International expert consensus recommendations for the diagnosis and treatment of Langerhans cell histiocytosis in adults

Langerhans cell histiocytosis (LCH) is a rare hematologic disorder that affects children and adults. LCH was one of the first histiocytic disorders to be recognized as a hematopoietic neoplasm by the World Health Organization (WHO) due to the establishment of clonality.¹ Subsequently, the discovery of BRAF-V600E mutation in about 50% of cases solidified the notion that LCH is a neoplastic disease.² LCH is included among the “L-group” of the revised classification of histiocytic disorders and is grouped along with Erdheim-Chester disease (ECD).³ Most of our current understanding of LCH is derived from pediatric studies, with a paucity of data examining adult counterparts. The exact incidence of adult LCH is undefined, estimated at 1 to 1.5 cases per million population per year.^4,5 Since the previous publication of adult LCH guidelines in 2013,⁶ several key advances have occurred in our understanding of the disease, including the discovery of MAPK pathway (RAS-RAF-MEK-ERK) mutations beyond BRAF-V600E^7,8 and efficacious treatments using kinase pathway inhibitors.^9,10 Tissue next-generation sequencing (NGS) studies have revealed that BRAF- wild-type LCH may harbor several other mutations in the MAPK-ERK pathway genes (eg, MAP2K1, MAP3K1, ARAF, NRAS, KRAS).⁷ Similar to ECD, kinase fusions (BRAF, ALK, NTRK1), BRAF- duplications, insertions, deletions, and rare mutations in the PI3K-AKT-mTOR pathway have been reported in LCH as well.^7,11,12 These molecular discoveries have implications for the diagnosis and management of LCH, and in conjunction with other recent studies on LCH in adults, have necessitated an update of the existing recommendations.

An international, multidisciplinary group of physicians and scientists convened at the annual Histiocyte Society meeting on 3 November 2019 to initiate the process of updating the adult LCH consensus recommendations. The group was comprised of experts from multiple subspecialties with experience in the clinical care/research of LCH and related histiocytic disorders (supplemental Table 1). The proposed update was based on 1) novel data since prior guideline publication in 2013 and 2) group members’ evolving experience with LCH in adults. We generated key recommendation statements (consensus statements or CS#1-32) (Table 1) and recorded the votes by coauthors as “agreement,” “disagreement,” and “not-qualified to answer” using a modified Delphi approach as previously described (supplemental Appendix 1).¹³ The degree of consensus for each statement was categorized into: A (strong consensus: ≥95%), B (consensus: 75-94%), and C (majority agreement: 50-74%).

Pulmonary LCH (PLCH) may occur as part of multisystem LCH or, more commonly, as an isolated disease (single-system PLCH).¹³ Single-system PLCH was initially thought to be an inflammatory/reactive disorder given the near-universal association with smoking¹⁴ and instances of remission upon smoking cessation.¹⁵ Earlier studies assessing clonality of single-system PLCH showed mixed results,^16,17 but more recent studies have identified recurrent MAPK pathway alterations in 85% of these lesions, including BRAF-V600E in 36% to 50% of cases.^18,19 The latter finding supports the contention that single-system PLCH is indeed a clonal process; more studies are needed to understand its natural history and the role of smoking in disease modulation.²⁰ 

Given the significant variability in clinical manifestations, various attempts have been undertaken to classify LCH in pediatric and adult populations.^3,21 To harmonize the definitions in light of recent evidence in adults with LCH, we propose the classification outlined in Table 2. It should be noted that our classification differs from the 2016 Histiocyte Society classification³ in that we do not differentiate risk-organ disease as a separate entity due to a lack of data on prognostic and therapeutic implications in adults in the era of targeted therapies. The proposed classification system differentiates between unifocal (solitary lesion) and single-system multifocal disease (>1 lesion). In multisystem LCH, ≥2 organs/organ systems are involved. We also propose a distinct category of single-system PLCH due to its unique etiopathogenesis and treatment. PLCH with extrapulmonary disease is classified under multisystem disease. Every attempt should be undertaken to appropriately define the disease extent using radiographic studies mentioned in the baseline evaluations section below.

