Eltrombopag-Induced Serum Pigmentation in Severe Aplastic Anemia: A Case Report Authors: Eva Desbrousses, Tifenn Charbonnel, Flavie Arbion, Hugo Alarcan, Nicolas Vallet Published: 2026 ABSTRACT AbstractIntroductionSevere aplastic anemia (SAA) is a bone marrow failure disorder characterized by profound pancytopenia associated with a high risk of infection and bleeding.Case PresentationWe report a case of a 19-year-old patient with a history of acute hepatitis who was admitted to the hospital with acquired SAA attested by 2% cellularity at bone marrow trephine biopsy. In the absence of matched sibling donor, immunosuppressive therapy with horse anti-thymocyte globulin, cyclosporine, and eltrombopag was initiated. Following centrifugation, serum obtained from blood samples addressed to the laboratory displayed a rare manifestation of brown color. Intravascular hemolysis was excluded based on normal haptoglobin and lactate dehydrogenase ranges, although bilirubin was mildly elevated. The brown staining disappeared after therapy discontinuation. Eltrombopag is a chromogenic molecule which absorbs visible light associated with serum coloration.ConclusionThis rare case highlights drug-induced serum pigmentation as a surprising manifestation in patients receiving eltrombopag and its normalization following drug withdrawal. Introduction Aplastic anemia is a rare disease characterized by hypoplastic bone marrow caused by the decrease of hematopoietic stem cells and pancytopenia in the absence of abnormal infiltrate or marrow fibrosis. The disease severity is defined according to the Camitta criteria [1]. The standard first-line therapy relies on allogeneic hematopoietic stem cell transplant from a matched sibling donor if available for younger patients. Otherwise, horse anti-thymocyte globulin (hATG) and cyclosporine with eltrombopag, a treatment broadly used for immune thrombocytopenia, is considered [1–3]. We report a striking serum coloration observed 34 days following the first-line treatment of a young patient who received eltrombopag for severe aplastic anemia (SAA) and its normalization following treatment withdrawal (online suppl. material; for all online suppl. material, see https://doi.org/10.1159/000553383). Patient Information A 19-year-old Caucasian woman was admitted to the clinical hematology department in the context of pancytopenia. The patient was born at 39 weeks of gestation, with a birth weight of 3,640 g. The patient reported a history of febrile hepatitis that occurred 2 months prior to hospitalization. Etiological investigations did not identify either a viral or an autoimmune or toxic cause. Liver function tests returned to normal spontaneously. The patient had no history of previous recurrent infectious events nor any organ dysfunctions. Previous blood counts had been within normal ranges. No significant medical history was identified in the parents or in the sibling. Clinical Findings Physical examination was normal, with no evidence of organomegaly, cutaneous abnormalities, intraoral malformations, or abnormalities of hand morphology; body weight and height were 65 kg and 173 cm, respectively. A summary of biological findings is shown in Table 1. Table 1.Peripheral blood characteristics at diagnosisBiological parametersUnitAt diagnosisLaboratory reference rangeHematology Hemoglobing/L113129–170 Mean corpuscular volumefL8180–100 Reticulocytes109/L25NA Absolute neutrophil count109/L0.381.5–7.5 Platelets109/L8150–400 Prothrombin time%98N/A Thrombin timeN/A1.08N/A Fibrinogeng/L2.852.0–4.0 Fetal hemoglobin%<1N/ABiochemistry Creatinineµmol/L6059–104 Estimated glomerular filtration rate (CKD-EPI formula)mL/min/1.73 m2127N/A Total bilirubinµmol/L62–24 Indirect bilirubinµmol/L12–17 Haptoglobing/L1.400.5–2.0 Lactate dehydrogenaseU/L156135–225 Ferritinµg/L15830–400 Reactive C proteinmg/L1.40.3–5.0 Vitamin B9nmol/L118.8–60.8 Vitamin B12pmol/L208145–569 Alpha-fetoproteinkU/L3.0<6Immunology CD3+ lymphocytes109/L1.492N/A CD3+CD4+ lymphocytes109/L1.035N/A CD3+CD8+ lymphocytes109/L0.383N/A CD19+ lymphocytes109/L0.183N/A CD3-CD56+CD16+ lymphocytes109/L0.043N/A Peripheral blood characteristics at diagnosis Diagnosis Assessment The bone marrow aspirate