BACKGROUND

Hereditary non-polyposis colorectal cancer (HNPCC) is one of the most common inherited disorders predisposing to cancer. The genes responsible for the disease have recently been cloned and characterised; their mutations induce a generalised genomic instability which is particularly evident at microsatellite loci (replication error (RER)+ phenotype).

AIMS

To investigate how to select individuals and families in the general population who should be screened for constitutional mutations predisposing to colorectal cancer.

PATIENTS/METHODS: Between 1984 and 1995, 1899 colorectal malignancies in 1831 patients were registered, and in 1721 of these (94%), family trees could be obtained. Patients and families were classified into five categories according to a more or less likely genetic basis: HNPCC; "suspected" HNPCC; juvenile cases; aspecific cancer aggregation; sporadic cases. In 18 families with HNPCC as well as in 18 with suspected HNPCC, microsatellite instability in tumour tissues and constitutional mutations of two DNA mismatch repair genes (MSH2 and MLH1) could be evaluated. RER status was studied with five markers (BAT40, D2S123, D18S57, D17S787, and BAT26) in paraffin embedded tissues. Germline mutations of MSH2 or MLH1 genes were assessed on DNA and RNA extracted from lymphomonocytic cells, using reverse transcription polymerase chain reaction, single strand conformation polymorphism analysis, and direct DNA sequencing.

RESULTS

HNPCC represented 2.6% and suspected HNPCC 4.6% of all registered colorectal neoplasms. Eleven out of 18 HNPCC families (61%) showed microsatellite instability as opposed to four (of 18) suspected HNPCC (22%; p<0.02). Three germline mutations (two in MSH2 and one in MLH1 gene) were found in three different large HNPCC families, whereas no mutations were detected in suspected HNPCC.

CONCLUSIONS

In this study of cancer genetic epidemiology, data from a tumour registry were analysed and this ultimately led to the identification and selection of families that should be tested for mutator gene mutations. With the use of a population based approach, the incidence of mutations was appreciably lower than previously reported and limited to families with full blown HNPCC. It is possible that in most families with a clinical spectrum of HNPCC (or suspected HNPCC) other DNA mismatch repair genes are involved in the pathogenesis of the disease.