Although LCH has distinct clinical and morphologic characteristics compared with ECD and Rosai-Dorfman disease (RDD), several reports of coexistence of these disorders as “mixed” or “overlap” histiocytosis exist. A case series noted LCH-like lesions cooccurring with RDD, most commonly in young patients and involving lymph nodes.²² In 90% of these instances, LCH constituted a minor component of the lymph node sections (<10%) compared with RDD. LCH/ECD overlap was reported in a French series of 23 patients, where the ECD component was found to occur either subsequently or concomitantly with LCH but never before it.²³ In contrast with the LCH/RDD overlap described earlier, the LCH and ECD lesions occurred less frequently in the same anatomic site (26%). These cases had a high frequency of BRAF-V600E mutations in the LCH (69%) and ECD (82%) lesions.²³ Pediatric reports have suggested that LCH may also occur concomitantly with juvenile xanthogranuloma.²⁴ These data highlight that the histiocytic disorders may potentially lie on a clinicopathologic spectrum; cognizance of the mixed histiocytoses may warrant further investigations, such as additional biopsies, especially at recurrence or unusual disease sites in order to diagnose these entities and tailor their management.

Several institutional and population-based studies have shown high incidence and prevalence of other malignancies in patients with LCH, including myeloid neoplasms (acute myeloid leukemia,⁴ myeloproliferative neoplasms, chronic myelomonocytic leukemia,^25,26 etc), lymphomas, and solid organ malignancies (especially lung and thyroid cancer).^27-31 LCH of the thyroid gland can cooccur with papillary thyroid cancer harboring BRAF-V600E mutations.^29,32 LCH foci have also been found incidentally around resected specimens of renal cell carcinoma.³⁰ Patients with PLCH, in particular, have a high incidence of concomitant lung carcinomas, likely due to their smoking history.³³ LCH-like lesions and aggregates of Langerhans cells (LCs) may cooccur in lymph node biopsies of lymphomas and likely represent a “bystander” phenomenon where the lymphoma subtype drives treatment.^34-36 Clinicians should be aware of the occurrence of second malignancies at any time point (before, during, or after) in relation to the LCH diagnosis as this may influence surveillance and management.

LCH has diverse manifestations in adults, with varying frequencies of disease sites based mostly upon institutional retrospective series (Figure 1; Table 3; supplemental Appendix 2).^{6,15,33,37-81} Clinicians should be familiar with the clinical and radiographic features of LCH and attempt to pursue comprehensive evaluation at baseline given the implications for diagnosis and management of the disease.

The diagnosis of LCH usually relies on a histopathologic review of lesional tissue. All attempts should be made to obtain a tissue biopsy, not only to achieve a diagnosis but also for mutational analysis that can direct therapy; however, in instances without classic pathologic findings or inaccessible/insufficient tissue specimen, a presumptive diagnosis can be made based on characteristic clinical, radiographic, or molecular features (Table 4; CS#1-2). Definitive diagnosis of single-system PLCH may be obtained through transbronchial, transthoracic, or, more commonly, by surgical lung biopsy.^82,83 In cases where a lung biopsy is not considered safe, a diagnosis of single-system PLCH can be rendered in the presence of typical radiologic findings on high-resolution computed tomography (HRCT) of the chest in a smoker after careful ruling out of other etiologies (Table 4; CS#1).^83,84 In such cases, circulating DNA analysis for MAPK-ERK mutations may help solidify the diagnosis. Among patients who have undergone FDG-PET/CT, the safely accessible sites of greatest FDG hypermetabolism should be targeted for biopsy. We recommend acquiring multiple (5-10) core biopsies (16-18 gauge) instead of fine-needle aspiration, given the variable cellularity of lesions. Bone specimens present unique challenges, including crushed samples and interference with mutational analysis from decalcification, including false-negative immunohistochemistry (IHC) results for BRAF-V600E. To circumvent this issue, it is recommended to either preserve a fragment of tissue without acid decalcification or use ethylenediaminetetraacetic acid (EDTA)-based decalcification to maintain DNA integrity.

