Eosinophilic Granuloma (Langerhans Cell Histiocytosis) 

Introduction and Nosology 

Langerhans Cell Histiocytosis (LCH) is a rare, heterogenous illness characterized by the proliferation of dendritic cells with Langerhans cell morphology. “LCH” refers to a spectrum of disease, from a localized lesion to diffuse multiorgan pathology. [1-3] LCH can affect any organ system; the most commonly affected are the skeletal system (80%), skin (33%), and pituitary (25%); others include the liver, spleen, lungs, and brain.[4] The most common named subtype, Eosinophilc Granuloma (EG), refers to a benign, localized LCH, most commonly of bone [2] Other named subtypes include Hand-Schuller-Christian disease, a multifocal LCH classically characterized by exophthalmos, diabetes insipidus, and osteolytic skull lesions and Letterer-Siwe disease, a diffuse systemic LCH clinically manifested by skin rash, hepatosplenomegaly, and pancytopenia. [5, 6] In 1953, Lichtenstein grouped all 3 of these diseases under the name “histiocytosis X”, a term which has since been replaced by Langerhans Cell Histiocytosis. [5, 7]

Epidemiology and Clinical Presentation 

Eosinophlic Granuloma (EG) that affects the bone is the most common subtype of LCH, representing an estimated 60-80% of cases; it can be single or multifocal and most commonly affects the calvarium but can also present in the vertebrae, ribs, long bones, and mandible [6, 8, 9] EG primarily affects children under the age of 15 with an estimated incidence of less than 1 per 100,000.[1, 3]

Clinically, EG presents as a painful soft tissue mass; radiography demonstrates sharply demarcated osteolytic lesions of the underlying bone. [2, 8] EG of the skull presents as a gradually enlarging scalp mass.[2] EG of the vertebral body in children most commonly presents with pain; neurologic deficit is uncommon even in cases of progression to vertebra plana. [10] [11] EG of the orbital bone may present as proptosis with an accompanying mass mimicking malignancy. [12] EG may even occur in the brain parenchyma with a tumor-like appearance; these lesions most commonly appear in the hypothalamic-pituitary axis and present with endocrinopathies. [13] Radiographically, LCH lesions are avidly contrast enhancing on both CT and MR, and demonstrate increased uptake in FDG-PET. [14] They appear hypointense on T1 MRI and iso to hyperintense on T2. [3, 15]

Etiology and Pathophysiology 

LCH is characterized by the proliferation of a myeloid-derived precursor dendritic cell with the characteristics of Langerhans cells (LC). [16] [17] One theory of the pathophysiology of LCH is based on recently published data that LCH cells express CD40 at high levels, leading to the overactivation of CD40L+ T-lymphocytes. The subsequent release of cytokines leads to further recruitment of LC progenitor cells, as well as to local destruction of bone, fibrosis, and necrosis. [18-20] 

The central debate of the pathogenesis of LCH is whether it is a reactive immune response or a neoplasm. Support for the neoplastic theory include the monoclonality of the pathologic cells (although this could also be seen with reactive immune processes) [21] [20], the up regulation of pro-cell proliferation and anti-apoptotic proteins (c-myc, h-ras, Bcl-2) in LCH samples, [22] the up regulation of Ki-67 (a marker of proliferation), [22] [23] the finding of activating BRAF mutations in a majority of samples and the clinical response of BRAF V600E+ disease to vemurafenib. [24] [25] Support for the theory that LCH represents a dysfunctional immune response include data showing the up-regulation of genes that lead to T-cell activation and recruitment in LCH lesions, [26] an increase of T-regulatory cells in the blood of LCH patients in comparison with controls, [27], and a lack of evidence of gross genomic or karyotypic aberration in LCH biopsy samples. [28] Of note, no antigenic trigger has ever been identified and recent evidence has mounted against a viral etiology. [29, 30] 

Workup & Diagnosis 

The differential for an osteolytic soft tissue mass is broad, and includes neoplastic processes: metastatic lesions, primary bone tumors such as osteosarcoma or Ewing’s sarcoma, neuroblastoma, rhabdomyosarcoma, lymphoma, PNET. Infections such as osteomyelitis or abscess. Fibrous dysplasia, cystic lesions (aneurysmal bone cyst, dermoid cyst), vascular processes (venous lakes, hemangioma, angiomatosis), and developmental anomalies (encephalocele). 

