Langerhans Cell Histiocytosis Treatment (PDQ®)
Last modified: 2016-09-23
Last downloaded: 2016-10-25
General Information About Langerhans Cell Histiocytosis (LCH)
The histiocytic diseases in children and adults include three major classes of disorders. Only Langerhans cell histiocytosis (LCH), a dendritic cell disorder, is discussed in detail in this summary. Erdheim-Chester disease (primarily found in adults) and juvenile xanthogranuloma (diagnosed in children and adults) are macrophage disorders. Other disorders of the macrophage/monocytoid lineages include Rosai-Dorfman disease and hemophagocytic lymphohistiocytosis. Malignant disorders include malignant histiocytosis of various histiocyte lineages (formerly called histiocytic sarcoma) and the monocytic or myelomonocytic leukemias.
LCH results from the clonal proliferation of immunophenotypically and functionally immature, morphologically rounded LCH cells along with eosinophils, macrophages, lymphocytes, and occasionally, multinucleated giant cells. The term LCH cells is used because there are clear morphologic, phenotypic, and gene expression differences between Langerhans cells of the epidermis (LCs) and those in LCH lesions (LCH cells). Controversy exists regarding whether the clonal proliferation of LCH cells results from a malignant transformation or is the result of an immunologic stimulus.
The recent discovery that approximately 60% of LCH biopsy specimens demonstrate the V600E mutation in the BRAF oncogene, regardless of stage or organ involvement, has led to the conclusion that LCH is a clonal neoplastic disorder. The same mutation has been found in other cancers, including malignant melanoma; however, V600E-mutated BRAF is also present in benign nevi, possibly indicating the need for additional mutations to render the cell malignant. This finding has raised the possibility of future targeted therapy with inhibitors already in use in melanoma, and several trials of BRAF inhibitors are open in adults and children with BRAF V600E mutated tumors, including LCH. Regardless of having a BRAF V600E mutation, nearly all lesions have been reported to show evidence of activated ERK downstream of BRAF; therefore, other mutations in genes that are part of the RAS-RAF-MEK-ERK pathway might also be identified. This has been shown with activating mutations that involve the CSF-1 receptor, RAS, and MAP2K1 (MEK) for a significant percentage of BRAF V600E–negative specimens. Regardless of the etiology of the clonal proliferation, the primary treatment is chemotherapy.
Langerhans cell histiocytosis is the terminology currently preferred over histiocytosis X, eosinophilic granuloma, Abt-Letterer-Siwe disease, Hand-Schuller-Christian disease, or diffuse reticuloendotheliosis. This is based on the observation that the pathologic histiocyte common to all of these diagnoses has the identical immunophenotypic characteristics including the presence of Birbeck granules identified by electron microscopy; in addition, the pathologic histiocyte or LCH cell has a gene expression profile more closely resembling a myeloid dendritic cell, raising the possibility that LCH cells arise from a circulating precursor cell rather than the skin LC. (Refer to the Cytogenetic and Genomic Studies section of this summary for more information.)
LCH may involve a single organ (single-system LCH), which may be a single site (unifocal) or involve multiple sites (multifocal); or LCH may involve multiple organs (multisystem LCH), which may involve a limited number of organs or be disseminated.
Back to Top
Children and adolescents with Langerhans cell histiocytosis (LCH) should be treated by a multidisciplinary team of health professionals who are experienced with this disease and its treatment. This multidisciplinary team approach incorporates the skills of the primary care physician, pediatric surgical subspecialists, radiation oncologists, pediatric medical oncologists/hematologists, rehabilitation specialists, pediatric nurse specialists, social workers, and others to ensure that children receive treatment, supportive care, and rehabilitation that will achieve optimal survival and quality of life.
Clinical trials organized by the Histiocyte Society have been accruing patients on childhood treatment studies since the 1980s. Information on centers enrolling patients on these trials can be found on the NCI website.
Children with high-risk or low-risk disease should be followed annually to document and attempt to correct adverse side effects of therapy or the disease. (Refer to the Late Disease and Treatment Effects of Childhood LCH section of this summary for more information about the incidence, type, and monitoring of late effects of childhood cancer and its therapy.)
The annual incidence of LCH has been estimated to be two to ten cases per 1 million children aged 15 years or younger. The male to female ratio (M/F) is close to one and the median age of presentation is 30 months. A report from Stockholm County, Sweden, described an annual incidence of 8.9 cases of LCH per 1 million children with a total of 29 cases in 10 years. Most of these cases were diagnosed between September and February (M/F = 1.2). A 4-year survey of 251 new LCH cases in France found an annual incidence of 4.6 cases per 1 million children younger than 15 years (M/F = 1.2). A survey of LCH in northwest England (Manchester) revealed an overall incidence of 2.6 cases per 1 million child-years.
Surveillance, Epidemiology, and End Results registry data from 2000 to 2009 were reviewed to identify high-risk LCH cases and assess various demographic variables. On the basis of 145 cases, the age-standardized incidence was 0.7 per 1 million children per year, with lower incidence in black patients (0.41 per 1 million) and higher incidence in Hispanic patients (1.63 per 1 million) younger than 5 years. Living in crowded conditions and lower socioeconomic circumstances were associated with a higher risk of LCH.
Identical twins with LCH, and non-twin siblings or multiple cases in one family, have been reported. Over 90% of adult pulmonary LCH occurs in young adults who smoke, often more than 20 cigarettes per day.
Although the following risk factors have been identified for LCH, strong and consistent associations have not been confirmed:
- Solvent exposure in parents.
- Family history of cancer.
- Personal or family history of thyroid disease.
- Perinatal infections.
- Parental occupational exposure to metal, granite, or wood dust.
- Ethnicity and race.
- Low socioeconomic status.
Prognosis is closely linked to the extent of disease at presentation when high-risk organs (liver, spleen, and/or bone marrow) are involved and to the response to initial treatment. The high-risk designation comes from the high mortality rate (35%) in those who did not respond well to therapy in the first 6 weeks. For many years, lung was thought to be a high-risk organ but isolated lung involvement in pediatric LCH is no longer considered to pose a significant risk of death. Because of treatment advances, the outcome for children with LCH involving high-risk organs has improved. Data from HISTSOC-LCH-III (NCT00276757) showed an 84% overall survival (OS) rate for patients treated for 12 months with systemic chemotherapy.
Patients with single-system disease and low-risk multisystem disease do not usually die from LCH, but recurrent disease may result in considerable morbidity and significant late effects. The major treatment challenge is to reduce the 20% to 30% incidence of recurrent lesions and the significant incidence of permanent consequences in this group of patients. HISTSOC-LCH-III data showed that there was a significant difference in reactivation for risk organ subjects when comparing 6 months versus 12 months of treatment, with 12 months being better (54% vs. 37%).
Prognostic factors for children with LCH have been identified and include the following:
- BRAF mutation: A study of 173 patients with the BRAF V600E mutation, and 142 without the mutation, revealed that the mutation occurred in 88% of patients with high-risk disease, 69% of patients with multisystem low-risk LCH, and 44% of patients with single-system low-risk LCH. The mutation was also found in 75% of patients with neurodegenerative syndrome and 73% of patients with pituitary involvement. Resistance to initial treatment and relapse were higher in patients with the mutation.
- Age at diagnosis: Although age younger than 2 years was once thought to portend a worse prognosis, data from the LCH-II study showed that patients aged 2 years or younger without high-risk organ involvement had the same response to therapy as older patients. By contrast, the OS was poorer in neonates with risk-organ involvement compared with infants and children with the same extent of disease when patients were treated for only 6 months.
- Response to treatment: Response to therapy at 6 to 12 weeks has been shown to be a more important prognostic factor than age. The overall response to therapy is influenced by the duration and intensity of treatment.
- Organ involvement: Involvement of craniofacial bones including orbital, mastoid, and temporal bones is associated with an increased risk of diabetes insipidus and an increased frequency of anterior pituitary hormone deficiencies and neurologic problems. (Refer to the Endocrine system subsection in the Multisystem Disease Presentation section of this summary for more information on diabetes insipidus.)
Back to Top
Histopathologic, Immunologic, and Cytogenetic Characteristics of LCH
Cell of Origin and Biologic Correlates
Modern classification of the histiocytic diseases subdivides them into dendritic cell–related, monocyte/macrophage-related, or true malignancies. Langerhans cell histiocytosis (LCH) is a dendritic cell disease. The Langerhans cells in LCH lesions (LCH cells) are immature dendritic cells making up less than 10% of the cells present in LCH lesions. Comprehensive gene expression array data analysis on LCH cells is consistent with the concept that the skin Langerhans cell (LC) is not the cell of origin for LCH. Rather it is likely to be a myeloid dendritic cell, which expresses the same antigens (CD1a and CD207) as the skin LC. This concept was further supported by a study reporting that the transcription profile of LCH cells was distinct from myeloid and plasmacytoid dendritic cells, as well as epidermal LCs.
LCH lesions also contain lymphocytes, macrophages, neutrophils, eosinophils, fibroblasts, and sometimes multinucleated giant cells. In the brain, the following three types of histopathologic findings have been described in LCH:
- Mass lesions in meninges or choroid plexus with CD1a-positive LCH cells and predominantly CD8-positive lymphocytes.
- Mass lesions in connective tissue spaces with CD1a-positive LCH cells and predominantly CD8-positive lymphocytes causing an inflammatory response and neuronal loss.
- Predominantly CD8+ lymphocyte infiltration with neuronal degeneration, microglial activation, and gliosis.
Normally, the LC is a primary presenter of antigen to naïve T-lymphocytes. However, in LCH, the pathologic dendritic cell does not efficiently stimulate primary T-lymphocyte responses. Antibody staining for the dendritic cell markers, CD80, CD86, and class II antigens, has been used to show that in LCH, the abnormal cells are immature dendritic cells that present antigen poorly and are proliferating at a low rate. Transforming growth factor-beta (TGF-beta) and interleukin (IL)-10 are possibly responsible for preventing LCH cell maturation in LCH. The expansion of regulatory T cells in patients with LCH has been reported. The population of CD4-positive CD25(high) FoxP3(high) cells was reported to comprise 20% of T cells and appeared to be in contact with LCH cells in the lesions. These T cells were present in higher numbers in the peripheral blood of patients with LCH than in controls and returned to a normal level when patients were in remission.
The etiology of LCH is unknown. Efforts to define a viral cause have not been successful. One study has shown that 1% of patients have a positive family history for LCH.
Cytogenetic and Genomic Studies
Studies published in 1994 showed clonality in Langerhans cell histiocytosis (LCH) using polymorphisms of methylation-specific restriction enzyme sites on the X-chromosome regions coding for the human androgen receptor, DXS255, PGK, and HPRT. Biopsies of lesions with single-system or multisystem disease were found to have a proliferation of LCH cells from a single clone. The discovery of recurring genomic alterations (primarily BRAF V600E) in LCH (see below) confirmed the clonality of LCH in children. Pulmonary LCH in adults is usually nonclonal and it is possible that this group represents a reactive process to smoking. However, a subset appeared to be clonal, as an analysis of BRAF mutations showed that a significant proportion of patients (25%–30%) have evidence for mutant BRAF V600E.
