A mini review on cancer of unknown primary site: A clinical puzzle for the oncologists

Cancer of unknown primary site (CUP) is a well recognised clinical disorder, accounting for 3-5% of all malignant epithelial tumours. CUP is clinically characterised as an aggressive disease with early dissemination. Diagnostic approaches to identify the primary site include detailed histopathological examination with specific immunohistochemistry and radiological assessment. Gene-profiling microarray diagnosis has high sensitivity, but further prospective study is necessary to establish whether patients' outcomes are improved by its clinical use. Metastatic adenocarcinoma is the most common CUP histopathology (80%). CUP patients are divided into subsets of favourable (20%) and unfavourable (80%) prognosis. Favourable subsets are mostly given locoregional treatment or systemic platinum-based chemotherapy. Responses and survival are similar to those of patients with relevant known primary tumours. Patients in unfavourable subsets are treated with empirical chemotherapy based on combination regimens of platinum or taxane, but responses and survival are generally poor. Copyright © 2012 Elsevier Ltd. All rights reserved.

Metastatic Cancer of Unknown Primary Site (CUP) accounts for approximately 3% of all malignant neoplasms and is therefore one of the 10 most frequent cancer diagnoses in man. Patients with CUP present with metastatic disease for which the site of origin cannot be identified at the time of diagnosis. It is now accepted that CUP represents a heterogeneous group of malignancies that share a unique clinical behaviour and, presumably, unique biology. The following clinicopathological entities have been recognised: (i) metastatic CUP primarily to the liver or to multiple sites, (ii) metastatic CUP to lymph nodes including the sub-sets involving primarily the mediastinal-retroperitoneal, the axillary, the cervical or the inguinal nodes, (iii) metastatic CUP of peritoneal cavity including the peritoneal papillary serous carcinomatosis in females and the peritoneal non-papillary carcinomatosis in males or females, (iv) metastatic CUP to the lungs with parenchymal metastases or isolated malignant pleural effusion, (v) metastatic CUP to the bones, (vi) metastatic CUP to the brain, (vii) metastatic neuroendocrine carcinomas and (viii) metastatic melanoma of an unknown primary. Extensive work-up with specific pathology investigations (immunohistochemistry, electron microscopy, molecular diagnosis) and modern imaging technology (computed tomography (CT), mammography, Positron Emission Tomography (PET) scan) have resulted in some improvements in diagnosis; however, the primary site remains unknown in most patients, even on autopsy. The most frequently detected primaries are carcinomas hidden in the lung or pancreas. Several favourable sub-sets of CUP have been identified, which are responsive to systemic chemotherapy and/or locoregional treatment. Identification and treatment of these patients is of paramount importance. The considered responsive sub-sets to platinum-based chemotherapy are the poorly differentiated carcinomas involving the mediastinal-retroperitoneal nodes, the peritoneal papillary serous adenocarcinomatosis in females and the poorly differentiated neuroendocrine carcinomas. Other tumours successfully managed by locoregional treatment with surgery and/or irradiation are the metastatic adenocarcinoma of isolated axillary nodes, metastatic squamous cell carcinoma of cervical nodes, or any other single metastatic site. Empirical chemotherapy benefits some of the patients who do not fit into any favourable sub-set, and should be considered in patients with a good performance status.

This bi-centric study aimed to determine the role of receptor PET/CT using (68)Ga-DOTA-NOC in the detection of undiagnosed primary sites of neuroendocrine tumours (NETs) and to understand the molecular behaviour of the primarily undiagnosed tumours. Overall 59 patients (33 men and 26 women, age: 65 + or - 9 years) with documented NET and unknown primary were enrolled. PET/CT was performed after injection of approximately 100 MBq (46-260 MBq) of (68)Ga-DOTA-NOC. The maximum standardised uptake values (SUV(max)) were calculated and compared with SUV(max) in known pancreatic NET (pNET) and ileum/jejunum/duodenum (SI-NET). The results of PET/CT were also correlated with CT alone. In 35 of 59 patients (59%), (68)Ga-DOTA-NOC PET/CT localised the site of the primary: ileum/jejunum (14), pancreas (16), rectum/colon (2), lungs (2) and paraganglioma (1). CT alone (on retrospective analyses) confirmed the findings in 12 of 59 patients (20%). The mean SUV(max) of identified previously unknown pNET and SI-NET were 18.6 + or - 9.8 (range: 7.8-34.8) and 9.1 + or - 6.0 (range: 4.2-27.8), respectively. SUV(max) in patients with previously known pNET and SI-NET were 26.1 + or - 14.5 (range: 8.7-42.4) and 11.3 + or - 3.7 (range: 5.6-17.9). The SUV(max) of the unknown pNET and SI-NET were significantly lower (p < 0.05) as compared to the ones with known primary tumour sites; 19% of the patients had high-grade and 81% low-grade NET. Based on (68)Ga-DOTA-NOC receptor PET/CT, 6 of 59 patients were operated and the primary was removed (4 pancreatic, 1 ileal and 1 rectal tumour) resulting in a management change in approximately 10% of the patients. In the remaining 29 patients, because of the far advanced stage of the disease (due to distant metastases), the primary tumours were not operated. Additional histopathological sampling was available from one patient with bronchial carcinoid (through bronchoscopy). Our data indicate that (68)Ga-DOTA-NOC PET/CT is highly superior to (111)In-OctreoScan (39% detection rate for CUP according to the literature) and can play a major role in the management of patients with CUP-NET.