Safety and Efficacy of Eltrombopag at Escalated Doses

Overview

Study rationale is based on the data that in previous clinical studies of eltrombopag in ITP there are some patients who have been reported as non responders at the maximal approved dose of 75 mg daily. The trend in both normal volunteers and in patients with ITP suggest and increasing response rate with increased doses of eltrombopag up to a dose of 75mg. Previously published data has shown no overt increase in toxicity in normal volunteers, oncology patients and aplastic anemia patients treated with escalated doses as high or higher than those proposed in this study. It therefore seems possible that in ITP patients who did not respond to a dose of 75mg daily, eltrombopag could be more effective at a higher dose. We propose a double blind randomized controlled trial in ITP patients who have been defined as non-responders at the maximum dose (75mg) of eltrombopag, assessing efficacy and toxicity at higher daily doses (100mg, 125 mg, 150 mg)

Preliminary data in pediatric patients (who seem to metabolize eltrombopag differently especially at younger ages) find that in the middle and lowest age cohort, even doses of 75 mg were often insufficient to obtain a response. Altogether, this data suggest both that a higher dose in ITP may be more efficacious in increasing the platelet count and that there is no overt increase in toxicity in normal volunteers, oncology patients and aplastic anemia patients treated with doses as high or higher than those proposed in this study. Double-blind, randomized, placebo-controlled trial in subjects with ITP who are 'non-responders' at the maximum package insert dose (75mg) of eltrombopag; non-responders are those with platelet counts \< 50,000 despite taking 75 mg/day of eltrombopag for at least 3 weeks. Study Design (First part): 8 Weeks Subjects will be randomly allocated in a two to one ratio to receive treatment or placebo. All subjects in the study will receive 75 mg eltrombopag and then be randomized to receive either an additional 25 mg of eltrombopag or matching placebo tablet dispensed by the research pharmacy. Subjects and investigators will be blinded to randomization. Randomization will be stratified according to splenectomy status. Randomization will be performed at the time of informed consent with a computer generated randomization table. Subjects and investigators will be blinded to assignment and treatment in this phase. Eltrombopag will be administered for 8 weeks or until the platelet count exceeds 100,000; at this point dosing will stop, subject will be considered a responder and the subject will eligible for entering Part 2 (the long term treatment part of the study) if desired and agreed to be appropriate by the subject giving consent, and the treating physician. The dose at which the subject achieved the primary endpoint (\> 50,000 and increase by \> 20,000) will be considered the dose of response. Even if a patient achieves a response, if the platelet counts remains \< 100,000, then the dose escalation schema in the figure above will continue. Dose escalation will continue, despite satisfaction of the primary endpoint of study (\> 50,000 and \> 20,000 above baseline), unless the platelet count reaches 100,000. The rationale for this approach is to maximize clinically relevant responses for the subjects and standardize data for analysis. Subjects will stop study medication if the platelet count is within the normal range, thereby minimizing any safety risk associated with elevated platelet count. By continuing to dose escalate subjects until the platelet count is \> 100,000, the maximum response and correlation of dose will be tracked allowing better understanding of sustained response and dose-dependent response, which can be applied to future consideration dosage for periods of short-term hemostasis ( i.e surgical procedures). The 8 weeks of data collected in enrolled subjects on the standardized protocol will have the balance of consistency of dosage and unequivocality of response while maintaining subject safety. Enrollment will be stratified to address concerns regarding the inclusion of pediatric subjects. Enrollment will begin with adult subjects ≥ 18 years of age, until there is safety information available for review from at least three adult patients who received active therapy for 8 weeks with exposure to the maximal proposed dose of 150 mg. After review of this unblinded safety information by independent Data Safety Monitoring Officer (DMSO) we will allow pediatric patients 12-18 years of age to enroll on the protocol. Long Term Treatment Extension (Part 2): (18 months) After 8 weeks or once dosing is stopped because the platelet count is \> 100,000, the subject will be unblinded, once the data forms are complete with the assistance of the research pharmacy. Subjects who received and responded to eltrombopag higher dose treatment will have the option to continue therapy with periodic monitoring and ongoing dose adjustment. Subject who had a platelet count \> 100,000 can enter Part 2 at the dose at which the subject met primary response criteria after the platelet count decreases to \< 100,000/uL (microliters) on serial monitoring in the immediate post-study monitoring period. Subjects randomized to the placebo group who did not respond will have the opportunity to receive open label escalated dose eltrombopag over 8 weeks following the study protocol as if they had been randomized to active drug in the study. This group will provide the "confirmatory group" to see if the rate of response in the randomized group can be confirmed in this group. Consent will be obtained for additional monitoring with bone marrow aspirate and biopsy at 1 year and additional ophthalmologic examinations at 6 months and 18 months. The trial will have an 80% statistical power at the 5% level of significance (two sided) to detect a difference in the proportion of subjects receiving increased dose of eltrombopag with a primary response (2 consecutive platelet counts of \> 50,000 and an increase of \> 20,000 from the study baseline with in the 8 week increased dose window not as a result of rescue treatment ) to subjects without a primary response in the placebo group assuming 50% of subjects randomized to active drug will have a response and 5% of subjects randomized to placebo would have a response. However an interim analysis would be performed after 18 subjects to test for futility and efficacy and also to re-estimate sample size. If neither futility or efficacy are met the trial would continue to enroll. Based on the observed treatment effect at the interim analysis the sample size would be allowed to expand to up to 60 subjects. Dose modification will be made on the basis of individual platelet response as detailed below to a maximal dose of 150mg.