Figure 1. Courtesy of Rikhia Chakraborty, Ph.D. Permission to reuse the figure in any form must be obtained directly from Dr. Chakraborty.
The genomic basis of LCH was advanced by a report in 2010 of an activating mutation of the BRAF oncogene (V600E) that was detected in 35 (57%) of 61 cases. Multiple subsequent reports have confirmed the presence of BRAF V600E mutations in 50% or more of LCH cases in children. Another BRAF mutation (BRAF 600DLAT) was identified, which resulted in the insertion of four amino acids and also appeared to activate signaling.ARAF mutations are infrequent in LCH, but when present, can also lead to RAS-MAPK pathway activation. No clinical characteristics associated with the BRAF V600E mutation have been identified.
The RAS-MAPK signaling pathway (Figure 1) transmits signals from a cell surface receptor (e.g., a growth factor) through the RAS pathway (via one of the RAF proteins [A, B, or C]) to phosphorylate MEK and then the extracellular signal-regulated kinase (ERK), which leads to nuclear signals affecting cell cycle and transcription regulation. The V600E mutation of BRAF leads to continuous phosphorylation, and thus activation, of MEK and ERK without the need for an external signal. Activation of ERK occurs by phosphorylation, and phosphorylated ERK can be detected in virtually all LCH lesions.
Because RAS-MAPK pathway activation can be detected in all LCH cases, but not all cases have BRAF mutations, the presence of genomic alterations in other components of the pathway was suspected. Whole-exome sequencing of BRAF-mutated versus BRAF-wild type LCH biopsies revealed that 7 of 21 BRAF wild-type specimens had MAP2K1 mutations, while no BRAF–mutated specimens had MAP2K1 mutations. The mutations in MAP2K1 (which codes for MEK) were activating, as indicated by their induction of ERK phosphorylation. Another study showed MAP2K1 mutations exclusively in 11 of 22 BRAF wild-type cases. Studies to date support the universal activation of ERK in LCH, with activation in most cases being explained by BRAF and MAP2K1 mutations.
The presence of BRAF V600E mutation in blood and bone marrow was studied in a series of 100 patients, of which 65% tested positive for the BRAF V600E mutation by a sensitive quantitative polymerase chain reaction technique. Circulating cells with the BRAF V600E mutation could be detected in all high-risk patients and in a subset of low-risk multisystem patients. The presence of circulating cells with the mutation conferred a twofold increased risk of relapse. The myeloid dendritic cell origin of LCH was confirmed by finding CD34+ stem cells with the mutation in the bone marrow of high-risk patients. Those with low-risk disease had more mature myeloid dendritic cells with the mutation, suggesting the stage of cell development is critical in defining the clinical characteristics of LCH, which can now be considered a myeloid neoplasia in most cases.
A study of 173 patients with the BRAF V600E mutation, and 142 patients without the mutation, revealed that the mutation occurred in 88% of patients with high-risk disease, 69% of patients with multisystem low-risk LCH, and 44% of patients with single-system low-risk LCH. The mutation was also found in 75% of patients with neurodegenerative syndrome and 73% of patients with pituitary involvement. Resistance to initial treatment and relapse were higher in patients with the mutation.
Clinical implications of the described genomic findings include the following:
- LCH joins a group of other pediatric entities with activating BRAF mutations, including select nonmalignant conditions (e.g., benign nevi) and low-grade malignancies (e.g., pilocytic astrocytoma). All of these conditions have a generally indolent course, with spontaneous resolution occurring in some cases. This distinctive clinical course may be a manifestation of oncogene-induced senescence.
- BRAF V600E mutations can be targeted by BRAF inhibitors (e.g., vemurafenib and dabrafenib) or by the combination of BRAF inhibitors plus MEK inhibitors (e.g., dabrafenib/trametinib and vemurafenib/cobimetinib). These agents and combinations are approved for adults with melanoma. Treatment of adults with combinations of a BRAF inhibitor and a MEK inhibitor showed significantly improved progression-free survival outcome compared with treatment using a BRAF inhibitor alone. The most serious side effect of BRAF inhibitors is the induction of cutaneous squamous cell carcinomas, with the incidence of these second cancers increasing with age; reduction of this side effect can occur with concurrent treatment with both BRAF and MEK inhibitors. Case reports have described activity of BRAF inhibitors against LCH in adult and pediatric patients, but there are insufficient data to assess the role of these agents in treatment of children with LCH.
- With further research, the observation of BRAF V600E (or potentially mutated MAP2K1) in circulating cells may become a useful diagnostic tool to define high-risk versus low-risk disease. Additionally, for patients who have a somatic mutation, persistence of circulating cells with the mutation may be useful as a marker of residual disease.
Cytokine Analysis by Immunohistochemical Staining and Gene Expression Array Studies
Immunohistochemical staining of LCH lesions has shown apparent upregulation of the chemokines CCR6 and possibly CCR7. In an analysis of gene expression in LCH by gene array techniques, 2,000 differentially expressed genes were identified. Of 65 genes previously reported to be associated with LCH, only 11 were found to be upregulated in the array results. The most highly upregulated gene in both CD207 and CD3-positive cells was osteopontin; other genes that activate and recruit T cells to sites of inflammation are also upregulated. The expression profile of the T cells was that of an activated regulatory T-cell phenotype with increased expression of FOXP3, CTLA4, and osteopontin. These findings support a previous report on the expansion of regulatory T cells in LCH. There was pronounced expression of genes associated with early myeloid progenitors including CD33 and CD44, which is consistent with an earlier report of elevated myeloid dendritic cells in the blood of patients with LCH. A model of "Misguided Myeloid Dendritic Cell Precursors" has been proposed, whereby myeloid dendritic cell precursors are recruited to sites of LCH by an unknown mechanism and the dendritic cells in turn recruit lymphocytes by excretion of osteopontin, neuropilin-1, and vannin-1.
Several investigators have published studies evaluating the level of various cytokines or growth factors in the blood of patients with LCH that have included many of the genes found not to be upregulated by the gene expression results discussed above. One explanation for elevated levels of these proteins is a systemic inflammatory response with the cytokines/growth factors being produced by cells outside the LCH lesions. A second possible explanation is that macrophages in the LCH lesions produce the cytokines measured in the blood or are concentrated in lesions.
IL-1 beta and prostaglandin GE2 levels were measured in the saliva of patients with oral LCH lesions or multisystem high-risk patients with and without oral lesions; levels of both were higher in patients with active disease and decreased after successful therapy.
Human Leukocyte Antigen (HLA) Type and Association With LCH
Specific associations of LCH with distinct HLA types and extent of disease have been reported. In a study of 84 Nordic patients, those with only skin or bone involvement more frequently had HLA-DRB1*03 type than those with multisystem disease. In 29 patients and 37 family members in the United States, the Cw7 and DR4 types were significantly more prevalent in Caucasians with single-bone lesions.
Back to Top
Presentation of LCH in Children
Langerhans cell histiocytosis (LCH) most commonly presents with a skin rash or a painful bone lesion. Systemic symptoms of fever, weight loss, diarrhea, edema, dyspnea, polydipsia, and polyuria, relate to specific organ involvement and single-system or multisystem disease presentation as noted below.
Specific organs are considered high-risk or low-risk when involved with disease presentation. Risk refers to the risk of mortality.
- High-risk organs include liver, spleen, and bone marrow.
- Low-risk organs include skin, bone, lung, lymph nodes, gastrointestinal tract, pituitary gland, and central nervous system (CNS).
Patients may present with a single organ (single-system LCH), which may involve a single site (unifocal) or multiple sites (multifocal). Bone is the most common single organ site. Less commonly, LCH may involve multiple organs (multisystem LCH), which may involve a limited number of organs or it may be disseminated. Treatment decisions for patients are based upon whether high-risk or low-risk organs are involved and whether LCH presents as single-system or multisystem disease. Patients can have LCH of the skin, bone, lymph nodes, and pituitary in any combination and still be considered at low-risk of death, although there may be relatively high-risk for long-term consequences of the disease.
Single-System Disease Presentation
In single-system LCH, as the name implies, the disease presents with involvement of a single site or organ, including skin and nails, oral cavity, bone, lymph nodes and thymus, pituitary gland, and thyroid.
Skin and nails
Infants: Seborrheic involvement of the scalp may be mistaken for prolonged cradle cap in infants. Infants with LCH may also present with a generalized skin rash, which may mimic many other skin disorders. Skin LCH in infants may be limited to skin (skin-only disease) or may be part of multisystem LCH. In a report of 61 neonatal cases from 1,069 patients in the Histiocyte Society database, nearly 60% had multisystem disease and 72% had risk-organ involvement. A retrospective analysis of 71 infants and children found that those older than 18 months were more likely to have multisystem involvement and often relapsed after treatment with vinblastine and prednisone. Eight of 11 patients in this category had circulating cells with the BRAF V600E mutation, compared with only 1 of 13 patients in the skin-only group. Patients younger than 1 year with skin-only disease, completely evaluated to exclude any other site of disease, had an 89% 3-year progression-free survival with initial therapy.
Skin-only LCH, which had historically been known as Hashimoto-Pritzer, may be self-limited as the lesions may disappear with no therapy during the first year of life. Therapy is used only for very extensive rashes, pain, ulceration, or bleeding. Importantly, these patients must be watched closely as skin-only LCH may also progress within weeks or months to high-risk multisystem disease, which may be life-threatening.
A review of patients presenting in the first 3 months of life with skin-only LCH compared the clinical and histopathologic findings in 21 children whose skin LCH regressed with ten children who did not regress. Patients with regressing disease had distal lesions that appeared in the first 3 months of life and were necrotic papules or hypopigmented macules. Patients with nonregressing disease who required systemic therapy were more often intertriginous. Immunohistochemical studies showed no difference in interleukin (IL)-10, Ki-67, E-cadherin expression, or T-reg number between the two clinical groups.
Children and adults: Children and adults may develop a red papular rash in the groin, abdomen, back, or chest that resembles a diffuse candidal rash. Seborrheic involvement of the scalp may be mistaken for a severe case of dandruff in older individuals. Ulcerative lesions behind the ears, involving the scalp, under the breasts, or genitalia or perianal region are often misdiagnosed as bacterial or fungal infections. Vesicular lesions may be seen and need to be differentiated from herpetic lesions.
Fingernail involvement is an unusual finding that may present as a single site or with other sites of LCH involvement. There are longitudinal, discolored grooves and loss of nail tissue. This condition often responds to the usual LCH therapies.
In the mouth, presenting symptoms include gingival hypertrophy and ulcers of the soft or hard palate, buccal mucosa, or on the tongue and lips. Hypermobile teeth (floating teeth) and tooth loss usually indicate involvement of underlying bone. Lesions of the oral cavity may precede evidence of LCH elsewhere.
LCH can occur in any bone of the body, although the hands and feet are often spared. Sites of LCH in children include the following:
- Lytic lesion of the skull: The most frequent site of LCH in children is a lytic lesion of the skull vault, which may be asymptomatic or painful. It is often surrounded by a soft tissue mass that may extend internally to impinge on the dura.
