• immune thrombocytopenia;
  • Eltrombopag;
  • thrombopoietin receptor agonists;
  • adverse effects;
  • cutaneous toxicity

Immune thrombocytopenia (ITP) is an autoimmune disorder characterized by thrombocytopenia <100 × 109/l without an evident cause (Neunert et al, 2011; Provan et al, 2010). Autoantibodies against platelet antigens and cellular immunity mediate increased platelet consumption and impaired thrombopoiesis (Cines et al, 2009). Mucocutaneous bleeding is frequent, while gastrointestinal or intracranial bleeding may occur (Cines & Bussel, 2005). As the risk of bleeding increases with lower platelet counts, treatment primarily aims at increasing platelets >30 × 109/l (Cines & Bussel, 2005; Neunert et al, 2011; Provan et al, 2010). Patients with chronic ITP (beyond 12 months) are frequently unresponsive to first-line treatment with corticosteroids and/or intravenous immunoglobulins (IVIG) and to second-line treatment with rituximab or other immunosuppressive agents (Cines & Bussel, 2005; Cines et al, 2009; Imbach & Crowther, 2011; Provan et al, 2010).

Thrombopoietin receptor (TPO-R) agonists represent novel therapeutics for chronic ITP. Two agonists binding to TPO-R on megakaryocytes were recently approved (Nieto et al, 2011). Eltrombopag is given orally once daily, Romiplostim subcutaneously once weekly. Interaction with the TPO-R transmembrane domain promotes proliferation and differentiation of megakaryocytes (Bussel et al, 2007, 2006; Imbach & Crowther, 2011; Nieto et al, 2011). Placebo-controlled trials have demonstrated effectiveness in 50–60% of patients. Tolerability was good with discontinuation of Eltrombopag due to adverse reactions equally frequent as with placebo (Bussel et al, 2007, 2009; Cheng et al, 2011). Concerns focus on hepatotoxicity and thromboembolic events, seen in 3·8–4% each (Bussel et al, 2007, 2009; Cheng et al, 2011; Nieto et al, 2011). Knowledge about cutaneous adverse reactions of Eltrombopag is scarce. A phase III study described rash in 3/88 patients on Eltrombopag versus 1/29 patients on placebo without specifying severity (Bussel et al, 2007). We describe three patients with significant Eltrombopag-induced cutaneous toxicity.

Patient 1 is a 81-year-old man with an 18-month-history of ITP initially responsive to prednisolone/IVIG. His first relapse at 4 months was treated with dexamethasone; deep vein thrombosis (DVT) occurred as platelets increased. A vitamin K antagonist was added for 6 months to long-standing acetylsalicylic acid due to coronary heart disease. A second relapse with epistaxis, petechiae and platelets of 30 × 109/l occurred 12 months after diagnosis. The patient refused splenectomy. Eltrombopag was initiated at a lower dose (25 mg) due to history of DVT. The patient responded with platelets of 300 × 109/l at 14 d but developed muscle vein thromboses of the lower legs. A vitamin K antagonist was re-introduced and Eltrombopag paused. Fourteen days later Eltrombopag was resumed at reduced dosing of 25 mg every 4 d and increased to daily dosing after another 14 d due to suboptimal response, while IVIG sustained platelet counts. Twelve days later, a pruritic, confluent maculo-papular exanthema developed, rapidly progressing to severe erythroderma (grade 3 toxicity according to common terminology criteria for adverse events). The patient chose to stop Eltrombopag. He underwent intense treatment with topical class IV corticosteroids. Symptoms were relieved within 24 h. Skin recovered over 4–5 weeks. Therapy was modified to Romiplostim 1 μg/kg body weight weekly, which normalized platelet counts without side effects during the four-month observation period (Fig 1).


Figure 1. Platelet counts and adverse event related to TPO-R agonists in Patient 1. ASS, acetylsalicylic acid; IVIG, intravenous immunoglobulins.

