|Genomic catalog of bladder tumors may aid diagnosis and therapy|
| By Marina
April 14, 2000
he latest installment of what will be a genome-wide catalog of genetic alterations associated with bladder cancer offers glimpses of the future potential for improving diagnosis and tailoring treatments.
One of the standard ways of diagnosing bladder cancer is to search a patient’s urine for cells that have sloughed off from the bladder wall and examine them for tell-tale changes in appearance. But at early stages in their progression, cancer cells often look identical to healthy cells and mild forms of cancer can look similar to cancers that later become invasive.
Several studies have identified key genetic changes that give pathologists molecular information to augment histological analysis. For instance, an alteration of the p53 gene is already helping physicians track the disease (a p53 mutation signals aggressive disease) and choose appropriate treatments.
Striving for a more comprehensive catalog of genetic alterations associated with bladder cancer, Bogdan Czerniak and his colleagues at the M.D. Anderson Cancer Center at the University of Texas in Houston have set about the arduous task of correlating genome-wide deletions with histological changes in the physical appearance of tumor cells. “I see this as a roadmap linking the human genome data with the histologic concept of urinary bladder cancer,” says Czerniak. The researchers have scanned 5 chromosomes for deletions, which are relatively easy to detect, and hope to finish scanning all 23 chromosomes within the next few months.
To build the roadmap, Czerniak and his team have scrutinized cancerous bladder tissue from five patients and discovered cells in various stages of malignant development throughout the bladder wall. Then they produced detailed maps showing where cells are located at various stages of development.
They looked for deletions in the DNA of cellshealthy looking or notand superimposed these findings onto the maps. Tracking more than 200 genome locations within the five chromosomes that are most frequently associated with bladder cancer, the researchers identified DNA alterations in 33 chromosome regions. A whopping 45 percent of the alterations occurred in healthy-looking cells, suggesting that cancer cells can go through many genetic changes before betraying themselves to the pathologist’s eye.
“I think it's the first demonstration where a simultaneous genetic and histological mapping was done in a whole bladder specimen,” says Dan Theodorescu, at the University of Virginia in Charlottesville, who studies bladder cancer progression. “That's the novelty here. I think it's quite exciting.”
The generality of Czerniak’s findings, however, is yet to be tested. “They only looked at five patients,” says Frederic Waldman at the Comprehensive Cancer Center at the University of California in San Francisco. “So one has to be cautious in interpreting the significance of these results in the whole population of patients with bladder cancer.”
To move beyond this limitation, Czerniak is looking forward to the completion of the catalog. Some of the newfound, early genetic modifications may be useful in screening patients in high risk groups, such as smokers. In addition, some of the modifications associated with later cancer stages could help physicians predict whether a tumor is likely to become invasive and to plan therapy accordingly.
Waldman thinks one of the most important next steps is to identify which of the chromosome alterations discovered by Czerniak are genetic hot spots. Not all of the alterations will necessarily be involved in cancer progressionfurther experiments should help pinpoint the important ones. In addition, an analysis of bladder tissues from patients at different stages in the progression of their symptoms could expand the database’s utility, says Waldman.
Although not everyone agrees that Czerniak’s casting of such a wide net is the best approach to understanding and diagnosing bladder cancer, his roadmap is likely to provide a powerful resource to complement other approaches. “Most common human cancers are genomic diseasesthey involve multiple chromosomes and they go through multiple complex steps,” says Czerniak. “So it seems logical that we use genomic approaches to build a description of the progression of human cancer.”
. . .