CONTEXT: Primary salivary clear cell tumors comprise an uncommonly encountered subgroup of salivary neoplasia. We hypothesize that clear cell carcinoma does not represent a "monomorphic" variant of epithelial-myoepithelial carcinoma, but is distinct in terms of histogenesis and tumor biology. OBJECTIVES: To compare the clinicopathologic features of 20 cases of salivary primary clear cell tumors, including 12 clear cell carcinomas (CCCs), 7 epithelial-myoepithelial carcinomas (EMECs), and 1 clear cell myoepithelial carcinoma (CCMEC); to investigate their interrelationship with regard to myoepithelial differentiation; and to offer a diagnostic approach for distinguishing between these entities. DESIGN: Retrospective and prospective identification and review of patients diagnosed with primary salivary clear cell neoplasia and review of the English language literature. SETTING: Three academic tertiary-care hospitals. PATIENTS: We identified 12 patients with CCC, 7 with EMEC, and 1 with CCMEC. Patients included 11 men and 9 women, aged 30 to 88 years (median 72.5 years). MAIN OUTCOME MEASURES: Immunohistochemical reactivity for S100, muscle-specific actin, and calponin; ultrastructural examination when feasible; review of patient charts; and telephone interviews to establish clinical outcome. RESULTS: Clear cell carcinoma has a predilection for intraoral sites, whereas EMEC has a predilection for the parotid. All 3 of the tumor types studied have a propensity for locoregional recurrence, which can manifest decades after initial surgery. There were no mortalities among patients with CCC, even after pulmonary metastasis in 2 patients, confirming the indolent nature of this group of clear cell tumors. A meta-analysis of reported cases of CCC, EMEC, and CCMEC confirmed that EMEC and CCMEC have a much greater propensity for locoregional recurrence than CCC, despite the predilection of both for a more surgically accessible site (parotid). We found no definitive evidence of myoepithelial differentiation in CCC, indicating that it is probably morphogenetically distinct from EMEC and CCMEC, both tumors with diagnostically requisite myoepithelial differentiation. CONCLUSIONS: The initial treatment of choice for CCC, CCMEC, and EMEC is surgical resection with negative margins. Locoregional recurrence should be treated aggressively, as it is still consistent with long disease-free intervals. The lack of myoepithelial differentiation in CCC is consistent with the concept that this tumor is histomorphogenically distinct from EMEC and that it is not merely a monomorphic variant.
CONTEXT: Primary salivary clear cell tumors comprise an uncommonly encountered subgroup of salivary neoplasia. We hypothesize that clear cell carcinoma does not represent a "monomorphic" variant of epithelial-myoepithelial carcinoma, but is distinct in terms of histogenesis and tumor biology. OBJECTIVES: To compare the clinicopathologic features of 20 cases of salivary primary clear cell tumors, including 12 clear cell carcinomas (CCCs), 7 epithelial-myoepithelial carcinomas (EMECs), and 1 clear cell myoepithelial carcinoma (CCMEC); to investigate their interrelationship with regard to myoepithelial differentiation; and to offer a diagnostic approach for distinguishing between these entities. DESIGN: Retrospective and prospective identification and review of patients diagnosed with primary salivary clear cell neoplasia and review of the English language literature. SETTING: Three academic tertiary-care hospitals. PATIENTS: We identified 12 patients with CCC, 7 with EMEC, and 1 with CCMEC. Patients included 11 men and 9 women, aged 30 to 88 years (median 72.5 years). MAIN OUTCOME MEASURES: Immunohistochemical reactivity for S100, muscle-specific actin, and calponin; ultrastructural examination when feasible; review of patient charts; and telephone interviews to establish clinical outcome. RESULTS:Clear cell carcinoma has a predilection for intraoral sites, whereas EMEC has a predilection for the parotid. All 3 of the tumor types studied have a propensity for locoregional recurrence, which can manifest decades after initial surgery. There were no mortalities among patients with CCC, even after pulmonary metastasis in 2 patients, confirming the indolent nature of this group of clear cell tumors. A meta-analysis of reported cases of CCC, EMEC, and CCMEC confirmed that EMEC and CCMEC have a much greater propensity for locoregional recurrence than CCC, despite the predilection of both for a more surgically accessible site (parotid). We found no definitive evidence of myoepithelial differentiation in CCC, indicating that it is probably morphogenetically distinct from EMEC and CCMEC, both tumors with diagnostically requisite myoepithelial differentiation. CONCLUSIONS: The initial treatment of choice for CCC, CCMEC, and EMEC is surgical resection with negative margins. Locoregional recurrence should be treated aggressively, as it is still consistent with long disease-free intervals. The lack of myoepithelial differentiation in CCC is consistent with the concept that this tumor is histomorphogenically distinct from EMEC and that it is not merely a monomorphic variant.