Cancer of unknown primary: clinicopathologic correlations
Article first published online: 16 DEC 2003
Volume 111, Issue 12, pages 1089–1094, December 2003
How to Cite
BLASZYK, H., HARTMANN, A. and BJÖRNSSON, J. (2003), Cancer of unknown primary: clinicopathologic correlations. APMIS, 111: 1089–1094. doi: 10.1111/j.1600-0463.2003.apm1111203.x
- Issue published online: 16 DEC 2003
- Article first published online: 16 DEC 2003
- Received February 2, 2003.; Accepted September 9, 2003.
- unknown primary;
Cancer of unknown primary origin (CUP) accounts for 5–10% of all malignant tumors at presentation and remains the death certificate diagnosis in 0.5–5% of patients. We investigated CUP patients whose primary site remained unknown throughout the entire clinical course. We reviewed 9,436 consecutive autopsies performed between 1984 and 1999 at the Mayo Clinic, matched with 177,167 cancer patients treated in the same time period. Sixty-four patients who died of CUP underwent postmortem examination. Antemortem pathologic diagnoses were obtained in 57 patients, agreed with postmortem diagnoses in 98%, and included adenocarcinoma (n=44), undifferentiated carcinoma (n=7), squamous cell carcinoma (n=3), and others (n=3). Autopsy located the primary site in 35 patients (55%). Common primary sites were lung (n=8), the pancreaticobiliary (n=13) and GI tracts (n=9). Of 43 patients evaluated for tumor-specific therapy, only six received no further oncologic treatment and untreated patients survived a median of 57 (range 10–280) days, compared with 225 (range 19–1,129) days for patients treated with chemotherapy and/or radiotherapy (n=37). Our findings show that (1) autopsy studies provide a valuable tool for quality control in the setting of CUP, and (2) treated patients have a small but significant survival benefit.