Treatment Patterns and Bleeding Events Among Patients with Immune Thrombocytopenia Based on Duration of Corticosteroid Treatment

Background: Oral corticosteroids (CS) are often the first line of therapy for immune thrombocytopenia (ITP). Recent guidelines recommend against prolonged CS use (>6 weeks) for adults with ITP and suggest switching to second-line treatments for non-responsive or CS-dependent patients. The duration and potential overuse of CS among newly treated with CS adult ITP patients has yet to be described in a real-world setting. This study quantified the real-world duration of CS use among adult ITP patients newly treated with CS and assessed the clinical and economic consequences of prolonged CS use.

Objectives: 1) To describe treatment patterns, bleeding events, and side effects of CS among patients with ITP; 2) To describe healthcare resource utilization (HRU) and costs among patients with ITP treated with CS.

Methods: A retrospective observational cohort study was conducted using claims contained in the IBM® MarketScan® Commercial and Medicare Databases. Adult patients with an ITP diagnosis between January 1, 2013 and June 30, 2018 were selected for the study. Patients were required to be newly treated with CS (prednisone, dexamethasone, methylprednisolone) (first CS claim=index date) with at least 12 months of continuous enrollment prior to and 12 months following index. Patients with a diagnosis for any other thrombocytopenia and evidence of causes of secondary thrombocytopenia were excluded.

Patients were categorized into cohorts based on CS duration during the follow-up period: CS days’ supply ≤6 weeks, >6 weeks to 8 weeks, or >8 weeks. Discontinuation was defined as a gap of >30 days of the index CS following the run-off date of the last observed claim. Bleeding events, incidence of CS side effects (defined as new events occurring during the follow-up period only), HRU and costs (2018 USD), and treatment patterns were described during the follow-up period.

Results: Among 1,720 patients who met study inclusion criteria, 65% had ≤6 weeks of CS use, 6% had >6-8 weeks, and 29% had >8 weeks. Median age among all patients was 48 years and majority were female (61%). About three quarters (71%) of patients started on prednisone as the index CS treatment, 16% on methylprednisolone, and 13% on dexamethasone. Patients with higher CS duration of use (>6-8 and >8 week) had a shorter time to CS treatment start following ITP diagnosis (mean 69 and 92 days, respectively; median 5 days for both) compared to the ≤6 week cohort (mean 306, median 152 days).

All but one patient discontinued their index CS therapy. Patients with >8 weeks of CS had higher CS restart rates following discontinuation (49%) compared to patients with ≤6 weeks (23%). Among patients who switched therapies following CS discontinuation, those in the ≤6 week cohort had higher utilization of second line ITP agents compared to the >8 week cohort, including eltrombopag (16% vs. 11%) and romiplostim (23% vs. 11%). Patients in the >8 week cohort had higher rates of switching to splenectomy (15%) and rituximab (33%) following CS discontinuation compared to the ≤6 cohort (6% and 26%).

About a fifth (19%) of patients experienced a bleeding event during the follow-up period. Bleeding event rates were higher for those with >8 weeks of CS than those with ≤6 weeks (23% vs. 17%). Nearly 40% of all patients experienced a CS side effect and rates were similar among the three cohorts.

Patients with greater use of CS (>8 weeks) had higher rates of inpatient admissions (34%) compared to the ≤6 week (22%) and >6-8 week (32%) cohorts. Mean annual all-cause costs were $24,800, $42,706, and $51,896 for the ≤6 week, >6-8 week, and >8 week cohorts, respectively.

Conclusions: Patients with ≤6 weeks of CS had the shortest switch time to a non-CS ITP treatment, with higher utilization of thrombopoietin receptor agonists, as compared to patients with greater CS use. Patients with a higher duration of CS use had significantly higher inpatient and outpatient HRU and more than double the healthcare costs compared to patients with ≤6 weeks of CS. The rate of bleeding events was lowest among patients with ≤6 weeks of CS use. Approximately 35% of patients in this study were treated with >6 weeks of CS, indicating a need for earlier intervention with more effective ITP therapies that may reduce CS use and risk of bleeding events, and subsequently decrease the economic burden incurred by patients with prolonged CS exposure. Further study is needed to assess the clinical and economic outcomes related to duration of CS use.