Update on romiplostim therapy for immune thrombocytopenic purpura

Romiplostim (Nplate) is a thrombopoietin receptor agonist that is currently indicated for the treatment of thrombocytopenia in patients with chronic immune (idiopathic) thrombocytopenic purpura (ITP) who have had an insufficient response to corticosteroids, immunoglobulins, or splenectomy. In a recently reported phase III trial,1 romiplostim treatment was associated with a number of benefits compared with the standard of care in nonsplenectomized ITP patients: higher platelet response rate, lower rates of treatment failure and splenectomy, fewer bleeding events, and fewer blood transfusions.

The safety and efficacy of romiplostim in children with ITP have not yet been established. A recently reported placebo-controlled trial indicates good response and good tolerability of romiplostim in children with chronic refractory ITP.2

Romiplostim versus standard of care in adult nonsplenectomized patients

In a 52-week, multicenter, openlabel trial,1 patients with ITP who had not undergone splenectomy, had received at least one prior treatment for ITP, and had a platelet count of less than 50 × 109/L were randomized to receive weekly SC injections of romiplostim (n = 157) or standard-ofcare treatment (n = 77). Romiplostim was started at a dose of 3 μg/kg, which could be increased up to a maximum of 10 μg/kg to achieve a target platelet count of 50–200 × 109/L. Standardof- care treatment was selected by the treating physician based on standard institutional practices or therapeutic guidelines. Throughout the study, patients in either treatment group could receive additional therapies for ITP, including short-term rescue therapy, such as IV immune globulin (IVIG), but excluding other thrombopoietin mimetics, as deemed medically necessary by investigators.

Study population, endpoints, other treatments

The median ages of patients in the romiplostim and standard-of-care groups were 58 and 57 years, respectively; 54% and 60% were women; the median duration since ITP diagnosis was 2.1 and 2.3 years; baseline median platelet counts were 33 × 109/L and 27 × 109/L; and 13% and 6% were receiving medications, primarily glucocorticoids (11% and 3%), for ITP at baseline. Splenectomy could be performed if study therapy was considered to be ineffective or was associated with severe side effects.

The primary endpoints of the trial were the incidence of splenectomy and the incidence of treatment failure, with treatment failure defined as a platelet count of 20 × 109/L or lower for 4 consecutive weeks, a major bleeding event, or a requirement for a change in therapy (including splenectomy) due to an adverse event or bleeding. In primary endpoint analyses, patients who received any study treatment and then discontinued study participation were counted as having both treatment failure and splenectomy.

During the study, ITP treatments other than romiplostim were received by 44% of patients in the romiplostim group and 79% of patients in the standard- of-care group, including glucocorticoids in 37% versus 63%, IVIG in 7% versus 33%, rituximab (Rituxan) in 1% versus 20%, azathioprine in 1% versus 9%, danazol in 2% versus 7%, other medications in 6% versus 19%, and platelet transfusions in 6% versus 16%.

Efficacy

The incidence of treatment failure was 11% in the romiplostim group versus 30% in the standard-of-care group (P < 0.001). Time to treatment failure was significantly prolonged in the romiplostim group (Figure 1a). The incidence of splenectomy was 9% in the romiplostim group versus 36% in the standard-of-care group (P < 0.001), and time to splenectomy was significantly prolonged in the include bleeding, thrombosis, hematologic cancer or myelodysplastic syndromes, and increased bone marrow reticulin.

After adjustment for duration of study-drug exposure, the romiplostim group had significantly fewer incidences of overall bleeding (P = 0.001) and grade 3 or higher bleeding (P = 0.02); no significant difference between the two treatment groups was observed for less severe bleeding. Overall, 260 bleeding events occurred in 80 patients (52%) in the romiplostim group, for a bleeding rate of 3.56 events/100 patient-weeks; 153 events occurred in 40 patients (53%) in the standard-of-care group, for a rate of 5.02 events/100 patient-weeks. Eight grade 3 or higher bleeding events occurred in five romiplostim-treated patients (3%), a rate of 0.11 events/100 patient weeks; 10 events occurred in five patients (7%) in the standard-ofcare group, a rate of 0.33 events/100 patient-weeks.

A total of 41 blood transfusions were given to 12 patients (8%) in the romiplostim group, whereas a total of 76 transfusions were given to 13 patients (17%) in the standard-of-care group. There was no significant difference between the two groups with regard to the occurrence of thrombotic events. A total of 11 thrombotic events occurred in six patients (4%) in the romiplostim group, yielding a rate of 0.15 events/100 patientweeks, whereas two events occurred in two patients (3%) in the standardof- care group, yielding a rate of 0.07 events/100 patient-weeks.

Two cases of hematologic cancer were observed, consisting of lymphoma and myelodysplastic syndrome in one patient each in the standard-ofcare group. Bone marrow reticulin was found in one romiplostim-treated patient during 6 months posttreatment follow up, with the level being within the normal range (grade 2).

Thrombopoietin mimetics challenge the conventional wisdom about controlling ITP

Chronic immune thrombocytopenic purpura (ITP) affects 60,000 adults in the United States and is associated with the risk of life-threatening hemorrhage. For decades, the standard treatments included glucocorticoids and splenectomy. The advent and subsequent regulatory approval of the thrombopoietin mimetics romiplostim (Nplate) and eltrombopag (Promacta) have not only added to the therapeutic arsenal for this orphan disease but also successfully challenged the conventional wisdom that the key to controlling ITP is to reduce platelet destruction. Instead, these agents work by increasing intramedullary platelet production and are thus able to outpace the rate of peripheral destruction.