On tissue biopsy, the LCH infiltrate is characterized by clusters of intermediate-sized cells with reniform nuclei, frequent longitudinal nuclear grooves, dispersed chromatin, and abundant eosinophilic cytoplasm.^3,48,70 Occasional multinucleated giant cells may be present. A reactive inflammatory cell background is common and includes eosinophils, small lymphocytes, and macrophages (Figure 2). In PLCH, the neoplastic LCs are preferentially located in and destroy the walls of distal bronchioles.⁸⁴ In certain anatomic sites, in particular bone and lung, there is an evolution of the lesions over time from cellular infiltrates to “burned-out” lesions with patchy cellularity, increased fibrosis, and foamy histiocytes.^3,85 Within skin lesions, the neoplastic cells of LCH are distinguished from reactive LC populations by their enlarged nuclei, sheet-like collections with involvement of the epidermis, and lack of dendritic processes.⁸⁶ In lymph nodes, LCH involves and expands the sinuses, with their microanatomic location being a critical distinction from the reactive LCs that are present in the paracortex of lymph nodes with dermatopathic change.⁷² 

The phenotype of LCH is similar to that of benign LCs, with the expression of CD68 (Golgi dot-like staining), S100 (nuclear and cytoplasmic), CD1a (surface), langerin/CD207 (cytoplasmic), and ZBTB46 (a marker of classical dendritic cells).⁸⁷ They are commonly negative for monocyte-macrophage markers CD14, CD163, and OCT2.⁸⁸ In contrast to reactive LCs, the neoplastic cells of LCH show more diffuse expression of cyclin D1.⁸⁶ To diagnose LCH, the minimal panel of antibodies should include CD1a and langerin/CD207, with the correct pathologic pattern and correlating clinical/radiographic findings.⁷⁰ In cases where BRAF-V600E is the driver mutation, this can be identified by IHC using a mutant-specific antibody clone VE1 with moderate to strong cytoplasmic staining (Figure 2).^89,90

LCH can be distinguished from other histiocytic neoplasms such as ECD and RDD by morphologic and immunophenotypic features (Table 5).^13,48 On morphology, ECD shows bland xanthogranulomatous inflammation and variable degrees of fibrosis,⁹¹ and RDD shows enlarged histiocytes with abundant pale-staining cytoplasm and emperipolesis. In contrast to LCH, the histiocytes of ECD and RDD express CD163 and lack expression of CD1a, langerin, and ZBTB46. ECD usually shows strong expression of factor XIIIa, and RDD is characteristically positive for S100, fascin, and OCT2.⁸⁸ Mixed histiocytoses can occur with both LCH and either ECD or RDD components present within the same biopsy or involving different sites in the same patient.⁹² Another rare differential diagnosis to consider is indeterminate cell histiocytosis, which is likely a distinct neoplasm with the absence of langerin/CD207 expression and primarily cutaneous involvement.⁹³ Recently, ETV3-NCOA2 translocations have been identified as a unique driver of this disease entity.⁹⁴ 

The distinction between LCH and Langerhans cell sarcoma (LCS) is currently based on morphologic criteria, with the WHO classification defining LCS as having “overtly malignant cytologic features,”^3,95 which may be subjective in some cases. Chromosome and molecular genetic studies may be helpful, as LCH generally has a normal karyotype and <5 somatic mutations,⁸ whereas malignant histiocytic neoplasms like LCS typically have frequent chromosomal gains or losses.⁹⁶ 

Multiple techniques are available for molecular analysis in LCH, and a stepwise approach may be the most practical for judicious use of limited material. Given its high frequency, we recommend BRAF-V600E mutation testing in all patients (CS#16). This can be pursued by IHC or molecular methods, including quantitative polymerase chain reaction (qPCR), droplet digital PCR (ddPCR), or NGS. Allele-specific qPCR and ddPCR offer much greater sensitivity than pyrosequencing, especially given the low variant allele fractions (<5%) seen in LCH.⁹⁷ The mutant-specific immunostain VE1 for BRAF-V600E has variable sensitivity and specificity in LCH tissue biopsies based on the tumor cellularity and intensity of staining (CS#17).^98,99 Therefore, we suggest confirming a negative or equivocal IHC test by one of the molecular methods. BRAF-V600E mutations may also be detected by ddPCR in bronchoalveolar lavage specimens in patients with PLCH.⁹⁸ For LCH specimens without BRAF-V600E, we recommend NGS for mutations and fusions in genes of the MAPK-ERK and related pathways (CS#18). In cases with insufficient tissue specimens, peripheral blood cell-free DNA testing is a reasonable alternative, bearing in mind that its correlation with tissue NGS may be higher for BRAF-V600E compared with other mutations (CS#19).^100-102 