Download figure to PowerPoint

Patient 2 is a 59-year-old man with ITP responsive to corticosteroids diagnosed 14 years ago. Platelet counts fluctuated between 20–100 × 109/l without bleeding. He had allergies to several drugs including amlodipin and cephalosporins and was on regular haemodialysis. Aggravated thrombocytopenia of 5 × 109/l 13 years after diagnosis showed some response to corticosteroids with platelets increasing to 40 × 109/l, but severe thrombocytopenia recurred upon tapering. Haematological investigation showed myelodysplastic syndrome (MDS) with refractory cytopenia and multilineage dysplasia. Despite MDS, an immune mechanism aggravating thrombocytopenia was supposed. Eltrombopag was initiated at 50 mg/d, increasing platelets to 45–55 × 109/l within 4 weeks. After 3 months of therapy a pruritic, rapidly progressing exanthema developed. The patient stopped Eltrombopag and presented shortly after with gastrointestinal bleeding and platelet count of 8 × 109/l. Romiplostim 150 μg weekly was initiated and increased to 250 μg. Platelets increased above 30 × 109/l while bleeding stopped.

Patient 3 is a 77-year-old woman with chronic ITP and history of hypersensitivity to acetylsalicylic acid, penicillins and cefuroxime. Mild thrombocytopenia was documented 5 years ago and ITP was diagnosed 2 years later. Corticosteroids were successfully initiated prior to peridural anesthesia for chronic back pain; platelet count was 10 × 109/l. Four months later, she presented with thrombocytopenia (11 × 109/l) prior to elective surgery. Corticosteroids were avoided and splenectomy not considered due to comorbidities. Eltrombopag was initiated at 50 mg daily. Exanthema, pruritus and enoral erythema developed after 2 d. The patient stopped Eltrombopag after 2 additional days due to progression. Treatment was reinitiated along with antihistamines despite persistent exanthema. Platelet counts increased to 200 × 109/l within 14 d, prompting reduction to 25 mg Eltrombopag daily. Pruritic exanthema slowly regressed over 4 weeks. Currently the patient has being 10 months under Eltrombopag and antihistamines, maintaining platelet counts at 100–200 × 109/l without further complications.

Following the registration of Eltrombopag in Switzerland (June 2010), 3 cases of significant cutaneous adverse events occurred at our centre ranging from mild exanthema to severe erythroderma. Two of these 3 patients had known polyallergy to other drugs. All three patients stopped Eltrombopag without counselling from their haematologist, highlighting the urgency of their symptoms. This is significant as platelets rapidly returned to baseline or below due to a potential rebound effect thus putting patients at increased risk of bleeding, particularly in the setting of concomitant antiplatelet or anticoagulant treatment (Neunert et al, 2011; Provan et al, 2010). Lower dosing of Eltrombopag (25 mg) apparently did not protect from cutaneous reactions. In the milder case, re-initiation of Eltrombopag along with an antihistamines was safe. In the 2 cases of severe exanthema with erythroderma, treatment was switched to Romiplostim without adverse skin reactions (Table 1).

Table 1. Characteristics of patients and cutaneous adverse reactions due to Eltrombopag
 Patient 1Patient 2Patient 3
  1. AE, adverse event; ITP, immune thrombocytopenia.

Initial dose (mg/d)255050
Adverse event (AE)Severe erythroderma, pruritusSevere exanthema, pruritusMild exanthema, enoral erythema, pruritus
Time to AE9 weeks3 months2 d
Management of AE

Eltrombopag stopped

Intense topical corticosteroids


Eltrombopag stopped


Alternative ITP therapyRomiplostimRomiplostimNone, Eltrombopag continued

Every new medication requires attentive post-marketing surveillance to optimize patients' safety. Randomized clinical trials observed rash as side effect in only 2–3% of patients (Bussel et al, 2007, 2009; Cheng et al, 2011). Relevant cutaneous toxicity has not been reported. To the best of our knowledge this is the first report of severe cutaneous toxicity induced by Eltrombopag.

