Urologic Oncology

SLIDES & TRANSCRIPTS
Friday, December 05, 2003

Bladder Cancer

Lee Moore, M.D.

Slide 1:

Hello. I am going to talk today and change a little bit from just straightforward epidemiology to talk about molecular epidemiology of bladder cancer and focus on some research that we recently conducted on two known bladder carcinogens, arsenic exposure, environmental arsenic exposure and tobacco

Slide 2:

and the type of study designs that were used for this research is the case-case study design using tumor tissue.

We use micro-dissected tumor tissue and they are very useful for etiologic studies when animal models are not appropriate and this is particularly the case with arsenic exposure because in arsenic exposure inorganic arsenic is not a known carcinogen in rats or mice but in humans it is carcinogenic at several sites, and they can also be used to determine if the genetic alterations observed are different, in this case the arsenic exposed group or tobacco-exposed group than unexposed cases thus strengthening the causal inference

Slide 3:

and the study I am going to talk about is these cases from an Argentine population-based study which was conducted in a region of Argentina where people had drinking water arsenic concentrations between 0 and 300 micrograms per liter and the second region of Chile in the region of Antofagasta where people were historically exposed for about 30 years to very high levels, very high concentrations. So, the arsenic exposure in these cases, the Chileans were all over 600 micrograms per liter and the Argentine group was between 50 and 300 for the exposed and then I had unexposed cases from each population that were exposed throughout their lifetime to less than 10 micrograms per liter, and just for those of you who aren't into public health the current maximum contaminant level now is 50 and it was recently lowered to 10 micrograms per liter.

So, these are very highly exposed people

Slide 4:

and in the study I used two bladder tumor markers. The first method was called comparative genomic hybridization which is a method of looking at the entire genome, gross chromosomal changes, gains and losses as well as amplifications and also p53 mutational spectrum analysis to look at one particular gene and look for mutational patterns within a gene and determine if those correlate with exposure.

Slide 5:

So, just to explain to you how comparative genomic hybridization works we extract DNA from tumor tissue and also normal tissue and you label the tumor tissue in this case with green, add normal tissue with red and then you co-hybridize the two DNA samples together and areas where you see more tumor DNA you see a green-to-red ratio of greater than one. If they are the same the ratio is one. In the area where there is a loss, where there is a deletion in the tumor you see a ratio of less than one,

Slide 6:

and this is just a picture of what it looks like and you use a digital camera to take pictures of your images and also to measure the amount of fluorescence which is emitted by either the green or red probes.

Slide 7:

So, this is the clinical variables for all bladder cancer patients but the groups are very similar except for gender. There were more females in the Chilean group

Slide 8:

and when we looked at stage and grade there were increases of CGH alterations with both stage and grade suggesting that there are more alterations with disease progression which is pretty well known and this has already been published, but with smoking we didn't see any type of difference in the number of CGH alterations.

Slide 9:

But when you look at arsenic exposure category we saw a positive trend with maximum 5-year peak arsenic concentration in drinking water and this is the number of CGH alterations per tumor.

Slide 10:

Then when we stratified the group by high-grade tumors versus low-grade tumors and there was virtually no trend in the low-grade tumors, but then in the high-grade tumors that is where we saw all of the association and what was very interesting is that even some of the tumors that were TA tumors which normally are superficial tumors and have between zero and two changes per case we had some with almost 27 alterations, and this is almost unheard of in non-arsenic exposed countries,

Slide 11:

and we also looked at particular chromosome alterations, gains that were associated with smoking history and arsenic exposure and you can see that there were particular gains here at 8q. I think that is cycline D1. A lot of the alterations correlated with grade and stage

Slide 12:

and with arsenic exposure the most strongly correlated loss was this loss of 17p.

So, we decided to look at the p53 gene and see if there was any type of a mutational pattern that was associated with exposure.

Slide 13:

So, just in conclusion patients with higher arsenic exposure showed higher levels of chromosomal instability with the exposure and that most arsenic associated changes were also associated with stage and grade and this could suggest that people who are exposed to arsenic may have more tumors that may behave more aggressively and perhaps that may explain some of the higher mortality that is seen in countries with arsenic exposure, and the strongest association was seen in the chromosome arm 17p.