A thorough history and physical examination are essential to assess potential sites of disease and symptomatology in LCH. We recommend FDG-PET/CT imaging (skull vertex to toes) in all patients with a diagnosis or suspicion of LCH to ascertain organ involvement and target disease sites for biopsy and response assessment (CS#3). Among patients with suspected single-system PLCH, HRCT can often be diagnostic and provides comprehensive architectural details of cystic lesions seen in advanced stages of the disease (CS#6). Among patients with dyspnea or disproportionately abnormal diffusing capacity for carbon monoxide (<60% predicted), screening for pulmonary hypertension should be conducted (CS#9,10). Additional baseline laboratory and organ-specific testing should be performed as outlined in Figure 3 (CS#11-14). Among patients with otherwise unexplained peripheral blood count abnormalities, a bone marrow biopsy is warranted to rule out a concomitant hematologic neoplasm (CS#15).

Unifocal LCH is often curable in adults, and local therapies may be sufficient in many cases (CS#20).¹⁰³ Surgical resection may be preferred depending on the disease location but should only be attempted if not heroic or mutilating. Most skull lesions can resolve after limited curettage without the need for prostheses. If surgical resection is successful, there is no established role for adjuvant treatments. Depending on the location, intralesional corticosteroid therapy (eg, methylprednisone or triamcinolone)¹⁰⁴ or, especially in the case of an accompanying soft tissue component, radiation therapy (eg, 10-20 Gy)^105,106 may offer an excellent and durable response. If symptoms disappear quickly after diagnostic biopsy, remission can occur even without further therapy.¹⁰⁷ Hence, in these cases, careful observation is an appropriate approach. Isolated skin nodules, lymph nodes, oral mucosal lesions, and isolated polyps of the gastrointestinal tract are usually completely resectable and require no further therapy; however, for anogenital and gingival lesions, oral methotrexate or hydroxyurea may be favored.^77,108-110 Symptomatic cutaneous LCH is generally treated with topical or oral therapy. The choice of therapy depends on the severity of the disease and patient/physician preference. Topical triamcinolone can be used for disease involving small areas of the skin. Locally extensive LCH of the skin is usually less sensitive to topical or systemic steroid therapy but usually responds to treatment with hydroxyurea or low-dose methotrexate.^110,111 In refractory cases, local therapy with imiquimod, 20% N-mustard, or irradiation can be effective.^{47,105,112,113} The treatment of unifocal LCH involving sites that are not easily amenable to local therapies (nervous system, pituitary, liver, spleen, heart, etc) is not well defined and likely warrants systemic therapies similar to multisystem disease (CS#21).

Management of patients with single-system PLCH should be individualized and take into consideration respiratory symptoms, the degree of lung function impairment, and the radiologic extent of disease. In all cases, smoking cessation (all forms, including marijuana and vaping) is mandatory, which remains difficult to achieve in practice (CS#23). It is often the only required intervention that results in complete remission.¹¹⁴ Combined inhaled corticosteroid and long-acting β2-agonist therapy may provide benefit to patients with wheezing and moderate obstructive lung disease. Pneumothorax management may be particularly difficult as it may recur in half of the patients.¹¹⁵ Surgical pleurodesis should ideally be avoided in these young patients who may need lung transplantation in the future. Lower respiratory tract infection is a common cause of deterioration for PLCH and should be promptly treated. Seasonal influenza, pneumococcal, and COVID-19 vaccinations are recommended for all patients barring contraindications. Systemic therapy should be instituted for patients with persistent or worsening symptoms who do not respond to smoking cessation or are unable to quit smoking (CS#24). Vinblastine has limited efficacy in PLCH.¹¹⁶ Cladribine (2-chlorodeoxyadenosine) use was associated with improved lung function in multiple studies and is the preferred systemic therapy.^117-120 Its long-term effects and tolerance are being evaluated in an uncontrolled phase 2 study (NCT01473797). The role of vasodilator therapy in severe PLCH with pulmonary hypertension is not well established and should be reserved for centers with expertise.¹²¹ Patients with advanced single-system PLCH may be candidates for lung transplantation, although the disease may relapse in the transplanted lungs (CS#25).¹²² The role of MAPK pathway inhibition in single-system PLCH warrants further exploration; review of existing literature showed a single case report of successful treatment using a MEK inhibitor, trametinib.¹²³ 

