Hormone Replacement Therapy: Why Can’t We Get It Right?

Hormone Replacement Therapy: Why Can’t We Get It Right?

So as some of you know, and some of you probably don’t, I’m not really a cancer guy but I’m a men’s health guy and I really came into the world of prostate cancer through the back door if I can use that expression here. Really working around testosterone and sexual medicine in the late 1980s, in the early 1990s, and the big concern of course with using testosterone was prostate cancer. So I started having these men who were responding well to testosterone treatment, and what we worried about was that we were going to precipitate terrible prostate cancer in them. And so a lot of my work over the years has been looking at that issue, and we’ve come a long way over the years around testosterone. But it’s always been a controversial topic. It’s probably been as controversial this past year as ever and the amazing thing is that as the concerns about prostate cancer seem to have diminished, just then we came in with issues around cardiovascular risk.

The big concern has been prostate cancer over the decades but now we’re talking about cardiovascular risks. And like many of you, this is what I learned about testosterone when I was in residency in the mid 1980s. I learned that high testosterone caused prostate cancer, low testosterone was supposed to be protective, men who were castrated early or eunuchs never got prostate cancer, that was my training. And if a physician were foolish enough to give testosterone to a man with existing prostate cancer that that cancer would grow like wild fire, and I’m sure that many of you heard the same expressions I did and probably have taught them as I did, that giving testosterone to prostate cancer, or to men with prostate cancer was like feeding a hungry tumor or pouring gasoline on a fire, right? We’ve all known this. And so testosterone is depicted as the devil. I finished my training in 1988, and in six years of general surgery, and then urology, I don’t think I treated a single person with testosterone.

But I had done some work on this as an undergraduate at Harvard in a lab that involved lizards. I’m not going to show you that work, but clearly testosterone was pretty good for lizards and sex, and women had told me my whole life that men are just like lizards or snakes and so I came out of training. We didn’t have that much to offer men with sexual issues, there was no Viagra, Cialis, we had implants. I started just testing testosterone levels in men who came in with mainly sexual complaints and was surprised at how often their levels were low.

When I started treating some of these men, I was impressed by how many of them did well. Sometimes in ways that I hadn’t even anticipated like they would say, my wife says I’m nicer to be around. I wake up in the morning, I take my turn around, I bring my legs down on the ground, I sit on the edge of the bed, and I’m optimistic about my day. Doctor, I haven’t been that way in many, many years. But the concern of course was prostate cancer and I would talk to these men about it, and the more of these guys I was doing, the more I felt like I really needed to protect my patients, and to some extent myself by excluding as best as possible the presence of prostate cancer so that it wouldn’t explode or grow like wild fire with treatment.

I started doing prostate biopsies in men with normal PSA, normal digital rectal exam solely because they had low levels of testosterone, and I wanted to make sure as best I could at that time that they didn’t have a hidden cancer in their prostate. Remember that our belief then was that high testosterone caused cancer and made it grow, and low testosterone was supposed to protect men from ever getting it. And surprisingly we found cancers right away.

In JAMA in 1996 we published our results showing, small series, but 11 cancers in the first 77 men we did this with, all with normal PSA, normal digital rectal exam. That cancer rate in the mid 1990s was the same as was being reported in large series for men who had elevations of PSA. In fact it looked at that time like having a low level of — these are sextant biopsies – low levels of testosterone was as risky for having prostate cancer as a PSA of between 4 and 10.

So after that, and we’ve gone on and we’ve published more or less the same results in a larger group of 345 men. I no longer believe that low testosterone was protective but I still did believe that high testosterone was a problem. I’m not even sure I can explain to you today why I believed it but I absolutely did. It’s what everybody believed; it’s what everybody was taught, every medical student around the world.

So I was privileged in 2004 together with my fellow put together for the New England Journal of Medicine a review article on the risks of testosterone. Testosterone was becoming a much more exciting topic. AndroGel had come in in the year 2000 and so the rates of treatment were going up. And Emani Rhoden was my fellow; some of you may have seen him. He’s a Brazilian urologist of some note and he’s a very confident man. We’ve pulled a couple of hundred articles.