As reported by Kuter and colleagues last November in The New England Journal of Medicine, romiplostim is safe, well tolerated, and highly effective in the adult population with chronic ITP, including individuals with an intact spleen. Among the remaining questions regarding its use are: Where in the sequence of treatments for ITP do thrombopoietin mimetics belong? What is the long-term safety of these agents, given the changes to the bone marrow microenvironment that they induce? And, what will the cost impact be of a treatment that is designed for disease maintenance as opposed to providing a cure.

Thrombopoietin mimetics have not been widely tested for the treatment of pediatric ITP. Two small trials have shown that in children with disease that has been refractory to all standard approaches (with the exception of splenectomy), romiplostim is well tolerated, effective, and results in a meaningful reduction of clinically significant bleeding episodes. These data are preliminary but nevertheless provocative, and they offer the potential for new hope to children with chronic refractory ITP. Additional experience will be required to prove long-term safety and tolerability in the pediatric population before these new agents will be adopted into standard practice.

— David M.J. Hoffman, MD, FACP
Tower Hematology Oncology Medical Group
Beverly Hills, CA

Three deaths occurred during the study treatment period, including one death due to pneumonia in one romiplostim-treated patient and two deaths due to hepatic failure and cardiorespiratory arrest, respectively, in two patients in the standard-of-care group. Three additional deaths due to metastatic lung cancer, left ventricular failure, and hepatic neoplasm occurred in the standard-of-care group during the 6-month posttreatment follow up. None of the deaths was attributed to study treatment. No neutralizing antibodies to romiplostim or thrombopoietin were detected.

Quality of life

Quality of life was assessed by the ITP Patient Assessment Questionnaire, consisting of 44 ITP-specific items on each of 10 scales ranging from 0 to 100 points each (with higher scores indicating better quality of life). Scores on two scales (Women’s Reproductive Health and Work Quality of Life) could not be assessed due to inadequacies of the statistical model used in the analysis. Of the eight scales assessed, clinically significant increases of 8–15 points were observed for both treatment groups on all but the Fatigue scale. The romiplostim treatment group showed statistically greater improvements on the Symptoms (P = 0.01), Bother (P = 0.008), Activity (P = 0.02), Psychological (P = 0.049), Fear (P < 0.001), Social Quality of Life (P = 0.002), and Overall Quality of Life (P = 0.02) scales compared with the standard-of-care group, although the between-group differences of 2–8 points on these scales are of uncertain clinical significance.

Romiplostim in children with chronic refractory ITP

Few data exist on the effects of romiplostim therapy in pediatric ITP patients. In a recent single-blind, placebo-controlled trial,2 18 patients aged 2.5 to 16 years with chronic refractory ITP (baseline platelet count < 20 × 109/L) who had not undergone splenectomy were randomized to receive weekly SC injections of romiplostim (n = 12) or placebo (n = 6) for 12 weeks. Romiplostim was started at a dose of 1 μg/kg, escalated to 5 μg/kg at 5 weeks, and tapered afterward.

All patients had either no response or failed to maintain a response to at least two prior treatment modalities for ITP. All had received prior steroid treatment; 44% had received corticosteroids, IVIG, and anti-D immunogloblin in combination or sequentially; and 22% had received cytotoxic or immune-modulating agents. All such treatments were stopped 2 weeks prior to the study.

For patients in the romiplostim versus placebo group, mean age was 9.5 versus 7.0 years, 10 (83%) versus 3 (50%) were male, and median disease duration was 2.3 versus 3.0 years. Median baseline platelet counts were 10.5 × 109/L in both groups.

The median platelet count in the romiplostim group was significantly higher (P = 0.039) than that in the placebo group within 1 week of the first dose of romiplostim (1 μg/kg) and remained so at 3 weeks after the end of treatment (15 weeks total; median platelet count of 47.5 × 109/L vs 19.0 × 109/L; P = 0.001). Changes in platelet count in the romiplostim group were dose-dependent, with a median peak platelet count of 73.5 × 109/L reached after 5 weeks, when the dose was 5 μg/kg. Eleven of the 12 patients in the romiplostim group (93%) reached the target range of greater than 50 × 109/L by the fifth week. At week 12 (end of treatment), 10 romiplostim patients (83%) versus no placebo patients were at target levels. Six romiplostim-treated patients (50%) maintained target platelet count levels at 3 weeks after treatment.

Adverse events occurred in 50% of patients in each group. The most frequent adverse events were headache, epistaxis, cough, and vomiting, which occurred in one patient each in the romiplostim group (8%) and placebo group (17%). Two romiplostim-treated patients (17%) developed a skin rash. None of the patients had thrombocytosis or rebound thrombocytopenia, and none of the romiplostim-treated patients developed bone marrow fibrosis by week 18 of follow up.

