Related resources
Full-text held externally
- PMID: 15467033
- UKPMCID: 15467033
- DOI: 10.1093/jnci/djh272
Search for item elsewhere
University researcher(s)
Academic department(s)
A molecular genetic and statistical approach for the diagnosis of dual-site cancers.
Brinkmann, Dirk; Ryan, Andy; Ayhan, Ayse; McCluggage, W Glen; Feakins, Roger; Santibanez-Koref, Mauro F; Mein, Charles A; Gayther, Simon A; Jacobs, Ian J
Journal of the National Cancer Institute. 2004;96(19):1441-6.
Access to files
Full-text and supplementary files are not available from Manchester eScholar. Full-text is available externally using the following links:
Full-text held externally
- PMID: 15467033
- UKPMCID: 15467033
- DOI: 10.1093/jnci/djh272
Abstract
BACKGROUND: Concurrent tumors can be synchronous, independently derived, non-metastatic tumors or metastatic tumors. The prognosis and clinical management of patients with these different concurrent tumor types are different. METHODS: DNA from normal and tumor tissues of 62 patients with synchronous endometrial and ovarian, bilateral ovarian, or endometrial and bilateral ovarian tumors was analyzed for loss of heterozygosity and microsatellite instability using eight polymorphic microsatellite markers at loci frequently deleted in ovarian and/or endometrial cancers. A statistical algorithm was designed to assess the clonal relationship between the tumors. RESULTS: The original histopathology reports classified 26 (42%) case patients with single primary tumors and related metastatic lesions and 21 (34%) with independent primary tumors; 15 (24%) were unclassified. Genetic data identified 35 (56%) case patients with single primary tumors and related metastatic lesions, 18 (29%) with independent primary tumors, and nine (15%) that could not be typed. Excluding case patients with histopathology reports for which a clonal relationship was uncertain or was not reported, there was 53% concordance between genetic and histopathology diagnoses. Increasing the stringency of the statistical analysis increased the number of uncertain diagnoses but did not affect the proportion of discordant genetic and histologic diagnoses. CONCLUSIONS: We have developed a rapid and robust combined genetic and statistical method to establish whether multiple tumors from the same patient represent distinct primary tumors or whether they are clonally related and therefore metastatic. For the majority of case patients, histopathology reports and genetic analyses were in agreement and diagnostic confidence was improved. Importantly, in approximately one-fourth of all case patients, genetic and histopathologic analyses suggested alternative diagnoses. The results suggest that genetic analysis has implications for clinical management and can be performed rapidly as a diagnostic test with paraffin-embedded tissues.