Long-term remissions can be achieved with multifocal and multisystem disease, and several therapeutic principles generally apply. For patients with multisystem LCH that is asymptomatic and without critical organ involvement (eg, brain, liver, lung) or end-organ dysfunction, an initial period of careful observation and surveillance is reasonable to determine the tempo of the disease. Treatment should be initiated when symptoms ensue or if there is impending organ compromise. The optimal treatment of single-system multifocal and multisystem disease remains undefined in adult LCH because of a paucity of prospective studies. Most retrospective studies have used chemotherapy regimens adopted from the pediatric LCH experience, and some existing treatment guidelines embody this approach. However, pediatric regimens may cause greater toxicities in adults, leading to discontinuation and subsequently diminished efficacy. Moreover, the codified risk stratification system¹²⁴ has not been fully validated in adults. Therefore, it is unclear whether the risk-based treatment strategy implemented in the pediatric setting can be applied to adults.

Treatment regimens that have been published in a series of ≥10 patients are listed in Table 6.^{110,116,120,125-130} In general, these can be categorized into 2 groups, vinca alkaloid/steroid-based and antimetabolite-based (cladribine, cytarabine) regimens. The vinca alkaloid/steroid-based regimens are similar to or adaptations of pediatric regimens. Although the response rates are generally high, between 70% and 80%, this treatment has been associated with high failure/relapse rates (40% to 80%).^116,126 While 2-drug regimens are relatively well tolerated, the more complex regimens with ≥4 or more drugs have, as expected, a higher toxicity profile. The only prospective trials available are for antimetabolite-based regimens. The largest trial investigated the methotrexate + cytarabine regimen, with a response rate of 88% and a 3-year progression rate of 32%. However, almost half (48%) developed febrile neutropenia while a third (33%) developed grade 3/4 thrombocytopenia.⁴⁴ The only other prospective trial used cladribine and showed a response rate of 75% with a median duration of 3 years and febrile neutropenia rate of 15%.¹²⁵ Recent retrospective studies have confirmed the high response rates (79% to 90%) among patients who received cladribine as frontline or at disease relapse.^120,127 Limiting cladribine treatment to 4 cycles may reduce the rate of severe cytopenias, and maximum response is attained usually within 4 cycles of treatment.¹²⁰ Another nucleoside analog, clofarabine, has demonstrated efficacy in refractory pediatric LCH¹³¹ and is being evaluated in a clinical trial that includes adults (NCT02425904).

Other systemic treatments, including bisphosphonates for bone-predominant disease,¹³² immunomodulatory imide drugs (IMiDs),^133-135 hydroxyurea, oral low-dose methotrexate, hematopoietic stem cell transplant (HSCT), and BRAF and MEK inhibitors, have been reported in smaller series or case reports. In a combined adult and pediatric series of patients with skeletal disease, the use of bisphosphonate therapy (alendronate, pamidronate, or zoledronate) was associated with durable resolution of bone pains as well as radiographic ossification and normalization of bone lesions in the majority of the cases (92%). About half of the patients received only 1 infusion of zoledronate, and the rest were treated over months to a few years.¹³² Of the IMiDs, both thalidomide and lenalidomide have been reported to elicit responses, with some lasting over a year. However, maintenance therapy was required, and relapses occurred almost uniformly after discontinuation of therapy.^133,134 Multifocal skin disease is usually responsive to hydroxyurea, IMiDs, oral low-dose methotrexate therapy with or without prednisone/6-mercaptopurine.^{110,111,133,136} Patients with asymptomatic lymph node-only disease can be monitored closely without treatment and may result in spontaneous remissions, especially in the case of cervical lymphadenopathy.¹⁰⁹ There are 6 published cases of adults with refractory LCH (age range 21-35 years) who underwent HSCT, with most cases achieving sustained remissions.^136-141 Data on the efficacy of BRAF and MEK inhibitors in frontline and relapsed setting are emerging with several published cases. The majority had BRAF-V600E mutation and received vemurafenib, although there are also reports using cobimetinib, trametinib, and dabrafenib + trametinib.^{9,10,136,142-146} Responses were near-universal regardless of the site of involvement, and most had a sustained complete remission as long as treatment continued.

