enzo123dog
New member
- Awards
- 0
The Memorial Sloan Kettering Experience
I was still giddy when I decided to look up the article detailing the experience of testosterone administration to men with metastatic disease from the Memorial Sloan Kettering Cancer Institute, published in 1981 by the urologic giant of his day, Willet Whitmore, and his colleague, Jackson Fowler. The short summary of the paper was quite damning. Over a course of eighteen years, fifty-two men with metastatic disease had undergone treatment with daily T injections, usually as a last-gasp treatment for their cancer. Of these fifty-two men, forty-five had experienced an “unfavorable response,” most within the first month of treatment.
This seemed pretty grim. Maybe Huggins had been right after all, despite basing his conclusions on a solitary patient. But then I discovered something equally shocking in the fine print of this article. Of the fifty-two men studied, all but four had already been treated with castration or estrogen treatment to lower testosterone. And of these four previously untreated men, one had an early, unspecified unfavorable response, while the remaining three men continued to receive daily T injections for 52, 55, and 310 days without apparent negative effects. In fact, one of these men was reported to have had a “favorable response” to T administration.
Drs. Fowler and Whitmore were impressed by the difference in outcomes for the untreated group of four men compared with the men who had already undergone hormonal treatment to lower testosterone. To explain the lack of negative effects on the untreated men, the authors postulated the following: “Normal endogenous testosterone levels may be sufficient to cause near maximal stimulation of prostatic tumors.” In other words, raising testosterone levels beyond the normal range did not seem to cause any increased cancer growth, even in men with metastatic disease!
This important concept was lost in the headline of the study, which clearly indicated that giving testosterone to men with prostate cancer was associated with rapid onset of negative consequences in most men. One had to read the article closely to learn that the headline applied only to men who had been previously castrated. Although this article has been cited for many years as evidence that T administration causes rapid and near-universal growth of prostate cancer (PCa), the authors in fact clearly made the point that the worrisome effects of T administration did not appear to occur in their small group of men without prior hormonal treatment.
Testosterone Flare
It had been an amazing day in the library, which had long since turned to night. My head was spinning, but I wanted to tackle the last hurdle, the problem of testosterone flare. In the early 1980s, medications were developed to replace the need for surgical removal of the testicles for men with advanced prostate cancer. These medications are called LHRH agonists, and they continue to be used to this day. LHRH injections cause T concentrations to increase by 50 percent or more for seven to ten days, after which testosterone levels fall rapidly to castrate levels. This transient rise in testosterone is called testosterone flare.
Not long after LHRH agonists began to be used, there were reports of complications occurring after men began these treatments, and these complications were attributed to testosterone flare causing rapid growth of prostate cancer. These complications included the inability to urinate, worsening of bone pain, or, in the most tragic cases, paralysis due to collapse of a vertebra in which the cancer had eaten away the bone. As a result, for the last twenty years, it has been routine to add medications to block testosterone flare when starting a patient on treatment with LHRH agonists.
That night in the basement of Countway Library, I pulled all the original studies I could find of LHRH agonists, as well as reports of bad outcomes due to the flare. As I read, two things became apparent. First, many of the bad outcomes attributed to testosterone flare occurred a month or more after initiation of treatment. This meant that these complications occurred not when testosterone levels were high, but when testosterone levels had already dropped for some time to castrate levels.
Second, out of the substantial literature on LHRH agonists and prostate cancer, I could find only two articles that actually measured and reported PSA levels during the time of the testosterone flare. And here was the kicker: both articles showed absolutely no change in mean PSA values during the time of the testosterone flare! Curiously, neither article so much as mentioned this result.
PSA is an excellent indicator of prostate cancer growth. The fact that PSA did not rise in these men during the testosterone flare strongly suggested that the cancers did not grow during this time. Perhaps the complications attributed to testosterone flare were nothing more than the cancer progression that would have happened without any treatment at all.