- Femur, ribs, humerus, and vertebra: Other frequently involved skeletal sites are femur, ribs, humerus, and vertebra. Spine lesions may involve any vertebra, although involvement of the cervical vertebrae is most common and spine lesions are frequently associated with other bone lesions. Spine lesions may result in collapse of the vertebral body (vertebra plana). Vertebral lesions with soft tissue extension often present with pain and may present with significant neurologic deficits, an indication for an urgent magnetic resonance imaging (MRI) scan.
CNS-risk bones: Proptosis from an LCH mass in the orbit mimics rhabdomyosarcomas, neuroblastoma, and benign fatty tumors of the eye.
Lesions of the facial bones or anterior or middle cranial fossae (e.g., temporal, orbit, sphenoid, ethmoid, zygomatic) with intracranial tumor extension comprise a CNS-risk group. These patients have a threefold increased risk of developing diabetes insipidus and other CNS disease. Because of the increased risk of diabetes insipidus, treatment is recommended for these patients.
Lymph nodes and thymus
The cervical nodes are most frequently involved and may be soft- or hard-matted groups with accompanying lymphedema. An enlarged thymus or mediastinal node involvement can mimic an infectious process and may cause asthma-like symptoms. Accordingly, biopsy with culture is indicated for these presentations. Mediastinal involvement is rare (<5%) and usually presents with respiratory distress, superior vena cava syndrome, or cough and tachypnea. The 5-year survival is 87%, with deaths mostly attributable to hematologic involvement.
The posterior part of the pituitary gland and pituitary stalk can be affected in patients with LCH, causing central diabetes insipidus. (Refer to the Endocrine subsection in the Multisystem Disease Presentation section of this summary for more information.) Anterior pituitary involvement often results in growth failure and delayed or precocious puberty. Rarely, hypothalamic involvement may cause morbid obesity.
Thyroid involvement has been reported in LCH. Symptoms include massive thyroid enlargement, hypothyroidism, and respiratory symptoms.
Multisystem Disease Presentation
In multisystem LCH, the disease presents in multiple organs or body systems including bone, abdominal/gastrointestinal system (liver and spleen), lung, bone marrow, endocrine system, eye, CNS, skin, and lymph nodes.
Bone and other organ systems
Patients with LCH may present with multiple bone lesions as a single site (single-system multifocal bone) or bone lesions with other organ systems involved (multisystem including bone). A review of patients with single-system multifocal bone presentation and patients with multisystem including bone presentation who were treated on the Japanese LCH study (JLSG-02) found that patients in the multisystem including bone group were more likely to have lesions in the temporal bone, mastoid/petrous bone, orbit, and zygomatic bone (CNS risk). Patients with multisystem including bone presentation had a higher incidence of diabetes insipidus, correlating with the higher frequency of lesions in the noted facial bones. There was no difference in the outcome of treatment, which was more intense in the JLSG-02 study than in the LCH-II study.
In LCH, the liver and spleen are considered high-risk organs, and involvement of these organs affects prognosis. Involvement in this context means the liver and spleen are enlarged from direct infiltration of LCH cells or as a secondary phenomenon of excess cytokines, which cause macrophage activation or infiltration of lymphocytes around bile ducts. LCH cells have a portal (bile duct) tropism that may lead to biliary damage and ductal sclerosis. A percutaneous (peripheral) liver biopsy may not be diagnostic of the infiltrate that tends to be more central in the liver, but will show the upstream obstructive effects of distal biliary occlusion. Hepatic enlargement can be accompanied by dysfunction, leading to hypoalbuminemia with ascites, hyperbilirubinemia, and clotting factor deficiencies. Sonography, computed tomography (CT), or MRI of the liver will show hypoechoic or low-signal intensity along the portal veins or biliary tracts when the liver is involved with LCH.
Liver (sclerosing cholangitis)
One of the most serious complications of hepatic LCH is cholestasis and sclerosing cholangitis. This usually occurs months after initial presentation, but on occasion may be present at diagnosis. The median age of children with this form of hepatic LCH is 23 months.
Patients with hepatic LCH present with hepatomegaly or hepatosplenomegaly, and elevated alkaline phosphatase, liver transaminases, and gamma glutamyl transpeptidase levels. While ultrasound and/or MRI-cholangiogram can be helpful in the diagnosis of this complication, liver biopsy is currently the only definitive way to determine whether active LCH or hepatic fibrosis is present. Biopsy results often show lymphocytes and biliary obstructive effects without LCH cells. Peribiliary LCH cells and, rarely, nodular masses of LCH, may also be present. It is thought that cytokines, such as transforming growth factor-beta (TGF)-beta, elaborated by lymphocytes during the active phase of the disease, leads to fibrosis and sclerosis around the bile ducts.
Seventy-five percent of children with sclerosing cholangitis will not respond to chemotherapy because the LCH is no longer active, but the fibrosis and sclerosis remain. Despite the limitations, liver biopsy may be the only way to distinguish active LCH from end-stage fibrosis. Liver transplantation is the only alternate treatment when hepatic function worsens. In one series of 28 children undergoing liver transplantation, 78% survived and 29% had recurrence of LCH, but only two cases of recurrent LCH occurred in the transplanted liver, although other cases have been reported since the initial data was published. If at all possible, active LCH should be under control before transplant. The patients who undergo liver transplant for LCH may have a higher incidence of posttransplant lymphoproliferative disease.
Massive splenomegaly may lead to cytopenias because of hypersplenism and may cause respiratory compromise. Splenectomy typically provides only transient relief of cytopenias, as increased liver size and reticuloendothelial activation result in peripheral blood cell sequestration and destruction. Although rare, LCH infiltration of the pancreas and kidneys has been reported. Splenectomy is performed only as a life-saving measure.
Other gastrointestinal manifestations
Patients with diarrhea, hematochezia, perianal fistulas, or malabsorption have been reported. Diagnosing gastrointestinal involvement with LCH is difficult because of patchy involvement. Careful endoscopic examination including multiple biopsies is usually needed.
In LCH, the lung is less frequently involved in children than in adults, because smoking in adults is a key etiologic factor. The cystic/nodular pattern of disease reflects the cytokine-induced destruction of lung tissue. Classically, the disease is symmetrical and predominates in the upper and middle lung fields, sparing the costophrenic angle and giving a very characteristic picture on high-resolution CT scan. Confluence of cysts may lead to bullous formation and spontaneous pneumothorax can be the first sign of LCH in the lung, although patients may present with tachypnea or dyspnea. Ultimately, widespread fibrosis and destruction of lung tissue may lead to severe pulmonary insufficiency. Declining diffusion capacity may also herald the onset of pulmonary hypertension. Widespread fibrosis and declining diffusion capacity are much less common in children. In young children with diffuse disease, therapy can halt progress of the tissue destruction and normal repair mechanisms may restore some function, although scarring or even residual nonactive cysts may continue to be visible on radiologic studies.
Pulmonary involvement is present in approximately 25% of children with multisystem low-risk and high-risk LCH. However, a multivariate analysis of pulmonary disease in multisystem LCH did not show pulmonary disease to be an independent prognostic factor, with a 5-year overall survival rate of 94% versus 96% for those with or without pulmonary involvement.
Most patients with bone marrow involvement are young children who have diffuse disease in the liver, spleen, lymph nodes, and skin who present with significant thrombocytopenia and anemia with or without neutropenia. Others have only mild cytopenias and are found to have bone marrow involvement with LCH by sensitive immunohistochemical or flow cytometric analysis of the bone marrow. A high content of bone marrow macrophages can obscure LCH cells. Patients with LCH who are considered at very high risk sometimes present with hemophagocytosis involving the bone marrow. The cytokine milieu driving LCH is probably responsible for the epiphenomenon of macrophage activation, which in the most severe cases, present with typical manifestations of hemophagocytic lymphohistiocytosis including cytopenias and hyperferritinemia.
Diabetes insipidus, caused by LCH-induced damage to the antidiuretic hormone-secreting cells of the posterior pituitary, is the most frequent endocrine manifestation in LCH. MRI scans usually show nodularity and/or thickening of the pituitary stalk and loss of the pituitary bright spot on T2-weighted images. Pituitary biopsies are rarely done and usually only when the stalk is greater than 6.5 mm or there is a hypothalamic mass. Pituitary disease is often associated with other sites of involvement; in order to avoid biopsy of the pituitary, these sites can be biopsied to establish the diagnosis.
Approximately 4% of LCH patients present with an apparently idiopathic presentation of diabetes insipidus before other lesions of LCH are identified. A review of pediatric patients presenting with idiopathic central diabetes insipidus showed that 19% eventually developed manifestations of LCH. Approximately 50% of patients who present with isolated diabetes insipidus as the initial manifestation of LCH either have anterior pituitary deficits at the time of diagnosis or develop them within 10 years of diabetes insipidus onset. These included secondary amenorrhea, panhypopituitarism, growth hormone deficiency, hypoadrenalism, and abnormalities of gonadotropins.
Patients with diabetes insipidus due to LCH have a 50% to 80% chance of developing other lesions diagnostic of LCH within 1 year of diabetes insipidus onset, including bone, lung, and skin. A study of 589 patients with LCH revealed the 10-year risk of pituitary involvement was 24%. No decrease in incidence of diabetes insipidus was seen in chemotherapy-treated patients, but this may reflect the length of the therapy and/or the number of drugs used. Using longer therapy and more drugs, the German-Austrian-Dutch (Deutsche Arbeits-gemeinschaft für Leukaemieforschung und-therapie im Kindesalter [DAL]) Group found the cumulative incidence to be 12%. Diabetes insipidus followed initial LCH diagnosis at a mean of 1 year and growth hormone deficiency occurred 5 years later. The incidence of diabetes insipidus was lower in patients treated with more intensive chemotherapy regimens on the LCH-III and JLSG-96 and JLSG-02 studies in Japan (8.9% for multisystem patients) than on the LCH-I and LCH-II studies (14.2%). Fifty-six percent of diabetes insipidus patients will develop anterior pituitary hormone deficiencies (growth, thyroid, or gonadal-stimulating hormones) within 10 years of the onset of diabetes insipidus. Diabetes insipidus occurs in 11% of patients treated with multiagent chemotherapy and in up to 50% of patients treated less aggressively.
Patients with multisystem disease and craniofacial involvement at the time of diagnosis, particularly of the orbit, mastoid, and temporal bones, carried a significantly increased risk of developing diabetes insipidus during their course (relative risk, 4.6), with 75% of patients with diabetes insipidus having these CNS-risk bone lesions. The risk increased when the disease remained active for a longer period of time or reactivated. The risk of diabetes insipidus development in this population was 20% at 15 years after diagnosis.
Although rare, there have been several cases of ocular involvement by LCH, sometimes leading to blindness. Patients may have other organ systems involved, and the ocular LCH may not respond well to conventional chemotherapy.
Central nervous system
CNS disease manifestations
Patients with LCH may develop mass lesions in the hypothalamic-pituitary region, the choroid plexus, the grey matter, or the white matter. These lesions contain CD1a-positive LCH cells and CD8-positive lymphocytes, and are, therefore, active LCH lesions.
