Identification of the occult tumor in cancer of unknown primary (CUP): A priority based on histology.

Abstract

1605 Background: The poor prognosis of cancer of unknown primary (CUP) patients is likely to improve with knowledge of the primary site. CUP patients form a diverse clinicopathological group; however, there remains doubt as CUP may indeed represent a unique clinical classification. Methods: We used the Ontario Cancer Registry to identify all patients diagnosed with cancer from January 2000 to December 2005. We linked these patients with the Canadian Institute of Health Information Same Day Surgery and Discharge Abstract Database to identify those who were initially diagnosed with a metastatic tumour. Information on clinical and pathological characteristics including age, gender, primary tumour site, histology, and second primary were available from the Ontario Cancer Registry. We stratified results according to primary tumour site and histology. Five-year survival data were available for all patients and were obtained from the Ontario Cancer Registry. Results: Of 52,619 patients diagnosed with metastatic tumour at the time of their initial cancer diagnosis, 4,866 (9.2%) patients were diagnosed with CUP and 47,753 (90.8%) patients were diagnosed with metastasis of known primary. The 5-year Kaplan-Meier estimate of CUP overall survival (OS) differed significantly with patients diagnosed with metastasis of known primary (median OS= 1.0 versus 10.4 months, respectively, log–rank test p<0.0001). In subgroup analyses, the 5-year OS of CUP patients with adenocarcinoma (n=1,389, median OS=1.83) was significantly worse when compared to metastatic adenocarcinoma of known primary (log-rank test p<0.0001). An identical result was obtained with CUP patients of undifferentiated histology (n=3,230). The 5-year OS of CUP patients with squamous cell carcinoma (n=247, median OS=18.5) did not differ significantly with those of other squamous cell carcinoma groups of known primary (log-rank test p>0.56). Conclusions: Although CUP patients as a whole have a poor prognosis compared to other metastatic patients of known primary, distinct subsets of CUP patients have similar prognosis. These data suggest that an intensive diagnostic approach for identification of the primary is not justified for all CUP patients.