented in Figure 3.

Across 2–24 weeks of romiplostim treatment, median platelet counts were >100 × 10⁹/L in patients with newly diagnosed, persistent, and chronic ITP (shown in Table 3; Fig. 4). An overall platelet response (≥50 × 10⁹/L) was achieved in all patients with newly diagnosed ITP or persistent ITP and almost all patients (96.2%) with chronic ITP. A durable platelet response (14–24 weeks) was achieved in 65.0–88.2% of patients across ITP phases. The median time from baseline to first platelet count ≥50 × 10⁹/L was similar in patients with newly diagnosed, persistent, and chronic ITP (approximately 1 week).

Following initiation of romiplostim treatment, at least one concomitant ITP therapy was administered in over half of the patients with newly diagnosed ITP (61.1%) and less than half in patients with persistent and chronic ITP (28.0% and 39.6%, respectively; Table 3). The most frequently administered medications to patients with newly diagnosed, persistent, and chronic ITP were corticosteroids and IVIg.

Six (33.3%), 6 (24.0%), and 19 (35.8%) patients with newly diagnosed, persistent, and chronic ITP, respectively, experienced a total of 38, 14, and 46 ADRs (ADRs were irrespective of causality to treatment; Table 4). Two (8.0%) and 2 (3.8%) patients with persistent and chronic ITP, respectively, discontinued romiplostim due to ADRs. The most common ADRs occurring in ≥5% of patients with newly diagnosed, persistent, and chronic ITP were fatigue, dizziness, and vomiting and diarrhoea in patients with chronic ITP only. Two patients (4.3%) with chronic ITP, both female and aged 67 years, experienced bone marrow fibrosis: 1 patient experienced myelofibrosis (CTCAE grade 1) with continuation of romiplostim and the second patient had severe myelofibrosis (CTCAE grade 3) with discontinuation of romiplostim. Neither patient had a record of a bone marrow biopsy before initiation of romiplostim. No thrombotic ADRs were reported.

In the 6 months prior to romiplostim therapy, bleeding events were reported in 30.2–40.0% of patients across ITP phases; 1 newly diagnosed patient experienced grade 3 or 4 bleeding events and required transfusion (Table 1). Following initiation of romiplostim therapy (through 2 years of follow-up), bleeding events occurred in 33.3–47.2% of patients across ITP phases. The majority of events were grade 1 or 2 in severity; 3 patients with newly diagnosed ITP and 1 with chronic ITP experienced grade 3 or 4 bleeding events and required transfusion (Table 3).

In the 6 months prior to romiplostim treatment, ITP-related hospitalizations were reported in 13.2–36.0% of patients across ITP phases (Table 1); after romiplostim treatment (during 2 years of follow-up), ITP-related -hospitalizations were reported in 8.0–16.7% of patients (Table 3).

The aim of this post hoc analysis of an observational study was to describe the population of adult patients with ITP who received romiplostim in routine clinical practice in Germany and to gain insights into the effectiveness and safety of romiplostim according to the phase of ITP in these patients. The results of our study showed that clinically meaningful platelet responses were achieved and sustained in the majority of romiplostim-treated patients in all ITP phases. Romiplostim was well tolerated regardless of ITP phase at romiplostim initiation, with minimal discontinuations due to ADRs.

Our study sample was representative of the general German ITP population, as shown by comparison to patients in a national survey (2016–2017) [25]. In both study cohorts, over 50% of patients had chronic ITP (>12 months); corticosteroids and IVIg were the most frequently used treatments prior to romiplostim initiation. Similarly, rates of splenectomy were relatively low across all ITP phases in this study and the national survey [25]. Over half of the patients in all ITP phases underwent bone marrow biopsy prior to romiplostim treatment. This is important to note, as patients with chronic ITP who undergo bone marrow biopsy after diagnosis have more severe disease and are at greater risk of potentially serious complications [26].

Notably, in the present study, romiplostim was initiated within the first year following ITP diagnosis in almost half of the patients, despite the chronic indication included in the European label at the time of the study. A substantial proportion of patients with newly diagnosed or persistent ITP have also been treated with romiplostim in other European countries, as demonstrated in real-world studies [17, 19, 27, 28]. This real-world usage in earlier treatment lines is aligned with Joint Working Group of European Haematology Societies in Germany, Austria, and Switzerland guidelines, which recommend the use of TPO-RAs in all patients who do not respond to first-line therapy, regardless of ITP stage [3], and with the American Society of Hematology guidelines, which recommend the use of TPO-RAs in adults with ITP for ≥3 months who are corticosteroid dependent or unresponsive to corticosteroids [8].