was hypocellular, with the absence of megakaryocytes and no evidence of dysplastic or abnormal cellular infiltration. The diagnosis was then assessed with trephine bone marrow biopsy which revealed a marked hypocellularity with less than 2% cellularity, replacement of hematopoietic tissue by adipose tissue without evidence of dysplasia or malignant infiltration or fibrosis (Fig. 1a). Case report overview. a Hematoxylin and eosin staining of the bone marrow trephine biopsy. b Serum exhibited reddish brown color during eltrombopag treatment. c Serum color normalization following eltrombopag discontinuation. Elapsed days from treatment start and from discontinuation are indicated on each sample. d Timeline illustration of the case report and findings. Flow cytometric analysis of peripheral blood identified a small type II paroxysmal nocturnal hemoglobinuria clone in neutrophils (0.3%) and monocytes (1.8%) [4]. The workup did not reveal signs of hemolysis or signs which may suggest a germline origin of the aplastic anemia (Table 1) [5]. Altogether the findings defined a post-hepatitis idiopathic SAA [1, 5]. Therapeutic Intervention Given the absence of HLA-identical sibling donor, a first-line therapy was started combining ciclosporin 200 mg/12 h, hATG 40 mg/kg/day, and eltrombopag 150 mg/day (day 1). Follow-Up and Outcomes On day 34, blood samples addressed to the laboratory displayed a brown color after centrifugation (Fig. 1b). On day 48, the patient presented with skin hyperpigmentation. The first hypothesis was that it may have been associated with hemolysis, rhabdomyolysis [6]. Intravascular hemolysis was excluded as haptoglobin and lactate dehydrogenase levels were within normal ranges, respectively, within 1.9–4.3 g/L and within 154–251 UI/L, although bilirubin was slightly elevated at 17 µmol/L. The patient presented a normal concentration of methemoglobin (4.2%) which might not explain the coloration of the serum [7]. Drug-induced hyperpigmentation was therefore suspected. Hyperpigmentation is a reported adverse effect in eltrombopag. However, eltrombopag-induced serum discoloration has been described only in a limited number of case reports. In fact, eltrombopag has an intrinsic pigment which absorbs visible light and mainly binds to albumin. It is widely distributed into tissues, including the skin and ocular tissues. This staining disappears after therapy discontinuation (Fig. 1c). During this event, total bilirubin was elevated with a peak at 34 μmol/L of mainly conjugated bilirubin (29 μmol/L). Still, the bilirubin dropped before discontinuation of the drug. Thus, bilirubin may not be associated with eltrombopag treatment (Fig. 1d). Due to first-line treatment failure illustrated by the persistence of pancytopenia, the patient subsequently underwent a 10/10 HLA-matched unrelated allogeneic hematopoietic stem cell transplant with bone marrow cells following conditioning with cyclophosphamide (300 mg/m2, day 6 to day 3), alemtuzumab (0.2 mg/kg, day 7 to day 3), fludarabine (30 mg/m2, day 6 to day 3), with graft-versus-host disease (GVHD) prophylaxis consisting of ciclosporin and methotrexate [8]. The patient achieved complete remission. However, 5 months after allogeneic hematopoietic stem cell transplant, the patient developed corticosteroid-refractory moderate chronic GVHD with skin (stage 1) and eye (stage 2) localization which required second-line ruxolitinib, leading to a complete response. At 13 months post-transplant, the patient successfully gave birth to a child, and at 23 months of follow-up, she remains alive with no evidence of active GVHD (Fig. 1d). Discussion Eltrombopag is a thrombopoietin receptor agonist used in combination with hATG and cyclosporine to treat patients diagnosed with SAA. High doses of eltrombopag are associated with skin hyperpigmentation and serum coloration. In this clinical case, the striking brown coloration of serum raised concern for hemolysis. However, the absence of biochemical evidence supporting hemolysis prompted further investigations. In fact, eltrombopag is reddish in aqueous solution, its color ranging from yellow to brown depending on the pH. At higher concentrations, this pigmentation becomes