Background: Immune Thrombocytopenia (ITP) is an autoimmune platelet disorder that can lead to serious morbidity caused by bleeding and therapy-associated toxicities. The sequence of first-line therapies is well-defined, but the sequence of second-line therapies lacks consensus. A description of real-world experience is necessary to understand practice patterns with respect to second-line therapies and identify gaps in care. The objective of this study was to describe the types and chronological sequences of second-line therapies in a large cohort of ITP patients.

Methods: This was a retrospective cohort study. The population was derived from the McMaster ITP Registry, a prospective longitudinal registry of patients presenting with thrombocytopenia (platelets <150 × 109/L) to an academic hematology clinic in Hamilton, Canada. Inclusion criteria were: 1) diagnosis of primary or secondary ITP at most recent follow-up; 2) received one or more second-line therapy since initial ITP diagnosis; and 3) ≥6 months of follow up. No exclusions were applied. The sequence of second-line therapies was determined by treatment start dates. For treatments that were administered before enrollment in the registry, the order of treatment was determined by the reported treatment start date. When the start date was not reported, the sequence of treatments was assumed to be the order in which they were mentioned in the medical records. Second-line therapies included: splenectomy; rituximab; danazol; dapsone; thrombopoietin receptor agonists (TPO-RAs): eltrombopag or romiplostim; and immunosuppressants: mycophenolate, cyclosporine, azathioprine or cyclophosphamide.

Results: From January 2010 to December 2017, 789 patients with thrombocytopenia were registered in the McMaster ITP Registry. Of those, 204 had ITP and received second-line therapy (57.4% female). Median duration of ITP from the initial presentation to the most recent diagnosis was 9 years (IQR, 4-19). The proportion of patients who received each type of second-line therapy at any time were immunosuppressants (n=106, 52.0%), splenectomy (n=106, 52.0%), TPO-RAs (n=75, 36.8%), danazol (n=73, 35.8%), rituximab (n=67, 32.8%), and dapsone (n=4, 2.0%). Overall, 69 unique treatment sequences were identified. For patients who received one second-line treatment only (n=88), the most common single treatment was splenectomy (n=28, 31.8%), followed by immunosuppressants (n=21, 23.9%) and danazol (n=18, 20.5%) (Figure 1). For patients who received more than one second-line therapy (n=116), the most common treatment sequences were splenectomy followed by immunosuppressants (n=7, 6.0%); immunosuppressants followed by rituximab (n=6, 5.0%), and immunosuppressants followed by danazol (n=5, 4.0%). In patients who received each of these therapies, the last second-line treatments received were TPO-RAs (52/75, 69.3%), immunosuppressants (45/106, 42.5%), rituximab (37/67, 55.2%), splenectomy (36/106, 33.9%), danazol (33/73, 45.2%), and dapsone (1/4, 25%).

Conclusion: The types and sequences of second-line therapies for ITP were variable in a large Canadian academic center, where access to certain treatments including TPO-RAs and rituximab is limited. In this setting, splenectomy and immunosuppressant medications were commonly used as early second-line therapies. TPO-RAs were most often the last second-line treatment in the sequence of therapies used. Drug accessibility and other factors related to the choice of second-line therapies require further evaluation.


Arnold:Bristol-Myers Squibb: Research Funding; Principia: Consultancy; Rigel: Consultancy, Research Funding; Novartis: Honoraria, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution