A heterogeneous group of cancers for which the site of origin remains occult after detailed investigations is defined as carcinomas of unknown primary origin (CUPs). Because patients with CUP have a dismal prognosis, we have analyzed CUPs to highlight the implication of clinicopathologic factors related with patient survival. A total of 106 consecutive cases of CUP were collected. A two-step strategy of immunohistochemistry to assess CUPs according NCCN Guidelines is used to separate carcinomatous tumors and subtype carcinomas. Median follow up of censored patients was 26 months. Median survival time of whole patients was 13 months (95% confidence interval [CI], 8.43 - 19.1 months), with one, two and fiveyear survival rate of 53.7%, 35.1%, and 30.5%, respectively. Factors related with shorter overall survival was adenocarcinoma histology (P=0.001), increased CA19- 9 (P=0.003), increased CEA (P=0.047), increased LDH (P < 0.001), CK20 positivity (P=0.002), presence of bone metastasis (P=0.017), metastasis not confined to the lymph nodes (P=0.015), unfavorable clinical group based predefined category (P=0.017), and patients with no treatment (P < 0.001). Multivariable analysis with cox regression model revealed factors related with overall survival; cases belonged to Culine's poor risk group (HR, 3.88; 95% CI, 1.75-8.64; P=0.001) and CK20 positivity (HR, 3.31; 95% CI, 1.42-7.70; P=0.005). In conclusion, the CK20 expression profile is a prognostic factor in patients with CUP and initial stratification of patient with Culine's model may provide a prognostic information in these patients. Assessment of clinical implication of these factors in the context of site specific therapy needs to be evaluated.
All Science Journal Classification (ASJC) codes