How I treat ITP Immune thrombocytopenic purpura (ITP) and its associated problems are frequently encountered by the practicing hematologist. When it comes treatment, there are several options:

• If the platelet count is at least 30,000/μL and there is no active bleeding, close observation is the most prudent approach. In anticipation of the possibility of splenectomy at some point in their course, patients are offered immunization against Streptococcus pneumoniae, Hemophilus influenza, and Neisseria meningitides. In addition, I screen for exposure to hepatitis B virus (HBV), because patients may at some point receive treatment with rituximab (Rituxan), which is associated with HBV reactivation.
• For patients with a platelet count below 30,000/μL, evidence of impending bleeding (such as wet purpura), or frank bleeding, treatment is indicated. Typically, this involves the use of glucocorticoids, often in combination with IV gamma globulin (IVIG) or, occasionally, anti-D immunoglobulin.
• For a platelet count below 10,000/μL, I give methylprednisolone, 1,000 mg/d IV, for 5 consecutive days with IVIG, 1 g/kg daily, for 2 consecutive days. I then shift the patient to oral prednisone, 1 mg/kg daily, for 1 week, followed by tapering the dose by 10 mg/wk, as allowed by the platelet count, which should remain at ≥ 30,000/μL. Failure of this approach, or the inability to reduce the prednisone dose to 10 mg/d or less, would raise the options of either giving rituximab, 375 mg/m2 per week, for 4 weeks or performing a therapeutic, laparoscopic-assisted splenectomy, the single intervention still the most likely to provide meaningful, long-lasting benefit.
• In the setting of asplenia (including the absence of accessory splenic tissue), I favor a trial of a thrombopoietin mimetic, either romiplostim (Nplate) or eltrombopag (Promacta).

In the truly refractory patient, other options include calcineurin inhibitors, cytotoxic agents, danazol, tumor necrosis factor-alpha inhibitors, and staphylococcal A column immunoadsorption.

— David M.J. Hoffman, MD, FACP

Rescue medication, consisting of IVIG 1 g/kg for two doses, was given to one romiplostim-treated patient (8%) due to head trauma and loss of consciousness and two placebo-treated patients (33%) during the 12-week study period, with no interruption of study drug being required. The number of romiplostim-treated patients with grade 3 bleeding decreased from four (33%) prior to the study to none during the study, and the number with grade 2 bleeding decreased from six (50%) to two (17%; P = 0.002).

These findings are similar to results observed in another small study of romiplostim in children with ITP.3,4 An open-label phase III study designed to study the long-term safety of romiplostim and the durability of platelet responses to the drug in pediatric patients currently is in the recruitment stage.5 In this extension trial, approximately 20 patients aged 1 to 18 years with ITP are to receive weekly SC injections of romiplostim, starting at 1 μg/kg (or prior dose) and escalated to 10 μg/kg (based on the platelet count) over a period of 3 years. The primary outcome measure is the incidence of adverse events, including significant changes in laboratory values and the incidence of antibody formation. Secondary outcome measures include the platelet response (> 50 × 109/L) in the absence of rescue medication and the need for concurrent ITP medication (corticosteroids, danazol, or azathioprine) over the duration of the study.

References

1. Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard of care in patients with immune thrombocytopenia. N Engl J Med 2010;363:1889–1899.
2. Elalfy MS, Abdelmaksoud AA, Eltonbary KY. Romiplostim in children with chronic refractory ITP: randomized placebo controlled study. Ann Hematol 2011;doi10.1007/s00277- 011-1172-9.
3. Buchanan GR, Bomgaars L, Bussel JB, et al. A randomized, double-blind, placebo-controlled phase 1/2 study to determine the safety and efficacy of romiplostim in children with chronic immune (idiopathic) thrombocytopenic purpura (ITP). Blood 2009;114(22):680.
4. Safety and efficacy study of AMG 531 to treat ITP in pediatric subjects. ClinicalTrials. gov Web site. http://clinicaltrials.gov/ct2/ show/NCT00515203. Accessed May 12, 2011.
5. A study evaluating the safety and efficacy of long-term dosing of romiplostim in thrombocytopenic pediatric subjects with immune (idiopathic) thrombocytopenia purpura. ClinicalTrials. gov Web site. http://clinicaltrials. gov/ct2/show/study/NCT01071954. Accessed May 12, 2011.

Romiplostim (Nplate) is a thrombopoietin receptor agonist that is currently indicated for the treatment of thrombocytopenia in patients with chronic immune (idiopathic) thrombocytopenic purpura (ITP) who have had an insufficient response to corticosteroids, immunoglobulins, or splenectomy. In a recently reported phase III trial,1 romiplostim treatment was associated with a number of benefits compared with the standard of care in nonsplenectomized ITP patients: higher platelet response rate, lower rates of treatment failure and splenectomy, fewer bleeding events, and fewer blood transfusions.

The safety and efficacy of romiplostim in children with ITP have not yet been established. A recently reported placebo-controlled trial indicates good response and good tolerability of romiplostim in children with chronic refractory ITP.2

Romiplostim versus standard of care in adult nonsplenectomized patients

In a 52-week, multicenter, openlabel trial,1 patients with ITP who had not undergone splenectomy, had received at least one prior treatment for ITP, and had a platelet count of less than 50 × 109/L were randomized to receive weekly SC injections of romiplostim (n = 157) or standard-ofcare treatment (n = 77). Romiplostim was started at a dose of 3 μg/kg, which could be increased up to a maximum of 10 μg/kg to achieve a target platelet count of 50–200 × 109/L. Standardof- care treatment was selected by the treating physician based on standard institutional practices or therapeutic guidelines. Throughout the study, patients in either treatment group could receive additional therapies for ITP, including short-term rescue therapy, such as IV immune globulin (IVIG), but excluding other thrombopoietin mimetics, as deemed medically necessary by investigators.

Study population, endpoints, other treatments

The median ages of patients in the romiplostim and standard-of-care groups were 58 and 57 years, respectively; 54% and 60% were women; the median duration since ITP diagnosis was 2.1 and 2.3 years; baseline median platelet counts were 33 × 109/L and 27 × 109/L; and 13% and 6% were receiving medications, primarily glucocorticoids (11% and 3%), for ITP at baseline. Splenectomy could be performed if study therapy was considered to be ineffective or was associated with severe side effects.

