In this blog PCRI presents an interview from our contributing partner, Prostatepedia.
Prostatepedia spoke at length with Dr. Laurence Klotz about why testosterone levels are significant in prostate cancer. Dr. Laurence Klotz is an esteemed Urological Oncologist from the University of Toronto. He is one of the driving forces behind the development of active surveillance as a viable approach for men with slow-growing, non-aggressive prostate cancer. Prostatepedia spoke with him about his 2015 study on the importance of testosterone levels during the first year on hormonal therapy. Here is the interview:
Can you walk us through your the study you published this past year in the Journal of Clinical Oncology on nadir testosterone during the first year on hormonal therapy?
Laurence Klotz: Before I begin, let me say that I was an early proponent of intermittent hormonal therapy. In fact, I was the first author on the first clinical series on intermittent therapy published in 1986, back when I was a Fellow.
In intermittent therapy, testosterone rises during the off-treatment interval. A lot of Phase 2 and some Phase 3 studies show that intermittent therapy for the majority of patients, particularly non-metastatic patients, works about as well as life-long continuous therapy.
On the basis of those studies and on some other evidence, I always believed that the testosterone level during ADT didn’t matter much. My impression was that the pharmaceutical companies in the LHRH field were trying to get a little bit of an edge over their competitors. I thought that the four or five LHRH agonist, in particular, were quite similar.
Then, beginning around five or six years ago, some articles were published, particularly from Europe, claiming that testosterone levels in men on LHRH agonist were important. There were three studies.
They were all retrospective and relatively small. The first one included only 73 patients. The other two included about 120 patients. All three showed that the men with lower levels of testosterone did a lot better. Frankly, I was originally very skeptical.
Around 15 years ago, we launched a prospective randomized trial led by Canada but supported by the US and the UK, of intermittent versus continuous hormone therapy in men with a rising PSA. It was a very successful trial. We accrued almost 1500 patients. The study was eventually published about two years ago in the New England Journal of Medicine. It showed absolutely no difference in overall survival between intermittent and life-long continuous hormonal therapy. It is considered a pivotal trial on the issue of intermittent therapy in non-metastatic patients, which is currently the most common indication for hormonal therapy.
There was still this issue of whether or not testosterone levels really matter. How could testosterone be so important if with intermittent therapy there is no significant difference, at least in terms of overall survival? There was a slight trend to a few more prostate cancer deaths in the intermittent arm, but it didn’t nearly reach statistical significance, and it was absolutely matched by more deaths in the continuous arm from other causes.
This question had always preyed on my mind. We had collected serum testosterone on these patients, particularly in the first few years. I thought: here is an opportunity to really answer this question. In contrast to the three earlier studies, which were small and retrospective, this was a large group of prospectively evaluated patients who were followed in a long-term study. We could now answer this question definitively.
We looked at serum testosterone, particularly in the first year. Most patients had had testosterone measured every two or three months. We had at least three testosterone readings on the vast majority of men—about 625 patients on the continuous arm.
We looked only at the continuous arm of that study, so as to not muddy the waters with patients coming off treatments and various rates of testosterone recovery and so on. To my absolute surprise, our findings completely supported the results of those smaller earlier studies. In fact, patients whose testosterone dropped below 20 nanograms/deciliter did much better than those whose median testosterone failed
In fact, there was a difference in time to progression and prostate cancer survival. The lower the testosterone, the better. The ones whose median testosterone was above 50 did considerably worse with time to castrate resistance—about three times greater rate of progression than the ones whose testosterone was suppressed below 20.
These results were published earlier this year in the Journal of Clinical Oncology. In my world of prostate cancer researchers and opinion leaders, it is considered to be a landmark study. This is probably the best data we’re ever going to get on this question of the importance of low testosterone.
How do you reconcile the success of intermittent therapy with the observations in this study? The implication is you need to hit the prostate cancer cells as hard as possible early on. You want to drive testosterone levels as low as you can and in that way kill off not just the very hormone-sensitive cells, but also the partially sensitive cells that only undergo apoptosis when the testosterone gets very low. Then you allow the more favorable cells to recover during the off- treatment interval. If you do it that way, the evidence seems to show that patients do better than if you don’t hit the cells as hard.
In fact, it’s quite plausible that low testosterone is important and that is why intermittent therapy can work as well as it does.
What should patients take away from all this? Should they insist their doctors do whatever they can to bring their testosterone levels as low as possible?
There are two points. First, the inference is that the testosterone levels should be monitored along with the PSA.
Isn’t that routinely done during hormonal therapy?
Well, it’ll be done more often now, but it certainly has not been widely done up until now.
Is there a consensus on what testosterone level doctors should aim for?
The ones who do the aiming aim for below 20 nanograms or 0.7 nanomolars. Definitely 20. That is what is achieved by surgical castration.
Another point is that LHRH agonist do not get testosterone levels that low in a considerable proportion of men. Levels often tend to be in the 20 to 50 range and occasionally higher than that.
What do you do then? Combine LHRH agonists with some other agent to drive testosterone lower?
The first thing is monitor the testosterone. The odd blip of testosterone above 20 doesn’t seem to be a problem, but if it is consistently above 20 something should be done. The options are to switch drugs. If the patient is on Lupron, try switching to Zoladex or Degarelix for example, or vice versa. There is one paper that suggests the benefit of anti-androgens is in patients whose testosterone is not adequately suppressed below 20. Another alternative if the man is not on an anti-androgen is to put him on one.
