Additional efficacy analysis of avatrombopag phase III data for the treatment of adults with immune thrombocytopenia

Abstract Avatrombopag is an oral thrombopoietin receptor agonist (TPO-RA) that was approved in the US in 2019 for treatment of chronic immune thrombocytopenia (ITP). This post hoc analysis of the pivotal phase III study (NCT01438840) of avatrombopag in adult patients with ITP evaluated platelet count response to avatrombopag during the core study in different subgroups, and durability of response data in patients who responded to avatrombopag treatment both during the core phase (total population) and during the core and extension phase (total population and by subgroup). Loss of response (LOR [platelet count <30 × 109/L]) was defined as LOR over two consecutive scheduled visits. The response was generally similar between subgroups though a few differences were observed. The durability of response analysis showed that avatrombopag-treated patients maintained their response for 84.5% of time on treatment during the core phase and 83.3% during the core and extension phase; 55.2% of patients in the core phase and 52.3% in the core and extension phase never experienced LOR. We conclude that the initial response to avatrombopag is both stable and durable. Plain Language Summary What is the context? Avatrombopag is a medicine that helps the body produce platelet cells, which are necessary for blood clotting. Avatrombopag is used to treat people with chronic immune thrombocytopenia (ITP); these patients have low numbers of platelet cells in their blood, so their blood may not clot efficiently, putting them at risk of uncontrolled bleeding. A phase III clinical study in patients with chronic ITP showed that platelet counts increased for most patients who were treated with avatrombopag; patients who had a platelet count ≥50 × 109/L were considered to have a response to avatrombopag treatment. The present study analyzes data from a phase III clinical study to determine whether there are any characteristics that make a patient more or less likely to have a loss of response (LOR) while taking avatrombopag, whether the initial response to avatrombopag is stable, and how long the response lasts. For this analysis, LOR is defined as platelet count <30 × 109/L for either four consecutive weeks or for two consecutive office visits while taking avatrombopag. What is new? In general, patient characteristics did not influence the likelihood of LOR. Patients who responded maintained their response for most of the time they were taking avatrombopag. Most patients did not experience LOR. What is the impact? The initial response to avatrombopag is stable and long-lasting in patients with chronic ITP. These findings indicate that patients with a variety of background characteristics can experience a durable platelet response with avatrombopag treatment.


Plain Language Summary
What is the context?
• Avatrombopag is a medicine that helps the body produce platelet cells, which are necessary for blood clotting.• Avatrombopag is used to treat people with chronic immune thrombocytopenia (ITP); these patients have low numbers of platelet cells in their blood, so their blood may not clot efficiently, putting them at risk of uncontrolled bleeding.
• A phase III clinical study in patients with chronic ITP showed that platelet counts increased for most patients who were treated with avatrombopag; patients who had a platelet count ≥50 × 10 9 /L were considered to have a response to avatrombopag treatment.• The present study analyzes data from a phase III clinical study to determine whether there are any characteristics that make a patient more or less likely to have a loss of response (LOR) while taking avatrombopag, whether the initial response to avatrombopag is stable, and how long the response lasts.
• For this analysis, LOR is defined as platelet count <30 × 10 9 /L for either four consecutive weeks or for two consecutive office visits while taking avatrombopag.
What is new?
• In general, patient characteristics did not influence the likelihood of LOR.
• Patients who responded maintained their response for most of the time they were taking avatrombopag.• Most patients did not experience LOR.
This article has been republished with minor changes.These changes do not impact the academic content of the article.
Correspondence: Shivi Jain, Division of Hematology, Oncology and Cell Therapy, Rush University Medical Center, 1725 West Harrison Street, Suite 809, Chicago, IL 60612, USA.E-mail: Shivi_Jain@rush.eduThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/ 4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent.