We’re getting ready to draft this article, and then he’s in front of me and he goes, Chief, and he looks nervous. He says, do you have the articles that show that testosterone is bad for prostate cancer? I said, I thought you had them. What we found in the PSA era which at that point was the mid 1980s when we wrote this, 2004, the number of articles that we could find, and there were quite a few that showed something worrisome about testosterone and prostate cancer. High grade, recurrence rates, bad outcomes, number of articles we found with the worrisome outcome was zero.

We didn’t say it didn’t exist. We just said we couldn’t find the data. And it’s interesting to read some of those articles from that time because everybody who writes an article of this type assumed that it should have been a slam dunk. And the discussions read one after the other, they say, we don’t know why we couldn’t show this relationship, we suggest that future studies look at X, Y and Z. And then the next study looks at X, Y and Z says, we don’t know why we didn’t find it. We suggest future studies look at A, B, C. It’s really interesting. So we couldn’t find this evidence.

Where did it come from? And of course, everything in prostate cancer pretty much goes back to Huggins, so 1941, Charles Huggins and his coauthor Clarence Hodges published their landmark study, a couple of studies actually, which demonstrated the benefits of androgen deprivation most by castration, but they also used exogenous estrogen. They also introduced acid phosphatase as a biochemical marker at that time, and basically what they wrote is they had these men with metastatic prostate cancer, they castrated them or they gave them estrogen, and they reported that acid phosphatase went down. They wrote that in every man in whom they gave testosterone to, acid phosphatase went up. And their conclusion therefore, it’s there on the slide is, “Cancer of the prostate is activated by testosterone injections.” And 25 years later, Huggins was awarded the Nobel Prize for this work, which basically ushered in even larger than just prostate cancer, the era and recognition of hormone sensitive cancers. Really interesting.

So as I’m going around on my lecture circuit and talking about this, and giving men testosterone, I pull up, I need to read this article so we didn’t have everything online at that time. I went to the basement of the Harvard Medical Library Countway, and I pulled out this paper from 1941. I read through it and I went, this is bad news, I’m giving testosterone to a lot of men, Huggins said it’s bad. But I read it again and I said how many guys did he actually give testosterone to? And the answer was three. Of those men, he actually only gave results for two. One of those men had already undergone castration before getting testosterone which today we would put in a separate category. We would call then androgen deprived.

In the end, the general rule that cancer of the prostate is activated by testosterone that Huggins put out was really based on a single patient, hormonally intact, who received testosterone injections for a total of 14 days. And his acid phosphatase curve goes up and down and frankly is uninterruptable. Yet, this is what the entire medical student global community learned as something that was dangerous.

What do modern data show? This is a pooled, 18 pooled studies, longitudinal studies looking at baseline blood test results for sex steroid hormones and subsequent development of prostate cancer, 3,000 men with cancer over 6,000 age matched controls. By quintiles for testosterone, free testosterone, DHT, there’s not relationship between blood levels of any of these androgens and the development of prostate cancer. Men with high testosterone had no greater risk than men with low testosterone.

In the placebo arm of the reduced trial, they had over 3,000 men, all of whom got prostate biopsies as part of the trial at year 2 and at year 4. Prostate cancer risk was not associated with serum levels of testosterone or DHT. And again men with higher serum levels had no greater risk than men with low serum levels of testosterone or DHT.

It almost looks like it’s unrelated doesn’t it? Recently it was published a metanalysis of 22 randomized control trials of testosterone versus placebo. Half the studies were less than a year, half of them were from between a year and three years, 2,300 men and they found no difference in prostate cancer rates for men who received testosterone versus men who received placebo.

How does that all make sense? We’ve just spent here, you guys have last couple of days lowing testosterone as much as you can, right? In every little bit you lower it, it gave you better and better results. How does it make any sense at all that raising it doesn’t seem to do much or that men with higher levels have no greater risk than lower levels? And the answer is, is that it’s very important. Androgens are important for the prostate, very much so, but all the action seems to happen as you go into and out of the castrate range.