Patients with large pituitary tumors (>6.5 mm) have a higher risk of anterior pituitary dysfunction and neurodegenerative CNS LCH. A retrospective study of 22 patients found that all had radiologic signs of neurodegenerative CNS LCH detected at a median time of 3 years and 4 months after LCH diagnosis and that it worsened in 19 patients. Five had neurologic dysfunction. Eighteen of 22 patients had anterior pituitary dysfunction and 20 had diabetes insipidus. Growth hormone deficiency occurred in 21 patients; luteinizing hormone/follicle-stimulating hormone deficiency occurred in ten patients; and thyroid hormone deficiency occurred in ten patients.
LCH CNS neurodegenerative syndrome
A chronic neurodegenerative syndrome that is manifested by dysarthria, ataxia, dysmetria, and sometimes behavior changes develops in 1% to 4% of patients with LCH. These patients may develop severe neuropsychologic dysfunction with tremor, gait disturbances, ataxia, dysarthria, headaches, visual disturbances, cognitive and behavioral problems, and psychosis. MRI scan results from these patients show hyperintensity of the dentate nucleus and white matter of the cerebellum on T2-weighted images or hyperintense lesions of the basal ganglia on T1-weighted images and/or atrophy of the cerebellum. The radiologic findings may precede the onset of symptoms by many years or be found coincidently. A study of 83 patients with LCH who had at least two MRI studies of the brain for evaluation of craniofacial lesions, diabetes insipidus, and/or other endocrine deficiencies of neuropsychological symptoms has been published. Forty-seven of 83 patients (57%) had radiological neurodegenerative changes at a median time of 34 months from diagnosis. Of the 47 patients, 12 (25%) developed clinical neurological deficits that presented 3 to 15 years after the LCH diagnosis. Fourteen of the 47 patients had subtle deficits in short-term auditory memory.
A study of CNS-related permanent consequences (neuropsychologic deficits) in 14 of 25 patients with LCH who were monitored for a median of 10 years has been published. Seven of these patients had diabetes insipidus and five patients had radiographic evidence of LCH CNS neurodegenerative changes. Patients with craniofacial lesions had lower performance and verbal intelligence quotient scores than those with other LCH lesions.
Histological evaluation of these neurodegenerative lesions shows a prominent T-cell infiltration, usually in the absence of the CD1a-positive dendritic cells along with microglial activation and gliosis. The neurodegenerative form of the disease has been compared to a paraneoplastic inflammatory response.
Back to Top
Diagnostic Evaluation of Childhood LCH
The complete evaluation of any patient, whether presenting with single-system or multisystem disease, should include the following:
- History and physical exam: A complete history and physical exam with special attention to the skin, lymph nodes, ears, oral pharynx, gingiva, tongue, teeth, bones, lungs, thyroid, liver and spleen size, bone abnormalities, growth velocity, and history of excessive thirst and urination.
Other tests and procedures include the following:
- Blood tests: Blood tests include complete blood count with leukocyte differential and platelet count, liver function tests (e.g., bilirubin, albumin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and prothrombin time/partial thromboplastin time in patients with hepatomegaly, jaundice, elevations of liver enzymes, or low albumin), and serum electrolytes.
- BRAF V600E: Although BRAF mutation assessment is not a required part of the workup for Langerhans cell histiocytosis (LCH), the BRAF mutation can be detected by either immunohistochemistry or molecular diagnostic methods.
- Urine tests: Urine tests include urinalysis and a water-deprivation test if diabetes insipidus is suspected. Water deprivation tests in very young children, especially infants, is performed under medical monitoring.
- Bone marrow aspirate and biopsy: The bone marrow aspirate and biopsy is indicated for patients with multisystem disease who have unexplained anemia or thrombocytopenia. The biopsy should be stained with anti-CD1a and/or anti-CD207 (langerin) and anti-CD163 immunostains to facilitate the detection of LCH cells.
Radiologic and imaging tests: Radiologic tests for the first level of screening include skeletal survey, skull series, bone scans, and chest X-ray. Newer diagnostic imaging modalities, such as somatostatin analog scintigraphy or fludeoxyglucose F 18 (18F-FDG) positron emission tomography (PET) scans, augment, but do not replace the standard tests.
- Computed tomography (CT) scan: CT scan of the head may be indicated if orbital, mastoid, or other maxillofacial involvement is suspected. Imaging tests may include magnetic resonance imaging (MRI) scan with gadolinium contrast of the brain for patients with diabetes insipidus or suspected brain or vertebral involvement.
CT scan of the lungs may be indicated for patients with abnormal chest X-rays or pulmonary symptoms. High-resolution CT scans may show evidence of pulmonary LCH when the chest X-ray is normal, thus in infants and toddlers with normal chest X-rays, a CT scan may be considered.
LCH causes fatty changes in the liver or hypodense areas along the portal tract, which can be identified by CT scans.
- 18F-FDG PET scan: 18F-FDG PET scan abnormalities have been reported in the brains of seven patients with LCH, with neurologic and radiographic signs of neurodegenerative disease. There was good correlation with MRI findings in the cerebellar white matter, but less so in the caudate nuclei and frontal cortex. It was suggested that PET scans of patients at high risk for developing neurodegenerative LCH could show abnormalities earlier than MRI. PET scans often demonstrate lesions not found by other modalities and show a decrease of activity after 6 weeks of therapy, thus providing a better assessment of response to therapy than bone scans or plain X-rays.
- MRI: MRI findings of patients with diabetes insipidus include thickening and nodularity of the pituitary stalk with loss of the posterior pituitary bright spot reflecting absence of anti-diuretic hormone. Later in the course, the stalk generally atrophies but this should not be used as evidence of response to therapy.
All patients with vertebral body involvement need careful assessment of associated soft tissue, which may impinge on the spinal cord.
MRI findings of central nervous system LCH include T2 fluid attenuated inversion recovery (FLAIR) enhancement in the pons, basal ganglia, white matter of the cerebellum, and mass lesions or meningeal enhancement. In a report of 163 patients, meningeal lesions were found in 29% and choroid plexus involvement in 6%. Paranasal sinus or mastoid lesions were found in 55% of patients versus 20% of controls, and accentuated Virchow-Robin spaces in 70% of patients versus 27% of controls.
- Computed tomography (CT) scan: CT scan of the head may be indicated if orbital, mastoid, or other maxillofacial involvement is suspected. Imaging tests may include magnetic resonance imaging (MRI) scan with gadolinium contrast of the brain for patients with diabetes insipidus or suspected brain or vertebral involvement.
Biopsy: Lytic bone lesions, skin, and lymph nodes are the most frequent sites biopsied for diagnosis of LCH. A liver biopsy is indicated when a child with LCH presents with hypoalbuminemia not caused by gastrointestinal LCH or other etiology. These patients usually have elevated levels of bilirubin or liver enzymes. An open lung biopsy may be necessary for obtaining tissue for diagnosis of pulmonary LCH when bronchoalveolar lavage is nondiagnostic.
A pathologic diagnosis is always required to make a definitive diagnosis. However, this may sometimes be difficult or contraindicated, such as in isolated pituitary stalk disease when the risk outweighs the benefit of a firm diagnosis. LCH cells are large cells with abundant pink cytoplasm and a bean-shaped nucleus on hematoxylin and eosin. LCH cells should stain positively with antibodies to CD1a and/or anti-langerin (CD207) to confirm the diagnosis of LCH.
Back to Top
Follow-up Considerations in Childhood LCH
Patients with diabetes insipidus and/or skull lesions in the orbit, mastoid, or temporal bones appear to be at higher risk for Langerhans cell histiocytosis (LCH) central nervous system (CNS) involvement and LCH CNS neurodegenerative syndrome. These patients should have magnetic resonance imaging (MRI) scans with gadolinium contrast at the time of LCH diagnosis and every 1 to 2 years thereafter for 10 years to detect evidence of CNS disease. The Histiocyte Society CNS LCH Committee does not recommend any treatment for radiologic CNS LCH of the neurodegenerative type if there is no associated clinical neurodegeneration. However, careful neurologic examinations and appropriate imaging with MRI is suggested at regular intervals. Brain stem auditory evoked responses should also be done at regular intervals to define the onset of clinical CNS LCH as early as possible, as this may affect response to therapy. When clinical signs are present, intervention may be indicated. Available studies of different forms of therapy for CNS neurodegeneration suggest that the neurodegenerative changes may be stabilized or improved, but only if therapy is started early. (Refer to the LCH CNS neurodegenerative syndrome section of this summary for more information.) Careful follow-up of patients at risk is critical.
For children with LCH in the lung, pulmonary function testing and chest computed tomography scans are sensitive methods for detecting disease progression.
In summary, many patients with multisystem disease will experience long-term sequelae due to their underlying disease and/or treatment. Endocrine and CNS sequelae are the most common. These long-term sequelae significantly affect health quality of life in many of these patients.[Level of evidence: 3iiiC] Specific long-term follow-up guidelines after treatment of childhood cancer or in those who have received chemotherapy have been published by the Children's Oncology Group and are available on their website.
Back to Top
Treatment of Childhood LCH
Depending on the site and extent of disease, treatment of Langerhans cell histiocytosis (LCH) may include surgery, radiation therapy, or oral, topical, and intravenous medication. The recommended duration of therapy is 12 months for patients who require chemotherapy for single-system bone, skin, or lymph node involvement. For patients with both high-risk and low-risk multisystem disease, the reactivation rate following 6 months of therapy was as high as 50% on the LCH-I and LCH-II trials.. Based upon data from the German-Austrian-Dutch (Deutsche Arbeits-gemeinschaft für Leukaemieforschung und-therapie im Kindesalter [DAL]) Group trials, which treated patients for 1 year and had fewer relapses (29%), the LCH-III trial was designed to give 12 months of chemotherapy for all high-risk multisystem patients and to randomly assign low-risk multisystem patients to either 6 months or 12 months of therapy. In patients with low-risk or high-risk disease who received 12 months of therapy, the reactivation rate was significantly reduced to approximately 30%. The LCH-IV trial will assess whether extending the duration of therapy further will reduce the incidence of reactivations and late effects. Although the LCH-IV study is open in several European centers, it is not open in the United States.
It is preferable that patients with LCH be enrolled in a clinical trial whenever possible so that advances in therapy can be achieved more quickly, utilizing evidence-based recommendations and to ensure optimal care. Information about clinical trials for LCH in children is available from the Histiocyte Society website.
Standard Treatment Options by Organ, Site or System Involvement for Pediatric Patients
The standard treatment for LCH is best chosen based on data from international trials with large numbers of patients. However, some patients may have LCH involving only the skin, mouth, pituitary gland, or other sites not studied in these international trials. In such cases therapy recommendations are based upon case series that lack the evidence-based strength of the trials.
Treatment of low-risk disease (single-system or multisystem)
Isolated skin involvement
- Observation. Observation is recommended for all pediatric patients with skin-only LCH. Therapy is suggested only for symptomatic disease such as extensive rashes, pain, ulceration, or bleeding.
- Topical steroids. Medium to high potency steroids are effective but the effect is usually not long lasting.
- Oral methotrexate (20 mg/m2) weekly for 6 months.