Introduction
3][4][5][6] Additionally, avatrombopag does not bind polyvalent cations, meaning that there are no food-type restrictions. 5,7he efficacy and safety of avatrombopag for adults with chronic ITP was previously demonstrated in a phase III study (NCT01438840) (Figure 1). 3 The study found that avatrombopag was superior to placebo in the primary endpoint of median cumulative number of weeks of platelet response (platelet count ≥ 50 × 10 9 /L; 12.4 versus 0 weeks; P < .0001)and the secondary endpoint of platelet response rate at day 8 versus placebo (65.63% versus 0%; P < .0001); the use of concomitant medications was reduced with avatrombopag versus placebo.3 Additional endpoints from that study were evaluated in a previously published post hoc analysis, which reported that more patients in the avatrombopag group than the placebo group achieved a complete response (platelet count ≥ 100 × 10 9 /L) at day 8 of the core study and had a platelet response or complete response at day 28 and month 6.A durable response (platelet count ≥ 50 × 10 9 /L) and clinically relevant response (platelet count ≥ 30 × 10 9 /L for 6 of the final 8 weeks of the core study) were achieved in 34.4% and 64.0% of patients in the avatrombopag group, respectively, and 0% (using either definition) of patients in the placebo group.4 During the extension phase, a respective 96.1% and 60.1% of the durable responders (patients not requiring rescue treatment at any point during the study and those who achieved a response in ≥6 weeks of the final 8 weeks of the core study) in the core phase achieved response and complete response.4 More than half of patients taking chronic corticosteroids at baseline were able to reduce or discontinue them. 4We conducted post hoc efficacy analyses of platelet count response to avatrombopag during the core phase of the phase III study (NCT01438840) in subgroups (by age, sex, number of prior ITP therapies, prior TPO-RA exposure, ITP disease duration, concomitant ITP medication at baseline, and baseline platelet count), and of the durability of response for all patients in the core phase, and for those who responded in the core and/or open-label extension phase (total population and by subgroup) to evaluate these outcomes in avatrombopag-treated patients.

Study design
These post-hoc analyses included patients enrolled in a multicenter, randomized, double-blind, parallel group phase III study evaluating the efficacy and safety of avatrombopag in adult patients with chronic ITP for 6 months (NCT01438840). 8ull details of the phase III study have been previously published. 3The core study design included a 6-week study drug titration period, a 12-week concomitant ITP medication reduction period, and an 8-week maintenance period.Patients who completed the maintenance phase of the core study or discontinued early because of lack of treatment effects were eligible for the extension phase, during which all patients received open-label avatrombopag once daily (20 mg initial dose that could be titrated from 5 to 40 mg) and underwent a 6-week conversion period and a 90-week maintenance period.During the core study, visits were weekly or biweekly depending on the phase of the study, with 21 visits occurring over 26 weeks.In the extension phase, visits were every 3-4 days, weekly, biweekly, or monthly, with 31 planned over 96 weeks.Overall, the core study enrolled 32 patients with ITP in the avatrombopag group and 17 in the placebo group and 39 were enrolled in the optional open-label extension phase (avatrombopag, n = 24; placebo, n = 15).

Analysis of response in patient subgroups
The response in patients who were treated with avatrombopag in the core study was analyzed according to age (<49 years/ ≥49 years), sex (female/male), number of prior ITP therapies (<3/≥3), prior TPO-RA exposure (yes/no), ITP disease duration (<7 years/≥7 years), concomitant ITP medication at baseline (yes/no), and baseline platelet count (<15 × 10 9 /L/ ≥15 × 10 9 /L).Given the small population size, the median value was used for demographic cutoff points.Descriptive evaluations of response included median cumulative number of weeks in which a response (platelet count 50 × 10 9 /L) was achieved and the proportion of patients achieving a response at least once, at least twice, and on day 8 of the core study.