One thing we don’t know, and my guess is we may never really know the answer, is whether an anti-androgen like Casodex confers the equivalent benefit of a low testosterone in someone whose testosterone is not fully suppressed. It may. There is one small study that suggests it does, but this is an unknown at the moment.
The other point is that if you accept that low testosterone is really important, it opens the door to using newer hormonal agents like Xtandi. These androgen receptor pathway destroyers are much more powerful than drugs we’ve had in the past. It’s possible, then, that these drugs may be more effective in those patients than agents we’ve traditionally used. There may be a role earlier on for these agents. The problem, of course, is these drugs are extremely expensive. Practically speaking, it is not currently feasible to use those drugs in hormone naïve patients. Down the road, though, we may find there is some benefit.
Your study looked at men dealing with a rising PSA after radiation and surgery or radiation alone, correct?
What about men who are on ADT as a primary treatment, or on ADT to shrink the prostate gland before radiation therapy? Do you think it is important to get testosterone that low in those scenarios?
I have no idea. Probably, but the idea behind using hormones before radiation therapy is not so much to shrink the gland, but that hormone therapy is synergistic with radiation and enhances the radiation response. So this is a complete unknown. There is no evidence that shrinking the gland before surgery is of any benefit to the patient. Outside of a clinical trial, I don’t think it should be encouraged.
That just doesn’t happen?
In metastatic disease, intermittent therapy is more controversial, but I think it still is the case that more aggressive hormonal therapy is probably better.
Another interesting point is that we also looked at the intermittent group to see if the testosterone during the period of induction when they were on treatment was also predictive. It was much more heterogeneous, because they had varying off treatment interval lengths and varying lengths of recovering testosterone.
Essentially, though, the trend was all in the same direction. The results just weren’t quite as consistent, undoubtedly reflecting the greater heterogeneity of treatment in these patients.
You looked at nadir testosterone in the first year on continuous ADT, but what about testosterone levels in subsequent years?
This was a very long-term study. It ran from 1997 until it was published in 2012. With a very long-term study, over time, compliance with long
term follow-up tends to falter. We didn’t have such gorgeous consistent evaluation of testosterone after the first few years and so we won’t be able to do as tight a study. It’s a good question, though. My own view is that the same principle likely applies and the lower the testosterone,
the better in the long run.
Some criticized that we didn’t look at testosterone levels over the full length of the trial, which was 15 years.
But no study is absolutely perfect.
We recently featured a series of conversations on erectile dysfunction and prostate cancer. One of the doctors we spoke with mentioned that once ADT is over, most but not all, men have their testosterone levels return to pretreatment levels. Can lowering testosterone as low as you’re talking about make it difFIcult for a man to eventually recover pretreatment testosterone levels?
The main predictor of the failure to achieve a low testosterone is young age. In other words, it's easier to get very low testosterone levels in older folks than in younger men in whom the pituitary gonadal axis is a little more resistant to suppression. The older men tend to have slower testosterone recovery; about 25% to 30% never actually recover a normal level. By six months, most patients recover at least 50% of their baseline testosterone, above the level where you’d expect them to be normally androgenized.
There is no evidence that more effective suppression results in slower recovery afterwards. There is no reason to think that is the case. It is very much age dependent and dependent on the duration of androgen deprivation therapy. Just to be clear: a testosterone of 50 is a low testosterone level
by any standard. It is just in the case of prostate cancer that it is better if it is even lower than that.
Is there anything else you think patients should know about your study?
I would add a cautionary note and say that what we showed is that lower levels of testosterone are associated with a better outcome. We did not show that taking men with a poorly suppressed testosterone, say in the range of 50, and lowering it to 20, resulted in a better outcome.
Skeptics will say, “You’ve shown that low testosterone on hormonal therapy is better, but perhaps there is some co-variant that goes along with low testosterone that results in a better outcome. It doesn’t necessarily mean that taking steps to lower testosterone further is necessarily going to result in a better outcome.”
But I think the skeptics are wrong and that taking steps to lower testosterone results in a better outcome. Mine is a reasonable inference, but from a rigorous scientific perspective, we haven’t shown that. We need a prospective randomized trial to show that, but such a trial would be almost impossible to carry out.
I’m sure a lot of patients will take from this discussion that they should ensure their testosterone levels are below 20 ng/ml during the First year on hormonal therapy.
I think they should take home that message, but it is a question of how much scientific rigor one demands. No one is doing long-term studies of hormone therapy alone at this point.
Because the interesting questions are elsewhere. For example, we’re now doing a study that asks the question: what is the optimal duration of induction therapy? Three or four months? Do men have to be on hormonal therapy for eight or ten months? No one has ever looked at that question, so we’re doing a 100 patient study. We wouldn’t be able to derive the same kind of information from that study that we can from this large-scale trial. At the moment, there is no study even on the horizon anywhere in the world that is going to give better data about this particular question than this study, even with all its limitations.
Prostatepedia, a division of Rivanna Health Publications, has a single mission: to help leading prostate cancer researchers explain their work and its implications to prostate cancer patients, activists, and health care providers. Formerly known as Prostate Forum, Prostatepedia, their monthly periodical, features informal conversations with leading experts.
Dr. Laurence Klotz, MD serves as Chief of Urology at Sunnybrook Health Sciences Centre, and is a globally-renowned clinician in the area of Uro-oncology. Dr. Klotz serves as a Member of Clinical Advisory Board at Profound Medical Inc. He serves as President of the Canadian Urological Association and the Canadian Urology Research Consortium, Chairman of the World Uro-Oncology Federation, and the Editor-in-Chief of the Canadian Journal of Urology.