The primary endpoints of the trial were the incidence of splenectomy and the incidence of treatment failure, with treatment failure defined as a platelet count of 20 × 109/L or lower for 4 consecutive weeks, a major bleeding event, or a requirement for a change in therapy (including splenectomy) due to an adverse event or bleeding. In primary endpoint analyses, patients who received any study treatment and then discontinued study participation were counted as having both treatment failure and splenectomy.

During the study, ITP treatments other than romiplostim were received by 44% of patients in the romiplostim group and 79% of patients in the standard- of-care group, including glucocorticoids in 37% versus 63%, IVIG in 7% versus 33%, rituximab (Rituxan) in 1% versus 20%, azathioprine in 1% versus 9%, danazol in 2% versus 7%, other medications in 6% versus 19%, and platelet transfusions in 6% versus 16%.

Efficacy

The incidence of treatment failure was 11% in the romiplostim group versus 30% in the standard-of-care group (P < 0.001). Time to treatment failure was significantly prolonged in the romiplostim group (Figure 1a). The incidence of splenectomy was 9% in the romiplostim group versus 36% in the standard-of-care group (P < 0.001), and time to splenectomy was significantly prolonged in the include bleeding, thrombosis, hematologic cancer or myelodysplastic syndromes, and increased bone marrow reticulin.

After adjustment for duration of study-drug exposure, the romiplostim group had significantly fewer incidences of overall bleeding (P = 0.001) and grade 3 or higher bleeding (P = 0.02); no significant difference between the two treatment groups was observed for less severe bleeding. Overall, 260 bleeding events occurred in 80 patients (52%) in the romiplostim group, for a bleeding rate of 3.56 events/100 patient-weeks; 153 events occurred in 40 patients (53%) in the standard-of-care group, for a rate of 5.02 events/100 patient-weeks. Eight grade 3 or higher bleeding events occurred in five romiplostim-treated patients (3%), a rate of 0.11 events/100 patient weeks; 10 events occurred in five patients (7%) in the standard-ofcare group, a rate of 0.33 events/100 patient-weeks.

A total of 41 blood transfusions were given to 12 patients (8%) in the romiplostim group, whereas a total of 76 transfusions were given to 13 patients (17%) in the standard-of-care group. There was no significant difference between the two groups with regard to the occurrence of thrombotic events. A total of 11 thrombotic events occurred in six patients (4%) in the romiplostim group, yielding a rate of 0.15 events/100 patientweeks, whereas two events occurred in two patients (3%) in the standardof- care group, yielding a rate of 0.07 events/100 patient-weeks.

Two cases of hematologic cancer were observed, consisting of lymphoma and myelodysplastic syndrome in one patient each in the standard-ofcare group. Bone marrow reticulin was found in one romiplostim-treated patient during 6 months posttreatment follow up, with the level being within the normal range (grade 2).

Thrombopoietin mimetics challenge the conventional wisdom about controlling ITP

Chronic immune thrombocytopenic purpura (ITP) affects 60,000 adults in the United States and is associated with the risk of life-threatening hemorrhage. For decades, the standard treatments included glucocorticoids and splenectomy. The advent and subsequent regulatory approval of the thrombopoietin mimetics romiplostim (Nplate) and eltrombopag (Promacta) have not only added to the therapeutic arsenal for this orphan disease but also successfully challenged the conventional wisdom that the key to controlling ITP is to reduce platelet destruction. Instead, these agents work by increasing intramedullary platelet production and are thus able to outpace the rate of peripheral destruction.

As reported by Kuter and colleagues last November in The New England Journal of Medicine, romiplostim is safe, well tolerated, and highly effective in the adult population with chronic ITP, including individuals with an intact spleen. Among the remaining questions regarding its use are: Where in the sequence of treatments for ITP do thrombopoietin mimetics belong? What is the long-term safety of these agents, given the changes to the bone marrow microenvironment that they induce? And, what will the cost impact be of a treatment that is designed for disease maintenance as opposed to providing a cure.

Thrombopoietin mimetics have not been widely tested for the treatment of pediatric ITP. Two small trials have shown that in children with disease that has been refractory to all standard approaches (with the exception of splenectomy), romiplostim is well tolerated, effective, and results in a meaningful reduction of clinically significant bleeding episodes. These data are preliminary but nevertheless provocative, and they offer the potential for new hope to children with chronic refractory ITP. Additional experience will be required to prove long-term safety and tolerability in the pediatric population before these new agents will be adopted into standard practice.

— David M.J. Hoffman, MD, FACP
Tower Hematology Oncology Medical Group
Beverly Hills, CA

Three deaths occurred during the study treatment period, including one death due to pneumonia in one romiplostim-treated patient and two deaths due to hepatic failure and cardiorespiratory arrest, respectively, in two patients in the standard-of-care group. Three additional deaths due to metastatic lung cancer, left ventricular failure, and hepatic neoplasm occurred in the standard-of-care group during the 6-month posttreatment follow up. None of the deaths was attributed to study treatment. No neutralizing antibodies to romiplostim or thrombopoietin were detected.