Here’s testosterone, here’s any measure you want of prostate cancer, PSA volume, etc. And you can do this in normal humans with normal prostates, you can do it in cancer cell lines, you can do whatever you want as you go from zero and you increase with androgens, you do get growth, but at some point that growth becomes maxed out. It reaches what’s called in biological terms a saturation point, and above that we find nothing. There’s no evidence anywhere that anybody can point to that men with very high levels of androgens have any worse outcomes or risks than men with middle levels, low normal levels, and there’s nothing that goes between these.

But there’s no question, and you guys all do this, that if you androgen deprive somebody, the PSA is going to drop. You can do that with a normal man with a normal prostate. You get down to 0.1 undetectable. And if you stop the LHRH agonist, and you let the testosterone rise, his PSA will rise too and his prostate will grow. But at some point everything is done that’s going to happen, it’s driven by androgens. We call that the saturation model and explains pretty much as far as we can tell all the observations we see where androgens do seem to matter and where they don’t seem to matter in terms of prostate or prostate cancer.

On a molecular basis, this is I think not the only explanation, but an important contribution to this idea, which is this. So the concept is saturation, androgen enters the cell, it’s called a prostate cell or a prostate cancer cell. Testosterone get’s reduced to DHT, it binds to the androgen receptor, and then the complex of androgen receptor and androgen come into the nucleus. They bind to androgen response elements and they do their tropic effects.

But notice that it’s not testosterone by itself. It’s not naked DHT that does it, and except in transformed cases, it’s usually not androgen receptor itself. The androgen driven part of this happens as a complex, and it turns out that the androgen receptor becomes saturated at very low levels in vitro and clinically where we also have sex hormone binding globulin that it accounts for about half of this. That number is about double. The saturation point clinically in men is about 250 ng/dl. And if you’re raising it above that, we see very little impact.

I don’t have the slide for it but this past year we’ve just published an AndroGel study that looked for six months versus placebo. Men who had testosterones less than 250 who got treated, did demonstrate a rise in PSA, modest. Those who started with a testosterone of 250 or higher and got treated showed no change in PSA at all.

So the old idea that testosterone is like food for a hungry tumor, I’d like to suggest it should be changed to like water for a thirsty tumor. What’s the difference? The difference is if you chugging water all day long, you’re just going to go to the washroom and get rid of it. Imagine a plant like this in the middle. It’s a houseplant, and think of this as cancer and think of water as androgen. If you deprive it of antigen, or water, it will shrink, it will lose matt, it will lose volume. If it’s still alive and you give it back water, it will grow. It will gain mass, it will gain volume. But at some point it becomes happy with how much water it’s got, and you could have a garden hose running into a 24 hours a day, that houseplant is never going to grow to be the size of a sequoia tree. It has satisfied its requirement for this particular let’s call it nutrient.

Some brave souls have started to give testosterone to men with known prostate cancer, histories, a couple of years ago together with my colleagues at Baylor. We published the first series in men who are on surveillance, so they were receiving no treatment for their prostate cancer. Most were Gleason 6, I think we had one Gleeson 7 in there. These men received testosterone for a median of 2.5 years. All of them had follow-up biopsies, an average of two sets of biopsies per individual. And what happens then, this is PSA time zero when they were diagnosed with their cancer. This is a 3, 6, 12 and 24 months after treatment with testosterone and PSA did not change. None of these men went on with clinical progression.

So nearly everything that we were tight about testosterone and prostate cancer in the old days was wrong. We have no evidence that high levels of testosterone contributes to prostate cancer. Low testosterone does not appear to be protective. We have certainly some ability of androgens that’s responsible for prostate cancer growth, but that concept is limited. There’s an N to it. What we do see is that testosterone provides benefits to some men, with or without prostate cancer, something that was not known or talked about when I was in training, but of course if you’re thinking about doing some of this you have to be very careful. Our safety data on this is very limited, and really consists of case series and I’m not aware of any placebo control trials. So be careful about it.

