Class 3 MAP2K1 mutations associate with relapse and multiple treatment lines in multisystem Langerhans cell histiocytosis

TO THE EDITOR:

Langerhans cell histiocytosis (LCH) is the most frequent histiocytic neoplasm in adults, with an incidence of 1 to 2 per million per year.^1,2 In the current World Health Organization classification, LCH is grouped under “Langerhans cell and other dendritic cell neoplasms,” separated from histiocyte/macrophage disorders, including Erdheim-Chester and Rosai-Dorfman diseases.³ It is characterized by an accumulation of neoplastic CD1a⁺ CD207⁺ Langerhans cells, leading to the destruction of affected tissues.^1,4,5 The discovery of alterations of the MAPK signaling pathway and the resulting proof of clonality established LCH as a neoplastic disease.⁶ Although some cases of LCH are self-limiting, others involve multiple organs and cause significant morbidity.⁷ The most frequently affected organ systems in adults are the bones, skin, lungs, and pituitary gland.^1,6, 7, 8

Data guiding clinical care of adults with LCH are mainly limited to case studies or are based on pediatric protocols. Current recommendations suggest first-line treatment with chemotherapy-based regimes such as cytarabine or cladribine.⁹ Increased understanding of MAPK (RAS-RAF-MEK–extracellular signal-regulated kinase [ERK]) pathway alterations as drivers of LCH has enabled the use of BRAF/MEK inhibitors.^6,9, 10, 11, 12

Next-generation sequencing (NGS) studies have revealed that BRAF wild-type LCH may harbor mutations in MAPK-ERK pathway genes such as MAP2K1.^10,13 Kinase fusions (BRAF, ALK, and NTRK1), BRAF duplications, insertions, deletions, and rare mutations in the phosphatidylinositol 3-kinase–protein kinase B (AKT)–mechanistic target of rapamycin (mTOR) pathway have been reported.¹⁴ BRAF^V600E mutations are found in 50% to 60% of the cases and allow successful treatment with BRAF inhibitors. MAP2K1 mutations are found in 10% to 15% of patients with LCH, permitting treatment with MEK inhibitors, particularly in patients who require rapid disease control.¹¹

Treatment with targeted therapy often leads to fast responses, but some patients show treatment resistance.¹⁵ Cytotoxic treatment can lead to long-term responses, whereas targeted kinase inhibitors have a high risk of disease recurrence after discontinuation.¹¹

Although the genomic landscape of LCH is increasingly understood, the therapeutic implications and prognostic impact of specific genetic alterations are still unclear. There is emerging evidence that class 3 MAP2K1 mutations are associated with an inferior response to currently available treatment approaches.^16,17 Here, we report 12 cases of MS-LCH, including clinical descriptions of 3 cases with a class 3 MAP2K1 mutation.

We analyzed 41 patients with LCH treated at our hospital since 2014 and included in the Swiss Histiocytosis Registry. All patients provided consent (local ethics committee approval number: 2024-02235). A total of 12 patients were diagnosed with multisystemic (MS)-LCH. Disease characteristics were extracted from the medical records. Testing for somatic mutations was performed using mutation-specific immunohistochemistry for BRAF^V600E and targeted DNA sequencing (Sanger sequencing and targeted NGS). NGS was performed with a commercially available amplicon-based panel (Oncomine Comprehensive Assay version 3; Thermo Fisher Scientific) or a custom hybrid capture panel designed for lymphoid and histiocytic neoplasms. Data analysis and visualization were performed in R version 4.4.2 software.

A total of 12 of the 41 (27 %) patients from our registry had MS-LCH, of which 3 showed a class 3 MAP2K1 mutation, all with p.E102_I103del. The median age of our population with MS-LCH was 45 years, and 75% were males. Clinical and treatment characteristics of patients are summarized in supplemental Table 1.

Figure 1 shows the follow-up of all 12 patients with MS-LCH who were analyzed. Of the 3 patients with class 3 MAP2K1 mutations (P01, P02, and P06), patient P01 was most intensively treated, receiving 4 treatment lines because of repeated disease relapses after different treatments (cytarabine, cobimetinib, hydroxyurea, and cladribine), leading to a complete response at the last follow-up. Patient P02 received 3 systemic treatment lines (cytarabine, cobimetinib, and hydroxyurea), achieving a complete response but relapsed 18 months later with a single bone lesion, which was treated with local therapy using triamcinolone and later radiotherapy. Patient P06 was initially treated with cytarabine. Because of progressive bone disease at a high-risk site (C1 vertebra), he underwent radiotherapy, and treatment was changed to cobimetinib and finally cladribine because of relapsing lung disease, leading to ongoing treatment response (Figure 1; supplemental Table 2). No patient received dual BRAF/MEK inhibitors or an ERK inhibitor. Overall, patients with class 3 MAP2K1 mutations relapsed sooner than patients without, with a median time to relapse after the first therapy of 397 days (interquartile range, 302-579) vs 507 days (interquartile range, 324-1160). All 3 eventually relapsed, whereas 37% of the patients with non–class 3 MAP2K1 showed durable responses and were more likely to experience indolent disease, although the median follow-up time was <3 years for 4 of the 9 patients in this group (median follow-up of 38 months in those with non–class 3 MAP2K1 vs 109 months in those with class 3 MAP2K1 mutations; Figure 2).

