The predictive value of imaging studies in evaluating regional lymph node involvement in Merkel cell carcinoma

Background: Merkel cell carcinoma (MCC) is a rare cutaneous neuroendocrine malignancy with high potential for nodal or distant metastatic spread. Little information exists on sensitivity and specificity of various imaging techniques when compared with the gold standard of histopathologic evaluation of the lymph node basin. Objective: We sought to further understand the value of various imaging modalities in the staging and initial workup of patients with MCC. Methods: Of 240 patients with primary MCC evaluated between 1981 and 2008, 99 had diagnostic imaging at initial presentation with biopsy-proven cutaneous MCC and had histopathologic nodal evaluation within 4 weeks of the initial scan. We conducted a retrospective chart review of these identified patients. Results: Computed tomography (n = 69) demonstrated a sensitivity of 47%, specificity of 97%, positive predictive value of 94%, and negative predictive value of 68% in detecting nodal basin involvement. Fluorine-18-fluorodeoxyglucose positron emission tomography scan (n = 33) demonstrated a sensitivity of 83%, specificity of 95%, positive predictive value of 91%, and negative predictive value of 91% in detecting nodal basin involvement. Magnetic resonance imaging (n = 10) demonstrated a sensitivity of 0%, specificity of 86%, positive predictive value of 0%, and negative predictive value of 67% in detecting nodal basin involvement. Limitations: This was a retrospective study with small sample size. Conclusion: Use of fluorine-18-fluorodeoxyglucose positron emission tomography in the evaluation of a regional lymph node basin in primary MCC is significantly more sensitive and equally specific when compared with traditional computed tomography. Both fluorine-18-fluorodeoxyglucose positron emission tomography and computed tomography are more sensitive than clinical examination alone.