Durability of response
Durability of the initial avatrombopag response (platelet count ≥ 50 × 10 9 /L) was assessed as the mean (standard deviation [SD]) number of days until loss of response (LOR), study completion, or study discontinuation; proportion of patients with no LOR, and mean (SD) proportion of time with response in responding patients (e.g., total time responding).Duration of response was analyzed for all patients achieving a response in the core phase (total population; divided by treatment group) and for all patients achieving a response to avatrombopag in the core and extension phase (total population and according to previously mentioned subgroups).LOR was defined as platelet count <30 × 10 9 /L over two consecutive scheduled visits.A broader definition of LOR for 4 consecutive weeks was also used (data not shown).The date of the first platelet count <30 × 10 9 /L was used as the starting point for the LOR calculation.If patients required corticosteroids or intravenous immunoglobulin as rescue therapy, they were considered non-responders for at least 8 weeks after the date the rescue therapy was administered.If patients received a platelet transfusion, they were considered non-responders for at least 1 week after the date the transfusion was administered.

Patients
The baseline characteristics of the patients included in the phase III study have been previously reported. 3The core-phase patient population for patients treated with avatrombopag according to the subgroups evaluated in this post hoc analysis are presented in Table I.

Avatrombopag efficacy and duration of response during the core phase (total population and according to subgroup
Among patients treated with avatrombopag, older patients (≥49 years) were more likely to achieve a response than younger patients (<49 years) (100% versus 82.4%), but the median cumulative weeks of response was over twice as long in younger patients (7.4 versus 15.3) (Table II).Males had a longer median cumulative duration of response than females (15.9 versus 9.6 weeks).Patients who did not receive concomitant ITP medication at baseline were more likely to achieve a response at least once, at least twice, and at day 8 of treatment and to have a longer median cumulative duration of response versus those who did.Additionally, those with a higher platelet count at baseline (≥15 × 10 9 /L) were more likely than those with a lower platelet count (<15 × 10 9 /L) to achieve a response at least once or at least twice and had a nearly fourfold higher median cumulative number of weeks with a response (19.2 versus 5.3 weeks).Other outcomes between subgroups of avatrombopag-treated patients were generally similar and there were few differences between the subgroups in patients treated with placebo.Efficacy of avatrombopag in ITP 3

Durability of initial platelet count response during the core phase
In the core phase, 29/32 patients in the avatrombopag group were responders, and 1/17 in the placebo group was a responder.Patients in the avatrombopag and placebo groups maintained their response for 84.5% and 32.9% of their remaining time in the study (i.e., percent of time in the study after their initial response), respectively (Table III).The time to first LOR, study discontinuation, or study completion was 119.1 days in the avatrombopag group and 28 days for the patient receiving placebo.In the avatrombopag group, 55.2% of patients never experienced an LOR; the placebo-treated patient experienced an LOR.

Durability of initial platelet count response during the extension phase
Of the 47 patients who received avatrombopag in the core and/or extension phase (core phase only: n = 8; extension phase only: n = 15, core and extension phase: n = 24), 44 (core phase only: n = 7; extension phase only: n = 13, core and extension phase: n = 24) responded and were evaluated in this post hoc analysis of initial platelet count durability.Seven of the patients who responded to avatrombopag during the core phase chose not to continue into the extension phase (reasons: withdrawal of consent/lost to follow-up, n = 4; adverse event, n = 3 [dizziness, acute gastritis, and chronic myelomonocytic leukemia]).Patients maintained their response for 83.3% of their remaining time in the study (Table IV lower (<15 × 10 9 /L) baseline platelet count (96.1% versus 73.0%).Patients with no concomitant ITP medications at baseline or with a higher baseline platelet count were also less likely to experience any LOR and maintained their response for longer versus those with concomitant ITP medications or lower platelet counts at baseline.Prior TPO-RA exposure had little effect on the durability of the response.

Discussion
In the open-label extension phase of this phase III study, patients treated with avatrombopag or placebo during the core phase who received avatrombopag during the extension phase maintained a response for 83.3% of the time on treatment and 55.2% of avatrombopag-treated patients in the core study never experienced LOR in the core phase; 52.3% never experienced LOR in the core and extension phase.The phase III study aimed to reduce both concomitant ITP medications and the dose of avatrombopag.The required starting dose of avatrombopag was 20 mg/day in the extension study, meaning that patients who were taking avatrombopag doses >20 mg/day in the core study were required to reduce their dose at the start of the extension study, and patients may not have responded as quickly in the extension phase given the low initial dose of avatrombopag.These data suggest that the initial response to avatrombopag is both stable and durable, despite the lower initial dosing in the extension phase.Among patients who did experience an LOR, many achieved platelet count responses again without a subsequent LOR.This report of a durable platelet response with avatrombopag treatment in patients with ITP is in line with the findings of previously reported studies of other TPO-RAs.A long-term study of eltrombopag in patients with ITP reported that patients had a platelet response (≥50 × 10 9 /L and 2× baseline) for a median of 44 cumulative weeks, demonstrating a stable and durable response. 9Similar results have also been reported for romiplostim.A long-term (up to 5 years) single-arm study reported that patients with ITP who were treated with romiplostim had a platelet response (platelet count ≥ 50 × 10 9 /L) for a median of 92% of their time on study. 10A post hoc analysis of the study found a negative correlation between baseline ITP duration (≥3 years versus <3 years) and median percentage of time on study with a platelet response. 10In this avatrombopag study, a longer disease duration at baseline (≥7 years) was associated with a greater likelihood of experiencing LOR versus a shorter disease duration (<7 years).The number of prior ITP therapies and whether a patient had prior TPO-RA exposure did not definitively predict platelet count response to avatrombopag, suggesting that it can be effective regardless of the number of ITP therapies previously received or previous TPO-RA exposure.However, patients with ≥ 3 prior ITP therapies were slightly more likely to have LOR versus those with <3 prior ITP therapies.This is generally consistent with findings from the EXTEND study in which a good response to eltrombopag was observed regardless of the number of prior ITP treatments, although the proportion of patients responding was numerically higher among those with fewer previous ITP treatments. 11Additionally, baseline characteristics of female sex, higher age (≥49 years), the presence of concomitant ITP medications at baseline, and a lower baseline platelet count (<15 × 10 9 /L) were associated with reduced durability of response to avatrombopag to some degree.To our knowledge, this is the first report to examine the association between these patient characteristics and duration of response to TPO-RA treatment in patients with ITP.
An important study limitation was the relatively small number of patients.Additionally, given that this was a post hoc analysis, the results should be interpreted with caution.
Overall, these post hoc analyses showed that the initial response to avatrombopag was both stable and durable.While some differences in duration of response were observed between subgroups, findings were generally similar, indicating that patients with a variety of background characteristics could experience a durable platelet response with avatrombopag treatment.
Table I.Baseline characteristics by subgroup of patients treated with avatrombopag in the core phase.Patients treated with avatrombopag in the core study (N = 32) Age Sex No. prior ITP treatments Prior TPO-RA use ITP disease duration Concomitant ITP medication ; ITP, immune thrombocytopenia; M, male; N, number of evaluable patients; n, number of patients in the indicated category; PC, platelet count; TPO-RA, thrombopoietin receptor agonist; y, years.DOI: https://doi.org/10.1080/09537104.2023.2195016

Table II .
Efficacy of avatrombopag during the core phase for the total population[3]and according to subgroups.
The number of prior ITP therapies and ITP disease duration had little effect on duration of response.A greater proportion of patients with <3 than with ≥3 prior ITP therapies never experienced LOR (72.2% versus 38.5%), as did a greater proportion of patients with a shorter ITP disease duration (<7 years versus ≥7 years, 63.6% versus 40.9%).Duration of response was greater for patients with no concomitant ITP medications at baseline versus those with (92.8% versus 72.8%) and for those with a higher (≥15 × 10 9 /L) versus