It had been quite a day and night in the Countway Library. I left with my head spinning and a feeling that I had stumbled onto something very important. It was like the children’s story The Emperor’s New Clothes—we see what we want to see. And for two-thirds of a century, it had been assumed that raising testosterone increased prostate cancer growth. But maybe the emperor was naked.
Even in men with metastatic disease, there was no evidence I could find that raising testosterone made prostate cancer grow more than it would have anyway. Shockingly, the very publications cited so regularly to demonstrate a dangerous relationship between testosterone and prostate cancer contained evidence that this was not true.
The Paradox Resolved
Still, I was worried, because there was a bothersome unresolved paradox to explain. For decades, the storyline was that lowering testosterone levels caused prostate cancer to shrink away and raising testosterone levels caused it grow. The second part of this story was now seriously in doubt, yet the first part was obviously correct. In my own practice, I had seen the beneficial effects of lowering testosterone levels many times over in men with advanced prostate cancer. This part of Dr. Huggins’s work was indisputable. But if lowering testosterone levels caused these cancers to shrink, how was it possible that raising testosterone levels did not cause the cancers to grow? This was a paradox that needed to be solved if physicians were to accept the possibility that testosterone therapy may not increase the risk of prostate cancer.
The answer turns out to be not all that complicated. All the reports of testosterone causing rapid growth of prostate cancer occurred in men who already had extremely low testosterone levels, due to castration or estrogen treatment. Once we get beyond the near-castrate range, it is hard to find any evidence that changes in T concentrations matter at all to prostate cancer. This is essentially what Drs. Fowler and Whitmore described in their 1981 article when they suggested that “near maximal” growth of prostate cancer is provided by naturally occurring T concentrations.
The experimental proof of this concept was provided by a landmark article published in 2006 using much more sophisticated means. In this study by Leonard Marks and colleagues, men with low testosterone received injections of testosterone or a placebo every two weeks for a total of six months. At the beginning and end of the study, measurements of testosterone and DHT (the more active form of testosterone within prostate tissue) were obtained from the blood and also from the prostate itself. The results showed that although blood concentrations of testosterone and DHT rose substantially in the T injection group, as expected, the concentration of testosterone and DHT within the prostate itself did not change at all and was similar to the group that received placebo injections. In addition, biochemical markers of prostate cell growth also did not change with T injections.
This study showed in elegant fashion that raising testosterone levels in the blood did not raise testosterone levels within the prostate. It is as if once the prostate has been exposed to enough testosterone, any additional testosterone is treated as excess and does not accumulate in the prostate. In technical terms, we say the prostate has been saturated with regard to testosterone. And it is this saturation that resolves the paradox of testosterone and prostate cancer.
Saturation explains the paradox in this way. At very low levels of T, near the castrate range, prostate growth is very sensitive to changes in T concentration. Thus, severely lowering testosterone will definitely cause prostate cancer to shrink; adding testosterone back will cause the cancer to regrow. However, once we get above the point where the prostate is saturated with testosterone, adding more testosterone will have little, if any, further impact on prostate cancer growth. Experimental studies suggest the concentration at which this saturation occurs is quite low.
In other words, the old analogy I learned in training was false. Testosterone is not like food for a hungry tumor. Instead, a much better analogy is, “Testosterone is like water for a thirsty tumor.” Once the thirst has been satisfied, prostate tumors have no use for additional testosterone. And the vast majority of men with low testosterone appear to have prostates that are not particularly thirsty.
A New Concern: Prostate Cancer and Low testosterone
I no longer fear that giving a man testosterone therapy will make a hidden prostate cancer grow or put him at increased risk of developing prostate cancer down the road. My real concern now is that men with low testosterone are at an increased risk of already having prostate cancer.
