Extranodal Rosai-Dorfman Disease in a Pediatric Patient: A Case Report

Introduction

Rosai-Dorfman disease (RDD) was first described in 1969 as sinus histiocytosis with massive lymphadenopathy (SHML) by Rosai and Dorfman [1]. It is a rare, non-Langerhans cell histiocytosis with a prevalence of 1: 200 000 and an estimated 100 new cases per year in the United States [2]. It is most common in children and young adults, with a mean age of onset of 21 years. In 2016, the Histiocyte Society reclassified RDD into the “R” group of histiocytoses, with further subclassification into familial, classical or nodal, extranodal, neoplasia-associated, and immune disease-associated RDD [2]. The clinical presentation of RDD is usually subtle and nonspecific. The most common initial presentation in patients with nodal RDD is painless, bilateral lymphadenopathy, usually cervical, accompanied by fever and weight loss [3]. Extranodal RDD presents either with a primary extranodal manifestation or in conjunction with nodal RDD. Laboratory results are variable and nonspecific. Common findings range from leukocytosis, and elevated erythrocyte sedimentation rate (ESR) to hypergammaglobulinemia [3]. Because of nonspecific clinical findings, the diagnosis of RDD should be confirmed by lesional tissue examination. The presence of large histiocytic cells with emperipolesis is the characteristic pathological feature. These cells stain positive for CD14, HLA-DR, CD68, CD163, S-100, and fascin, and negative for CD1a, differentiating RDD from Langerhans cell histiocytosis (LCH) [2].

The objective of this paper is to report an unusual case of RDD, which will improve clinicians’ and pathologists’ understanding of RDD in the diagnosis of radiographically aggressive bone lesions in children.

Case Report

This 16-month-old boy presented to the Emergency Department for distal, right-forearm pain, following a right-wrist injury while playing with a sibling. He was born full term and was healthy prior to this incident. On physical examination, all findings were normal other than a swollen right wrist, which caused him to avoid using his right arm. X-ray of the right forearm revealed a lytic lesion with a wide zone of transition, cortical destruction, and poorly defined periosteal reaction in the distal radius, which was concerning for an aggressive primary bone lesion (Figure 1A). He was admitted for further evaluation by a pediatric oncologist. Subsequent magnetic resonance imaging (MRI) of the right arm confirmed a destructive lesion in the distal radius with a prominent adjacent soft-tissue component. The imaging evaluation favored an aggressive bone lesion, such as Ewing sarcoma, given the patient’s age. Laboratory results were normal for leukocyte counts, ESR, and immunoglobulins.

Three days later, a core-needle biopsy was performed for diagnosis. The biopsy material was scant and fragmented. It was composed of reactive lamella bone with acute and chronic inflammatory infiltration. There was no evidence of malignancy. Examination of the scant biopsy material suggested a diagnosis of osteomyelitis. The patient received cephalexin for 5 days, without improvement of symptoms. Subsequently, on hospital day 8, he underwent curettage with washout for the management of presumed osteomyelitis, which was unresponsive to antibiotics, and to obtain more material to rule out malignancy. The curettage material showed mixed inflammatory infiltration with a prominent population of large histiocytic cells containing abundant foamy cytoplasm, and showing emperipolesis (Figure 1B). Immunohistochemistry revealed histiocytic cells positive for CD68 and S-100 (Figure 1C) and negative for CD1a. The fungal, mycobacterial, and bacterial cultures, as well as PCR for Kingella kingae, were negative. A diagnosis of extranodal osseous RDD was made. An external expert bone pathology consultant confirmed the diagnosis. The patient was released from the hospital 2 days after the surgery, without further treatment. Radiologic and pathologic features for the differential diagnosis in this case are summarized in Table 1.

