Analysis of a diagnostic strategy for patients with suspected tumors of unknown origin.

PURPOSE Diagnostic strategies designed to identify the underlying primary malignancies in patients with unknown primary tumors (UPTs) have relied on retrospective analyses. We analyzed 879 consecutive patients referred with suspected UPTs to determine the yield and cost of a limited diagnostic evaluation, assess the contribution of specific studies to diagnosis, and analyze the survival patterns of patients in whom the primary tumor was diagnosed. PATIENTS AND METHODS Data from patients with a suspected UPT were entered into a computerized data base, and the patients underwent a predefined limited diagnostic evaluation. Primary malignancies were diagnosed by pathologic review alone or by pathologic criteria plus a physical or radiographic finding. Survival was measured from diagnosis, estimated using the Kaplan-Meier method, and compared using the Cox-Mantel log-rank test. RESULTS A primary tumor was found in 179 of 879 patients (20%). The survival duration of patients in whom the primary tumor was diagnosed was superior to that of patients in whom the primary tumor remained unknown. Specific patient subsets contributed most to the improved survival duration of the group in which the primary tumor was found, including lymphoma patients diagnosed solely by pathologic criteria and female patients with primary breast or ovarian cancer. The cost of diagnosis was mostly due to the extensive use of computed tomography. Except for ovarian cancer, computed tomography rarely identified treatable primary tumors. CONCLUSION The limited diagnostic evaluation used in this study identified patients with treatable malignancies and increased the survival duration of a population of suspected UPT patients. Primary malignancies with the best survival can be diagnosed through careful pathologic review and focused evaluations for breast and ovarian cancer in women.