En bloc resection for extra-nodal Rosai-Dorfman disease in the anterior mediastinum: a case report Authors: Ling You, Nanlin He, Lei Liu, Xiaoming Qiu Published: 2025 ABSTRACT AbstractRosai-Dorfman disease (RDD) is a sinus histiocytosis with an undefined etiology and usually presents with painless massive lymphadenopathy. RDD is a benign histiocytic proliferative disease with extremely low incidence and commonly in young patients. Isolated intrathoracic RDD is comparatively rare in clinical practice with non-specific clinical manifestations, and can only be diagnosed after histological confirmation. Here, we reported a 28-year-old female who complained of a one-year cough with unknown causes, and chest contrast-enhanced computed tomography showed a soft-tissue density and slightly strengthened mass with 6.8 cm * 2.4 cm detected in the anterior-upper mediastinum. We suggested that en bloc resection of mediastinal RDD with potential adjacent tissue invasion was dependable and could lead to a favored prognosis. This case emphasizes the importance of considering RDD in the differential diagnosis of anterior mediastinal masses.Keywords: Rosai-Dorfman disease, Anterior mediastinum, En bloc resection, Surgical resection Introduction Rosai-Dorfman disease (RDD), first reported by Rosai and Dorfman in 1969, is a sinus histiocytosis with an undefined etiology and usually presents with painless massive lymphadenopathy [1]. RDD is a benign histiocytic proliferative disease with extremely low incidence (~ 1/200 000) and commonly in young patients [1]. It is a self-limited disease and could be classified into two subtypes conformed to the disease involvement, nodal RDDs and extra-nodal RDDs [2, 3]. Isolated intrathoracic RDD showed non-specific presentation, including cough, dyspnea, and chest pain, with fever and weight loss could be noted in several cases [2–5]. RDD can only be diagnosed after histological confirmation and lack of standard treatment. Here, we reported our experience of en bloc resection for extra-nodal RDD located in the anterior mediastinum and surrounded by cardiac great vessels, and we proposed that curative intervention was feasible and with favored prognosis. Case presentation A 28-year-old female complained of a one-year cough with unknown causes, and she received chest computed tomography (CT) which showed an anterior mediastinal mass. Then she received a puncture biopsy, and pathological examination suggested mesenchymal neoplasms with focal positivity of IgG4 for immunohistochemistry staining. Thereafter, she reported having received hormone therapy at another hospital without otherwise specified, and the mass got a bit smaller in the follow-up. She was admitted to our center, and chest contrast-enhanced CT showed a soft-tissue density and slightly strengthened mass with 6.8 cm * 2.4 cm detected in the anterior-upper mediastinum. The mass showed ambiguous boundaries and the left brachiocephalic vein was compressed with slight lumen stenosis (Fig. 1A). Laboratory investigations didn’t show specific presentations. A Preoperative contrast-enhanced computed tomography of the 28-year-old female patient showed a soft-tissue density and slightly strengthened mass with 6.8 cm * 2.4 cm detected in the anterior-upper mediastinum; B the hematoxylin-eosin staining of the mediastinal Rosai-Dorfman disease (RDD) and the black arrow indicates the emperipolesis; C-I the immunohistochemistry staining of the RDD for S-100 (+; C), CD38 (+; D), IgG4 (+; E), MUM1 (+; F), Ki-67 (3–5%; G), PCK (-; H), and EMA (-; I) Then we performed mid-sternal thoracotomy for the anterior mediastinal mass under general anesthesia, during which we found that it invaded the pericardium and wrapped the brachiocephalic vein. We carefully ligated and cut the feeder vessels, and removed the involved pericardium without pericardium reconstruction. It was a solid, malleable mass with intact capsule, 9.5 cm * 7.5 cm * 4 cm. The histopathological examination indicated focal lymphadenosis and plasma cell infiltration among the fibroblastic proliferation and hyalinization contained with fusiform cells and foam-like tissue (Fig. 1B). In HE staining, the lymphocytes and plasma cells stretch into the histiocytes. Immunohistochemistry staining indicated the positivity of S-100, cluster of differentiation 38 (CD38), multiple myeloma oncogene 1 (MUM1), IgG4 (12/high power field [HPF]), Ki-67 (MIB-1, 3–5%, in the lymphocyte), and the negativity of pan-cytokeratin, epithelial membrane antigen (EMA/MUC-1), CD34, CD99, BCL2, STAT6, desmin, smooth muscle actin (SMA), and anaplastic lymphoma kinase 1 (ALK-1) (Figs. 1C-I). Incorporated with the morphology and molecular biology findings, it was finally diagnosed with extra-nodal RDD. After an eight-year follow-up, the patient reported durable cough relief and was asymptomatic. Discussion RDD, first described by Destombes in 1965 and named by Rosai and Dorfman in 1969, was also termed Rosai-Dorfman Destombes Disease [1]. RDD is a benign and limited