Introduction: Immune thrombocytopenia (ITP) is a rare autoimmune disorder characterized by low platelet counts, leading to a wide spectrum of clinical manifestations including bleeding. Current 1st-line (1L) therapies include corticosteroids (CS) and/or intravenous immunoglobulin (IVIG). ASH guidelines recommend limiting high dose CS use to <6 weeks. However, information regarding the timing of 2nd-line (2L) therapies is less prescriptive. This retrospective study sought to evaluate the frequency and timing of treatment transitions beyond initial therapies among adults with ITP in the real-world setting.

Methods: This noninterventional retrospective cohort study used records from the large US-based Optum electronic health record (EHR) database that were dated January 1, 2016, through December 31, 2024. Adults (≥18 years) with ≥2 distinct diagnoses for ITP (ICD-10-CM D69.3) were included; those with secondary ITP were excluded. Patients were required to have been clinically active in the 12 months prior to the initial diagnosis and through the time of initiation of their 1L and/or 2L treatment (index date). Data on patient demographics, clinical characteristics, treatment patterns including discontinuations, and clinical outcomes were collected. Patients received oral CS, IVIG, anti-D, or rituximab as 1L and TPO-RA, rituximab, fostamatinib, splenectomy, or immunosuppressants as 2L treatments.

Results: Among 3287 eligible patients diagnosed with ITP who initiated a 1L treatment, median (IQR) age at diagnosis was 64.6 (49.8-75.4) years, 59.4% were female, and median (IQR) time from ITP diagnosis to index date was 0.1 (0-6.9) months. In the 1L treatment cohort, 3106 patients (94.5%) used oral CS, including 82.2% as monotherapy and 11.7% in combination with IVIG. The median (95% CI) duration on 1L treatment was 2.0 (1.8-2.1) months overall, and 2.1 (1.9, 2.3) months for oral CS monotherapy.

Of the 3287 patients, 1302 (39.6%) did not initiate subsequent treatment after 1L discontinuation and 748 (22.7%) did not initiate guideline recommended 2L treatment but used oral CS after 1L discontinuation. 745 patients (22.7%) remained on 1L treatment (mainly CS) without transitioning to 2L therapy, though the follow-up times were short (median follow-up: 5.1 months for those on oral CS and 2.7 months for those on non-oral CS). 492 patients (15.0%) initiated a guideline-recommended 2L treatment (median follow-up: 26.6 months), of whom 200 (40.7%) received TPO-RA–based regimens, 198 (40.2%) received rituximab-based regimens, and 24 (4.9%) received combination rituximab + TPO-RA.

A total of 1108 1L initiators (33.7%) had bleeding-related events (BREs) during 1L follow-up period and 145 (4.4%) patients had severe BREs. Patients with ≥1 BRE had a mean (SD) of 2.6 (3.4) events; those with ≥1 severe BRE had 1.3 (0.9) events. During 1L follow-up, 775 patients (23.6%) used rescue therapy (mean [SD] number of events, 3.8 [6.0]), most commonly injectable corticosteroids (89.2%) and platelet transfusions (18.2%).

Among patients with available baseline platelet measures within the month prior to 1L initiation, the median (IQR) average platelet count was 71.6×109/L (28.9-125.5×109/L), indicating substantial variability in disease severity. The median (IQR) change from baseline during the 1L follow-up was 22.0×109/L (−4.2 to 65.5×109/L), reflecting heterogeneous treatment response.

Conclusions: This contemporary study from US EHRs confirmed that oral CS were the main 1L therapy for ITP and revealed that a substantial proportion of patients either discontinued or recycled oral CS without initiating guideline-recommended 2L treatment. During 1L follow-up, approximately one-third experienced ≥1 BRE and nearly one-quarter used ≥1 rescue therapy. Substantial variations in baseline and follow-up platelet counts were observed. Additional analyses stratified by baseline disease severity are warranted. Overall, this retrospective study identifies critical gaps in transitioning patients with ITP from initial (CS) to later lines of therapy. Additional data are needed regarding causes of suboptimal treatment transitions in this population.

This content is only available as a PDF.
2025
Sign in via your Institution