[Molecular changes in development and progression of urothelial carcinoma].

The multistep development of malignant tumors with increasing accumulation of genetic alterations from preneoplastic lesions to invasive carcinoma is an accepted model of carcinogenesis. Urothelial carcinoma of the bladder and upper urinary tract is an interesting model system to study tumor development and progression. There is both clinical and molecular evidence that urothelial carcinoma can be divided in two groups with different characteristics: 1) well differentiated genetic stable and mostly superficial papillary tumors with frequent recurrence and low progression risk and 2) poorly differentiated mostly solid and invasive tumors with a high number of genetic alterations. The aim of the studies summarized in this manuscript were: 1) to identify genetic changes with importance for urothelial carcinogenesis by investigation of preneoplastic and early neoplastic urothelial lesions, 2) to define molecular markers for progression of papillary carcinoma, and 3) to investigate the importance of microsatellite instability and mismatch repair defects for development of tumors of the upper urinary tract which are frequently found within the HNPCC syndrome. The investigation of urothelial hyperplasias, dysplasias and carcinoma in situ by deletion mapping (LOH analysis), FISH, CGH and mutation detection revealed that urothelial hyperplasias are precursors of papillary bladder tumors and flat dysplasias can be regarded as precursors of solid bladder cancers. In bladder cancer patients, there are genetic alterations already detectable in histologically inconspicous urothelium. The investigation of papillary bladder cancers for progression-related genetic alterations showed that mutations in the wnt pathway genes APC and beta-Catenin do not play an important role in urothelial carcinogenesis. Instead, the expression of the antagonistic wnt-related genes WIF-1 and sFRPI is strongly reduced in bladder cancer and associated with poor prognosis in papillary tumors. Loss of sFRP1 expression is not due to gene mutation but to epigenetic inactivation by promoter hypermethylation and is related to deletions at chromosome 8p12. In contrast to bladder cancers, tumors of the ureter and renal pelvis develop through a different genetic pathway in 30% of cases. The loss of mismatch repair proteins (hMSH2, hMLH1 or hMSH6) leads to a mutator phenotype with accumulation of genetic alterations in multiple repetitive sequences (microsatellite instability, MSI). MSI-positive tumors were predominantly located in the ureter and showed a lower tumor stage and grade and papillary and frequently inverted growth pattern. They were more frequent in females and younger patients and had a higher incidence of colorectal carcinomas and other tumors in the family. Chromosome 9 deletions, a hallmark of urothelial carcinomas, and the number of chromosomal alterations as detected by CGH analysis were significantly less frequent in these tumors. These data strongly suggest a distinct molecular pathway in the development of upper urinary tract tumors with mutator phenotype.