Due to the limited number of patients, we can only hypothesize about risk factors for Eltrombopag-induced skin toxicity. Two of the 3 patients had hypersensitivity to other drugs, a risk factor for further drug hypersensitivity. As two patients tolerated Romiplostim without adverse reactions, cutaneous toxicity related to TPO-R agonists does not seem to represent a class effect.

In patients with mild skin rash, continuation of Eltrombopag along with antihistamines can be attempted. However, in severe cases with erythroderma, Eltrombopag should be stopped and alternative modalities including Romiplostim considered. Longer observational studies are warranted to characterize risk factors for TPO-R agonist-induced skin toxicity.


  1. Top of page
  2. References
  • Bussel, J.B., Kuter, D.J., George, J.N., McMillan, R., Aledort, L.M., Conklin, G.T., Lichtin, A.E., Lyons, R.M., Nieva, J., Wasser, J.S., Wiznitzer, I., Kelly, R., Chen, C.F. & Nichol, J.L. (2006) AMG 531, a thrombopoiesis-stimulating protein, for chronic ITP. New England Journal of Medicine, 355, 16721681.
  • Bussel, J.B., Cheng, G., Saleh, M.N., Psaila, B., Kovaleva, L., Meddeb, B., Kloczko, J., Hassani, H., Mayer, B., Stone, N.L., Arning, M., Provan, D. & Jenkins, J.M. (2007) Eltrombopag for the treatment of chronic idiopathic thrombocytopenic purpura. New England Journal of Medicine, 357, 22372247.
  • Bussel, J.B., Provan, D., Shamsi, T., Cheng, G., Psaila, B., Kovaleva, L., Salama, A., Jenkins, J.M., Roychowdhury, D., Mayer, B., Stone, N. & Arning, M. (2009) Effect of eltrombopag on platelet counts and bleeding during treatment of chronic idiopathic thrombocytopenic purpura: a randomised, double-blind, placebo-controlled trial. Lancet, 373, 641648.
  • Cheng, G., Saleh, M.N., Marcher, C., Vasey, S., Mayer, B., Aivado, M., Arning, M., Stone, N.L. & Bussel, J.B. (2011) Eltrombopag for management of chronic immune thrombocytopenia (RAISE): a 6-month, randomised, phase 3 study. Lancet, 377, 393402.
  • Cines, D.B. & Bussel, J.B. (2005) How I treat idiopathic thrombocytopenic purpura (ITP). Blood, 106, 22442251.
  • Cines, D.B., Bussel, J.B., Liebman, H.A. & Luning Prak, E.T. (2009) The ITP syndrome: pathogenic and clinical diversity. Blood, 113, 65116521.
  • Imbach, P. & Crowther, M. (2011) Thrombopoietin-receptor agonists for primary immune thrombocytopenia. New England Journal of Medicine, 365, 734741.
  • Neunert, C., Lim, W., Crowther, M., Cohen, A., Solberg, Jr, L. & Crowther, M.A. (2011) The American Society of Hematology 2011 evidence-based practice guideline for immune thrombocytopenia. Blood, 117, 41904207.
  • Nieto, M., Calvo, G., Hudson, I., Feldschreiber, P., Brown, D., Lee, C.C., Lay, G., Valeri, A., Abadie, E., Thomas, A. & Pignatti, F. (2011) The European Medicines Agency review of eltrombopag (Revolade) for the treatment of adult chronic immune (idiopathic) thrombocytopenic purpura: summary of the scientific assessment of the Committee for Medicinal Products for Human Use. Haematologica, 96, e33e40.
  • Provan, D., Stasi, R., Newland, A.C., Blanchette, V.S., Bolton-Maggs, P., Bussel, J.B., Chong, B.H., Cines, D.B., Gernsheimer, T.B., Godeau, B., Grainger, J., Greer, I., Hunt, B.J., Imbach, P.A., Lyons, G., McMillan, R., Rodeghiero, F., Sanz, M.A., Tarantino, M., Watson, S., Young, J. & Kuter, D.J. (2010) International consensus report on the investigation and management of primary immune thrombocytopenia. Blood, 115, 168186.