When my colleagues and I published our results in 1996 from prostate biopsies in men with low testosterone and PSA of 4.0 ng/mL or less, the 14 percent cancer rate was several times higher than any published series of men with normal PSA. In 2006, Dr. Rhoden and I published a larger study of prostate biopsies performed in 345 men. The cancer rate of 15 percent in this group was very similar to the first study. But whereas the cancer rate in 1996 was much higher than anything published to that date in men with PSA of 4.0 ng/mL or less, in 2006 the perspective had changed due to an important study called the Prostate Cancer Prevention Trial.
In that study, the cancer rate among men with a PSA of 4.0 ng/mL or less was also 15 percent. Because this value is identical to what we had found in our patients with low testosterone, it was suggested that the cancer rate in men with low testosterone is the same as the normal population—neither higher nor lower. However, the average age of men in our study was a decade younger than the men studied in the Prostate Cancer Prevention Trial (fifty-nine versus sixty-nine years). Almost half the men in the other study were seventy years or older, and age is the greatest risk factor we know for prostate cancer. The way I look at these numbers is that men with low testosterone have a cancer rate as high as men with normal T who are a decade older.
More importantly, in our study of 345 men, we found that the degree of testosterone deficiency correlated with the degree of cancer risk. Men whose testosterone levels were in the bottom third of the group were twice as likely to have cancer diagnosed on biopsy as men in the upper third. This finding adds to the concern that low testosterone is a risk factor for prostate cancer.
There is now additional data from around the world associating low testosterone and worrisome features of prostate cancer. For example, low testosterone is associated with more aggressive tumors. In addition, men with low testosterone appear to have a more advanced stage of disease at the time of surgical treatment.
Whereas I originally began to perform prostate biopsies in men with low testosterone because I was worried that treatment might cause a hidden cancer to grow, I now perform biopsies in these men because I am concerned they might have an increased risk of cancer. This risk is approximately one in seven for men with PSA values less than 4 ng/mL.
Because prostate cancer tends to be curable when caught early, I feel I’ve done these men a service by finding their cancers before they have an abnormal PSA or DRE. With today’s ability to monitor men with prostate cancer, not all of these men will necessarily require treatment. But the ones who have evidence of more aggressive tumors should definitely have an advantage by having their diagnosis made early.
I was still giddy when I decided to look up the article detailing the experience of testosterone administration to men with metastatic disease from the Memorial Sloan Kettering Cancer Institute, published in 1981 by the urologic giant of his day, Willet Whitmore, and his colleague, Jackson Fowler. The short summary of the paper was quite damning. Over a course of eighteen years, fifty-two men with metastatic disease had undergone treatment with daily T injections, usually as a last-gasp treatment for their cancer. Of these fifty-two men, forty-five had experienced an “unfavorable response,” most within the first month of treatment.
This seemed pretty grim. Maybe Huggins had been right after all, despite basing his conclusions on a solitary patient. But then I discovered something equally shocking in the fine print of this article. Of the fifty-two men studied, all but four had already been treated with castration or estrogen treatment to lower testosterone. And of these four previously untreated men, one had an early, unspecified unfavorable response, while the remaining three men continued to receive daily T injections for 52, 55, and 310 days without apparent negative effects. In fact, one of these men was reported to have had a “favorable response” to T administration.
Drs. Fowler and Whitmore were impressed by the difference in outcomes for the untreated group of four men compared with the men who had already undergone hormonal treatment to lower testosterone. To explain the lack of negative effects on the untreated men, the authors postulated the following: “Normal endogenous testosterone levels may be sufficient to cause near maximal stimulation of prostatic tumors.” In other words, raising testosterone levels beyond the normal range did not seem to cause any increased cancer growth, even in men with metastatic disease!
This important concept was lost in the headline of the study, which clearly indicated that giving testosterone to men with prostate cancer was associated with rapid onset of negative consequences in most men. One had to read the article closely to learn that the headline applied only to men who had been previously castrated. Although this article has been cited for many years as evidence that T administration causes rapid and near-universal growth of prostate cancer (PCa), the authors in fact clearly made the point that the worrisome effects of T administration did not appear to occur in their small group of men without prior hormonal treatment.