Tissue biopsy is necessary for definitive diagnosis: Histopathologically, LCH lesions show a proliferation of LC-type cells in a milieu of lymphocytes, macrophages, and eosinophils. In 1987, the histiocyte society established diagnostic criteria for LCH: the identification of CD1a on immunohistochemistry or of Birbeck granules on electron microscopy. [5] Immunhistochemistry for Langerin, a protein enriched in and necessary for the formation of Birbeck granules, is an acceptable modern alternative to electron microscopy. [4, 20] Once mysterious, recent work suggests Birbeck granules are part of the eadosomal recycling system, and may be involved with loading CD1a (a protein similar to MHC-I) with glycoprotein antigens for presentation to T-cells. [31] 

Workup should include a systemic survey to identify any other potential sites of involvement, as management recommendations and prognosis vary depending on the number and type of organs involved in LCH.[32] The Euro Histio network published guidelines for the workup of LCH in 2013 based on the available review of the literature, and recommend the following labs: CBC with differential, CMP (electrolytes, BUN/creatinine and liver function tests), ESR, coagulation studies (PT/INR, PTT, fibrinogen), chest x-ray and radiographic skeletal survey. [4] The histiocyte society recommends the same labs, with the addition of urine specific gravity and abdominal ultrasound.[33] Endocrine labs (FSH/LH, TSH, GH, cortisol) are indicated if there is any suspicion of pituitary involvement. MRI is useful for the detection of bone marrow or soft tissue involvement, and a fast whole-body T2 STIR protocol is one possible screening test. [34, 35] PET has been shown to be superior to bone scans for lesion detection, but exposes children to much higher radiation doses [35]. A recent study by Mueller et al found MRI had a higher sensitivity (81%) but FDG-PET had a higher specificity (76%) for the detection of LCH lesions. [36] 

Treatment

Solitary EG of the calvarial vault without invasion into neurological structures has a good prognosis. Historically managed with surgical curettage or excision, [4] [17] [37] recent reports by Oliveira et al (n=4) and by De Angulo et al (n=8) found that these lesions fully resolved without intervention after 6-19 months. [17] [38] The major limitation of these studies is the lack of diagnostic certainty in the absence of tissue analysis. There is no class I evidence for the management of these lesions, and treatment decisions must be made on a case by case basis. On the one hand, overly aggressive surgical excision may be associated with prolonged healing and deformity, particularly with large lesions; the European Histiocyte Society recommends avoidance of surgical excision in lesions > 5 cm in children. [4] On the other hand, observation alone may be a risky decision, as the differential diagnosis of LCH includes aggressive cancers and infection, and the disease may progress to pathologic fracture or encroachment on neurological structures. [39] One middle ground option is to perform a biopsy or partial resection for definitive diagnosis, followed by observation or a conservative treatment option- such as intralesional injection of steroids or inteferon, or systemic treatments with indomethacin or bisphosphonates. These treatments have been shown to be efficacious in the literature, [40] [41] [42] however the results must be interpreted in the context of a disease that is potentially self-resolving. 

The histiocyte society recommends more aggressive treatment in cases of multifocal bone disease or disease that involves “CNS-risk” sites (odontoid peg, vertebrae with intraspinal soft tissue extension, facial bones, skull base, orbit, oral cavity). [20] [33] In these types of LCH, the risk of recurrence is high (30-50%) as is the risk of disease invasion into neurological tissue or the development of neurological sequeale (40%) such as endocriniopathies, diabetes insipidus, and parenchymal brain disease. [33] For LCH of this type, the histiocyte society recommends systemic treatment with prednisone and vinblastine for 12 months. [33] There is no class I evidence or guideline recommendations regarding surgical treatment for these lesions. Historically, lesions that invade the dura or compress neurological tissues and that are located in accessible sites have undergone surgical excision. [6, 43] [44] In cases where surgical excision would be impossible or excessively morbid, systemic chemotherapy and/or injections of steroids or interferon may be a preferential alternative. [39] In the past, radiation therapy was used as an adjunctive treatment, however it is no longer recommended except in emergent cases of compression of critical neurological structures, such as the optic nerve or spinal cord, out of concern for the possible effects of radiotherapy on the developing brain and spine. [4] [20] 

LCH of the spine is rare. Treatment patterns in the literature have varied from observation to complete surgical excision with fixation to radiotherapy. Bertram et al in a review of the literature of spine EG (n=53) found that most cases resolved without treatment, and recommend immobilization and observation only in cases without spinal instability or neurological deficit. [10]  Similar findings have been reported by Raab et al (n=14) and Yeom et al (n=23). [45, 46] LCH spares the endochondral plates, and so recovery of vertebral heights is possible if the endochondral tissues are not disturbed with surgery or radiation. [45] [37] Surgical excision, followed by segmental fusion and internal fixation, is indicated in cases of spinal instability, neurological deficit from compression of the spinal cord or spinal nerves, or in patients unable or unwilling to comply with prolonged immobilization or external bracing. [37] As in cranial lesions, radiotherapy is no longer recommended for spine LCH except in emergent cases of spinal cord compression. [4] [20] 

Prognosis 

The prognosis for LCH is good. In a large study from South Korea (n=603), 5-year overall survival in those with single-system LCH was 99.8%, for multisystem LCH without risk-organ involvement was 98.4%, for multisystem LCH with risk-organ involvement was 77%. [47] Similar findings have been found by other groups in Japan (n=91) and Italy (n=121). [48] [49] In a large study from the Mayo Clinic (n=263), recurrence rates in those with single-bone LCH was 7%, and the appearance of new bone lesions was 14%. [44] As aforementioned, the Histiocyte Society reports a high rate of recurrence (30-50%) among those with “CNS-risk” lesions. [33]