sinus histiocytosis, while mediastinal RDD is quite rare in clinical practice. It is prevalent in any age and comparatively frequent in young patients with a mean age of around 20 years old [1]. Extra-nodal RDD is reported in 47% of the cases, and it is generally distributed within the skin and soft tissues [2]. Only 2% of the patients were described as intrathoracic RDD, and clinical manifestations were quite various, including chronic dry cough, progressive dyspnea, and pleural effusions [6, 7]. Hence, in this case, it might be first considered as other common mediastinal neoplasms. Clinical presentation of mediastinal RDD is often non-specific, and laboratory findings are typically unremarkable. In our case, only a cough with unknown causes was identified in the 28-year-old patient, which might be explained by the compressive effect on the thorax or endocrine disorders. The pathogenesis remained to be confused and it might be associated with viral infection and autoimmune disorders. MAPK pathway activation contributed most to the known driver kinase mutations, including MAP2K1, KRAS, NRAS, BRAF, and ARAF mutations [8–10]. The diagnosis is more dependent on the histopathological evaluations, with genetic sequencings being more informative [11]. RDD shares comparable morphological patterns with Langerhans cell histiocytosis and features by nonspecific reactive histiocytes. The abnormal histiocytes primarily occupy the sinuses, and nodal enlargement results from the collective hyperplasia of histiocytes, lymphocytes, and plasma cells within the parenchyma [12]. RDD was commonly distributed in a reactive inflammatory background, and IgG4-positive plasma infiltration could be identified [13]. Among the numerous plasma cells, an elevated rate of plasma IgG/IgG4 > 40% is highly indicative [14]. The emperipolesis phenomenon that integrated inflammation cells or erythrocytes could be detected in part of the histiocytes, which was also positive in this case and in line with prior reports. Notably, emperipolesis was characteristic, but not a specific finding [13]. What’s more, it was suggested that the RDD cells showed no cytologic atypia with positive expression of S-100, fascin, and macrophage-specific markers (CD14, CD33, or CD68), and negative expression of proliferation antigens (Ki-67 or CD25), with controversial expression of DC markers (CD1a) [15]. Clinical observation is appropriate in most cases due to the self-limited property, and 20–50% of the patients with nodal or cutaneous involvement were likely to be associated with disease relief [11, 16]. Corticosteroids were reported partial response for the central nervous system and bone RDD, while some reports claimed no response for orbital, tracheal, renal, and soft tissue RDD [11, 17–19]. In this case, the patient also received hormone therapy and the mass became smaller during the follow-up. Surgery is more recommended in unifocal diseases or symptomatic diseases, and adjuvant treatment might be considered in case of the disease recurrence [11, 17]. While clinical diagnosis through imaging examination might not provide enough directional evidence, we recommend to conduct en bloc resection for mediastinal RDD, and reconstructive surgery could be considered when part of the adjacent tissues was involved. Chemotherapy and immunotherapy could be an alternative for severe or refractory diseases [11]. While target therapy was also promising that cobimetinib, a MEK inhibitor, was effective in blocking the RAS–MAP2K1 pathway, a pivotal pathogenic pathway in RDD [20]. Extra-nodal RDD in the anterior mediastinum was uncommon in clinical practice, and we suggested that en bloc resection of mediastinal RDD with potential adjacent tissue invasion and without elsewhere involvement was dependable and could lead to a favored prognosis. Authors’ contributions LY and NH contributed to the conceptualization of the study and drafting of the manuscript and took full responsibility for the content. LY and LL contributed to the data curation. LY and XQ contributed to the revision of the manuscript. XQ contributed to supervision and project administration. LY, NH, LL, and XQ contributed to the approval of the final manuscript. Funding The study was supported by the 1·3.5 project for disciplines of excellence, West China Hospital, Sichuan University (No. ZYAI24064). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate The study was approved by the Ethics Committee on Biomedical Research, West China Hospital of Sichuan University (2022 − 1758), and informed consent was obtained from the participant. Consent for publication Written informed consent for publication of both Case 1 and Case 2 s clinical details and/or clinical images was obtained directly from both of the patients. A copy of this consent to publish is available for review by the editor of the journal. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Ling You and Nanlin He contributed equally to this work. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.