Quality of life

Quality of life was assessed by the ITP Patient Assessment Questionnaire, consisting of 44 ITP-specific items on each of 10 scales ranging from 0 to 100 points each (with higher scores indicating better quality of life). Scores on two scales (Women’s Reproductive Health and Work Quality of Life) could not be assessed due to inadequacies of the statistical model used in the analysis. Of the eight scales assessed, clinically significant increases of 8–15 points were observed for both treatment groups on all but the Fatigue scale. The romiplostim treatment group showed statistically greater improvements on the Symptoms (P = 0.01), Bother (P = 0.008), Activity (P = 0.02), Psychological (P = 0.049), Fear (P < 0.001), Social Quality of Life (P = 0.002), and Overall Quality of Life (P = 0.02) scales compared with the standard-of-care group, although the between-group differences of 2–8 points on these scales are of uncertain clinical significance.

Romiplostim in children with chronic refractory ITP

Few data exist on the effects of romiplostim therapy in pediatric ITP patients. In a recent single-blind, placebo-controlled trial,2 18 patients aged 2.5 to 16 years with chronic refractory ITP (baseline platelet count < 20 × 109/L) who had not undergone splenectomy were randomized to receive weekly SC injections of romiplostim (n = 12) or placebo (n = 6) for 12 weeks. Romiplostim was started at a dose of 1 μg/kg, escalated to 5 μg/kg at 5 weeks, and tapered afterward.

All patients had either no response or failed to maintain a response to at least two prior treatment modalities for ITP. All had received prior steroid treatment; 44% had received corticosteroids, IVIG, and anti-D immunogloblin in combination or sequentially; and 22% had received cytotoxic or immune-modulating agents. All such treatments were stopped 2 weeks prior to the study.

For patients in the romiplostim versus placebo group, mean age was 9.5 versus 7.0 years, 10 (83%) versus 3 (50%) were male, and median disease duration was 2.3 versus 3.0 years. Median baseline platelet counts were 10.5 × 109/L in both groups.

The median platelet count in the romiplostim group was significantly higher (P = 0.039) than that in the placebo group within 1 week of the first dose of romiplostim (1 μg/kg) and remained so at 3 weeks after the end of treatment (15 weeks total; median platelet count of 47.5 × 109/L vs 19.0 × 109/L; P = 0.001). Changes in platelet count in the romiplostim group were dose-dependent, with a median peak platelet count of 73.5 × 109/L reached after 5 weeks, when the dose was 5 μg/kg. Eleven of the 12 patients in the romiplostim group (93%) reached the target range of greater than 50 × 109/L by the fifth week. At week 12 (end of treatment), 10 romiplostim patients (83%) versus no placebo patients were at target levels. Six romiplostim-treated patients (50%) maintained target platelet count levels at 3 weeks after treatment.

Adverse events occurred in 50% of patients in each group. The most frequent adverse events were headache, epistaxis, cough, and vomiting, which occurred in one patient each in the romiplostim group (8%) and placebo group (17%). Two romiplostim-treated patients (17%) developed a skin rash. None of the patients had thrombocytosis or rebound thrombocytopenia, and none of the romiplostim-treated patients developed bone marrow fibrosis by week 18 of follow up.

How I treat ITP Immune thrombocytopenic purpura (ITP) and its associated problems are frequently encountered by the practicing hematologist. When it comes treatment, there are several options:

• If the platelet count is at least 30,000/μL and there is no active bleeding, close observation is the most prudent approach. In anticipation of the possibility of splenectomy at some point in their course, patients are offered immunization against Streptococcus pneumoniae, Hemophilus influenza, and Neisseria meningitides. In addition, I screen for exposure to hepatitis B virus (HBV), because patients may at some point receive treatment with rituximab (Rituxan), which is associated with HBV reactivation.
• For patients with a platelet count below 30,000/μL, evidence of impending bleeding (such as wet purpura), or frank bleeding, treatment is indicated. Typically, this involves the use of glucocorticoids, often in combination with IV gamma globulin (IVIG) or, occasionally, anti-D immunoglobulin.
• For a platelet count below 10,000/μL, I give methylprednisolone, 1,000 mg/d IV, for 5 consecutive days with IVIG, 1 g/kg daily, for 2 consecutive days. I then shift the patient to oral prednisone, 1 mg/kg daily, for 1 week, followed by tapering the dose by 10 mg/wk, as allowed by the platelet count, which should remain at ≥ 30,000/μL. Failure of this approach, or the inability to reduce the prednisone dose to 10 mg/d or less, would raise the options of either giving rituximab, 375 mg/m2 per week, for 4 weeks or performing a therapeutic, laparoscopic-assisted splenectomy, the single intervention still the most likely to provide meaningful, long-lasting benefit.
• In the setting of asplenia (including the absence of accessory splenic tissue), I favor a trial of a thrombopoietin mimetic, either romiplostim (Nplate) or eltrombopag (Promacta).

In the truly refractory patient, other options include calcineurin inhibitors, cytotoxic agents, danazol, tumor necrosis factor-alpha inhibitors, and staphylococcal A column immunoadsorption.

— David M.J. Hoffman, MD, FACP

Rescue medication, consisting of IVIG 1 g/kg for two doses, was given to one romiplostim-treated patient (8%) due to head trauma and loss of consciousness and two placebo-treated patients (33%) during the 12-week study period, with no interruption of study drug being required. The number of romiplostim-treated patients with grade 3 bleeding decreased from four (33%) prior to the study to none during the study, and the number with grade 2 bleeding decreased from six (50%) to two (17%; P = 0.002).