- Oral thalidomide 50 mg to 200 mg nightly. Oral thalidomide may be effective for both pediatric and adult patients.
- Topical application of nitrogen mustard can be effective for cutaneous LCH that is resistant to oral therapies, but not for disease involving large areas of skin.
- Psoralen and long-wave ultraviolet A radiation (PUVA) and UVB can be effective in skin LCH but its use is limited by the potential for late skin cancers, especially in patients with light skin tones.
Single skull lesions of the frontal, parietal, or occipital regions, or single lesions of any other bone
- Curettage only is the recommended therapy, when possible for isolated bone lesions; curettage plus injection of methylprednisolone may be used. Low-dose radiation therapy is effective but its use is limited in pediatric patients to lesions that threaten organ function.; [Level of evidence: 3iiiA] LCH bone lesions may not need complete excision, because this may increase healing time and the risk of long-term complications.
Skull lesions in the mastoid, temporal, or orbital bones
The purpose of treating patients with isolated skull lesions in the mastoid, temporal, or orbital bones is to decrease the chance of developing diabetes insipidus and other long-term problems. Comparison of diabetes insipidus incidence with no systemic therapy (40%) versus 6 months of vinblastine/prednisone (20%) strongly supports treatment of the central nervous system (CNS)–risk bones even when it occurs in a single site. However, the efficacy of therapy and the optimal length of therapy have yet to be proven in a prospective trial.
- Twelve months of vinblastine and prednisone as per the LCH-III study results: Weekly vinblastine (6 mg/m2) for 7 weeks then every 3 weeks for good response. Daily prednisone (40 mg/m2) for 4 weeks then tapered over 2 weeks. Afterward prednisone is given for 5 days at 40 mg/m2 every 3 weeks with the vinblastine injections.
- There is some controversy about whether systemic therapy is required for the first presentation with unifocal bone LCH even in the CNS risk bones. Ear, nose, and throat surgeons have reported a series of patients with orbital or mastoid lesions who received only surgical curettage. None of these patients developed diabetes insipidus. However, when comparing the incidence rates of diabetes insipidus in patients who received little or no chemotherapy (20%–50% incidence of diabetes insipidus) versus diabetes insipidus incidence rates reported by the German-Austrian-Dutch (DAL) Group HX-83 trial (10% incidence of diabetes insipidus in patients treated for LCH), it appears that the weight of evidence from the DAL HX-83 trial supports chemotherapy treatment to prevent diabetes insipidus in patients with LCH of the mastoid, temporal, or orbital bones. It should be noted, however, that the DAL HX studies used more drugs and treated patients for a duration of 12 months.
Vertebral or femoral bone lesions at risk for collapse
- A single vertebral body lesion without soft tissue extension to the extradural space may be observed only.
- Low-dose radiation therapy may be used to try to promote resolution in an isolated vertebral body lesion or a large femoral neck lesion at risk for fracture, where chemotherapy is not usually indicated (single bone lesion). Despite the low dose required (700–1,000 cGy), radiation therapy should be used with caution in the area of the thyroid gland, brain, or any growth plates.
- Patients with soft tissue extension from vertebral lesions are often treated successfully with chemotherapy,[Level of evidence: 3iiDiii] but prolonged therapy does not appear to be needed beyond the period required to reduce the mass and any risk to the spinal cord. The risk of reactivation of a single bone lesion was only 9% in one large retrospective series.
- When instability of the cervical vertebrae and/or neurologic symptoms are present, bracing, or rarely, spinal fusion may be needed. Patients with soft tissue extension from the vertebral lesions are often treated successfully with chemotherapy.[Level of evidence: 3iiDiii]
Multiple bone lesions (single-system multifocal bone)
- The most commonly used systemic chemotherapy regimen is the combination of vinblastine and prednisone. Based on the results of the HISTSOC-LCH-III (NCT00276757) trial, 12 months of treatment with weekly vinblastine (6 mg/m2) for 7 weeks then every 3 weeks is used for good responders. Prednisone (40 mg/m2) is given daily for 4 weeks then tapered over 2 weeks. Afterwards prednisone is given for 5 days at 40 mg/m2 every 3 weeks with the vinblastine injections. A short (<6 months) treatment course with only a single agent (e.g., prednisone) is not sufficient, and the number of relapses is higher. A reactivation rate of 18% was reported with a multidrug treatment regimen that was used for 6 months versus a historical reactivation rate of 50% to 80% with surgery alone or with a single-drug treatment regimen.
Multiple bone lesions in combination with skin, lymph node, or diabetes insipidus (low-risk multisystem LCH)
- Vinblastine and prednisone in combination. Based on the results of the randomized HISTSOC-LCH-III (NCT00276757) trial, the same chemotherapy regimen of vinblastine and prednisone as described above is used for 12 months. Patients without risk-organ involvement who were randomly assigned to 12 months of vinblastine/prednisone had a lower 5-year reactivation rate (37%) than did patients who received only 6 months of treatment (54%; P = .03) and patients treated with historical 6-month schedules (52% [LCH-I] and 48% [LCH-II]; P < .001). Most disease reactivations were in bone, skin, or other nonrisk locations.
- Other chemotherapy regimens have also been effective, including the following:
- Vincristine, cytosine arabinoside, and prednisone in combination. This combination has been proven to be an effective frontline or salvage therapy. However, prednisone is given for a much shorter duration than was originally published; currently, prednisone is given for 4 to 6 weeks during the induction phase and then for 5 days every 3 weeks with a single dose of vincristine and 5 days of cytosine arabinoside during maintenance.
- Cladribine. Cladribine given at 5 mg/m2/day for 5 days every 3 weeks for two to six cycles can be an effective salvage therapy for recurrent bone or low-risk multisystem disease. More than six cycles is not recommended because of the risk of cumulative cytopenias.
- Pamidronate can also be effective for treating LCH bone lesions. A nationwide survey from Japan described 16 children treated with bisphosphonates for bone LCH. All had bone disease; none had risk-organ disease. Most patients received six cycles of pamidronate at 1 mg/kg/course given at 4-week intervals. In 12 of 16 patients, all active lesions including skin (n = 3) and soft tissues (n = 3) resolved. Eight remained disease free at a median of 3.3 years. Other bisphosphonates, such as zoledronate and oral alendronate, have been used to successfully treat bone LCH.
Treatment of high-risk multisystem disease
Spleen, liver, and bone marrow (may or may not include skin, bone, lymph node, lung, or pituitary gland)
- The standard therapy length recommended for LCH involving the spleen, liver, or bone marrow (high-risk organs) is now 12 months based upon the DAL-HX-83 and HISTSOC-LCH-III (NCT00276757) studies. In the Histiocyte Society LCH-II and LCH-III studies, the standard arm consisted of vinblastine and prednisone as described above under multifocal bone, but 6-mercaptopurine was added to the continuation phase of the protocol.
- The LCH-II study was a randomized trial to compare treatment of patients with vinblastine, prednisone, and mercaptopurine or vinblastine, prednisone, mercaptopurine, and etoposide.[Level of evidence: 1iiA]
There was no statistical significance in outcomes (response at 6 weeks, 5-year probability of survival, relapses, and permanent consequences) between the two treatment groups. Hence, etoposide has not been used in subsequent Histiocyte Society trials. Late review of the results, however, reported reduced mortality of patients with risk-organ involvement in the etoposide arm. Although controversial, a comparison of patients in the LCH-I trial with patients in the LCH-II trial suggested that increased treatment intensity promoted additional early responses and reduced mortality.
It is important to note that those studies included lungs as risk organs. However, subsequent analyses have shown that lung involvement lacks prognostic significance.
- The LCH-III study randomly assigned risk organ–affected patients to either vinblastine/prednisone/6-mercaptopurine or vinblastine/prednisone/6-mercaptopurine plus methotrexate (intravenous during the induction phase and oral in the continuation phase). The response rates at 6 and 12 weeks and overall survival were not improved; however, there were significantly increased grade 3 and grade 4 toxicities in patients who received methotrexate.
An important finding of the LCH-III study was that the mortality of patients with high-risk LCH on both arms of the study was significantly reduced compared with the earlier LCH-II study, even though the standard arm utilizes the same drugs. Possible explanations for reduced mortality include the following:
- A second 6-week induction phase of weekly vinblastine with prednisone given for 3 days per week. This reinduction phase was given to all patients who did not achieve a status of no active disease by the end of the 6-week induction phase, before going onto the every-3-weeks maintenance courses. The rate of no active disease increased after the second induction phase and this course may have played a significant role in the reduced mortality rate.
- Better supportive care.
- Earlier change to an effective salvage strategy for nonresponsive lesions.
It should be noted that although survival was improved in the LCH-III study, only 60% of patients had no active disease in risk organs after a year of therapy and 25% to 29% of patients relapsed.
- The Japan LCH Study Group (JLSG) reported 5-year response and overall survival rates of 78% and 95%, respectively, for patients with multisystem disease treated on the JLSG-96 trial (6-week induction regimen of cytosine arabinoside, vincristine, and prednisolone followed by 6 months of maintenance therapy with cytarabine, vincristine, prednisolone, and low-dose intravenous methotrexate). If patients had a poor response to the initial regimen, they were switched to a salvage regimen of intensive combination doxorubicin, cyclophosphamide, methotrexate, vincristine, and prednisolone.
Similar to the LCH-III study, the important finding of this study was the decreased mortality compared with previous JLSG studies and to the LCH-II study. This was attributed to the early change to a more effective salvage therapy for patients with nonresponsive disease, as well as better supportive care.
- Some patients develop a “macrophage activation” of their marrow. This may be confusing to clinicians who may think the patient has hemophagocytic lymphohistiocytosis and LCH. The best therapy for this life-threatening manifestation is not clear, because it tends not to respond well to standard hemophagocytic lymphohistiocytosis therapy. Clofarabine, anti-CD52 antibody alemtuzumab, or reduced-intensity allogeneic stem cell transplant could be considered.
Treatment of CNS disease
There are three types of LCH CNS lesions:
- Mass lesions or tumors in the cerebrum, cerebellum, or choroid plexus.
- Mass lesions of the hypothalamic-pituitary axis that are always associated with diabetes insipidus and are often associated with other endocrinopathies.
- Neurodegenerative syndrome. T2 fluid attenuated inversion recovery (FLAIR) hyperintense signals are present, most often in the cerebellar white matter, pons, basal ganglia, and sometimes, in the cerebrum.
Drugs that cross the blood-brain barrier, such as cladribine, or other nucleoside analogs, such as cytarabine, are used for active CNS LCH lesions.
- Treatment of mass lesions with cladribine has been effective in 13 reported cases.; [Level of evidence: 3iiiDiii] Mass lesions included enlargement of the hypothalamic-pituitary axis, parenchymal mass lesions, and leptomeningeal involvement. Doses of cladribine ranged from 5 mg/m2 to 13 mg/m2, given at varying frequencies.[Level of evidence: 3iiiDiii]
- Patients with LCH and mass lesions in the hypothalamic-pituitary region, the choroid plexus, the grey matter, or the white matter, may also respond to standard LCH chemotherapy. Treatment with vinblastine with or without corticosteroids for patients with CNS mass lesions (20 patients; mainly pituitary) demonstrated an objective response in 15 patients, with 5 of 20 patients demonstrating a complete response and 10 of 20 patients demonstrating a partial response.