Figure 1.

Treatment course of all patients with MS-LCH: class 3 MAP2K1 mutation vs non–class 3 MAP2K1 mutation. MTX, methotrexate.

Figure 2.

Cumulative incidence of relapse according to class 3 MAP2K1 mutation upon first-line treatment.

In this study, we report a cohort of 12 patients with MS-LCH, including 3 patients harboring the class 3 MAP2K1 p.E102_I103del mutation. This specific deletion, located in exon 3, affects both the autoregulatory domain and catalytic core of MAP2K1, resulting in constitutive kinase activation.¹⁵ Class 3 MAP2K1 mutations are characterized by RAF-independent autophosphorylation of MEK and are associated with elevated downstream ERK phosphorylation.15, 16, 17

These mutations are defined according to RAF dependency. RAF-dependent class 1 mutants (D67N, P124L/S, and L177V) are phosphorylated and activated by RAF. RAF-regulated class 2 mutants (F53_Q58del, F53L, Q56P, K57E/N, C121S, L177M, and E203K) are partially phosphorylated by RAF but retain their kinase activity independent of RAF. RAF-independent class 3 mutants (I98_I103del, I99_K104del, E102_I103del, and I103_K104del) autophosphorylate and activate downstream signals.¹⁵ These mechanisms may contribute to differences in response to treatment, depending on the underlying mutation. A summary of all MAP2K1 mutations found in histiocytic disorders is presented in supplemental Table 3.

Our findings support the challenging nature of treating patients with class 3 MAP2K1 mutations. All 3 patients carrying the p.E102_I103del mutation showed progression after first-line therapy, whereas disease control was achieved in 4 of the other 6 patients who received systemic treatment. Two other patients with non–class 3 mutations demonstrated favorable outcomes with either local treatment or watchful waiting approaches. Only 1 patient (P04), notably carrying an atypical BRAF mutation (p.Asn486_Pro490del), required 3 lines of systemic therapy. The treatment resistance observed in patients with class 3 MAP2K1 mutations may be attributed to the reduced binding affinity of current allosteric inhibitors. Given the elevated ERK phosphorylation observed in these cases, ERK inhibitors could represent a promising alternative therapeutic strategy for relapsing patients. This approach is supported by recent findings from Diamond et al showing an objective response upon ERK inhibition in class 3 MAP2K1 mutant mouse models and individual patients with MEK inhibitor–resistant disease.¹⁸ Some data suggest that class 3 MAP2K1 mutated disease may respond to allosteric MEK inhibitors such as cobimetinib, in a dose-dependent manner.^19,20 In our cohort, 2 patients (P01 and P06) could not be treated with the full dose of cobimetinib (60 mg/d) because of toxicity (supplemental Table 2). Patient P02 tolerated the full dose, but treatment was stopped after 6 months because of progressive disease.

Current pathophysiologic models emphasize the importance of the differentiation stage of hematopoietic precursor cells at which driver mutations occur.^21,22 The heterogeneity in disease course in our population with MS-LCH suggests that factors beyond the cell of origin may influence disease progression. The tumor microenvironment may play a crucial role, potentially varying among patients with different phenotypes, and might be more dependent on mutation type than the cell of origin.^21,23,24 Recent findings by van Haalteren et al regarding lymphoid cells carrying driver mutations in histiocytic disorders have provided new insights into the complex pathobiology of these rare diseases.²³ Future research should focus on identifying additional factors that determine the extent and severity of histiocytic neoplasms, including the role of the tumor microenvironment, epigenetic changes, immune response modulation, and the impact of specific genetic alterations on treatment response. Understanding these complex interactions may lead to more effective therapeutic strategies for patients with resistant disease.

Despite the limitations of our small sample size, reflecting the rarity of MS-LCH, the marked difference in long-term remission rates between patients with non–class 3 vs class 3 MAP2K1 mutations underscores the critical role of genetic alterations in determining treatment response and selection in MS-LCH. These findings also emphasize the importance of comprehensive genetic profiling in adult MS-LCH, particularly through NGS in BRAF-WT cases, to identify actionable therapeutic targets.

Conflict-of-interest disclosure: M.H. reports grants from GE HealthCare, the CRPP AI Oncological Imaging Network of the University of Zurich, and funds from the Alfred and Annemarie von Sick legacy for translational and clinical cardiac and oncological research. E.L.D. reports unpaid editorial support from Pfizer Inc and serves on an advisory board for Opna Bio, all outside the submitted work. The remaining authors declare no competing financial interests.