FDG PET/CT and Conventional Imaging Methods in Cancer of Unknown Primary: an Approach to Overscanning / 대한핵의학회잡지

PURPOSE: To compare the performance of fluorine-18 fluorodeoxyglucose positron emission tomography and computed tomography (FDG PET/CT) with conventional imaging methods (CIM), including computed tomography (CT), magnetic resonance imaging (MRI), and mammography (MMG) in cancer of unknown primary (CUP).METHODS: A total of 36 patients with CUP, who referred to our clinic for a FDG PET/CTscan, were enrolled in this study. Thirty of the patients were also examined through either diagnostic CT/MRI and/or MMG. The diagnostic performance of both methods for the primary cancer location was analyzed. The results of FDG PET/CT and CIM were compared based on the standard reference of the histopathology and/or clinical and laboratory follow-up.RESULTS: The primary cancer locations were detected in 24 patients (66.6%, 24/36) by FDG PET/CT, whereas CIM identified the locations in 16 patients (53.3%, 16/30). Sensitivity, specificity, PPV, NPV, and accuracy rates of the detection of the primary tumor localizations were as follows: 83, 70, 89, 58, and 79% for FDG PET/CT; 70, 62, 84, 42, and 68% for CIM, respectively. There was no statistical significance between modalities regarding any of the categories in 30 patients.CONCLUSION: FDG PET/CT detected the primary tumors of the patients with CUP more than CIM did. However, the difference between them was not found to be statistically significant. It may be considered that FDG PET/CT scan can be performed as a first-line tool in the initial diagnosis of the patients with CUP and to add radiodiagnostic imaging in selective cases.We conclude that if the first-line examination of a CUP patient has been already performed by a CIM and the result was negative or inconclusive, FDG PET/CT can be considered to avoid unnecessary imaging procedures.

A Case of Squamous Cell Carcinoma of Thymus with Negative CD5, Cytokeratin 7 and Positive Cytokeratin 13 / 결핵

The most common anterior mediastinal tumors originate from the thymus. Among them, thymic carcinomas occur as an early local invasion and wide spread metastases. However, when squamous cell carcinoma in the thymus or mediastinum is identified, an occult primary lung cancer must be excluded because the histologic types resemble those found more typically in the lung. CD5 and cytokeratin immunohistochemical staining is useful in evaluating biopsy samples from those tumors. Spuamous cell carcinoma of an unknown primary origin in the mediastinum is a rare occurrence and there are only a handful of case reports. Here we describe a case with an anterior mediastinal mass of squamous cell carcinoma with unknown primary origin. A resection of the mediastinal mass without an association with the lung was performed. Immunohistochemical stainings were positive using cytokeratin 13, and negative using CD5 and cytokeratin 7. This was followed by chemotherapy for presured thymic carcinoma.

The Efficacy of PEEL Chemotherapy and Identification of Favoranble Subgroups in Patients with Carcinomas of Unknown Primary Origin / 대한암학회지

PURPOSE: In order to evaluate the efficacy of PEFL (cisplatin, etoposide, 5-fluorouracil and leucovorin) chemotherapy and to identify favorable subsets, we conducted a phase II trial of PEFL regimen for patients with carcinomas of unknown primary origin (CUPO). MATERIALS AND METHODS: A total of 38 patients was enrolled in this study between May 1995 and September 1997. CUPO was defined as the presence of metastatic cancer documented in the absence of an identifiable primary site. All entered patients were treated with PEFL combination chemotherapy (cisplatin 20 mg/m(2)/day i.v, days 1-5, etoposide 100 mg/m(2)/day i.v. days 1, 3 & 5, 5-fluorouracil 800 mg/m(2)/day continuous infusion days 1-5, and leucovorin 20 mg/m(2)/day i.v, days 1-5; repeated every 4 weeks). The end points of this study were response and survival. To identify favorable subsets, univariate and multivariate analyses were perfonned. RESULTS: Among 38 patients, 29 had measurable lesions. Three (11%) out of 27 evaluable patients had a complete response and 7 (26%) had a partial response (response rate 37%; 95% confidence interval 19~55%). The median survival of the total 38 enrolled patients was 9.1 (range; 1~21.9+) months. The median progression-free survival of the 27 evaluable patients was 5.3 (range 0~ 16.0) months. Among total 132 cycles of chemotherapy, leukopenia of grade II or more was observed in 15% and thrombocytopenia of grade I in 4%. There was no treatment-related death. Main non-hematologic toxicities were nausea/vomiting (79%), stomatitis (70%), and neurotoxicity (33%). The prognostic factor analyses identified 2 favorable subgroups; One was the patient group whose disease had poorly differentiated histology and presented in cervical lymph node. This group of patients had better response rate than other patients (response rate; 71% vs 25%, p=0.02). The other was the patient group who had normal tumor markers (CEA, CA 125 and CA 19-9). This group of patients had better survival than other patients(median survival; 14.8 vs 8.4 months, p=0.05). CONCLUSION: PEFL chemotherapy seemed to be moderately active and tolerable in patients with CUPO. Among heterogenous patients with CUPO, the subset with cervical lymph node and poorly differentiated histology responded better to the chemotherapy and those with normal tumor markers tended toward longer survival.