As an outpatient he was staged with imaging studies 1 month after the diagnosis, which confirmed the isolated right radius bone lesion without enlarged lymph nodes or any other organ or tissue involvement. Since the curettage removed the lesional tissue, additional surgery was deemed unnecessary. The patient remained disease free by imaging and clinical status. The decision was made to perform observation management, with surveillance imaging every 6 months. Follow-up computed tomography (CT) scans of the right forearm over the next 36 months showed bone sclerosis and remodeling, which was consistent with callus formation without recurrent disease. To date, he is healthy, without pain, and has normal right-wrist function.

Discussion

A standard treatment protocol for RDD is not available due to the wide spectrum of clinical manifestations, especially since up to 50% lesions regress spontaneously [4]. Observation is the preferred option in most cases [5,6]. Chemotherapy, immunotherapy, and targeted therapy are used for symptomatic systemic RDD diseases [7]. In case of bone lesions, surgery is the choice for pain control and prevention of fracture [6]. In our case, observation without further therapy was chosen because there was no residual bone disease after curettage and there was no evidence of systemic disease.

The pathogenesis of RDD is not well-defined, and, based on the phenotypes, is likely variable. RDD is a non-neoplastic disease. Although recent studies identified gene mutations, such as point mutation in KRAS and MAP2K1 in patients with RDD, none could prove the neoplastic nature of RDD cells [8]. Rare familial cases were reported, but our patient had no family history of predisposing RDD.

Most patients with RDD present with enlarged lymph nodes. Up to 40% of patients with RDD manifested extranodal involvement, with or without coexisting lymphadenopathy [9]. The most common extranodal site is the skin and soft tissue [10]. Bone involvement occurs in 5–10% of RDD cases but is usually associated with or preceded by lymph node manifestations [11]. RDD with isolated bone involvement occurs in only 2% of patients [12]. The most common bone manifestations involve the cranium, facial bones, and tibia [13]. RDD affects metaphysis or diaphysis of long bones, with features of an osteolytic process. The radiographic and histologic differential diagnosis is broad, ranging from chronic osteomyelitis, to lymphoma, LCH, and Ewing sarcoma. When a patient presents with bone RDD without lymph node involvement, it is important to keep the rare disease aspect of RDD in mind, otherwise it can be challenging to make the correct diagnosis. Our review of the literature found 2 reports in adult patients with isolated osseous RDD (Table 2). Both cases were misdiagnosed as chronic inflammation upon clinical, radiographic, and histologic examination [14,15]. As in our case, isolated osseous RDD was only diagnosed on the second biopsy. Clinical, radiological, and histological features of these 2 cases are compared with our case in Table 2.

Among the rare presentations of RDD exclusively involving bone, only 15 pediatric cases have been reported in the existing literature [12]. The age range of these pediatric patients was broad, although most cases involved children 10 years of age or older. Reports of young patients with isolated osseous RDD are exceedingly rare. We only found 2 reported osseous RDD cases in children 2 years old or younger. In 1998, it was reported that a 2-year-old child presented with bone lesions in the left cuboid and right calcaneus. An excisional biopsy of cuboid revealed osseous RDD, but further work-up discovered hilar lymph node involvement [16]. Similar to our case, a case of a 17-month-old boy with talus RDD without lymph node involvement was recently described [17]. The diagnosis was made on the curettage specimen and the bone healed 1 year after the surgery. Our case is likely the second published case of RDD with isolated bone involvement in a child younger than 2 years of age. Table 3 compares clinical and pathologic features of these 2 cases to our case. The addition of this case to the existing literature will raise the awareness of RDD with unusual clinical presentation in young patients. Clinicians of all specialties should keep extranodal RDD in mind to make the correct diagnosis in managing these patients, especially in the pediatric population.

Conclusions

Primary osseous RDD in children is exceedingly rare and can present a diagnostic challenge for clinicians. Clinically, radiographically, and histologically, osseous RDD can mimic common bone diseases in children, such as osteomyelitis, LCH, and Ewing sarcoma. Our case serves as a reminder that despite the diagnostic challenge, keeping RDD in the differential diagnosis, especially in young patients, can avoid misdiagnosis and mismanagement. A better understanding of the pathophysiology of RDD could improve the accuracy of RDD diagnosis.