Slide 14:

So, next we looked at the p53 gene and the p53 gene is a good candidate for looking at mutational spectrum. It is mutated in between 30 and 50 percent of bladder cancers and some suggest that in cancer in general that the pathway is altered in about 90 percent of cancers and most of the mutations are observed at CPG sites.

These are the three hot spots in cancer in general. These are CPG sites. They have lots of C's and G's but the cytosines that are heavily methylated are the site of what is considered mostly endogenous mutations, C to T mutations caused by deamination of cytosine and where environmental carcinogens tend to form bulky adducts is at the guanine next to the methylated cytosine. Nobody knows why but what happens is that during the DNA repair sometimes there are errors because there is this big bulky adduct sitting on the guanine.

So, in lung cancer for example, PAH exposure which causes adducts causes primarily D to T mutations and in other types of cancer like bladder you see more G to A transitions,

Slide 15:

and this just shows you that in smokers we found an elevated hot spot at codon 273 which was found only in smokers

Slide 16:

and when we looked at the number of transitions which could be considered, well, looking at smoking there is basically, well, with grade there was an increase in the number of transitions and that is a G to A transition. That was also seen with pack years and possibly difference with smoking but when we looked at these G to A mutations at CPG sites they were not found associated with grade, using that as a mark, or stage using that as a marker of progression but they were observed with the exposure and similarly this codon 273 mutation was observed only in smokers but was not observed with grade suggesting that this type of mutation is not caused by progression of disease but may be an early marker of exposure to tobacco.

Slide 17:

So, in conclusion the prevalence of p53 mutations increased with stage, grade and smoking and primarily there is a preference for G to A mutations at CPG sites. This finding was recently corroborated by a study with Jack Taylor's group at NIEHS and arsenic exposure, I didn't show you that data but it was not associated with any p53 mutations.

Slide 18:

So, what I have been doing in this research is combining markers, gross markers like CGH changes and p53 mutations and looking at those that are associated with disease progression and then looking at different exposures and what you can see is that the majority of changes are associated with progression, disease progression but with arsenic only a subset was also associated with progression and those were the gross chromosomal changes whereas with tobacco more of the mutational changes were associated with the exposure and there were some changes that were not associated with progression but they were observed with the exposure, suggesting that these are areas to look at as early markers in etiologic studies.

So, in the future I plan on conducting more studies focusing on chromosome 9q. Also, 9p is not up here which is frequently lost in bladder cancer, but that is because it was lost all the way across the board, and it didn't have any relationship with exposure, but we plan to continue this work with additional markers and also working on looking at additional exposures.

Slide 19:

In future studies, right now I am working on two studies, a Spanish bladder cancer study at NIH. Deborah Silverman and Mustaf Hydrosanichi are the PIs in our Occupational and Environmental Epidemiology Branch and this study has a heavy focus on occupational exposures and also environmental exposures. So, we are going to be looking at disinfection by-products, exposures in the textile industry, smoking, urinary pH, urinary stasis.

The newest studies, and now this study has 1200 cases and 1200 controls; the study I just showed you had 114 case-control pairs. So, these studies are much bigger to look at a number of genetic susceptibility markers. We are looking at one carbon metabolism pathway, other types of metabolic genes and the part that I am primarily going to be working on and building a collaboration with Spanish collaborators is looking at epigenetic changes which are some of the earlier changes that don't associate with disease progression but may correlate directly with specific early exposures as well as genetic tumor marker studies looking very closely at chromosome 9p and 9q and I am also working on a New England bladder cancer study which has a heavy environmental component. This is a population-based study, and one hypothesis for the excess incidence in bladder cancer in the New England states is environmental arsenic exposure.

The levels are not as high as what I have shown here, but they are still elevated. Also, looking at tumor markers and genetic susceptibility enzymes

Slide 20:

and this is just a list of all my collaborators.