For unifocal disease, local therapy such as surgical excision may be curative if it can be safely undertaken. Unifocal disease not amenable to local therapies (pituitary, CNS, heart) should be treated similar to multisystem disease with systemic therapies (CS#21). Radiation therapy is an effective option for symptomatic bone-only disease with a limited number of lesions (<3). Other treatment options for multifocal cutaneous or bone disease include immunosuppressive agents like low-dose methotrexate, hydroxyurea, 6-mercaptopurine, or IMiDs. Bisphosphonates are an option for bone-only disease without the risk of structural compromise or instability (CS#26-27). For disease recurrence not amenable to local therapies or unresponsive to immunosuppressive agents, systemic chemotherapy is recommended (Table 6; CS#28). The treatment of single-system PLCH often requires collaboration between pulmonologists and oncologists. For multifocal and multisystem disease, we propose the treatment approach using the algorithm provided in Figure 4. Among systemic treatment options for LCH, chemotherapy using either cladribine or cytarabine is preferred because of relatively high overall response rates and the potential for long-term remissions with limited cycles of treatment (CS#28). Vinblastine + prednisone is a reasonable alternative, although it is not the preferred choice of therapy in adults given the high risk of relapse and potential for peripheral neuropathy. Data on BRAF and MEK inhibitors are still evolving, but advantages include near-universal and quick responses, making them a preferred treatment for patients who require a rapid reversal of organ compromise. Similar to the experience in ECD, discontinuation of these agents will likely result in disease relapse, necessitating prolonged therapy.¹⁴⁷ However, carefully selected patients may be good candidates for low-dose chronic administration or treatment breaks. Because most patients will require a dose reduction and lower doses of BRAF and MEK inhibitors are quite effective, it is reasonable to start at half of the dose currently used to treat patients with melanoma. With the advent of targeted inhibitors, the role of HSCT in adults with refractory LCH is unclear.

For patients with CNS disease, cladribine, higher doses of cytarabine, IV methotrexate-based regimens, or kinase inhibitors are preferred (CS#29,30). The treatment data on adults with neurodegenerative disease are lacking, but BRAF/MEK–inhibitor- or cytarabine-based chemotherapy may be preferred based on limited pediatric experience.^148,149 For LCH-associated sclerosing cholangitis, the optimum therapy is unknown. Due to poor outcomes with chemotherapy, targeted agents are preferable. Prolonged remissions following liver transplant have been reported^50,150-152; therefore, early transplant referral should be undertaken in otherwise fit patients. In a study of mostly pediatric patients with LCH, disease relapse in the transplanted liver was seen in 8% of cases.¹⁵² For disease that is relapsed or refractory following chemotherapy, it is reasonable to consider an alternate chemotherapy agent or a kinase inhibitor based on patient-specific factors and drug availability (CS#30).

Response assessment and surveillance strategies in LCH vary based on multiple factors: organ involvement at diagnosis, type of treatment used, and patient preference (CS#31). For patients with abnormal FDG-PET/CT findings at diagnosis, the first response assessment should occur in 2 to 3 months of initiation of therapy and every 3 to 6 months thereafter (CS#32). In select cases, organ-specific imaging (CT, HRCT, ultrasound, or magnetic resonance imaging) may be necessary to assess for a response. Additional testing may be necessary based on clinical manifestations and organs involved, with specific tailoring for single-system PLCH (Table 8). There are no established response criteria for adults with LCH, and pediatric criteria set forth by the International LCH Study Group¹⁵³ may not be applicable in adults. Recent adult histiocytosis clinical trials and retrospective studies have used clinical and radiographic criteria that are preferred in adult LCH (Table 7).^9,10,154 Any degree of improvement in clinical or radiographic findings is considered a response. It may take several months to attain the best response, especially with oral therapies. Once the best response is achieved and disease has stabilized, the frequency of imaging studies can be individualized. As endocrinopathies involving the pituitary or hypothalamus may not resolve with LCH-specific therapy, it is necessary to monitor the hormone levels and continue replacement, preferably in collaboration with an endocrinologist.

In the collective experience of the authors, patients with LCH frequently endorse disabling chronic musculoskeletal pain, fatigue, and mental fogging that diminish their quality of life. In some instances, previously healthy adults who develop a discrete single system lesion of LCH find that they never return to full functional status. In the experience of the authors, the site of a bone lesion remains an enduring focus of pain in up to half of patients despite the resolution of metabolic activity on PET/CT imaging. In many cases, the constellation of symptoms satisfies the criteria for the diagnosis of chronic fatigue syndrome^155-157 or fibromyalgia.¹⁵⁸ Such patients may benefit from referral to specialized services, if available.

