Introduction
Histiocytic diseases include Langerhans cell histiocytosis (LCH) and non-LCH entities such as Erdheim–Chester disease (ECD). Histiocytic disorders are classified into five groups: among these, LCH and ECD belong to the L-group, cutaneous juvenile xanthogranuloma (JXG) to the C-group, and Rosai–Dorfman disease (RDD) to the R-group, with additional two groups. Extracutaneous or disseminated JXG with identifiable genetic mutations has been proposed to be ECD within the L-group.1 LCH, the most common form of histiocytosis, is now recognized as an inflammatory myeloid neoplasia.2
The majority of LCH and ECD cases harbor somatic oncogenic mutations in the MAPK pathway genes (RAS–RAF–MEK–ERK), most commonly involving BRAF and MAP2K1 mutations.3,4 In both pediatric and adult LCH, the skull bones are the most frequently affected sites.5–7 Most parietal bone lesions resolve with limited curettage alone, without the need for prosthetic reconstruction or adjuvant therapy.4 However, temporal and facial bone LCH carries a substantially higher risk of late relapse; in pediatric LCH, the relapse rate has been reported to be 43.2%. Sakamoto et al. further noted that craniofacial bone lesions in children may relapse up to the sixth occurrence, whereas spinal LCH typically relapses only up to the third episode.5 The most common CNS manifestation, central diabetes insipidus resulting from pituitary stalk involvement in LCH, may precede the eventual diagnosis of LCH by less than one year or by several years.8,9 The paradigms in the management of LCH in adults are mostly derived from pediatric experiences.4
In addition, “mixed histiocytosis,” defined as the presence of two or more histiocytic neoplasms in a single patient, either concurrently or asynchronously, has attracted increasing attention.10,11 Among these, LCH/ECD mixed histiocytosis is particularly notable, as it has been hypothesized to arise from a common progenitor cell.12 Here, we report an elderly patient who developed BRAF-positive relapsed cranial bone LCH, followed by a subsequent relapse involving the L2 vertebra, in which biopsy confirmed mixed histiocytosis.
Case presentation
In 2024, a 76-year-old woman presented with lumbago, and computed tomography revealed a lesion involving the L2 spine, which was diagnosed as the third relapse of BRAF V600E-positive LCH. Her medical history included bronchial asthma and a prior transient ischemic attack; however, these comorbidities did not influence the treatment strategy for LCH.
Regarding LCH, as summarized in Figures 1a and b1-4, the patient had first experienced polyuria and polydipsia in 2010 but attributed these symptoms to aging and did not seek medical attention. In 2015, she was diagnosed with unifocal LCH (pathologically, CD1a+, S100+) of the left temporal skull, which was totally excised without adjuvant therapy. In 2017, LCH relapsed in the right parietal skull, in which the BRAF V600E gene mutation was positive in the biopsied tissue, using the BRAF V600E mutation analysis kit (real-time polymerase chain reaction assay; BRAF RT-50, EntroGen Inc.). At that time, central diabetes insipidus related to LCH was also confirmed from a thickened pituitary stalk on magnetic resonance imaging. She was treated with 10 courses of the Special C regimen [vinblastine 0.1 mg/kg (max. 6 mg), day 1; prednisolone 2.0 mg/kg (max.60 mg), days 1–5; methotrexate 20 mg/m2, day 15; 6-mercaptopurine 1.5 mg/kg, days 1–28] q4 weeks,13 along with nasal desmopressin (1-deamino-8-D-arginine vasopressin). A second relapse occurred in the left parietal skull and was managed with 10 courses of a modified Special C regimen [methotrexate was replaced with cytosine arabinoside (60 mg/m2), day 15] q4 weeks. After the treatment, complete remission was achieved, and chemotherapy was discontinued.
A third relapse occurred at the L2 spine, nine years after the initial chemotherapy for skull LCH and three years after cessation of therapy. The lesions were visualized on 18F-fluorodeoxyglucose–positron emission tomography/computed tomography (PET/CT) (Fig. 1b4). A biopsy of the left paravertebral soft tissue adjacent to the L2 lesion was performed. Low-power histology revealed mixed areas of proliferating CD1a-positive cells consistent with LCH, alongside areas consisting of predominantly foamy histiocytes with CD1a-negative and CD68-positive staining, including Touton-like multinucleated giant cells, indicative of ECD (Fig. 2a–d). High-power examination confirmed this admixture, showing both CD1a-positive cells (Fig. 2e) and CD68-positive histiocytes, including multinucleated cells (Fig. 2f), within the same lesion.