I show this just for fun. Canadians are the best aren’t they? They did a story on testosterone. Here this is Toronto life. You know you have magazines in different cities like Denver Life or something, and here’s a testosterone time line and the best is they put me here right between Viagra and AndroGel. I only wish my mother were around to see that. She’d like that.

All right, it’s prostate cancer. So we stopped doing this. When I ask for a show of hands, five, seven years ago I’d say how many of you are treatment with prostate cancer with testosterone? After radical, and it used to be no hands, there may be a couple of hands, and now it’s probably if we have a group of clinicians here, it’s probably more than half of you, which means that there is a level of comfort that we never had before and it’s new. And just as this has gone down, pow, we’ve got stories that testosterone is causing heart attacks, strokes, and deaths. November 6, 2013 came a study in JAMA, follow up two months later. This was about mortality, myocardial infarction and stroke. Two months later, about non-fatal MIs and men getting testosterone prescriptions, two days after this article comes out covered by the media, the FDA announces it’s going to investigate cardiovascular risks. The world went crazy.

New York Times not an opid [phonetic], but their actual editorial board combined the rise in prescriptions with these new cardiovascular studies and came up with a title to an editorial in which they criticized men for being unable to face aging properly and they called it overselling testosterone dangerously. The danger has to do with cardiovascular. It was on the news, and then all of you saw the legal stuff. Are you on testosterone? Have you had a heart attack? Have you had a stroke? Call us. Lawyers went berserk. And of course our patients who were on testosterone stopped. And they stopped because they thought something bad was going to happen to them. I had patients who were called by their cardiologist. These men had been on testosterone for 10 years some of them, they said you better get off that stuff.

This is the JAMA article. I want you to see what this is all about. This is the one that started it. This was what was presented. Rates of death, myocardial infarction, and stroke in men treated with testosterone and untreated men. This is a retrospective study, all these men were entered into the study because they had coronary angiography through the VA system, 8,700 of these guys. Some of them went on to get testosterone, and some did not, so that’s our two groups.

What they reported was that at one year after angiography, two years, three years, men who got testosterone had 25.7% rate of these bad events compared to 19.9% for the others. Okay, well that’s interesting until you actually look at the abstract and something doesn’t make sense.

They give you all the information you need. This many men did not receive testosterone, this many died, this many had MIs, this many had strokes. This was the group that did take testosterone, this many men, this many died, and they write that the absolute rate of events were 19.9% in the no testosterone group, what I just showed you, and 25.7% in the testosterone group. This is JAMA national attention.

Now what’s absolute rate of events? It’s the simplest calculation in medicine. It’s the number of bad things that happened divided by the number of people in their group. That’s all it is. So watch. If you take their numbers and you put them here, the no T group had an absolute rate of events of 21%, and the testosterone group had an adverse absolute rate of events of 10%. I didn’t get those numbers wrong. These are their numbers, 681, 420, 486. Twenty-one percent events in the no testosterone group, 10%, half, in the testosterone group.

What happened? This is a group of epidemiologists and statisticians who got this term wrong, absolute rate of events. But that’s what got carried through everywhere. The reviewers missed it. The authors put it out this way. And people believed that it was real. In fact, what it was was Kaplan-Meier estimated cumulative probability of events. Sound statistical? It is. That’s what they did. They used a statistical methodology that’s never been reported before for the presentation of results called Kaplan-Meier, I’m sorry, stabilized inversed propensity of treatment waiting, based on Kaplan-Meier curves. They adjusted for more than 50 variables, most of which have nothing to do with cardiovascular risk, and in the end as the authors explained in a letter, an event in the no testosterone group ended up counting as less than one event, and an event in the testosterone group ended up counting as three events. That’s how you get an absolute rate of event where the testosterone group was better by half, and making it so that you reverse the results.

