Expression of TLE1 in Malignant Melanoma With Spindle Cell Morphology: A Potential Diagnostic Pitfall

Introduction

Transducer-like enhancer of split 1 (TLE1) immunohistochemistry (IHC) is widely used as a biomarker of synovial sarcoma.^1-6 There are 4 TLE genes that have all been highly expressed in synovial sarcoma. ⁵ TLE1 IHC has been reported to be a sensitive diagnostic marker for synovial sarcoma. However, its specificity has recently been questioned as it has been found to be positive in other neoplasms.^1,5-8

The most common subtype of synovial sarcoma is monophasic, consisting of monomorphic spindle cells.^5,8 Given this subtype resembles other malignant spindle cell tumors, the diagnosis can be morphologically challenging and the differential may be broad including malignant peripheral nerve sheath tumor, dermatofibrosarcoma protuberans, and hemangiopericytoma/solitary fibrous tumor, among others.^5,6,8 The histologic appearance of spindle cell or desmoplastic melanoma can mimic synovial sarcoma, and so far, its TLE1 expression has not been fully elucidated. The aim of this study was to investigate the expression of TLE1 in melanomas, particularly with spindle cell morphology.

Materials and Methods

The study protocol was approved by the University of Massachusetts Medical School Institutional Review Board. A search of the surgical pathology files from 2009 to 2017 resulted in 57 cases of melanomas diagnosed with a spindle cell or desmoplastic component. After review, 8 cases had no definitive dermal spindle cell component and 7 cases had insufficient tissue remaining and were excluded from the study. A total of 42 melanomas with a spindle cell or desmoplastic dermal/subcutis component were evaluated by TLE1 IHC.

Immunohistochemical studies were performed on 5-µm sections of formalin-fixed, paraffin-embedded tissue using the Ventana Ultra (Ventana Medical Systems, www.ventana.com) staining instrument. Deparaffinization and antigen retrieval were done using Cell Conditioner #1 applied for 64 minutes. Primary antibody TLE-1 Mouse Monoclonal Ready-to-Use (Cell Marque cat. # 401M-18, clone 1F5) was incubated for 32 minutes. Ventana UltraView DAB (Ventana Medical Systems) reagents were applied for detection, visualization, and counterstaining. Sections of synovial tissue with known positivity for the target protein were used as positive controls.

TLE1 positivity was defined as dark brown staining confined to the nucleus and graded as 3+ (strong staining of >50% of cells identified at 4×), 2+ (moderate staining of 10% to 50% of cells at 4× or >50% of cells staining above background at 10×), and 1+ (weak staining of <50% of cells above background at 10×). No visible staining was scored as 0. Staining for TLE1 was considered positive for 2 to 3+ and negative for 0 to 1+. ¹

Results

Nuclear TLE1 expression was identified in 24 (57%) of the 42 melanoma cases with spindle cell morphology, with 10 cases (24%) graded as 3+ and 14 cases (33%) graded 2+ (Table 1 and Figure 1). TLE1 was considered negative in 18 cases (43%), of which most (n = 14 [33%]) exhibited 1+ staining and only a small subset did not show any staining (0, n = 4 [10%]).

OPEN IN VIEWER

Table 1.

TLE1 Expression in Melanoma With Spindle Cell or Desmoplastic Morphology (n = 42).

Negative [0 to 1+] (43%)		Positive [2+ to 3+] (57%)
0 (No staining)	1+ (Weak staining in <50% of cells at 10×)	2+ (Moderate staining in 10% to 50% of cells at 4× or >50% of cells at 10×)	3+ (Strong staining in >50% of cells at 4×)
4 (10%)	14 (33%)	14 (33%)	10 (24%)

OPEN IN VIEWER

Figure 1.

