TPO-RAs and ITP remission: cause or coincidence?

In this issue of Blood, Guillet et al report high rates of sustained platelet response following discontinuation of thrombopoietin receptor agonist (TPO-RA) therapy in adult patients with persistent or chronic immune thrombocytopenia (ITP) and a documented complete response (platelet count > 100 × 10⁹/L).¹ 

Second-line treatment options for ITP include rituximab, splenectomy, and TPO-RAs. In the absence of clinical trials directly comparing these strategies, the choice of treatment is guided by patient values and preferences. Patients who value avoidance of surgery and/or immunosuppression and do not mind taking a regular medication can be considered for TPO-RA therapy.

TPO-RAs work by binding to the thrombopoietin receptor on megakaryocytes to increase production of mature platelets. TPO-RAs are associated with a platelet response in >70% of patients who have failed ≥1 prior therapy, with ≥40% of patients achieving durable platelet responses on therapy.² In addition to the ability of TPO-RAs to stimulate thrombopoiesis, treatment with TPO-RAs has been associated with immunologic alterations in several studies. In 1 study, treatment with TPO-RA resulted in improved regulatory T-cell activity with concurrent reduction in interleukin-2–producing T helper cells and an increase in circulating transforming growth factor-β1 (TGF-β1), changes that correlated with the degree of platelet count improvement.³ In a second study, patients with chronic ITP who responded to TPO-RA treatment were found to have an increase in CD19⁺ regulatory B cells.⁴ More recently, in a study of 21 patients with chronic ITP, treatment with the TPO-RA eltrombopag resulted in an increase in Fcγ receptor (FcγR) IIb with a decrease in FcγRI and IIa levels.⁵ In addition, TGF-β1 concentrations increased and monocyte and macrophage phagocytic capacity decreased in responding patients but not in TPO-RA nonresponders.

Immune modulation with TPO-RAs raises the question of whether treatment with these agents can induce sustained improvement in the platelet count, even after discontinuation of TPO-RA therapy. In the present study, Guillet et al found that of patients with persistent or chronic ITP who achieved a complete response (platelet count >100 × 10⁹/L) on eltrombopag or romiplostim, 56.2% were able to sustain a platelet response (platelet count >30 × 10⁹/L and no bleeding) following discontinuation of TPO-RA. Most of these patients (25/27) maintained a response at 52 weeks, and 29.2% maintained a complete response over this interval. No clinical predictors of sustained response off treatment were identified; however, patients demonstrating relapse after discontinuation of TPO-RA were found to have overexpression of CD69 on CD8⁺ T cells at baseline compared with those with sustained response.

Although the sustained response rates off treatment reported by Guillet et al are impressive, they must be interpreted in the context of the patient population and eligibility criteria. Rather than enrolling consecutive or unselected patients with persistent or chronic ITP on TPO-RA therapy, the authors included only those with a stable complete response (platelet count >100 × 10⁹/L for >2 months) on treatment. In so doing, they selected patients with the mildest and/or most treatment-responsive disease. Given that only 90 of the 388 patients (23%) screened for the study met eligibility criteria, the authors’ findings may be applicable to only a minority of TPO-RA–treated patients. Other studies with broader inclusion criteria have demonstrated more modest sustained treatment-free response rates following TPO-RA discontinuation of ∼20% to 30%.⁵ 

This study also does not address the etiology of the sustained responses observed following TPO-RA discontinuation. Other medical therapies including corticosteroids, IV immunoglobulin (IVIG), and rituximab are also associated with long-term treatment-free response rates of ∼30% (see table), begging the question of whether these treatments are inducing long-term remissions or ITP is destined to remit in approximately one-third of adults, irrespective of treatment.⁹ It would be interesting to test whether the same immunologic changes that predict treatment-free remission following discontinuation of TPO-RAs also predict long-term responses after treatment with other medical therapies. It should also be noted that some patients, even those with chronic and highly refractory ITP, will eventually experience spontaneous remission in their disease.¹⁰ 

Regardless, the findings of Guillet et al are promising and offer additional evidence to support a trial of TPO-RA discontinuation in ITP patients who are responding well to therapy as well as a role for TPO-RAs in immune modulation of patients with persistent/chronic ITP. More research is needed to determine whether TPO-RAs play a role in inducing sustained treatment-free responses or whether they merely serve as a bridge for patients whose disease is destined to ultimately improve on its own.

Conflict-of-interest disclosure: K.S. has served as a consultant for Health Services Advisory Group, has provided expert case review for Quinn Johnston, and receives funding from the National Heart, Lung and Blood Institute, National Institutes of Health (1R01HL166386-01) and the American Society of Hematology (ASH) (ASH Scholar Award). A.C. has served as a consultant for Synergy and New York Blood Center and has received authorship royalties from UpToDate.