Enhanced Patient Satisfaction and Stability of Platelet Counts Following a Switch from Eltrombopag (ELT) or Romiplostim (ROMI) to Avatrombopag (AVA) in Adult Idiopathic Thrombocytopenic Purpura (ITP): Post-Hoc Analyses from a Prospective Phase 4 Study

Background: Thrombopoietin receptor agonists (TPO-RAs) are indicated for treating patients with ITP. AVA, widely approved, is orally administered without food-type restrictions which differentiates it from other TPO-RAs.

Patient satisfaction to AVA following a switch from another TPO-RA has not been previously reported in a prospective manner.

A recent prospective, phase 4 study [top-line results reported separately at this congress] in adult ITP patients after switching to AVA from ELT or ROMI evaluates the safety, efficacy, and patient-reported outcomes of treatment satisfaction and convenience.

Expanding on the topline results, these analyses aim to present patient satisfaction and platelet counts (PC) to AVA treatment based on the specific prior TPO-RA, including exploring the influence of prior TPO-RA dose on endpoints.

Methods: A prospective, open-label Phase 4 study (NCT04638829) enrolled patients from 2021-2023 who had been receiving ELT or ROMI for ≥90 days prior to study entry with a prior response (PC ≥50x10⁹/L) to the individual TPO-RA. Patients were switched to AVA at a starting dose of 20 mg daily and followed for 90 days.

The primary endpoint was adverse events and serious adverse events.

Secondary endpoints included measures of PC along with treatment satisfaction and convenience via the Treatment Satisfaction Questionnaire for Medication (TSQM v1.4).

We report post-hoc analyses of TSQM domain score (convenience, effectiveness, global satisfaction, side effects) changes from baseline (BL) to Day 90 (D90) and PC changes from BL to D90, stratified by BL dose of the prior TPO-RA [≤25mg (Low Dose, LD), 26-50mg (Mid Dose, MD), and ≥51mg (High Dose, HD) for ELT-switchers and ≤1.5mg/kg (LD), 1.6-3.0mg/kg (MD), and ≥3.1mg/kg (HD) for ROMI-switchers].

Results: For TSQM analyses at D90, stratification by BL dose of prior therapy yielded 10 LD,19 MD and 7 HD evaluable patients for ELT-switchers and 5 LD,7 MD and 7 HD evaluable patients for ROMI-switchers. For PC analyses at D90, 10 LD, 20 MD and 5 HD ELT-switchers and 5 LD, 7 MD, 7 HD ROMI-switchers were evaluable.

For ELT-switchers, the BL median scores for both convenience and global satisfaction were 66.7 and 64.3, 69.4 and 67.9, and 77.8 and 78.6, respectively, in LD, MD, and HD. Scores increased at D90 regardless of BL dose and were increased by 16.6 and 14.3 points, 8.4 and 17.8 points, and 22.2 and 3.4 points, respectively, in LD, MD, and HD.

The median domain BL scores for effectiveness were 77.8, 66.7, and 77.8 for LD, MD, HD and rose at D90 by 4.5 points, 16.6 points, and 5.5 points, respectively. For side effects, the BL score was 100 (highest possible score) for all dose groups and remained at 100 on D90.

For PC on AVA, BL median PCs were ≥50x10⁹/L for each ELT dose group and increased to ≥100x10⁹/L in all groups at D90. The median PCs at D90 were 182.5x10⁹/L, 155.5x10⁹/L, and 138.0x10⁹/L for the LD, MD, and HD groups respectively.

For ROMI-switchers, the BL median score for both convenience and effectiveness were 77.8 for both, 52.8 and 66.7, and 72.2 and 77.8, respectively, for LD, MD, and HD. Scores increased at D90 by 5.5 points for convenience and were unchanged for effectiveness in LD; increased by 36.1 points and 16.6 points, respectively, in MD; and increased by 27.8 points and 5.5 points, respectively in HD.

The median domain BL scores for global satisfaction were 78.6, 78.6, and 71.4 for LD, MD, and HD, respectively, and rose at D90 by 14.3 points, 21.4 points, and 21.5 points, respectively. For side effects, the median BL score was 100 in the LD and HD groups and 90.6 in the MD group and increased to, or remained at, 100 on D90 for all BL doses.

For PC on AVA, BL median PCs on ROMI were ≥100x10⁹/L in all groups, fluctuated at different timepoints, but maintained ≥75x10⁹/L in all groups at D90. The median PCs were 184.0x10⁹/L, 153.0x10⁹/L, and 75.0x10⁹/L for the LD, MD, and HD groups respectively at D90.

Summary/Conclusion: Patients generally responding to their prior TPO-RA who switched to AVA reported higher scores from BL across multiple domains of the TSQM with PCs either improved or maintained in a clinically acceptable range, irrespective of the prior TPO-RA experience. When stratified by BL dose, no observable pattern of median change from BL was observed across the TSQM domains for either ELT or ROMI. These results further reinforce that patients may experience sustained effectiveness paired with enhanced treatment satisfaction when switching from ELT or ROMI to AVA.