SLIDES & TRANSCRIPTS
Saturday, December 14, 2002

Late Hormonal Therapy for Biochemical Failure

Edward Messing, M.D.

And at the original request we were asked to actually talk about this, and then it's been expanded to early versus late hormonal therapy, both because Marty has talked about a lot of the issues, and secondly, for purposes of time, I'm going to restrict it to post-prostatectomy patients.

But I would add one thing, now that I have seen Marty's presentation. If you look at the patients who received radiation therapy, and I agree there are four or five studies which indicate a benefit, with the exception of one which is a little bit confusing, the Granfors study, which had only 100 patients in it or less, there is one of the major studies that actually indicates a true survival advantage. All the others don't, and they require subset analyses to sort of tease out people who may have been benefited.

They all demonstrate a tremendous local control advantage. And so when Marty said that he thinks this is because they treated metastases and not local disease, I'm not entirely sure of that. And indeed, the studies that were effective in the radiation therapy group tended to all have to give hormonal therapy at the beginning of radiation, or before it to be effective.

Otherwise, those that combined radiation therapy with hormonal therapy, where the hormonal therapy was started after the completion of radiation, or at the end of radiation, tended not to be positive at all.

So I just bring that up. I'm not going to discuss that further. But primarily, we'll talk about again, the definition of biochemical failure, because I think that's a major issue when deciding in the post-prostatectomy patients, who really is a failure, and who needs treatment, when and how to treat.

TOP

I think I do a pretty good prostatectomy. I think Herb Lepor, who was the senior author on this paper, does a good prostatectomy. It is clear we don't always get all the prostate out. We know we don't get all the cancer out sometimes. But we don't get all the prostate out not infrequently. And this is someone who does 250 or 300 prostatectomies a year.

So it's something where we don't always get it out. There has been evidence for the past decade that we don't always get it out. That in people who are PSA failures afterwards, you often find benign tissue left, and so on and so on - and there is ectopic prostatic tissue. I think the issue is sometimes people who begin to become PSA failures after prostatectomy may not be cancer failures.

TOP

Again, this is Chris Amling’s work. I met Chris for the first time yesterday, but I had read his paper before, and was impressed by it. This comes from the overwhelming, abundant data from the Mayo Clinic. And indeed, if you have a PSA that was only in the 0.2 range or so, you tended to only have a 50 percent chance of progressing over the next 7-10 years. That's progressing biochemically, let alone progressing of course clinically.

So it's not clear that whatever you did with these people, you would see the PSA drop, but what good you would be doing them. They recommended again this level. I think in these days of micro-assays for PSA, where at our institution it's 0.05, in others it's 0.01, it's very hard to make sense out of a lot of this; what we are calling biochemical failures.

TOP

The Pound data was brought up before, and I think stands as a landmark, because these people weren't treated with other things. But you are all familiar with the group that is going to do very badly -- high Gleason grades, relatively rapid PSA failures after prostatectomy, fairly brief PSA doubling times.

TOP

And just as an example to show that, if you were a Gleason 8 tumor and you recurred within two years, you had a very high chance of progressing. And this represented all the people who died in that study. Versus if you had just as bad cancer, but if you recurred after two years, you had a longer time before you recurred; quite a bit longer, double the length of time.

And you could get groups of patients, including the major group that I think we see, which is here, where only a small percentage really progressed during that time. Now, in a 50-year-old, that's a serious issue. But is it a serious issue in a 74-year-old?

TOP

There are certainly alternatives to hormonal therapy for the PSA failure after prostatectomy, however you are going to define the PSA failure. And indeed, these alternatives don't work that badly. Now, it depends exactly how critical you are of how you see they don't work. But if you start the treatment early enough, if you don't have very extensive disease, node-positive, or seminal vesical positive disease, you can get reasonable response rates.

There are too many things to list all the references, so I just had names and years. But several papers really concur with this finding. Now, one could be cynical about this, and claim that they are using the sort of spooky definition of success. They also used projected statistics. And of course if some of these people were in the group where you had benign tissue left, that might explain the 50 percent success rate.