apparent after centrifugation. Beyond its color, eltrombopag may interfere with laboratory assays. Spectrophotometric methods used for bilirubin measurement are particularly susceptible to interference because eltrombopag absorbs light at overlapping wavelengths, potentially leading to falsely elevated total bilirubin values. Eltrombopag inhibits organic anion‐transporting polypeptide (OATP1B1) and uridine diphosphoglucuronosyltransferase (UGT1A1) which are involved in hepatic bilirubin uptake and conjugation [6, 9]. This pharmacologic interaction may contribute to mild indirect hyperbilirubinemia, further complicating interpretation of hemolysis workup. Automated hemolysis and icterus indices may therefore be misleading in patients receiving eltrombopag. An interference is described for total cholesterol, triglycerides, and lipoprotein cholesterol. Here, elevation of conjugated bilirubin levels was observed during the treatment, but its lowering before the discontinuation of eltrombopag is not in favor of an impact of the drug on bilirubin levels. To our knowledge, no significant drug interactions have been described with the high doses of eltrombopag used in SAA, particularly with cyclosporine. The strengths of the report are the clinical history, including a time course of images demonstrating plasma color before and after eltrombopag discontinuation. Eltrombopag has a reported elimination half-life ranging from approximately 21–32 h. However, it is highly protein-bound (>99%), predominantly to albumin, and exhibits marked tissue distribution, including cutaneous and ocular tissues [10, 11]. In addition, eltrombopag was shown to be associated with interference in chemistry testing, notably bilirubin [12]. Consequently, the persistence of visible serum pigmentation may therefore not directly correlate with elevated bilirubin but with persistence of the drug. Also, the skin pigmentation could be related to the bilirubin levels observed in our case, but skin jaundice usually occurs when bilirubin reaches 50 μmol/L, lower than what was found here [13]. Only a few cases of serum hyperpigmentation attributed to eltrombopag have been reported. One article described a 78-year-old man with SAA treated with cyclosporine 125 mg/12 h, eltrombopag 150 mg/24 h, and allopurinol 300 mg/24 h who presented with brown serum [6]. Another report described a 52-year-old woman with aplastic anemia receiving eltrombopag to 150 mg/24 h who also presented “red serum” [14]. Herein, this case report describes the change in colors after eltrombopag discontinuation. As serum coloration is not routinely evaluated. Therefore, serum pigmentation may have occurred several days before it was first documented. In addition, there may be more patients concerned by this observation which may warrant further study to understand the characteristics of concerned patients and whether this manifestation may be associated with the outcome of the patient. The dose of treatment is significantly associated with hyperpigmentation and serum of patients, thus lower dose for immune thrombocytopenia might be not be associated with hyperpigmentation [3, 12, 15]. This case describes a rare hyperpigmentation of the serum observed induced by eltrombopag. The analytical interference of bilirubin and lipid exploration resulted in conflicting results, depending on instrumentation and assays, but this had no impact on treatment management and was reversible upon drug discontinuation. Statement of Ethics This study has been granted an exemption from requiring ethics approval by the Helsinki Committee. Written informed consent was obtained from the patient for publication of the details of their medical case and any accompanying images. Conflict of Interest Statement The authors have no conflicts of interest to declare. Funding Sources This study was not supported by any sponsor or funder. Author Contributions N.V. and H.A. conceptualized the study. T.C. and H.A. analyzed the serum samples. E.D. and N.V. wrote the original draft of the manuscript. E.D., N.V., T.C., and H.A. ensured data collection. F.A. provided illustration of the bone marrow biopsy. All authors critically reviewed and edited the final manuscript.