Measuring success by different rulers: Heterogeneity in definition of outcomes in immune thrombocytopenia trials Free

Background:

Immune thrombocytopenia (ITP) is an acquired autoimmune disorder characterized by low platelet counts and increased bleeding risk. Although various therapies—such as glucocorticoids, thrombopoietin receptor agonists (TPO-RAs), rituximab, among others—have been evaluated in clinical trials, the lack of uniformly employed outcome definitions in clinical trials hampers the interpretation, cross-trial comparison of findings, and applicability of results. Recognizing this gap, consensus definitions were established by the International Working Group (IWG, 2009) and later by the American Society of Hematology (ASH, 2019) in their guidelines. This study aimed to evaluate the extent to which phase 2/3 and phase 3 ITP trials adhered to outcome definitions recommended by the ASH 2019 and IWG 2009 guidelines

Methods:

A systematic search was conducted on ClinicalTrials.gov up to June 30, 2025. Filters included interventional phase 2/3 or 3 trials, adults aged ≥18 years, and conditions restricted to ITP. Trials with overlapping cytopenias or autoimmune conditions were excluded, as well as early phase trials. Only trials with start dates posterior to the publication of the earliest guidelines (IWG 2009) were included in this study. Clinical trials for any type of ITP (newly diagnosed, persistent, and chronic) were included. The primary objective was to assess the proportion of trials that defined their primary outcomes in accordance with the IWG 2009 or ASH 2019 guidelines. Secondary outcomes included disease definition for enrollment, time since diagnosis acceptable for accrual, platelet count thresholds for enrollment and response, and whether a longitudinal assessment of platelet counts was assessed as part of the response criteria or not. A descriptive analysis was conducted using R.

Results:

A total of 79 trials met inclusion criteria. Most trials were phase 3 (92.4%, n=73), randomized (68.4%, n=54), industry-funded (55.7%, n=44), and conducted in China (31.6%, n=25). The most studied condition was chronic ITP (40.5%, n=32). However, inclusion criteria were frequently underdefined: 41.8% (n=33) of trials did not specify time from diagnosis, and 82.3% (n=65) lacked diagnostic platelet count thresholds. Only 22.8% (n=19) referenced national or international diagnostic criteria.

Among the included studies, 120 primary outcomes were identified (median per trial: 1, IQR 1-3, range 1-5); 78.3% (n = 94) were platelet-related, while the remainder focused on safety or exploratory measures. Notably, 79.7% (n=154) of these outcomes did not align with definitions provided by either the IWG or ASH guidelines, while only 20.3% (n=16) conformed to at least one of the recommended definitions. Most guideline-compliant outcomes referenced IWG criteria (19.0%, n=15), and none used ASH 2019 terminology directly, but the definition was consistent with said guidelines in 1 trial. The most frequently matched outcome definition was the IWG's “No Response” (8.8%, n=7). Many non-compliant trials relied on non-standard terminology, including terms such as “Overall response rate”, “Duration Response”, “Effective rate of maintenance treatment”, “Platelet maintenance response time”, “Percentage of patients with stable platelet response”, and “Extent of disease control“.

One-third (n=31) of platelet-based outcomes required achieving a platelet count >50×10⁹/L, while 11.6% (n=11) focused on thresholds <30×10⁹/L, often related to relapse or treatment failure. Twenty-nine outcomes incorporated bleeding events into their composite endpoint. In most trials (63.3%, n=50), determining a patient's outcome classification necessitated at least two separate platelet count assessments. The most common follow-up interval for outcome evaluation was 3 months (15%, n=18), with notably some outcomes defined by follow-up periods as short as 9 days.

Conclusion:

Despite two international consensus documents aiming to standardize ITP outcome reporting, adherence to such recommendations among phase 3 trials remains suboptimal. The limited use of guideline-compliant definitions highlights ongoing variability in how outcomes are defined and reported. This variability can reduce comparability and generalizability, slow evidence synthesis, and hinder patient-centered decision-making. Broader adoption of standardized definitions could enhance trial design, improve interpretability, and increase the real-world relevance in ITP research.