SLIDES & TRANSCRIPTS
Friday, December 5, 2003

Markers of Bladder Cancer

Edward Messing, M.D.

After following these two very high-tech talks, I feel a little embarrassed with what I am going to say, but it deals with what we as clinicians can do with certain markers right now, and if one is talking about markers and bladder cancer, I think realistically there are numerous places where they are potentially useful.

TOP

I am going to focus on just one, just for purposes of time, and this is one that is quite relevant to what we all do, that is can the use of the markers that are available reduce the likelihood that we can do cystoscopy in the myriad of patients we now see with superficial bladder cancer who have recurrences,

TOP

and I guess to start off with there are several reasons why we do as frequent cystoscopies as we do in monitoring the patients with superficial bladder cancer, but clearly the first one is the major reason or at least the major motivation for doing that.

TOP

If we are going to talk about markers that may assist us in this regard to reduce the frequency of cystoscopy, and I know you are all familiar with this. So, for this group I don't have to mention it, positive predictive value besides its definition you need to remember is associated with a prevalence of the disease in the population you are looking at with the given sensitivity and specificity of the test.

TOP

In looking at the markers that are currently available in the US market and the ones that some of which may soon be available, I am going to focus on the ones that are available, not that these are not more important and potentially far more accurate but because they are not here yet and we want to know what can be done with our current situation and then throw in the values here and see if we could change it.

TOP

Briefly in the next 2 minutes I will just describe what is known about these markers or at least for the pertinence we are going to talk about. We all know the BTA stat test. It is a human complement factor 8. Its sensitivities vary. It is grade dependent. Specificities are around in the 60 to 70 percent range in most of the studies published and there is a significant false-positive rate with people with inflammatory conditions which becomes an issue in doing it in a lot of patients with bladder cancer histories.

TOP

NMP 22 is a quantitative test although a point of service test is now available. It is a nuclear matrix protein which is important in regulating the spindle and mitosis. Its sensitivities in most studies has been in the 70 percent range. It is probably a little bit better than BTA for low-grade tumors. There was one study from the Cleveland Clinic that I really cannot understand in which 18 out of 18 patients had their cancers diagnosed by this although these were primarily big and high-grade cancers. Its specificity has again been in the 60 to 70 percent range mostly.

TOP

The other two tests that are primarily available although not used very often are the UroVysion FISH test; these are both tests that use fluorescence. This is a test as you know that uses DNA probes to the centromeric forces of these three chromosomes and the portion where p16 is on 9p and again you could detect deletions and additions to these chromosomal fractions.

TOP

and finally ImmunoCyt is using a cocktail of antibody, monoclonal antibodies to three antigens expressed on TCC as well as cytology and intriguingly the sensitivities have been pretty consistent in most studies, and 67 percent in the study I did had only six high-grade tumors. So, it detected four out of the six and you can't really make much out of that but these other studies had much larger numbers and we have much larger numbers here and again it is quite sensitive for low-grade tumors. Its specificities have been in the 70 to 80 percent range as well.

The one study here from Toronto there may have been technical issues in, and I won't go into that further but again most of the studies have had pretty consistent results.

TOP

Well, if one is going to use this test to try to prevent or catch tumors that are going to be muscle invading before they are as a way of replacing cystoscopy then we need to know something about which tumors are likely to become muscle invading. Of course, we know a lot about this. This comes from a study published about 7 or 8 years ago when patients managed by endoscopic means alone without intravesicle therapy and the patients could be listed based on tumor characteristics of grade, the number of recurrences and the size of the tumor into likelihood of recurrence and clearly we are going to primarily focus on this group of patients which represents the majority of bladder cancer patients seen and followed.

TOP

If we get patients who have low-grade tumors, and we can define much of the risk of those, who are very unlikely to progress to muscle invasion can we reduce the frequency of cystoscopy

TOP

and if so we need a sensitive test for low-grade tumors. We need it to be fairly specific because if we have a high false-positive rate we are not going to do any good. We are still going to do just as many cystoscopies as before, but it must be incredibly sensitive for the high-grade tumors because that is the thing we are really afraid of missing.

TOP

Again, do these tests meet those parameters? I think the fluorescence tests clearly meet it more than the other two tests we primarily use now but all the tests are less sensitive for recurrent cancer than they are for newly diagnosed cancer, and we will see why in a second.

TOP

Dr.Lotan is here in the audience I believe but he and Dr. Roehrborn did an analysis that was published about a year ago where they were trying to model whether you could replace cystoscopy in the monitoring of superficial bladder cancer patients.

The way they did it was alternate marker tests, then the next 3 months cystoscopy and cytology, marker tests the following 3 months and so on, alternating back and forth and one could argue with that pattern as a matter of fact I will but for a relatively broad range of recurrences, broad range of progression they felt that the new policy with a reasonably priced test would reduce by about 10 to 12 percent the cost of evaluation over the ensuing 2 years.

TOP

There are problems with this. One of them is that they predicted or assumed a linear recurrence rate which of course in bladder cancer does not occur and we will show that in a minute. They, also, did not really take into consideration the differences of grade, stage, size, multiplicity of the tumors, factors which we all know and have been shown to be very important in predicting recurrence and progression rates, particularly.

They didn't take into account whether the tumors were progressing to a smaller than muscle-invading tumor but still one that required a lot of additional treatment beyond just the cystoscopy.

TOP

and a group from Sweden has tried to look at the fact of can these markers, they picked three up here, which were commercially available in Sweden, be useful in these recurrent bladder cancer patients, and again their final conclusion was that size does matter, that the recurrent tumors because we are doing an interval cystoscopy before they present with clinical symptoms tend to be smaller, and none of these tests even in combination was really sufficiently sensitive for the high-grade cancers to replace cystoscopy now because I don't think any of us would feel comfortable in missing a Grade 3 tumor and having the risk of that happen.

TOP

This is just to remind you. This comes from a series of Japanese studies of intravesicle therapy and this is what happens to the control group and this is just to remind you that there is not a linear recurrence rate over 2 years or 5 years, that it really does fluctuate with peak recurrences at 3 months and then at various times after based on tumor characteristics

TOP

and so if one is going to try to do this one has to develop a model of picking really low-risk cancers that are not likely to progress and then test this. This is the model that I thought up. I talked to a couple of people who are in the audience and they have thought of similar plans, but really the first cystoscopy at 3 months out I think we really need to do. We cannot replace that.

Then you could replace probably the next two, cystoscope them again at a year because again that is the point of another peak incidence and then miss the next three check cystoscopies and then at 2 years out and then thereafter you could follow them with the markers.

TOP