In new patients, caution should be exercised in promising complete resolution of symptoms, and timely preparation for lifelong symptom management is warranted. Patients with persistent pain will typically achieve unsatisfactory responses to most available analgesic drugs. Some patients may notice a relief of fatigue with methylphenidate, which appears to be safe for use in the long term. Optimum management of concomitant depression and anxiety is also recommended and may improve quality of life, fatigue, and the burden of chronic pain. The experience of the authors is that, in many instances, unmanageable symptoms of patients are incorrectly viewed as a result of a maladaptive psychological state, while the reverse is, in reality, more likely. Physicians specializing in histiocytosis should acknowledge and validate the patient’s experience of these problems as “real” (ie, in some way related to disease) even if they are difficult to fully palliate. This is often met with lasting appreciation even if therapeutic options are limited. Lack of recognition of the severity of chronic ill health is a problem for many patients who outwardly look well; a reminder that LCH is classified as a hematologic neoplasm is often useful when empathy or services are not forthcoming. Early referral for psychological support and counseling for coping with chronic pain is recommended and may be the most beneficial strategy for many patients.

The prognosis of LCH depends on the extent of organ involvement and the response to treatment. Unifocal LCH has an excellent prognosis, with 5-year overall survival of >90%. In a series of 44 unifocal LCH cases, 20% had disease recurrence at another site but were able to be successfully treated.¹⁰³ The natural history of single-system PLCH is variable and difficult to predict; ∼50% of patients experience a decrease of pulmonary function at 5 years of follow-up, and some will develop obstructive lung disease.¹⁵⁹ The 5-year overall survival of single-system PLCH was reported to be around 70% to 80% in earlier studies,^33,41 and seems more favorable in the current era.^13,160 Approximately half of the deaths in PLCH occur due to respiratory failure,³³ and follow-up testing is essential to identify patients who may develop progressive disease. The prognosis of multisystem/disseminated LCH is generally considered good with ∼90% 5-year overall survival rates.^4,41 A recent phase 2 trial of methotrexate and cytarabine suggested that liver and spleen involvement may be associated with poorer prognosis in adults, although none of the patients received BRAF or MEK inhibitors.⁴⁴ The prognosis of rare phenotypes such as sclerosing cholangitis and neurodegenerative LCH is guarded but may improve in the era of targeted therapies. As the LCH-specific survival of adult patients improves with novel therapeutics, it is critical to evaluate and monitor them for long-term effects of therapies and disease, including new-onset morbidities and second cancers that may result in premature mortality.

This work was supported by the Walter B. Frommeyer, Jr., Fellowship Award in Investigative Medicine, University of Alabama at Birmingham (G.G.).

Contribution: All authors participated in outlining the manuscript and providing expert recommendation grading; G.G. wrote the manuscript draft with input from R.S.G., K.L.R., J.L.P., R.V., J.V.L., M.L.H., X.-X.C., P.M., G.K., J.H., M.C., and K.L.M., E.L.D., and M.G., ; J.R.Y., A.T., and C.W.C. provided the clinical and radiographic images; and all authors then contributed to the editing of the manuscript.

Conflict of interest disclosures: M.L.H. is a member of the external advisory board at Novartis and Sobi Corporation; P.M. received fees for lectures/advisory boards and research grants from Amgen and fees for lectures/advisory boards from Glaxo, Lilly, UCB, Elpen, and Vianex. G.K. received honorarium and research grants from IPSEN, Novartis, Pfizer, Novo, and Shire. K.L.M. is a member of the medical advisory board for SOBI Corporation. E.L.D. received editorial support from Pfizer, Inc, and paid advisory board membership with Day One Biopharmaceuticals. All remaining authors declare no competing financial interests.

Correspondence: Eli L. Diamond, Department of Neurology, Memorial Sloan-Kettering Cancer Center, 160 East 53rd St, 2nd Floor Neurology, New York, NY 10022; e-mail: diamone1@mskcc.org; and Michael Girschikofsky, Internal Medicine I (Hemostasis, Hematology and Stem Cell Transplantation and Medical Oncology), Ordensklinikum Linz Elisabethinen, Fadingerstr 1, 4020 Linz, Austria; e-mail: michael.girschikofsky@ordensklinikum.at.