Abstract

PURPOSE
Unknown primary carcinoma takes up approximately 0.5-10% of the oncology patients evaluated, and the patients have poor survival of between 3 to 11 months. Despite the short survival, certain clinically defined subsets of patients were reported to have a better prognosis. Thus, the objective of this study was to identify prognostic factors.
MATERIALS AND METHODS
The present study was con ducted with 103 patients who were referred from January 1988 to July 1999. The primary end point was survival. The survival curves were estimated using the Kaplan-Meier method and compared using the Log-rank test and Cox's proportional hazards regression analysis.
RESULTS
Most patients had histologic evidence of ade nocarcinoma or squamous cell carcinoma. Univariate and multivariate analyses identified good prognostic factors including performance status (grade 0-2), female and adenocarcinoma with more than moderate level of differentiation. The responders of chemotherapy in squamous cell carcinoma and lung, breast, ovary -estimated- cancer showed good survival rates.
CONCLUSION
Unknown primary carcinoma tended to show a poor prognosis. However, when treatment modality of unknown primary carcinoma is to be determined through the prognostic factors, the patients quality of life can be improved through reducing the treatment side effects and economic burden on the patients.

• Keywords: Unknown primary carcinoma;Prognosis;Survival rate;Chemotherapy