As if that isn’t bad enough, in a response to a challenge to my fellow coworkers and me, we questioned why they had excluded a group of men. It doesn’t matter why, 1,132 men. The authors wrote that they looked at that group and they made an incorrect notation. It shouldn’t have been 1,132, it was really only 128 which is an error of more than 1,000 people. They took 900 of them and they said, we’re going to put them in another category and by the way, when we looked at this group, we found 100 women which is 9% of this all male population.

Imagine you had a prostate cancer study and it turned out that early 10% of the population were women, and you didn’t know where they came from in your dataset. Would you believe those data? So 29 groups said the data no longer credible, and called for retraction which JAMA has not done. Not because we disagreed with the study, but just that the data is no longer credible.

The truth is that none of this actually ever made much sense. The data has been progressing for more than 20 years in a way that suggested that the testosterone if anything is beneficial for cardiovascular issues.

This observational study, there are a few of them; I’m just going to go through them quickly. This is in a VA hospital. This is survival so people are dying as the curve goes from left to right. The lowest group here, men with two levels of testosterone less than 300, these are two groups where men had levels above 300 on two occasions. This is intermediate, low testosterone is associated with increased mortality.

Here’s another one looking at cardiovascular mortality based on quartiles, large group, 11,000 men. Men with low levels of testosterone died sooner.

Here’s another one divided by bioavailable T, men with lower levels died sooner. There’s no question about this shown in metanalyses now. Low testosterone by itself, causal or just associated? We don’t know. Is associated with worse overall longevity.

We’ve actually have a number of trials where they’ve given testosterone to me with known cardiovascular disease. Here’s one 46 men with known angina, randomized to testosterone gel or placebo. Twelve week trial. These men were out on treadmills, I mm of ST depression was used to indicate ischemia. Those who got treated with testosterone were able to exercise longer than men on placebo until the time that they developed ischemic changes. We’ve seen the same thing in men with congestive heart failure.

This is my thought. Testosterone has always been controversial. A lot of what we react to has nothing to do with the science of testosterone but has to actually do with our thoughts about testosterone as a hormone, as a male hormone, as something that has to do with sexuality. We’ve dealt with the prostate cancer issue for 70 years. I think that story is changing and now we’re dealing with the cardiovascular stuff. Once there’s a dominant narrative that reaches the media, it becomes very difficult to change.

And it’s interesting if you’re interested in this, you’re interested in men’s health, and you look at these stories, you will see a few themes. One is that the symptoms of testosterone deficiency are minimized. So people almost make fun of it. These guys have low sex drive or they’re tired. Well, they’re older. It’s like come on, get a grip. In fact my patients, many of whom are older think that their sexuality is about as important to them now as it was throughout the rest of their lives. It’s part of what makes men feel like men. The benefits are denied.

At the FDA meeting which I attended and made a presentation, there was a nice presentation by Adrianne Dobbs, part of the writing group of the Endocrine Society guidelines where she showed with metanalyses the benefits level evidence for benefits of testosterone in a variety of ways. And as soon as she stopped speaking the FDA panel starts talking amongst themselves how the benefits are unproven. It makes no sense. The indications are ridiculed. This is all supposed to be about anti-aging and body building and certainly that influences how we think about these things and the risks are exaggerated.

So how did we get here? I love this quote, “The public will believe anything, so long as it is not founded on truth.” And even better I think for Mark Twain, “If you don’t read the newspaper you are uninformed. If you do read the newspaper you are misinformed.”