Testosterone Flare
It had been an amazing day in the library, which had long since turned to night. My head was spinning, but I wanted to tackle the last hurdle, the problem of testosterone flare. In the early 1980s, medications were developed to replace the need for surgical removal of the testicles for men with advanced prostate cancer. These medications are called LHRH agonists, and they continue to be used to this day. LHRH injections cause T concentrations to increase by 50 percent or more for seven to ten days, after which testosterone levels fall rapidly to castrate levels. This transient rise in testosterone is called testosterone flare.
Not long after LHRH agonists began to be used, there were reports of complications occurring after men began these treatments, and these complications were attributed to testosterone flare causing rapid growth of prostate cancer. These complications included the inability to urinate, worsening of bone pain, or, in the most tragic cases, paralysis due to collapse of a vertebra in which the cancer had eaten away the bone. As a result, for the last twenty years, it has been routine to add medications to block testosterone flare when starting a patient on treatment with LHRH agonists.
That night in the basement of Countway Library, I pulled all the original studies I could find of LHRH agonists, as well as reports of bad outcomes due to the flare. As I read, two things became apparent. First, many of the bad outcomes attributed to testosterone flare occurred a month or more after initiation of treatment. This meant that these complications occurred not when testosterone levels were high, but when testosterone levels had already dropped for some time to castrate levels.
Second, out of the substantial literature on LHRH agonists and prostate cancer, I could find only two articles that actually measured and reported PSA levels during the time of the testosterone flare. And here was the kicker: both articles showed absolutely no change in mean PSA values during the time of the testosterone flare! Curiously, neither article so much as mentioned this result.
PSA is an excellent indicator of prostate cancer growth. The fact that PSA did not rise in these men during the testosterone flare strongly suggested that the cancers did not grow during this time. Perhaps the complications attributed to testosterone flare were nothing more than the cancer progression that would have happened without any treatment at all.
It had been quite a day and night in the Countway Library. I left with my head spinning and a feeling that I had stumbled onto something very important. It was like the children’s story The Emperor’s New Clothes—we see what we want to see. And for two-thirds of a century, it had been assumed that raising testosterone increased prostate cancer growth. But maybe the emperor was naked.
Even in men with metastatic disease, there was no evidence I could find that raising testosterone made prostate cancer grow more than it would have anyway. Shockingly, the very publications cited so regularly to demonstrate a dangerous relationship between testosterone and prostate cancer contained evidence that this was not true.
The Paradox Resolved
Still, I was worried, because there was a bothersome unresolved paradox to explain. For decades, the storyline was that lowering testosterone levels caused prostate cancer to shrink away and raising testosterone levels caused it grow. The second part of this story was now seriously in doubt, yet the first part was obviously correct. In my own practice, I had seen the beneficial effects of lowering testosterone levels many times over in men with advanced prostate cancer. This part of Dr. Huggins’s work was indisputable. But if lowering testosterone levels caused these cancers to shrink, how was it possible that raising testosterone levels did not cause the cancers to grow? This was a paradox that needed to be solved if physicians were to accept the possibility that testosterone therapy may not increase the risk of prostate cancer.
The answer turns out to be not all that complicated. All the reports of testosterone causing rapid growth of prostate cancer occurred in men who already had extremely low testosterone levels, due to castration or estrogen treatment. Once we get beyond the near-castrate range, it is hard to find any evidence that changes in T concentrations matter at all to prostate cancer. This is essentially what Drs. Fowler and Whitmore described in their 1981 article when they suggested that “near maximal” growth of prostate cancer is provided by naturally occurring T concentrations.
The experimental proof of this concept was provided by a landmark article published in 2006 using much more sophisticated means. In this study by Leonard Marks and colleagues, men with low testosterone received injections of testosterone or a placebo every two weeks for a total of six months. At the beginning and end of the study, measurements of testosterone and DHT (the more active form of testosterone within prostate tissue) were obtained from the blood and also from the prostate itself. The results showed that although blood concentrations of testosterone and DHT rose substantially in the T injection group, as expected, the concentration of testosterone and DHT within the prostate itself did not change at all and was similar to the group that received placebo injections. In addition, biochemical markers of prostate cell growth also did not change with T injections.