CNS neurodegenerative syndrome
Drugs used in active LCH, such as dexamethasone and cladribine, along with other agents, such as retinoic acid, intravenous immunoglobulin (IVIg), infliximab, and cytarabine with or without vincristine have been used in small numbers of patients with mixed results. Many of these agents may result in the complete or partial resolution of radiographic findings, but definitive clinical response rates have not been rigorously defined.; [Level of evidence: 3iiiDiii]
- Retinoic acid was given at a dose of 45 mg/m2 daily for 6 weeks, then 2 weeks per month for 1 year. Patients were reported to have stable clinical status.
- IVIg (400 mg/m2) was given monthly with chemotherapy consisting of oral prednisolone with or without oral or intravenous methotrexate and oral 6-mercaptopurine for at least 1 year. Magnetic resonance imaging (MRI) findings were stable but clinical efficacy was difficult to judge because patients were reported to have no progression in their neurologic symptoms.
- A study using cytarabine with or without vincristine for up to 24 months reported improvement in the clinical and MRI findings in some patients and stabilization of disease in the others.[Level of evidence: 3iiiC] Seven of eight patients have been followed for more than 8 years after stopping therapy and have had stable neurologic and radiographic findings.
- In the Japan LCH Study Group-96 Protocol, cytarabine failed to prevent the onset of neurodegenerative syndrome. Patients received cytarabine 100 mg/m2 daily on days 1 to 5 during induction and 150 mg/m2 on day 1 of each maintenance cycle (every 2 weeks for 6 months). Three of 91 patients developed neurodegenerative disease, which is similar to the rate experienced on the Histiocyte Society studies.
Perhaps the most important aspect of therapy for neurodegenerative disease is the early recognition of clinical neurodegeneration and institution of therapy. Studies combining MRI findings together with cerebrospinal fluid markers of demyelination, to identify patients who require therapy, even before onset of clinical symptoms, are currently underway in several countries.
Treatment Options for Childhood LCH No Longer Considered Effective
Treatments that have been used in the past but are no longer recommended for pediatric patients with LCH in any location include cyclosporine and interferon-alpha. Extensive surgery is also not indicated. Curettage of a circumscribed skull lesion may be sufficient if the lesion in not in the temporal, mastoid, or orbital areas (CNS risk). Patients with disease in these particular sites are recommended to receive 6 months of systemic therapy with vinblastine and prednisone. For lesions of the mandible, extensive surgery may destroy any possibility of secondary tooth development. Surgical resection of groin or genital lesions is contraindicated as these lesions can be healed by chemotherapy.
Radiation therapy use in LCH has been significantly reduced in pediatric patients, and even low-dose radiation therapy should be limited to single-bone vertebral body lesions or other single-bone lesions compressing the spinal cord or optic nerve that do not respond to chemotherapy.
Assessment of Response to Treatment
Response assessment remains one of the most difficult areas in LCH therapy unless there is a specific area that can be followed clinically or with sonography, computed tomography (CT), or MRI scans of areas such as the skin, hepato/splenomegaly, and other mass lesions. Clinical judgment including evaluation of pain and other symptoms remains important.
Bone lesions may take many months to heal and are difficult to evaluate on plain radiographs, although sclerosis around the periphery of a bone lesion suggests healing. CT or MRI scans are useful in assessing response of a soft tissue mass associated with a bone lesion, but is not particularly helpful in lytic bone lesions. Technetium bone scans remain positive in healing bone. Positron emission tomography (PET) scans may be helpful in following the response to therapy since intensity of the PET image diminishes with response of lesions and healing of bone.
For children or adults with lung LCH, pulmonary function testing and high resolution CT scans are sensitive methods for detecting disease progression. Residual interstitial changes reflecting residual fibrosis or residual inactive cysts must be distinguished from active disease and somatostatin analogue scintigraphy may be useful in this regard.
Current Clinical Trials
Check the list of NCI-supported cancer clinical trials that are now accepting patients with childhood Langerhans cell histiocytosis. The list of clinical trials can be further narrowed by location, drug, intervention, and other criteria.
General information about clinical trials is also available from the NCI website.
Back to Top
Treatment of Recurrent, Refractory, or Progressive Childhood LCH
Recurrent Low-Risk Organ Involvement
Reactivation of single-system and multisystem LCH
Reactivation of Langerhans cell histiocytosis (LCH) after complete response has been reported; usually occurring within the first 9 to 12 months after stopping treatment. The percentage of patients with reactivations was 9% to 17.4% for single-site disease; 37% for single-system, multifocal disease; 46% for multisystem (nonrisk organ) disease; and 54% for patients with risk-organ involvement. Forty-three percent of reactivations were in bone, 11% in ears, 9% in skin, and 7% developed diabetes insipidus; a lower percentage of patients had lymph node, bone marrow, or risk-organ relapses. The median time to reactivation was 12 to 15 months in nonrisk patients and 9 months in risk patients. One-third of patients had more than one reactivation varying from 9 to 14 months after the initial reactivation. Patients with reactivations were more likely to have long-term sequelae in the bones, diabetes insipidus, or other endocrine, ear, or lung problems.
A comprehensive review of the German-Austrian-Dutch (Deutsche Arbeits-gemeinschaft für Leukaemieforschung und-therapie im Kindesalter [DAL]) and Histiocyte Society clinical trials revealed a reactivation rate of 46% at 5 years for patients with multisystem LCH, with most reactivations occurring within 2 years of first remission. A second reactivation occurred in 44%, again within 2 years of the second remission. Involvement of the risk organs in these reactivations occurred only in those who were initially in the high-risk group (meaning they had liver, spleen, or bone marrow involvement at the time of original diagnosis).[Level of evidence: 3iiiDiii] Most reactivations, even in patients with high-risk disease who initially responded to therapy, were in bone, skin, or other nonrisk locations.
The percentage of reactivations in multisystem disease was 45% in the Japanese trial [Level of evidence: 1iiA] and 46% in the LCH-II trial. There was not a statistically significant difference in reactivations between the high-risk and low-risk groups. Both the DAL-HX and Japanese studies concluded that intensified treatment increased rapid response, particularly in young children and infants younger than 2 years, and together with rapid switch to salvage therapy for nonresponders, reduced mortality for patients with high-risk multisystem LCH.
Treatment of low-risk organ involvement
The optimal therapy for patients with relapsed or recurrent LCH has not been determined. Several regimens exist, including the following:
- Patients with recurrent bone disease who recur months after stopping vinblastine and prednisone can benefit from treatment with a reinduction of vinblastine weekly and daily prednisone for 6 weeks. If there is no active disease or very little evidence of active disease, treatment can be changed to every 3 weeks, with the addition of oral mercaptopurine nightly.
- An alternative treatment regimen employs vincristine, prednisone, and cytosine arabinoside. The prednisone dose is reduced from the dose used in the original publication.
- Bisphosphonate therapy is also effective for treating recurrent LCH bone lesions. In a survey from Japan, bisphosphonate therapy successfully treated the bone lesions in 12 of 16 patients. Skin and soft tissue LCH lesions also resolved in the responding patients. None of the patients had risk-organ disease. Most patients received six cycles of pamidronate at 1 mg/kg/course, given at 4-week intervals. Eight of the 12 patients remained disease free at a median of 3.3 years. Other bisphosphonates, such as zoledronate and oral alendronate, have also been successful in treating bone LCH.
- Cladribine at 5 mg/m2/day for 5 days per course has also been shown to be effective therapy for recurrent low-risk LCH (multifocal bone and low-risk multisystem LCH) with very little toxicity. Cladribine therapy should, if possible, be limited to a maximum of six cycles to avoid cumulative and potentially long-lasting cytopenias.
- A phase II trial of thalidomide for patients with LCH (ten low-risk patients; six high-risk patients) who failed primary and at least one secondary regimen demonstrated complete (four of ten) and partial (three of ten) responses for the low-risk patients. Complete remission was defined as healing of bone lesions on plain radiographs (n = 3) or complete resolution of skin rash (n = 4, including 3 with bone lesions that had complete resolution). Partial response was defined as healing of bone lesion, but then worsening of a skin rash that was partially resolved. However, dose-limiting toxicities, such as neuropathy and neutropenia, may limit the overall usefulness of thalidomide. This agent is not used in pediatric patients to a significant degree.
- Clofarabine is a proven effective therapy for patients with multiple relapses of low-risk or high-risk organs.
Refractory High-Risk Organ Involvement
A new treatment plan is indicated when a patient with multisystem involvement shows progressive disease after 6 weeks of standard treatment, or has not had a partial response by 12 weeks. Data from the DAL Group studies have shown that these children have only a 10% chance of surviving. Results from the LCH-II trial revealed that patients treated with vinblastine/prednisone who did not respond well by 6 weeks had a 27% chance of survival, compared with 52% for good responders.[Level of evidence: 1iiA] All studies suggest that patients with poorly responsive disease need to be changed early to salvage strategies at 6 weeks for progressive disease and no later than 12 weeks for those without at least a good response.
Cladribine and 2’-deoxycoformycin have been tested as salvage therapies for LCH.; [Level of evidence: 3iiiDiv] A case series reported that patients with multiple reactivations or high-risk disease could be effectively treated with continuous-infusion cladribine for 3 days. Seven of ten patients on this trial required no more therapy.[Level of evidence: 3iiiDii]
Ten patients with refractory high-risk organ (liver, spleen, or bone marrow) involvement and resistant multisystem low-risk organ involvement have been treated with an intensive acute myeloid leukemia–like protocol. Prompt change of therapy to cladribine and cytosine arabinoside appeared to provide an improvement in overall survival (OS).; [Level of evidence: 3iiiDii]; [Level of evidence: 3iiiDiv] A prospective trial of the cladribine and cytosine arabinoside protocol (N = 27) showed a progression-free survival rate of 63% and a 5-year OS rate of 85%. However, all patients developed grade 4 hematologic toxicity, and five of these patients had severe sepsis.
Six patients with multiorgan LCH resistant to other agents, including cladribine, were reported to respond to treatment with clofarabine.; [Level of evidence: 3iiiDii] An additional 11 patients with recurrent multisystem high-risk and low-risk disease had a 90% OS. If confirmed in prospective trials, the reduced toxicity of this regimen, compared with the cladribine/cytarabine combination, could be advantageous despite the cost of the drug.
Hematopoietic stem cell transplantation (HSCT) has been used in patients with multisystem high-risk organ involvement that is refractory to chemotherapy. Reduced-intensity conditioning provides no OS advantage over myeloablative conditioning before HSCT for LCH patients. However, the relapse rate after reduced-intensity conditioning was significantly higher (28%) than the relapse rate after myeloablative conditioning (8%). Many of the reduced-intensity conditioning patients who relapsed were successfully re-treated with chemotherapy alone.