Endogenous Hormones and Prostate Cancer Collaborative Group, Roddam AW, Allen NE, et al. Endogenous sex hormones and prostate cancer: a collaborative analysis of 18 prospective studies. J Natl Cancer Inst. 2008 Feb 6;100(3):170-83.
http://www.ncbi.nlm.nih.gov/pubmed/18230794

English KM, Steeds RP, Jones TH, et al. Low-dose transdermal testosterone therapy improves angina threshold in men with chronic stable angina: A randomized, double-blind, placebo-controlled study. Circulation. 2000 Oct 17;102(16):1906-11.
http://www.ncbi.nlm.nih.gov/pubmed/11034937

Finkle WD, Greenland S, Ridgeway GK, et al. Increased risk of non-fatal myocardial infarction following testosterone therapy prescription in men. PLoS One. 2014 Jan 29;9(1):e85805.
http://www.ncbi.nlm.nih.gov/pubmed/24489673

Friedman AE. Re: Abraham Morgentaler, Abdulmaged M. Traish. Shifting the paradigm of testosterone and prostate cancer: the saturation model and the limits of androgen-dependent growth. Eur Urol 2009;55:310-21. Eur Urol. 2009 Jul;56(1):e4; author reply e5.
http://www.ncbi.nlm.nih.gov/pubmed/19349108

Huggins C, Hodges CV. Studies on prostatic cancer: I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. 1941. J Urol. 2002 Jul;168(1):9-12.
http://www.ncbi.nlm.nih.gov/pubmed/12050481

Khaw KT, Dowsett M, Folkerd E, et al. Endogenous testosterone and mortality due to all causes, cardiovascular disease, and cancer in men: European prospective investigation into cancer in Norfolk (EPIC-Norfolk) Prospective Population Study. Circulation. 2007 Dec 4;116(23):2694-701.
http://www.ncbi.nlm.nih.gov/pubmed/18040028

Malkin CJ, Pugh PJ, Morris PD, et al. Low serum testosterone and increased mortality in men with coronary heart disease. Heart. 2010 Nov;96(22):1821-5.
http://www.ncbi.nlm.nih.gov/pubmed/20959649

Morgentaler A, Bruning CO 3rd, DeWolf WC.. Occult prostate cancer in men with low serum testosterone levels. JAMA. 1996 Dec 18;276(23):1904-6.
http://www.ncbi.nlm.nih.gov/pubmed/8968017

Morgentaler A. Testosterone and prostate cancer: an historical perspective on a modern myth. Eur Urol. 2006 Nov;50(5):935-9.
http://www.ncbi.nlm.nih.gov/pubmed/16875775

Morgentaler A, Lipshultz LI, Bennett R, et al. Testosterone therapy in men with untreated prostate cancer. J Urol. 2011 Apr;185(4):1256-60.
http://www.ncbi.nlm.nih.gov/pubmed/21334649

Muller RL, Gerber L, Moreira DM, et al. Serum testosterone and dihydrotestosterone and prostate cancer risk in the placebo arm of the Reduction by Dutasteride of Prostate Cancer Events trial. Eur Urol. 2012 Nov;62(5):757-64.
http://www.ncbi.nlm.nih.gov/pubmed/22658758

Rhoden EL, Morgentaler A. Risks of testosterone-replacement therapy and recommendations for monitoring. N Engl J Med. 2004 Jan 29;350(5):482-92.
http://www.ncbi.nlm.nih.gov/pubmed/14749457

Shores MM, Matsumoto AM, Sloan KL, et al. Low serum testosterone and mortality in male veterans. Arch Intern Med. 2006 Aug 14-28;166(15):1660-5.
http://www.ncbi.nlm.nih.gov/pubmed/1690880

Tindall DJ, Rittmaster RS. The rationale for inhibiting 5alpha-reductase isoenzymes in the prevention and treatment of prostate cancer. J Urol. 2008 Apr;179(4):1235-42.
http://www.ncbi.nlm.nih.gov/pubmed/18280514

Vigen R, O’Donnell CI, Barón AE, et al. Association of testosterone therapy with mortality, myocardial infarction, and stroke in men with lowtestosterone levels. JAMA. 2013 Nov 6;310(17):1829-36.
http://www.ncbi.nlm.nih.gov/pubmed/24193080

Vigen R, O’Donnell CI, Barón AE, et al. Association of testosterone therapy with mortality, myocardial infarction, and stroke in men with low testosterone levels. JAMA. 2013 Nov 6;310(17):1829-36.
http://www.ncbi.nlm.nih.gov/pubmed/24193080