These findings are similar to results observed in another small study of romiplostim in children with ITP.3,4 An open-label phase III study designed to study the long-term safety of romiplostim and the durability of platelet responses to the drug in pediatric patients currently is in the recruitment stage.5 In this extension trial, approximately 20 patients aged 1 to 18 years with ITP are to receive weekly SC injections of romiplostim, starting at 1 μg/kg (or prior dose) and escalated to 10 μg/kg (based on the platelet count) over a period of 3 years. The primary outcome measure is the incidence of adverse events, including significant changes in laboratory values and the incidence of antibody formation. Secondary outcome measures include the platelet response (> 50 × 109/L) in the absence of rescue medication and the need for concurrent ITP medication (corticosteroids, danazol, or azathioprine) over the duration of the study.

References

1. Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard of care in patients with immune thrombocytopenia. N Engl J Med 2010;363:1889–1899.
2. Elalfy MS, Abdelmaksoud AA, Eltonbary KY. Romiplostim in children with chronic refractory ITP: randomized placebo controlled study. Ann Hematol 2011;doi10.1007/s00277- 011-1172-9.
3. Buchanan GR, Bomgaars L, Bussel JB, et al. A randomized, double-blind, placebo-controlled phase 1/2 study to determine the safety and efficacy of romiplostim in children with chronic immune (idiopathic) thrombocytopenic purpura (ITP). Blood 2009;114(22):680.
4. Safety and efficacy study of AMG 531 to treat ITP in pediatric subjects. ClinicalTrials. gov Web site. http://clinicaltrials.gov/ct2/ show/NCT00515203. Accessed May 12, 2011.
5. A study evaluating the safety and efficacy of long-term dosing of romiplostim in thrombocytopenic pediatric subjects with immune (idiopathic) thrombocytopenia purpura. ClinicalTrials. gov Web site. http://clinicaltrials. gov/ct2/show/study/NCT01071954. Accessed May 12, 2011.

Romiplostim (Nplate) is a thrombopoietin receptor agonist that is currently indicated for the treatment of thrombocytopenia in patients with chronic immune (idiopathic) thrombocytopenic purpura (ITP) who have had an insufficient response to corticosteroids, immunoglobulins, or splenectomy. In a recently reported phase III trial,1 romiplostim treatment was associated with a number of benefits compared with the standard of care in nonsplenectomized ITP patients: higher platelet response rate, lower rates of treatment failure and splenectomy, fewer bleeding events, and fewer blood transfusions.

The safety and efficacy of romiplostim in children with ITP have not yet been established. A recently reported placebo-controlled trial indicates good response and good tolerability of romiplostim in children with chronic refractory ITP.2

Romiplostim versus standard of care in adult nonsplenectomized patients

In a 52-week, multicenter, openlabel trial,1 patients with ITP who had not undergone splenectomy, had received at least one prior treatment for ITP, and had a platelet count of less than 50 × 109/L were randomized to receive weekly SC injections of romiplostim (n = 157) or standard-ofcare treatment (n = 77). Romiplostim was started at a dose of 3 μg/kg, which could be increased up to a maximum of 10 μg/kg to achieve a target platelet count of 50–200 × 109/L. Standardof- care treatment was selected by the treating physician based on standard institutional practices or therapeutic guidelines. Throughout the study, patients in either treatment group could receive additional therapies for ITP, including short-term rescue therapy, such as IV immune globulin (IVIG), but excluding other thrombopoietin mimetics, as deemed medically necessary by investigators.

Study population, endpoints, other treatments

The median ages of patients in the romiplostim and standard-of-care groups were 58 and 57 years, respectively; 54% and 60% were women; the median duration since ITP diagnosis was 2.1 and 2.3 years; baseline median platelet counts were 33 × 109/L and 27 × 109/L; and 13% and 6% were receiving medications, primarily glucocorticoids (11% and 3%), for ITP at baseline. Splenectomy could be performed if study therapy was considered to be ineffective or was associated with severe side effects.

The primary endpoints of the trial were the incidence of splenectomy and the incidence of treatment failure, with treatment failure defined as a platelet count of 20 × 109/L or lower for 4 consecutive weeks, a major bleeding event, or a requirement for a change in therapy (including splenectomy) due to an adverse event or bleeding. In primary endpoint analyses, patients who received any study treatment and then discontinued study participation were counted as having both treatment failure and splenectomy.

During the study, ITP treatments other than romiplostim were received by 44% of patients in the romiplostim group and 79% of patients in the standard- of-care group, including glucocorticoids in 37% versus 63%, IVIG in 7% versus 33%, rituximab (Rituxan) in 1% versus 20%, azathioprine in 1% versus 9%, danazol in 2% versus 7%, other medications in 6% versus 19%, and platelet transfusions in 6% versus 16%.

Efficacy

The incidence of treatment failure was 11% in the romiplostim group versus 30% in the standard-of-care group (P < 0.001). Time to treatment failure was significantly prolonged in the romiplostim group (Figure 1a). The incidence of splenectomy was 9% in the romiplostim group versus 36% in the standard-of-care group (P < 0.001), and time to splenectomy was significantly prolonged in the include bleeding, thrombosis, hematologic cancer or myelodysplastic syndromes, and increased bone marrow reticulin.