Transducer-like enhancer of split 1 (TLE1) grading. (A) Spindle cell melanoma with (B) negative staining for TLE1 (0) (100×, hematoxylin-eosin [H&E] and TLE1 stain, respectively). (C) Spindle cell melanoma with (D) weak staining for TLE1 in <50% of cells (1+) (100×, H&E and TLE1 stain, respectively). (E) Spindle cell melanoma with (F) weak staining in >50% of nuclei noticed at 100× (2+) (100×, H&E stain and TLE1 stain, respectively). (G) Spindle cell melanoma with (H) strong staining of >50% of tumor nuclei identified from low power (100×, H&E and TLE1 stain, respectively).

Discussion

Synovial sarcoma is characterized by a diagnostic translocation that produces the fusion of SYT-SSX oncogenes.^4,7,8 The Wnt/β-catenin signaling pathway is strongly associated with synovial sarcoma and TLE1 is an important protein in this pathway signaling.^5,7,8 Therefore, TLE1 is highly expressed in synovial sarcoma and is another diagnostic biomarker, which is particularly useful when molecular studies to identify the translocation cannot be performed.^5,7,8

The TLE1 gene plays a role in epithelial and neuronal cell differentiation.^9,10 A recent study of TLE1-deficient mice found evidence that TLE1 plays a role as a modulator of NF-κB-mediated inflammation as well as systemic and local cytokine and chemokine expression. ¹¹ A relationship with cancer progression was also seen and using an orthotopic transplant model and TLE1-deficient mice exhibited increased melanoma tumor growth compared with wild-type mice. ¹¹

Although it has been found helpful in the diagnosis of synovial sarcoma with a sensitivity of 78% to 100%, TLE1 labeling has been noted in other spindle cell tumors, questioning its specificity.^1,4-8,12 TLE1 expression has been reported, most often as weak or focal in rare or a small subset of malignant peripheral nerve sheath tumor, hemangiopericytoma/solitary fibrous tumor, fibrosarcoma, pleomorphic sarcoma, acral myxoinflammatory fibroblastic sarcoma, endometrial stromal sarcoma, epithelioid sarcoma, leiomyosarcoma, liposarcoma, myxofibrosarcoma, rhabdomyosarcoma, Ewing sarcoma, chondrosarcoma, carcinosarcoma, and gastrointestinal stromal tumor, albeit there are also cases reported with strong labeling.^1,4-8,12 Additionally, TLE1 expression has recently been identified in invasive breast cancer with high expression significantly associated with HER2 positivity. ¹³ Furthermore, certain populations of normal cells are variably labeled by TLE1, including endothelium, mesothelium, perineurium, basal keratinocytes, and adipocytes. ⁷ Indeed, in the present study, TLE1 was noted to also highlight sebaceous glands and follicular epithelium.

Terry et al ¹ evaluated 11 melanomas for TLE1 expression and found 4 (57%) cases stained weakly (1+), defined as weak to moderate nuclear staining in <50% of cells and overall considered negative, which is the same scoring system we used in our study. Another study included 2 melanoma cases but did not report the results. ² In the present study, 57% of melanomas with a spindle cell or desmoplastic component were positive for TLE1. As the study’s cutoff for negative staining included those graded as 1+, a total of 38 (90%) spindle cell or desmoplastic melanomas actually showed at least weak staining (<50% of cells above background at 10×). Although in most instances differentiating melanoma from synovial sarcoma can be done based on the site of involvement and morphology, the frequent expression of TLE1 may pose as a diagnostic pitfall, particularly in a subcutaneous/soft tissue malignant spindle cell neoplasm in which a metastatic melanoma and synovial sarcoma are both being considered in the histologic differential. Expression of TLE1 may result in a misdiagnosis of synovial sarcoma, and melanoma should be ruled out on other morphological and immunohistochemical grounds. Additionally, molecular studies confirming the diagnostic translocation (X;18) for synovial sarcoma should be performed.

Conclusion

In conclusion, TLE1 is a sensitive biomarker for synovial sarcoma, but also identifies a significant proportion of melanomas, acknowledging an imperfect specificity. Furthermore, more than half of the spindle cell or desmoplastic melanomas displayed strong (2-3+) TLE1 labeling, which may pose as a diagnostic pitfall when both melanoma and synovial sarcoma are being considered.