But all in all, I think that there is some data that there are alternative treatments for PSA failures after prostatectomy. As long as the people have regained continence already, the complication rate isn't terribly high.

TOP

With a hefty dose of radiation, you can avoid these complication in a vast majority of patients. I'm not sure how this study particularly really assessed potency. I think that is questionable. But I think the other complications are certainly in a reasonable range for these types of failures.

TOP

And I think even more importantly, you can predict which patients won't respond to radiation therapy pretty quickly. And if you look at these three, they are just about identical to the Pound criteria. So you know in advance who is not going to respond to radiation. They are the same people who are going to do miserably with observation alone. Again, you should treat pretty early, and again, you should avoid very extensive disease. There are no surprises here, but you can pick the people who are likely to respond to radiation.

TOP

I will not talk about these. We had a really fabulous discussion in the first section this morning, talking about many of these. I found them fabulous. Again, these are now experimental treatments. They are all ranging from phase I to a few phase II studies. There are some phase III studies that will be in there, such as this one. But I think right now these are experimental. We really don't know their results. I think if we can find patients to go on them, we all should be encouraged to do so.

TOP

TOP

What did the study we did in ECOG and SWOG, the inter-group study demonstrate? Well, I think first of all you must remember, as Dave mentioned and as Marty mentioned before, all these people are really at enormous risk to recur. They all had node-positive disease. They all of course, underwent prostatectomy. Most had high grade tumors; two-thirds had Gleason 7 or above, about 30 percent had Gleason 8 or above. They had huge volume tumors, 14 ccs on the average, most locally extensive, extraprostatic disease locally. And they virtually all, 80 percent has undetectable PSAs at the time of entry into their study. So they had tiny quantities of cancer remaining.

TOP

Marty went through the curves, so I won't repeat them. At ten years out now, median follow-up, the curves are staying identical. So the differences are being held, and median survival has been reached, and the median overall survival has been reached in observed men, and it is still identical differences.

TOP

TOP

TOP

Again, then these people were at truly enormous risks for failing. I mean the vast majority has PSA biochemical failures within two years. This says by seven years, but most of these failed within two or three years, and most less than two. If you have biochemical failure, three-quarters were metastatic disease. And almost half of those succumbed to their cancer.

So these were at truly enormous risk for failing. They recurred well less than the two years in terms of this is when they started hormonal therapy, 20 months out. So their PSAs became detectable at say 5 or 6 months out. And we started hormonal therapy at this level, and that was when almost all had metastases.

TOP

What we learned from that, well, hormonal therapy is a tremendously beneficial thing in people at truly enormous risk, and when you have extraordinarily low volumes of cancer. How low a volume of cancer do you need to have a detectable PSA, however, is another issue. Remember, there is tumor xenograft model from UCLA in which somewhere around 1 in 100,000 to 1 in a million cancer cells are androgen-naive, ostensibly de novo.

If you start treating people with undetectable PSAs, you might actually have below that number. So you have almost no androgen-independent cells. If you start waiting until the PSA becomes detectable in this range, or 1 or 2, how many cells you have is obviously anyone's guess, and probably varies per tumor.

TOP

So to summarize, I think that early hormonal therapy is probably justified in the absence of a clinical trial, which I would certainly prefer, in people at very high risk, who are biochemical failures within two years, who have the high Gleason grades, who have rapid PSA doubling times.

I think it's perfectly reasonable to wait a while to assess the PSA doubling time, not just treat when the PSA is 0.1 or something before you have seen a trend. I think that you can add these factors to it, because they are certainly unlikely to respond to the other standard therapy available. And I certainly would agree with Marty that that group of patients are going to do terribly with our currently available modalities, and hormonal therapy is likely to extend their lives.

TOP

Hormonal therapy obviously comes at some costs, and Marty has discussed these already, including the potential beneficial impact of intermittent or anti-androgen alone therapy, so I won't discuss that.

TOP