After adjustment for duration of study-drug exposure, the romiplostim group had significantly fewer incidences of overall bleeding (P = 0.001) and grade 3 or higher bleeding (P = 0.02); no significant difference between the two treatment groups was observed for less severe bleeding. Overall, 260 bleeding events occurred in 80 patients (52%) in the romiplostim group, for a bleeding rate of 3.56 events/100 patient-weeks; 153 events occurred in 40 patients (53%) in the standard-of-care group, for a rate of 5.02 events/100 patient-weeks. Eight grade 3 or higher bleeding events occurred in five romiplostim-treated patients (3%), a rate of 0.11 events/100 patient weeks; 10 events occurred in five patients (7%) in the standard-ofcare group, a rate of 0.33 events/100 patient-weeks.

A total of 41 blood transfusions were given to 12 patients (8%) in the romiplostim group, whereas a total of 76 transfusions were given to 13 patients (17%) in the standard-of-care group. There was no significant difference between the two groups with regard to the occurrence of thrombotic events. A total of 11 thrombotic events occurred in six patients (4%) in the romiplostim group, yielding a rate of 0.15 events/100 patientweeks, whereas two events occurred in two patients (3%) in the standardof- care group, yielding a rate of 0.07 events/100 patient-weeks.

Two cases of hematologic cancer were observed, consisting of lymphoma and myelodysplastic syndrome in one patient each in the standard-ofcare group. Bone marrow reticulin was found in one romiplostim-treated patient during 6 months posttreatment follow up, with the level being within the normal range (grade 2).

Thrombopoietin mimetics challenge the conventional wisdom about controlling ITP

Chronic immune thrombocytopenic purpura (ITP) affects 60,000 adults in the United States and is associated with the risk of life-threatening hemorrhage. For decades, the standard treatments included glucocorticoids and splenectomy. The advent and subsequent regulatory approval of the thrombopoietin mimetics romiplostim (Nplate) and eltrombopag (Promacta) have not only added to the therapeutic arsenal for this orphan disease but also successfully challenged the conventional wisdom that the key to controlling ITP is to reduce platelet destruction. Instead, these agents work by increasing intramedullary platelet production and are thus able to outpace the rate of peripheral destruction.

As reported by Kuter and colleagues last November in The New England Journal of Medicine, romiplostim is safe, well tolerated, and highly effective in the adult population with chronic ITP, including individuals with an intact spleen. Among the remaining questions regarding its use are: Where in the sequence of treatments for ITP do thrombopoietin mimetics belong? What is the long-term safety of these agents, given the changes to the bone marrow microenvironment that they induce? And, what will the cost impact be of a treatment that is designed for disease maintenance as opposed to providing a cure.

Thrombopoietin mimetics have not been widely tested for the treatment of pediatric ITP. Two small trials have shown that in children with disease that has been refractory to all standard approaches (with the exception of splenectomy), romiplostim is well tolerated, effective, and results in a meaningful reduction of clinically significant bleeding episodes. These data are preliminary but nevertheless provocative, and they offer the potential for new hope to children with chronic refractory ITP. Additional experience will be required to prove long-term safety and tolerability in the pediatric population before these new agents will be adopted into standard practice.

— David M.J. Hoffman, MD, FACP
Tower Hematology Oncology Medical Group
Beverly Hills, CA

Three deaths occurred during the study treatment period, including one death due to pneumonia in one romiplostim-treated patient and two deaths due to hepatic failure and cardiorespiratory arrest, respectively, in two patients in the standard-of-care group. Three additional deaths due to metastatic lung cancer, left ventricular failure, and hepatic neoplasm occurred in the standard-of-care group during the 6-month posttreatment follow up. None of the deaths was attributed to study treatment. No neutralizing antibodies to romiplostim or thrombopoietin were detected.

Quality of life

Quality of life was assessed by the ITP Patient Assessment Questionnaire, consisting of 44 ITP-specific items on each of 10 scales ranging from 0 to 100 points each (with higher scores indicating better quality of life). Scores on two scales (Women’s Reproductive Health and Work Quality of Life) could not be assessed due to inadequacies of the statistical model used in the analysis. Of the eight scales assessed, clinically significant increases of 8–15 points were observed for both treatment groups on all but the Fatigue scale. The romiplostim treatment group showed statistically greater improvements on the Symptoms (P = 0.01), Bother (P = 0.008), Activity (P = 0.02), Psychological (P = 0.049), Fear (P < 0.001), Social Quality of Life (P = 0.002), and Overall Quality of Life (P = 0.02) scales compared with the standard-of-care group, although the between-group differences of 2–8 points on these scales are of uncertain clinical significance.

Romiplostim in children with chronic refractory ITP

Few data exist on the effects of romiplostim therapy in pediatric ITP patients. In a recent single-blind, placebo-controlled trial,2 18 patients aged 2.5 to 16 years with chronic refractory ITP (baseline platelet count < 20 × 109/L) who had not undergone splenectomy were randomized to receive weekly SC injections of romiplostim (n = 12) or placebo (n = 6) for 12 weeks. Romiplostim was started at a dose of 1 μg/kg, escalated to 5 μg/kg at 5 weeks, and tapered afterward.

All patients had either no response or failed to maintain a response to at least two prior treatment modalities for ITP. All had received prior steroid treatment; 44% had received corticosteroids, IVIG, and anti-D immunogloblin in combination or sequentially; and 22% had received cytotoxic or immune-modulating agents. All such treatments were stopped 2 weeks prior to the study.

For patients in the romiplostim versus placebo group, mean age was 9.5 versus 7.0 years, 10 (83%) versus 3 (50%) were male, and median disease duration was 2.3 versus 3.0 years. Median baseline platelet counts were 10.5 × 109/L in both groups.