Treatment Options Under Clinical Evaluation
New targeted therapies under investigation include the following:
- RAS pathway inhibitors: The discovery that most patients with LCH have BRAF V600E or other mutations that result in activation of the RAS pathway, suggests that new therapies that target molecules within this pathway will become an important part of LCH therapy in the near future. NCT01677741 is currently open in the United States and Canada for adults with LCH and children with recurrent LCH. Although three adults with LCH and Erdheim-Chester disease had favorable responses to vemurafenib, the 30% incidence of squamous cell carcinoma in melanoma patients receiving this drug needs to be taken into account with moderate- to long-term use in children.
- Tyrosine kinase inhibitors: Several reports suggested that LCH, including central nervous system LCH, could be successfully treated with the tyrosine kinase inhibitor imatinib, while other reports failed to show an effect.
Back to Top
Late Disease and Treatment Effects of Childhood LCH
The reported overall frequency of long-term consequences of Langerhans cell histiocytosis (LCH) has ranged from 20% to 70%. The reason for this wide variation is due to case definition, sample size, therapy used, method of data collection, and follow-up duration. Of note, in one study of the quality of life of long-term survivors of skeletal LCH, the quality-of-life scores were not significantly different from healthy control children and adults. In addition, the quality-of-life scores were very similar between those with and without permanent sequelae. In another study of 40 patients who were carefully screened for late effects, adverse quality-of-life scores were found in more than 50% of patients. Seventy-five percent of patients had detectable long-term sequelae—hypothalamic/pituitary dysfunction (50%), cognitive dysfunction (20%), and cerebellar involvement (17.5%) being the most common.
Children with low-risk organ involvement (skin, bones, lymph nodes, or pituitary gland) have an approximately 20% chance of developing long-term sequelae. Patients with diabetes insipidus are at risk for panhypopituitarism and should be monitored carefully for adequacy of growth and development. In a retrospective review of 141 patients with LCH and diabetes insipidus, 43% developed growth hormone (GH) deficiency. The 5-year and 10-year risks of GH deficiency among children with LCH and diabetes insipidus were 35% and 54%, respectively. There was no increased reactivation of LCH in patients who received GH compared with those who did not.
Growth and development problems are more frequent because of the young age at presentation and the more toxic effects of long-term prednisone therapy in the very young child. Patients with multisystem involvement have a 71% incidence of long-term problems.
Hearing loss has been found in 38% of children who were treated for LCH. Seventy percent of patients with LCH in this study had ear involvement which included aural discharge, mastoid swelling, and hearing loss. Of those with computed tomography or magnetic resonance imaging (MRI) abnormalities in the mastoid, 59% had hearing loss.[Level of evidence: 3iiiC]
Neurologic symptoms secondary to vertebral compression of cervical lesions have been reported in 3 of 26 patients with LCH and spinal lesions. Central nervous system (CNS) LCH occurs most often in children with LCH of the pituitary or CNS-risk skull bones (mastoid, orbit, or temporal bone). Significant cognitive defects and MRI abnormalities may develop in some long-term survivors with CNS-risk skull lesions. Some patients have markedly abnormal cerebellar function and behavior abnormalities, while others have subtle deficits in short-term memory and brain stem–evoked potentials.
Orthopedic problems from lesions of the spine, femur, tibia, or humerus may be seen in 20% of patients. These problems include vertebral collapse or instability of the spine that may lead to scoliosis and facial or limb asymmetry.
Diffuse pulmonary disease may result in poor lung function with higher risk for infections and decreased exercise tolerance. These patients should be monitored with pulmonary function testing, including the diffusing capacity of carbon monoxide and ratio of residual volume to total lung capacity.
Liver disease may lead to sclerosing cholangitis, which rarely responds to any treatment other than liver transplantation.
Dental problems characterized by loss of teeth have been significant for some patients, usually related to overly aggressive dental surgery.
Bone marrow failure secondary to LCH or from therapy is rare and is associated with a higher risk of malignancy. Patients with LCH have a higher-than-normal risk of developing secondary cancers. Leukemia (usually acute myeloid) occurs after treatment, as does lymphoblastic lymphoma. Concurrent LCH/malignancy has been reported in a few patients, and some patients have had their malignancy first, followed by development of LCH. Three patients with T-cell acute lymphoblastic leukemia (T-ALL) and aggressive LCH were reported and, as with all histiocytic disorders associated with or following lymphoblastic malignancies, the same genetic changes were found in both diseases, suggesting a shared clonal origin. One study reported two cases in which clonality with the same T-cell receptor gamma genotype was found. The authors of this study emphasized the plasticity of lymphocytes developing into Langerhans cells. In the second study, one patient with LCH after T-ALL who had the same T-cell receptor gene rearrangements and activating mutations of the NOTCH1 gene was described.
An association between solid tumors and LCH has also been reported. Solid tumors associated with LCH include retinoblastoma, brain tumors, hepatocellular carcinoma, and Ewing sarcoma.
Back to Top
The natural history of disease in adult Langerhans cell histiocytosis (LCH), with the exception of pulmonary LCH, is unknown. It is unclear whether there are significant differences from childhood LCH, although it appears that multisystem-risk LCH is less aggressive than childhood high-risk disease. The risk of reactivations is unknown.
A consensus group reported on the evaluation and treatment of adult patients with LCH. However, discussion continues, particularly regarding optimal first-line therapy.
It is estimated that one to two adult cases of LCH occur per 1 million population. The true incidence of this disease is not known, however, because published studies are mostly nonpopulation based and the disorder is likely to be underdiagnosed. A survey from Germany reported that 66% of the patients with LCH were women, with an average age of 43.5 years for all patients.
Presentation of adult LCH by organ, site, or system
Adult patients with LCH may have symptoms and signs for many months before a definitive diagnosis and treatment. LCH in adults is often similar to that in children, and appears to involve the same organs, although the proportions may be different. There is a predominance of lung disease in adults, usually occurring as single-system disease and closely associated with smoking and with some unique biologic characteristics. Most adult lung LCH cases are polyclonal and possibly reactive, while the minority of lung LCH cases are monoclonal. A German registry with 121 registrants showed that 62% had single-organ involvement and 38% had multisystem involvement, while 34% of the total had lung involvement. The median age at diagnosis was 44 years ± 12.8 years. The most common organ involved was lung followed by bone and skin. All organ systems found in childhood LCH were seen, including endocrine and central nervous system, liver, spleen, bone marrow, and gastrointestinal tract. The major difference is the much higher incidence of isolated pulmonary LCH in adults, particularly in young adults who smoke. Other differences appear to be the more frequent involvement of genital and oral mucosa. There may possibly be a difference in the distribution of bone lesions, but both groups suffer reactivations of bone lesions and progression to diabetes insipidus, although the exact incidence is unknown in adults.
Presenting symptoms from published studies are (in order of decreasing frequency) dyspnea or tachypnea, polydipsia and polyuria, bone pain, lymphadenopathy, weight loss, fever, gingival hypertrophy, ataxia, and memory problems. The signs of LCH are skin rash, scalp nodules, soft tissue swelling near bone lesions, lymphadenopathy, gingival hypertrophy, and hepatosplenomegaly. Patients who present with isolated diabetes insipidus should be carefully observed for onset of other symptoms or signs characteristic of LCH. At least 80% of patients with diabetes insipidus had involvement of other organ systems, including bone (68%), skin (57%), lung (39%), and lymph nodes (18%).
Skin and oral cavity
Thirty-seven percent of adults with LCH have skin involvement which usually occurs as part of multisystem disease. Skin-only LCH occurs but it is less common in adults than in children. The prognosis of adults with skin-only LCH is excellent, with 100% probability of 5-year survival. The cutaneous involvement is clinically similar to that seen in children and may take many forms. Infra-mammary and vulvar involvement may be seen in adult women with skin LCH.
Many patients have a papular rash with brown, red, or crusted areas ranging from the size of a pinhead to a dime. In the scalp, the rash is similar to that of seborrhea. Skin in the inguinal region, genitalia, or around the anus may have open ulcers that do not heal after antibacterial or antifungal therapy. The lesions are usually asymptomatic but may be pruritic or painful. In the mouth, swollen gums or ulcers along the cheeks, roof of the mouth, or tongue may be signs of LCH.
Diagnosis of LCH is usually made by skin biopsy performed for persistent skin lesions.
The relative frequency of bone involvement in adults differs from that in children: mandible (30% vs. 7%) and skull (21% vs. 40%). The frequency of vertebrae (13%), pelvis (13%), extremities (17%), and rib (6%) lesions in adults are similar to those found in children.
Pulmonary LCH in adults is usually single-system disease, but in a minority of patients other organs may be involved, including bone (18%), skin (13%), and diabetes insipidus (5%).
Pulmonary LCH is more prevalent in smokers than in nonsmokers and the male to female ratio may be near unity depending on the incidence of smoking in the population studied. Patients with pulmonary LCH usually present with a dry cough, dyspnea, or chest pain, although nearly 20% of adults with lung involvement have no symptoms. Chest pain may indicate a spontaneous pneumothorax (10%–20% of adult pulmonary LCH cases). Pulmonary LCH can be diagnosed by bronchoscopy in about 50% of adult patients, as defined by characteristic CD1a immunostaining cells of greater than or equal to 5% of cells observed.
The LCH cells in adult lung lesions were shown to be mature dendritic cells expressing high levels of the accessory molecules CD80 and CD86, unlike Langerhans cells (LCs) found in other lung disorders. In addition, pulmonary LCH in adults appears to be primarily a reactive process, rather than a clonal proliferation as seen in childhood LCH. However, BRAF V600E mutations have been demonstrated in pulmonary LCH lesions in adults, suggesting a clonal process in a significant proportion (25%–30%) of patients.
The course of pulmonary LCH in adults is variable and unpredictable. Fifty-nine percent of patients do well with either spontaneous remission with cessation of smoking, or with some form of therapy. Adults with pulmonary LCH who have minimal symptoms have a good prognosis, although some have steady deterioration over many years. Age older than 26 years and lower forced expiratory volume/forced vital capacity (FEV1/FVC) ratio and higher residual volume/total lung capacity (RV/TLC) ratio are adverse prognostic variables. About 10% to 20% have early severe progression to respiratory failure, severe pulmonary hypertension, and cor pulmonale. Adults who have progression with diffuse bullae formation, multiple pneumothoraces, and fibrosis have a poor prognosis. The remainder have a variable course, with stable disease in some patients and relapses and progression of respiratory dysfunction in others, some after many years. One study reported that smoking cessation did not increase the longevity of adults with pulmonary LCH, apparently because the tempo of disease is so variable. Patients receiving lung transplantation for treatment of pulmonary LCH have a 77% survival rate at 1 year and 54% survival rate at 10 years, with a 20% chance of LCH recurrence. A natural history study of 58 LCH patients with pulmonary involvement found that 38% of patients had deterioration of lung function after 2 years. The most significant adverse prognostic variables were positive smoking statuses and low PaO2 levels at the time of inclusion.
