Long-term outcomes of eltrombopag plus cyclosporine A in eltrombopag-exposed patients with glucocorticoid-resistant or -dependent primary immune thrombocytopenia: a prospective, multicenter, single-arm, phase II study with extended follow-up.

Background: Primary immune thrombocytopenia (ITP) is an autoimmune disorder characterized by thrombocytopenia, and approximately 20-30% of patients are refractory to or dependent on first-line glucocorticoid therapy. Our previous prospective phase II study demonstrated promising short-term efficacy of eltrombopag (ELT) combined with cyclosporine A (CsA); however, the durability of response, long-term safety, and patterns of treatment failure remain incompletely characterized. Methods: We conducted a long-term follow-up analysis of patients enrolled in a multicenter, single-arm, prospective phase II trial (ChiCTR2000040991). In the original study, adult patients with primary ITP who were resistant to or dependent on glucocorticoids were enrolled between December 2020 and December 2023. All patients had previously received ELT monotherapy without an adequate or durable response; therefore, no ELT-naïve patients were included in this cohort. Patients received ELT (initial dose: 50 mg/day) combined with CsA (3 mg/kg/day, target trough concentration: 150-250 μg/L) for 8 weeks, and responders continued maintenance therapy. The follow-up period for the present analysis was extended until January 1, 2026. Safety and long-term Kaplan-Meier analyses included all 28 patients according to the intention-to-treat (ITT) principle, whereas the 8-week efficacy analysis included the 25 patients who completed 8 weeks of treatment. The three patients who discontinued within the first 8 weeks were included in the Kaplan-Meier analysis and counted as treatment-failure events at the time of discontinuation. The primary endpoints were the overall response rate (ORR) and complete response (CR) rate at 8 weeks, and the long-term sustained remission rate. Secondary endpoints included duration of response and the incidence of adverse events. Univariate logistic regression was performed as an exploratory analysis of associations between baseline variables and treatment failure. Results: A total of 28 patients were included, with a median follow-up of 32.46 months. Of these, 17 (60.7%) had shown previous ELT non-response and 11 (39.3%) had achieved a previous response followed by relapse. Previous ELT response status was not significantly associated with treatment failure (OR, 3.810; 95% CI, 0.738-19.663; Conclusions: ELT plus CsA demonstrated high early response rates and durable disease control in a subset of ELT-exposed patients with glucocorticoid-resistant or -dependent ITP. AE-related treatment discontinuation was an important contributor to treatment failure, and the occurrence of two fatal infections highlights the need for careful patient selection and close infection monitoring. The re-challenge and baseline bleeding-score findings were exploratory and require validation in larger, controlled studies.