This study showed in elegant fashion that raising testosterone levels in the blood did not raise testosterone levels within the prostate. It is as if once the prostate has been exposed to enough testosterone, any additional testosterone is treated as excess and does not accumulate in the prostate. In technical terms, we say the prostate has been saturated with regard to testosterone. And it is this saturation that resolves the paradox of testosterone and prostate cancer.
Saturation explains the paradox in this way. At very low levels of T, near the castrate range, prostate growth is very sensitive to changes in T concentration. Thus, severely lowering testosterone will definitely cause prostate cancer to shrink; adding testosterone back will cause the cancer to regrow. However, once we get above the point where the prostate is saturated with testosterone, adding more testosterone will have little, if any, further impact on prostate cancer growth. Experimental studies suggest the concentration at which this saturation occurs is quite low.
In other words, the old analogy I learned in training was false. Testosterone is not like food for a hungry tumor. Instead, a much better analogy is, “Testosterone is like water for a thirsty tumor.” Once the thirst has been satisfied, prostate tumors have no use for additional testosterone. And the vast majority of men with low testosterone appear to have prostates that are not particularly thirsty.
A New Concern: Prostate Cancer and Low testosterone
I no longer fear that giving a man testosterone therapy will make a hidden prostate cancer grow or put him at increased risk of developing prostate cancer down the road. My real concern now is that men with low testosterone are at an increased risk of already having prostate cancer.
When my colleagues and I published our results in 1996 from prostate biopsies in men with low testosterone and PSA of 4.0 ng/mL or less, the 14 percent cancer rate was several times higher than any published series of men with normal PSA. In 2006, Dr. Rhoden and I published a larger study of prostate biopsies performed in 345 men. The cancer rate of 15 percent in this group was very similar to the first study. But whereas the cancer rate in 1996 was much higher than anything published to that date in men with PSA of 4.0 ng/mL or less, in 2006 the perspective had changed due to an important study called the Prostate Cancer Prevention Trial.
In that study, the cancer rate among men with a PSA of 4.0 ng/mL or less was also 15 percent. Because this value is identical to what we had found in our patients with low testosterone, it was suggested that the cancer rate in men with low testosterone is the same as the normal population—neither higher nor lower. However, the average age of men in our study was a decade younger than the men studied in the Prostate Cancer Prevention Trial (fifty-nine versus sixty-nine years). Almost half the men in the other study were seventy years or older, and age is the greatest risk factor we know for prostate cancer. The way I look at these numbers is that men with low testosterone have a cancer rate as high as men with normal T who are a decade older.
More importantly, in our study of 345 men, we found that the degree of testosterone deficiency correlated with the degree of cancer risk. Men whose testosterone levels were in the bottom third of the group were twice as likely to have cancer diagnosed on biopsy as men in the upper third. This finding adds to the concern that low testosterone is a risk factor for prostate cancer.
There is now additional data from around the world associating low testosterone and worrisome features of prostate cancer. For example, low testosterone is associated with more aggressive tumors. In addition, men with low testosterone appear to have a more advanced stage of disease at the time of surgical treatment.
Whereas I originally began to perform prostate biopsies in men with low testosterone because I was worried that treatment might cause a hidden cancer to grow, I now perform biopsies in these men because I am concerned they might have an increased risk of cancer. This risk is approximately one in seven for men with PSA values less than 4 ng/mL.
Because prostate cancer tends to be curable when caught early, I feel I’ve done these men a service by finding their cancers before they have an abnormal PSA or DRE. With today’s ability to monitor men with prostate cancer, not all of these men will necessarily require treatment. But the ones who have evidence of more aggressive tumors should definitely have an advantage by having their diagnosis made early.