Platelet responses to romiplostim were achieved rapidly and consistently by patients in all ITP phases. The time to first platelet response was generally consistent across ITP phases, ranging from 7 days in patients with persistent or chronic ITP to 8.5 days in those with newly diagnosed ITP. Moreover, patients with newly diagnosed and persistent ITP achieved high rates of overall platelet responses (100%) and high rates of durable platelet responses (65.0–88.2%), suggesting effectiveness of romiplostim within 1 year of ITP diagnosis, as well as in patients with chronic disease. Our findings were similar to those reported from clinical studies, which indicate that romiplostim can lead to sustained platelet responses in the majority of patients regardless of whether they have newly diagnosed, persistent, or chronic ITP [11, 13, 14, 20]. Platelet response definitions can be highly variable between different ITP studies; however, the same platelet count threshold of ≥50 × 10⁹/L used in this study has also been used in previous studies of romiplostim and other ITP therapies [11, 13, 14, 20, 29-32]. Although neither a dosing algorithm nor a target platelet count was pre-defined in this observational study, the majority of patients across all ITP phases were treated with an initial dose of 1 µg/kg in line with the recommended dose for patients from the European label [22] (noting that a substantial proportion, 15–24%, started above this dose). Furthermore, the median weekly romiplostim dose during the first 24 weeks of romiplostim treatment (1.4–2.4 µg/kg) corresponds well to the median dose used in the registrational trials (i.e., 2–3 µg/kg) [22].

The results from this study confirm the established favourable safety profile of romiplostim from previous clinical studies [11-14] and real-world studies [17-19]. The proportion of patients with an ADR leading to romiplostim withdrawal was low (4%), no patients experienced thrombotic ADRs, and bone marrow-related ADRs were reported in only 2 patients (both with chronic ITP). In the overall study population, the rate of time-adjusted bleeding events of at least grade 3 in severity prior to romiplostim was 7.2/100 patient-years [23]. This time-adjusted rate decreased to 4.0/100 patient-years following initiation of romiplostim [23]. In contrast, corticosteroid use is associated with a number of side effects, such as weight gain, hypertension, increased risk of infection, and osteoporosis [3]. Despite guidelines warning that first-line corticosteroids should be used for a limited period of time (e.g., to a maximum of 6 weeks for prednisolone) [3, 8, 9], there is an over-reliance on corticosteroids in routine clinical practice [25]. Therefore, the earlier second-line use of romiplostim may help to avoid side effects associated with corticosteroid use and decrease the need for other subsequent therapies, such as splenectomy, that may have long-term health risks.

A limitation of the study was the observational design. There are no data in this study on the severity of ITP in patients at the point of romiplostim initiation; however, baseline platelet count and prior bleeding events were similar irrespective of ITP duration, suggesting that patients with chronic ITP did not have more severe disease than patients with newly diagnosed or persistent ITP. Non-random selection of sites, investigators, and patients may have led to some degree of bias; however, sites were selected to ensure a good geographical spread throughout Germany, and restrictions on the number of participants enrolled per investigator likely reduced any potential investigator bias. Additionally, the collection of data through standardized eCRFs aimed to minimize misclassification or measurement error, and eCRFs were regularly verified to ensure data quality. A further potential limitation of the present study was that any potential pre-existing bone marrow diseases could not be excluded, as patients are not mandated to have a bone marrow biopsy prior to romiplostim initiation in clinical practice in Germany.

This study provides an overview of romiplostim utilization across a wide spectrum of patients with ITP undergoing treatment in routine clinical practice in Germany. Our findings confirm the effectiveness and safety of romiplostim across all phases of ITP, thereby supporting the use of romiplostim in adults with newly diagnosed, persistent, or chronic ITP who are refractory to other treatments, such as corticosteroids and IVIg.

We thank all patients and investigators involved in the study. Medical writing support, including development of a draft outline and subsequent drafts in consultation with the authors, collating author comments, copyediting, fact checking, and referencing, was provided by Heather Davies, PhD, and Ryan Woodrow, BSc (Hon), CMPP, at Aspire Scientific Limited (Bollington, UK). Funding for medical writing support for this article was provided by Amgen (Europe) (Rotkreuz, Switzerland).

All patients provided informed consent. The Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, BfArM, Germany) was notified of this non-interventional (observational) study on November 24, 2009 (Notification No. 281), and the competent Ethics Committee (State Medical Association of Baden-Württemberg, Germany) was notified and did not raise any objections on December 15, 2009 (file number 2009-150-f). The non-interventional study was conducted in accordance with the local law.

M. Reiser, K.M. Josten, and H. Dietzfelbinger have nothing to declare. A. Seesaghur, M. Schill, and J. Hippenmeyer are employees and shareholders of Amgen. M. Welslau reports advisory or expert activity for Amgen, Bristol Myers Squibb, Celgene, Gilead, Hexal, Janssen, Lilly, medac, Novartis, Roche, and Sanofi, and receipt of honoraria from Amgen, Astellas, AstraZeneca, Celgene, Gilead, Hexal, Janssen, Lilly, Novartis, Roche, and Sanofi.

This study was funded by Amgen (Europe) GmbH. The study sponsor, Amgen (Europe) GmbH (München, Germany), played a role in the study design, data collection and analysis, decision to publish, and preparation of the manuscript.

M.R., K.M.J., H.D., and M.W. collected study data. All authors contributed to the analysis or interpretation of the data, revision of the manuscript, and approved the final version.

Qualified researchers may request data from Amgen clinical studies. Complete details are available at the following: https://www.amgen.com/science/clinical-trials/clinical-data-transparency-practices/clinical-trial-data-sharing-request/.