The most frequent pulmonary function abnormality finding in patients with pulmonary LCH is a reduced carbon monoxide diffusing capacity in 70% to 90% of cases. A high-resolution computed tomography (CT) scan, which reveals a reticulonodular pattern classically with cysts and nodules, usually in the upper lobes and sparing the costophrenic angle, is characteristic of LCH. Despite the typical CT findings, most pulmonologists agree that a lung biopsy is needed to confirm the diagnosis. The presence of cystic abnormalities on high-resolution CT scans appears to be a poor predictor of which patients will have progressive disease. A study correlating lung CT findings and lung biopsy results in 27 patients with pulmonary LCH has shed some light on pulmonary LCH. Thin-walled and bizarre cysts had active LCs and eosinophils. Fifty-two percent of patients improved, most with smoking cessation, and some with steroid treatment within 14 months of diagnosis. Four patients (15%) were stable, and nine patients (33%) progressed over 22 months.
Liver involvement in adults has been reported in 27% of a series of adult patients with LCH and multiorgan disease. Hepatomegaly (48%) and liver enzyme abnormalities (61%) were present. CT and ultrasound imaging abnormalities are often found. The early histopathologic stage of liver LCH includes infiltration of CD1a+ cells and periductal fibrosis with inflammatory infiltrates with or without steatosis. The late stage is biliary tree sclerosis and treatment with ursodeoxycholic acid is suggested.
In a large series of patients from the Mayo Clinic, 31% had multisystem LCH compared with 69% registered on the Histiocyte Society adult registry; this likely reflects referral bias. In the adult multisystem patients, the organs involved included the following:
- Skin (50%).
- Mucocutaneous (40%).
- Diabetes insipidus (29.6%).
- Hepatosplenomegaly (16%).
- Hypothyroidism (6.6%).
- Lymphadenopathy (6%).
Back to Top
Treatment of Adult LCH
Standard Treatment Options
The lack of clinical trials limits the ability to make evidence-based recommendations for adult patients with Langerhans cell histiocytosis (LCH).
Most investigators have previously recommended treatment according to the guidelines given above for standard treatment of children with Langerhans cell histiocytosis. It is unclear, however, whether adult LCH responds as well as the childhood form of the disease. In addition, the drugs used in the treatment of children are not as well-tolerated when used in adults. Excessive neurologic toxicity from vinblastine, for example, prompted closure of the LCH-A1 trial.
A consensus opinion reported on the evaluation and treatment of adult patients with LCH. Discussion continues, however, particularly with regard to optimal first-line therapy with some experienced clinicians preferring to start with vinblastine and prednisone and others with alternative therapy, such as single-agent cytosine arabinoside or cladribine.[Level of evidence: 3iiiC]
Treatment of pulmonary LCH
It is difficult to judge the effectiveness of various treatments for pulmonary LCH because patients can recover spontaneously or have stable disease without treatment. Smoking cessation is mandatory because the apparent causal effect of smoking in pulmonary LCH. It is not known if steroid therapy is efficacious in the treatment of adult pulmonary LCH because reported case series did not control for smoking cessation. Most adult patients with LCH have gradual disease progression with continued smoking. The disease may regress or progress with the cessation of smoking. Some patients have been reported to respond to cladribine therapy.
Lung transplantation may be necessary for adults with extensive pulmonary destruction from LCH. This multicenter study reported 54% survival at 10 years posttransplant, with 20% of patients having recurrent LCH that did not impact survival; longer follow-up of these patients is needed. Another study confirmed an approximate 50% survival at 10 years and improved hemodynamic changes associated with pulmonary arterial hypertension, but did not alter pulmonary function testing or incidence of pulmonary edema. The best strategy for follow-up of pulmonary LCH includes physical examination, chest radiographs, lung function tests, and high-resolution computed tomography (CT) scans.
Treatment of bone LCH
As in children, adults with single-bone lesions should undergo curettage of the lesion followed by observation, with or without intralesional corticosteroids. Extensive or radical surgery leading to loss of function and disfigurement is contraindicated at any site, including the teeth or jaw bones. Systemic chemotherapy will cause bone lesions to regress and the involved teeth and jaw bones cannot reform. For those failing chemotherapy, low-dose radiation therapy may be indicated and should be tried before any radical surgery that leads to extensive loss of function and disfigurement. Radiation therapy is also indicated for impending neurological deficits from vertebral body lesions or visual problems from orbital lesions. A German cooperative radiation therapy group reported on a series of 98 adult patients with LCH, most of whom (60 of 98) had only bone lesions, and 24 had multisystem disease including bone, treated with radiation therapy.[Level of evidence: 3iiiDiv] Of 89 evaluable patients, 77% achieved a complete remission, 9% developed an infield recurrence, and 15.7% (14 of 89) experienced a progression outside the radiation field(s). A retrospective analysis of 80 patients treated with radiation therapy alone reported a 77% complete remission rate and a 12.5% partial remission rate, with 80% long-term control noted in adults. No adverse late effects were reported.
A variety of chemotherapy regimens, including cladribine, have been published in the treatment of a relatively limited number of patients. (Refer to the Chemotherapy section of this summary for more information.)
Anecdotal reports have described the successful use of the bisphosphonate pamidronate in controlling severe bone pain in patients with multiple osteolytic lesions. Successful use of oral bisphosphonates have also been described and may be a useful and relatively less toxic way of treating adult bone LCH. Because of the increased toxicity of chemotherapy in adults, bisphosphonate therapy could be used before chemotherapy in multifocal bone disease. Response of other organs, such as skin and soft tissue, to bisphosphonate therapy has been reported.
Another approach using anti-inflammatory agents (pioglitazone and rofecoxib) coupled with trofosfamide in a specific timed sequence was successful in two patients who had disease resistant to standard chemotherapy treatment.
Treatment of single-system skin disease
- Localized lesions can be treated by surgical excision, but as with bone, mutilating surgery, including hemivulvectomy, should be avoided unless the disease is refractory to available therapy.
- Topical therapies are described in greater detail in the childhood isolated skin involvement section of this summary and include topical or intralesional corticosteroid, topical tacrolimus, imiquimod, psoralen and long-wave ultraviolet A radiation (PUVA), and UVB. Therapies such as PUVA/UVB may be more useful in adults when consideration of long-term toxicity may be less.
- Systemic therapy for severe skin LCH includes oral methotrexate, oral thalidomide, oral interferon-alpha, or combinations of interferon and thalidomide. Recurrences after stopping treatment may occur but may respond to re-treatment.
- Oral isotretinoin has achieved remission in some refractory cases of skin LCH in adults.
Chemotherapy for the treatment of single-system and multisystem disease
Chemotherapy is generally used for skin LCH associated with multisystem disease in adults.
- A single-center, retrospective review of 58 adult patients with LCH reported on the efficacy and toxicities of treatment with vinblastine/prednisone, cladribine, and cytarabine. Patients treated with vinblastine/prednisone had the worst outcome, with 84% not responding within 6 weeks or relapsing within a year. The no-response/relapse rate was 59% for cladribine and 21% for cytarabine. Grade 3 or 4 neurologic toxic effects occurred in 75% of patients treated with vinblastine. Grade 3 or 4 neutropenia occurred in 37% of patients treated with cladribine and in 20% of patients receiving cytarabine.
- Etoposide has been used with some success in single-system and multisystem LCH. Use of prolonged oral etoposide in adults with skin LCH has been reported with minimal toxicity, while 3-day courses of intravenous etoposide (100 mg/m2/day) achieved complete remission in a small number of patients with resistant single-system and multisystem disease. Another study at the same center found that azathioprine was the most successful drug for localized disease in adults, with the addition of etoposide for refractory and multisystem disease.
- For patients who do not respond to front-line therapy with etoposide, cladribine is effective for adults with skin, bone, lymph node, and probably pulmonary and central nervous system (CNS) disease. The first study that used cladribine to treat refractory and recurrent skin LCH disease reported on three patients (aged 33, 51, and 57 years) who received two to four courses of cladribine at 0.7 mg/kg intravenously over 2 hours/day for 5 days. In a series of five adults (one untreated and four with refractory LCH treated with cladribine at the same dose noted above), three patients achieved a complete remission and two patients achieved a partial remission.
- An adult lymphoma treatment regimen, methotrexate, doxorubicin, cyclophosphamide, vincristine, prednisone and bleomycin (MACOP-B), was used in three patients with multisystem LCH and four with single-system multifocal bone LCH from 1995 to 2007. Total duration of therapy was 12 weeks; response was seen in all patients, two with partial response and five with complete response. Three recurrences were seen after stopping therapy. Despite the small number of patients and the retrospective nature of the study, MACOP-B may be useful as salvage therapy in adult patients with LCH and deserves further study.
- Anecdotal reports have described the successful use of the bisphosphonate pamidronate in controlling severe bone pain in patients with multiple osteolytic lesions.
- A case report suggests some benefit to treating neurodegenerative CNS LCH disease with infliximab, a tumor necrosis factor-alpha inhibitor.
- A report of stereotactic radiosurgery for the treatment of pituitary LCH in adults showed efficacy in reducing the masses. However, radiation therapy is not considered the standard of care for children with pituitary involvement. Systemic chemotherapy with cytarabine and cladribine have been the preferred treatments.
Targeted therapies for the treatment of single-system and multisystem disease
New targeted therapies under investigation include the following:
- Tyrosine kinase inhibitors: Imatinib mesylate has been effective in the treatment of four adult patients with LCH who had skin, lung, bone, and/or CNS involvement. Another adult patient with LCH did not respond to imatinib mesylate.
- RAS pathway inhibitors: The finding that most patients with LCH have BRAF and other RAS pathway mutations has led to several anecdotal reports of responses to vemurafenib, a BRAF V600E inhibitor, in adult patients with LCH, Erdheim-Chester (ECD) disease, or mixed ECD/LCH. A number of clinical trials of BRAF and other RAS pathway inhibitors in adults and children with LCH are ongoing. NCT01677741 is one such trial.
Current Clinical Trials
Check the list of NCI-supported cancer clinical trials that are now accepting patients with adult Langerhans cell histiocytosis. The list of clinical trials can be further narrowed by location, drug, intervention, and other criteria.
General information about clinical trials is also available from the NCI website.
Back to Top
Changes to This Summary (09/23/2016)
The PDQ cancer information summaries are reviewed regularly and updated as new information becomes available. This section describes the latest changes made to this summary as of the date above.
Added BRAF mutation as a prognostic factor for children with LCH (cited Héritier et al. as reference 19).
Added text about a study that reported that BRAF mutation correlates with high-risk LCH, increased resistance to initial treatment, and higher relapse rate (cited Héritier et al. as reference 25).
Added text to state that a natural history study of 58 LCH patients with pulmonary involvement found that 38% of patients had deterioration of lung function after 2 years. The most significant adverse prognostic variables were positive smoking statuses and low PaO2 levels at the time of inclusion (cited Tazi et al. as reference 20).
This summary is written and maintained by the PDQ Pediatric Treatment Editorial Board, which is editorially independent of NCI. The summary reflects an independent review of the literature and does not represent a policy statement of NCI or NIH. More information about summary policies and the role of the PDQ Editorial Boards in maintaining the PDQ summaries can be found on the About This PDQ Summary and PDQ® - NCI's Comprehensive Cancer Database pages.
Back to Top Source: The National Cancer Institute's Physician Data Query (PDQ®) Cancer Information Summaries (http://www.cancer.gov/cancertopics/pdq)