PF710 LONG‐TERM OVERUSE OF CORTICOSTEROIDS IN THE REAL‐WORLD TREATMENT OF PATIENTS WITH IMMUNE THROMBOCYTOPENIA IN THE US EXPLORYS AND MARKETSCAN DATABASES

Background: Corticosteroids (CSs) are standard first‐line (1L) therapy for immune thrombocytopenia (ITP), but prolonged exposure leads to detrimental effects. Multiple 2L treatment options exist; however, there is little consensus on the preferred sequence. Aims: To assess real‐world ITP treatment patterns using the US IBM ® Explorys and MarketScan databases. Methods: This retrospective, observational cohort analysis used claims in the US IBM ® Explorys and MarketScan databases (2011–2017) of adult patients newly diagnosed with ITP (pts). Included patients were limited to those with commercial or private health coverage and proxies of eligibility within the Electronic Health Record (EHR) system in MarketScan and Explorys, respectively. Selection criteria: Adults with ≥1 primary ITP diagnosis (ICD‐9‐CM 287.31 or ICD‐10‐CM D69.3); ≥12 months’ data available prior to first ITP diagnosis; ≥1 ITP treatment; and ≥1 month of data available post‐first ITP treatment. Patients with hepatitis and HIV were excluded. Line of treatment was considered continuous until one of the following: patient discontinued all treatments for ≥60 days, switched or augmented treatment (addition of a new ITP agent ≥28 days after the treatment line start date), dis‐enrolled, died, or was censored due to study end. Data were collected and analyzed separately for each database. Descriptive statistics are presented by ITP therapy up to 7L. Results: 4066 of 22900 (18%) and 7837 of 50428 (16%) ITP patients were eligible (majority of patients did not meet criteria for newly diagnosed ITP and ≥1 month of data post‐first ITP treatment) in Explorys and MarketScan, respectively. Mean ± SD age (years) at start of treatment was 61 ± 18 and 56 ± 18, respectively. Most patients received ≥2L in both Explorys (54%) and MarketScan (63%). Surprisingly, >90% of patients in both databases received CSs, not only as part of 1L therapy, but also as part of all subsequent lines of treatment (Fig). Standard 2L treatments were used less frequently in Explorys vs MarketScan: rituximab (12% vs 25%), eltrombopag (5% vs 6%), romiplostim (7% vs 12%), and splenectomy (3% vs 8%). Repeated courses of CSs were common. The prednisone–prednisone–prednisone sequence was the most common treatment sequence across 1–3L (Explorys vs MarketScan: 33% vs 18%). Eltrombopag use increased from 1L onwards (Fig). From 2015 until 2017, the proportion of patients treated with eltrombopag and rituximab as 2L and 3L in the Explorys sample increased (eltrombopag: 5–14%; rituximab: 8–29%). Prednisone use as 2L and 3L therapy decreased from 62% to 43% but remained the most widely used treatment over the period covered in the data analysis. Summary/Conclusion: Data from Explorys (US EMR database) and MarketScan (US claims database) showed that long‐term CS use is highly prevalent in ITP patients, not only at 1L, but also at all subsequent lines. Thrombopoietin receptor agonists, rituximab, and splenectomy were used much less frequently than expected. These data suggest the need for improved interventions to reduce long‐term CS use and increase the use of other effective therapies currently available. Further validation in clinical practice may be required considering these analyses were limited to <20% of ITP patients in each database.