Clinical characteristics and outcomes of Langerhans cell histiocytosis at a single institution in Thailand: a 20-year retrospective study

In the present study, we describe the clinical characteristics, imaging modalities, treatment regimens, and outcomes for children with LCH in a large tertiary care center for pediatric oncology in Bangkok, Thailand over a 20-year period. The incidence of LCH is approximately <6 per million persons per year in Thailand [8]. This is consistent with other countries in Asia with 3 patients per year [27] and 5–9 per million persons per year in Europe [28]. During the 20-year period of the study to the end of 2016, we found only 14 LCH patients, due to the rarity of disease. Only 14% of patients had DI, consistent with a related report of from 2% to 25% of patients with LCH in developed countries [12, 29]. However, 2 patients with LCH presenting with DI had been classified as RO− MS-LCH and 1 of these experienced MFB involvement. A related report by Imashuku et al. [19] showed MS-LCH with MFB was associated with a significantly higher incidence of DI.

The most common organ involvements for LCH are bone [12] and skin [30]. Therefore, plain X-ray radiographs [31, 32], skeletal surveys, and Tc-99m bone imaging [33] are crucial modalities and investigations of choice for patients with LCH to explore systematic involvement of the disease. We found that almost 80% of patients with LCH had been diagnosed after plain X-ray radiographs of the skull. In addition, we found Tc-99m bone imaging showed abnormal results in almost 90% of patients, and 100% for FDG-PET-CT. Tc-99m bone imaging is currently substituted by FDG-PET-CT scan [34] because it is more highly sensitive for detecting positive lesions with a very low false-positive rate among patients with LCH [14].

Interestingly, patient No. 4 with RO+ MS-LCH was 15 years old at diagnosis. He had liver, spleen, lung, vertebral, and multiple bone involvement indicated by CT and MRI. As the age is uncommon for LCH and CT/MRI was unable to distinguish active LCH lesions and other causes, FDG-PET-CT was also performed. The result showed hypermetabolic activities in active lesions, which were not clearly indicated by MRI or CT (Figure 3). In our present study, 3 patients received both Tc-99m bone imaging and FDG-PET-CT determining parallel positive results in both studies. This implied that the sensitivity of the FDG-PET-CT is very high for detecting LCH lesions compared with Tc-99m bone imaging and plain X-ray radiographs as described by Phillips et al. [34]. Moreover, FDG-PET-CT can be more useful to monitor and evaluate activity of the disease after treatment than Tc-99m bone imaging because the radiotracer uptake in Tc-99m bone imaging might persist over months to years even without active disease. However, only a small number of patients underwent FDG-PET-CT. Further exploration of the effectiveness of this imaging is warranted.

Figure 3

Treatment of LCH is tailored by organ system involvement. SS-LCH would be cured by local treatment such as curettage or surgery [35]. Low-intensity CMT in MFB SS-LCH [36] and RO− MS-LCH [18] have shown excellent outcomes. By contrast, disease remission and prognosis of RO+ MS-LCH is still unsatisfactory even using high-intensity therapy and prolonged therapy regimens [18].

The outcome of pediatric LCH in Thailand is consistent with that in developed countries in the Asia–Pacific such as Taiwan. Five-year EFS was 100% in SS-LCH for both countries. However, 5-year EFS of MS-LCH was lower in Thailand (50%) than it was in Taiwan (85.7%) [27]. Outcomes for RO+ MS-LCH are inferior to RO− MS-LCH and SS-LCH [18] especially correlated with the number of RO+. Hematopoietic involvement was associated with a worse prognosis in RO+ [37]. From our present study, patients with MS-LCH having 2 RO+; liver and spleen involvement (patient Nos. 3 and 4; Table 2) were still alive with evidence of disease reactivation, while patients with 3 RO+ (Nos. 1 and 5) died from disease progression. Patient No. 2 with 3 RO+ was ANED. This patient had not undergone a bone marrow examination, but still met the criteria of hematopoietic involvement due to anemia and thrombocytopenia. Bicytopenia might be due to other causes, i.e., an infectious process, which might not be truly explained by bone marrow involvement in LCH. We found patients with RO+ MS-LCH (Nos. 1 and 5), who had hypoalbuminemia and/or conjugated hyperbilirubinemia, but had disease progression and subsequently died from the disease. Conjugated hyperbilirubinemia could be a sign of sclerosing cholangitis with greater mortality for MS-LCH [12, 38]. Braier et al. [37] reported that hypoalbuminemia was associated with inferior prognosis in MS-LCH. By contrast, patients having disease remission, with LFT, had only abnormal unconjugated hyperbilirubinemia.

