DIAGNOSIS
Langerhans Cell Sarcoma
DISCUSSION
Langerhans Cell Sarcoma (LCS) is a very rare malignancy both in the head and neck and other sites. It is thought to arise from Langerhans cells: antigen-presenting cells of the mucous membranes, dermis, lymph nodes, and thymus. Langerhans cells are members of the histiocyte family (Howard et al., 2016), and neoplasms of Langerhans cells are divided into the relatively more indolent Langerhans Cell Histiocytosis (LCH) and the more aggressive LCS. LCS may arise de novo, from LCH, or from/in association with a variety of hematolymphoid malignancies (Pan and Xu, 2019; Facchetti et al., 2017; Howard et al., 2016). LCS patients tend to be older than those with LCH (Pan and Xu, 2019). Some reports suggest males are diagnosed with LCS approximately twice as frequently as females, a pattern inverse to LCH (Nakamine et al., 2016).
In the head and neck, the median age of presentation for LCS is 56.5, with only a minority of patients presenting with localized disease (Howard et al., 2016). The upper aerodigestive tract is the third most commonly affected site (behind lymph nodes and skin) (Howard et al., 2016). Since many patients present with disseminated disease, identification of a precise single site of origin can be challenging.
Histiocytic and dendritic cell neoplasms demonstrate overlapping features (Pan and Xu, 2019). Specifically, LCS must be distinguished from LCH. Both entities contain large, epithelioid cells with grooved (coffee bean-like) nuclei. The reactive eosinophils characteristic of LCH may or may not be present in LCS. Tumor cells of LCS demonstrate greater pleomorphism and mitotic activity than LCH and both atypical mitoses and necrosis may be present. Generally >50 mitoses/10 high-powered fields can be identified, although this is not a strict diagnostic criterion (Swerdlow et al., 2017). Immunohistochemically, LCS and LCH both demonstrate S100, CD1a, and Langerin positivity, although the degree to which these may be expressed in LCS varies (Pan and Xu, 2019).
Other dendritic/histiocytic lesions to consider in the differential diagnosis of LCS include: histiocytic sarcoma (positive for CD68 and CD163, negative for S100, CD1a, and Langerin), indeterminate cell sarcoma (positive for S100 and CD1a, negative for Langerin), interdigitating dendritic cell sarcoma (S100 positive, negative for CD1a and Langerin, generally arises in a lymph node with characteristic paracortical growth) and follicular dendritic cell sarcoma (originates in cervical lymph nodes and expresses CD21, CD23, and CD35) (Pan and Xu, 2019; Facchetti et al., 2017.). Non-dendritic/histiocytic lesions to be excluded in the differential diagnosis of LCS in the head and neck include malignant melanoma and sarcomatoid (spindle cell) squamous cell carcinoma. Birbeck granules are characteristic of LCH and LCS upon ultrastructural examination (Facchetti et al., 2017).
Mutations in the RAS-MAPK and MTOR signaling pathways underlie a large proportion of histiocytic-dendritic cell neoplasms (Shanmugam et al, 2019). BRAF V600E mutations have been described in 50-60% of LCH cases and have also been described in LCS (Pan and Xu, 2019). A possible putative role for Merkel cell polyomavirus has been posited in LCS and LCH for at least a proportion of cases (Murakami et al., 2014).
Interestingly, there is an association between hematolymphoid neoplasms and LCS/LCH which includes but extends beyond co-occurrence. The possibility of morphologic "transdifferentiation" of a variety of leukemias/lymphomas into LCS and/or LCH has been well described (Chen et al., 2013; Chang et al., 2013; Skala et al., 2019). In some cases, there has been molecular confirmation of a common clonal origin to both neoplasms, illustrating the underlying relationship between the lesions in a single patient (Nakamine et al., 2016). This remains a possibility in the patient presented here, as he has a history of CLL, though the relevant and necessary molecular studies to confirm transdifferentiation have not yet been performed. The clinical and therapeutic implications of distinguishing a transdifferentiated neoplasm from a second primary tumor are uncertain.
LCS is frequently lethal, especially when detected with disseminated disease. Five-year disease specific survival is reported as 61%, 24%, and 0% for patients with local, local-regional, and disseminated disease, respectively. Multiple different therapeutic modalities are available for LCS and include surgery, chemotherapy, and radiotherapy. Rarely, bone marrow transplants may be considered (Howard et al., 2016; Chung et al., 2013). BRAF V600E targeted therapies have been explored in LCS with variable success (Mourah et al, 2015; Lorillon et al., 2016). Disease specific and disease-free survival are not significantly different for patients with head and neck LCS versus LCS of other sites.
This patient has subsequently undergone selective neck dissection (right levels II-V) which revealed metastatic LCS in one lymph node (level II) as well as extensive involvement with CLL/SLL.
REFERENCES
Contributed by Vincent Cracolici, MD and Raja R. Seethala, MD