After receiving two courses of modified CHOP (cyclophosphamide 10 mg/kg, day 1; doxorubicin 35 mg/m2, day 1; vincristine 0.05 mg/kg, day 1; prednisolone 30 mg/day, days 1–5) q4 weeks for the mixed LCH/ECD relapse,14 PET/CT revealed a poor response, with persistence of the L2 lesion (Fig. 3a) and emergence of new lesions in the left 6th rib (Fig. 3b) and the proximal metaphysis of the left femur (data not shown). In addition, the patient achieved only a partial response after five courses of cladribine (0.09 mg/kg, days 1–5) q4 weeks, prompting a switch to reduced doses of targeted therapy with daily oral dabrafenib (50 mg/day) and trametinib (1 mg/day) due to cladribine-induced myelosuppression. Ultimately, all fluorodeoxyglucose-avid lesions disappeared on PET/CT following targeted therapy. Continuation of dabrafenib/trametinib with an increase in doses is planned to maintain remission, given reports of disease recurrence upon interruption of targeted therapy.15,16
Discussion
Although mixed histiocytosis has been well recognized recently,10,11 it has not been clarified whether it was noted at the initial diagnosis of histiocytosis or developed after treatment for LCH or for non-LCH histiocytosis. This report is novel because we have described that the first skull masses were LCH and the subsequently relapsed spinal L2 lesion was mixed histiocytosis. In addition, despite mixed histiocytosis (LCH/ECD) being noted more frequently in the long bones,11 it was observed in the spinal bone in this case.
In this case, mixed histiocytosis (LCH/ECD) developed nine years after the initial chemotherapy for skull LCH. At the time of spinal recurrence, pathological examination of the lesion revealed features of both LCH (CD1a+, S100+) and JXG or ECD, which showed foamy CD68-positive histiocytes associated with multinucleated giant Touton cells, as noted in ECD (Fig. 2).17,18 The pathology of JXG and ECD can mimic each other, making histopathological distinction challenging; however, we diagnosed this case as mixed LCH/ECD based on the criteria,1 though the patient lacked typical clinical features of ECD, such as long bone pain, retroperitoneal fibrosis, hairy kidneys, and coated aorta. Analysis of reported mixed histiocytosis cases revealed LCH/ECD in 19, LCH/RDD in 1, ECD/RDD in 6, and ECD/RDD/LCH in 1 case, predominantly in adults, with pediatric cases being rare.10 In the context of LCH and JXG, a pediatric case demonstrating coexistence of both LCH and JXG cell populations within cutaneous JXG lesions has been reported.19 Additionally, disseminated JXG occurring after treatment for LCH has been described in pediatric cases20,21; however, those cases did not exhibit a mixed phenotype within the involved lesions.
In 27 of 69 reported cases of mixed histiocytosis (LCH/ECD), the diagnosis of mixed ECD was delayed, with a median interval of 4 years after the initial LCH diagnosis.11 These cases often demonstrate poor responsiveness to conventional therapy but show marked sensitivity to targeted treatments.10 Given the presence of BRAF mutations in LCH/ECD,22 molecularly targeted agents, such as BRAF and/or MEK inhibitors, could be effective.23 In the present case, the relapsed spinal lesion exhibiting mixed histiocytosis was refractory to modified CHOP therapy, whereas partial remission was achieved with a cladribine regimen. Complete metabolic remission on PET/CT was subsequently obtained with combined BRAF and MEK inhibition. However, the contribution of cladribine to the therapeutic response cannot be excluded, as the two regimens were administered consecutively and targeted therapy was initiated at low doses, as shown in the Case presentation. Despite the efficacy of targeted therapies, Pegorano et al. reported that, after a median follow-up of 71 months, 24 of 69 patients (35%) had died, indicating that outcomes in mixed histiocytosis are poorer than in LCH or ECD alone,11 suggesting that longer follow-up of our case is needed. Furthermore, important clinical aspects of the transition from LCH to mixed histiocytosis remain unknown. Unresolved questions regarding mixed histiocytosis after LCH treatment include how frequently it might be detected in older patients with careful evaluation, the role of prior chemotherapy for LCH in this transition, and whether early introduction of targeted therapy could reduce its incidence.
This report has several limitations. Although cases of mixed histiocytosis developing after treatment for LCH have rarely been described, this is a single-case report. In addition, BRAF mutation analysis was available only on the first relapsed skull LCH lesion. No comprehensive genetic analyses could be performed on other lesions, such as the further relapsed skull lesion and the spinal mixed lesion. In addition, the follow-up period for the mixed histiocytosis was relatively short.
Conclusions
This case is instructive in several respects. First, when patients initially present with symptoms of polyuria and polydipsia, a thorough search for LCH lesions is warranted. Although these symptoms are characteristic of LCH, the diagnosis may be delayed for several years. In the present case, LCH was diagnosed five years after symptom onset. Second, LCH lesions involving the skull are known to have a relatively high relapse rate. Therefore, even when skull lesions achieve remission, the possibility of subsequent recurrence in other skeletal sites cannot be excluded. Third, when LCH recurs at any site, mixed histiocytosis should be considered in the differential diagnosis. Based on these findings, we recommend routine PET/CT for systemic disease screening, along with thorough pathological examination and genetic testing, including BRAF mutation analysis of all lesions during the management of patients with LCH.
Declarations
Acknowledgement
The authors thank all the staff members who cared for the patient.
Ethical statement
The study was conducted in accordance with the ethical standards of our affiliated institution and the principles of the Declaration of Helsinki (as revised 2024). Written informed consent was obtained from the patient for publication of this case and the accompanying images.
Funding
None.
Conflict of interest
The authors declare no conflicts of interest.
Authors’ contributions
Manuscript drafting (TH, MK, SI), patient management (TH, MK, NA, SI), and pathological diagnosis (AF, NN). All authors have made significant contributions to this study and have approved the final manuscript.