Historical treatments such as the DAL-Hx83 protocol, conducted between 1983 and 1991 in Europe, were an effective regimen especially for RO+ MS-LCH due to its high intensity [25, 26]. Currently, standard treatment of RO+ MS-LCH for Thai children has been adapted from the Histiocyte Society LCH studies. However, the survival rate for RO+ MS-LCH in our present study was approximately 20% by comparison with 84% in the LCH-III study [12, 18]. This might be the result of the small number of 5 patients in our study compared with the 285 patients in the LCH-III study as well as limited resources for complex supportive care in patients receiving high-intensity CMTs with vital organ damage in RO+ MS-LCH.

Disease reactivation is common in patients with LCH patients. Patients with MFB disease reactivation usually respond well to second-line therapy including oral 6-MP and MTX [15] as did patient No. 3 (Table 3). By contrast, prognosis is dismal in patients with disease progression in risk organs during therapy especially in those with hematopoietic and liver involvement. More intensified CMT with high-dose Ara-C may be effective for achieving disease remission [39] as it was in patient No. 4 (Table 3). Moreover, BRAF inhibitors are promising therapeutic regimen for this group of patients [40].

BRAF^V600E activating mutation has been identified in 70% of patients with LCH of myeloid origin [12]. Therefore, BRAF^V600E activating mutation is an option to confirm diagnosis of LCH in equivocal settings. Because a BRAF inhibitor demonstrated efficacy in LCH [41], detected BRAF^V600E activating mutation in LCH may facilitate the feasibility of using a BRAF inhibitor as a second-line treatment. In our setting, BRAF^V600E activating mutation was analyzed in 2 patients with RO+ MS-LCH and detected in 1 (patient No. 5) who had progressive disease despite using intensive CMT. This finding corresponded with those of Bhatia et al. who reported that BRAF^V600E activating mutation in childhood LCH correlated with MS-LCH and poor survival [5]. A multicenter prospective study combining genetic testing of BRAF^V600E activating mutation could be modified with clinical correlation particularly in an MS-LCH population. BRAF inhibitors would provide a new generation treatment to improve outcomes in this group of patients.

To our knowledge, this study is the first to record the clinical characteristics and outcomes of LCH in Thailand over 20 years. Considering the rarity of disease, variety of clinical presentations, and serious comorbidities in RO+ MS-LCH, it is our opinion that a uniformed CMT protocol and centralized care by disease-specific professionals would facilitate an optimal standard of care that would result in better outcomes.

This was a retrospective study in which some data was unavailable and might not represent the entire patient population of our institute. Moreover, the study duration was short for some patients. Patients who had not been followed at the Division of Hematology–Oncology, such as those with skin-only LCH, who were treated at the Dermatology Department, were not included in the study. These might have an effect on outcomes for the small sample size. A multicenter prospective study would improve the power and quality of data, which would more truly represent the LCH disease situation in Thailand.

LCH is a rare disease with some patients having DI. FDG-PET-CT is considered as an investigation of choice and has greater accuracy than conventional plain radiographs and Tc-99m bone imaging to detect sites of active disease and for disease monitoring after treatment. SS-LCH and RO− MS-LCH have been successfully treated using local treatment or low-intensity CMT. However, patients with RO+ MS-LCH have often encountered disease reactivation or progressive disease with a dismal prognosis. Hematopoietic system and organ of risk involvement were associated with worse prognosis. Hypoalbuminemia and conjugated hyperbilirubinemia may predict a greater mortality in RO+ MS-LCH.