The median platelet count in the romiplostim group was significantly higher (P = 0.039) than that in the placebo group within 1 week of the first dose of romiplostim (1 μg/kg) and remained so at 3 weeks after the end of treatment (15 weeks total; median platelet count of 47.5 × 109/L vs 19.0 × 109/L; P = 0.001). Changes in platelet count in the romiplostim group were dose-dependent, with a median peak platelet count of 73.5 × 109/L reached after 5 weeks, when the dose was 5 μg/kg. Eleven of the 12 patients in the romiplostim group (93%) reached the target range of greater than 50 × 109/L by the fifth week. At week 12 (end of treatment), 10 romiplostim patients (83%) versus no placebo patients were at target levels. Six romiplostim-treated patients (50%) maintained target platelet count levels at 3 weeks after treatment.

Adverse events occurred in 50% of patients in each group. The most frequent adverse events were headache, epistaxis, cough, and vomiting, which occurred in one patient each in the romiplostim group (8%) and placebo group (17%). Two romiplostim-treated patients (17%) developed a skin rash. None of the patients had thrombocytosis or rebound thrombocytopenia, and none of the romiplostim-treated patients developed bone marrow fibrosis by week 18 of follow up.

How I treat ITP Immune thrombocytopenic purpura (ITP) and its associated problems are frequently encountered by the practicing hematologist. When it comes treatment, there are several options:

• If the platelet count is at least 30,000/μL and there is no active bleeding, close observation is the most prudent approach. In anticipation of the possibility of splenectomy at some point in their course, patients are offered immunization against Streptococcus pneumoniae, Hemophilus influenza, and Neisseria meningitides. In addition, I screen for exposure to hepatitis B virus (HBV), because patients may at some point receive treatment with rituximab (Rituxan), which is associated with HBV reactivation.
• For patients with a platelet count below 30,000/μL, evidence of impending bleeding (such as wet purpura), or frank bleeding, treatment is indicated. Typically, this involves the use of glucocorticoids, often in combination with IV gamma globulin (IVIG) or, occasionally, anti-D immunoglobulin.
• For a platelet count below 10,000/μL, I give methylprednisolone, 1,000 mg/d IV, for 5 consecutive days with IVIG, 1 g/kg daily, for 2 consecutive days. I then shift the patient to oral prednisone, 1 mg/kg daily, for 1 week, followed by tapering the dose by 10 mg/wk, as allowed by the platelet count, which should remain at ≥ 30,000/μL. Failure of this approach, or the inability to reduce the prednisone dose to 10 mg/d or less, would raise the options of either giving rituximab, 375 mg/m2 per week, for 4 weeks or performing a therapeutic, laparoscopic-assisted splenectomy, the single intervention still the most likely to provide meaningful, long-lasting benefit.
• In the setting of asplenia (including the absence of accessory splenic tissue), I favor a trial of a thrombopoietin mimetic, either romiplostim (Nplate) or eltrombopag (Promacta).

In the truly refractory patient, other options include calcineurin inhibitors, cytotoxic agents, danazol, tumor necrosis factor-alpha inhibitors, and staphylococcal A column immunoadsorption.

— David M.J. Hoffman, MD, FACP

Rescue medication, consisting of IVIG 1 g/kg for two doses, was given to one romiplostim-treated patient (8%) due to head trauma and loss of consciousness and two placebo-treated patients (33%) during the 12-week study period, with no interruption of study drug being required. The number of romiplostim-treated patients with grade 3 bleeding decreased from four (33%) prior to the study to none during the study, and the number with grade 2 bleeding decreased from six (50%) to two (17%; P = 0.002).

These findings are similar to results observed in another small study of romiplostim in children with ITP.3,4 An open-label phase III study designed to study the long-term safety of romiplostim and the durability of platelet responses to the drug in pediatric patients currently is in the recruitment stage.5 In this extension trial, approximately 20 patients aged 1 to 18 years with ITP are to receive weekly SC injections of romiplostim, starting at 1 μg/kg (or prior dose) and escalated to 10 μg/kg (based on the platelet count) over a period of 3 years. The primary outcome measure is the incidence of adverse events, including significant changes in laboratory values and the incidence of antibody formation. Secondary outcome measures include the platelet response (> 50 × 109/L) in the absence of rescue medication and the need for concurrent ITP medication (corticosteroids, danazol, or azathioprine) over the duration of the study.

References

1. Kuter DJ, Rummel M, Boccia R, et al. Romiplostim or standard of care in patients with immune thrombocytopenia. N Engl J Med 2010;363:1889–1899.
2. Elalfy MS, Abdelmaksoud AA, Eltonbary KY. Romiplostim in children with chronic refractory ITP: randomized placebo controlled study. Ann Hematol 2011;doi10.1007/s00277- 011-1172-9.
3. Buchanan GR, Bomgaars L, Bussel JB, et al. A randomized, double-blind, placebo-controlled phase 1/2 study to determine the safety and efficacy of romiplostim in children with chronic immune (idiopathic) thrombocytopenic purpura (ITP). Blood 2009;114(22):680.
4. Safety and efficacy study of AMG 531 to treat ITP in pediatric subjects. ClinicalTrials. gov Web site. http://clinicaltrials.gov/ct2/ show/NCT00515203. Accessed May 12, 2011.
5. A study evaluating the safety and efficacy of long-term dosing of romiplostim in thrombocytopenic pediatric subjects with immune (idiopathic) thrombocytopenia purpura. ClinicalTrials. gov Web site. http://clinicaltrials. gov/ct2/show/study/NCT01071954. Accessed May 12, 2011.