Hypogonadism in Elderly Men — What to Do Until the Evidence Comes
http://www.100md.com
《新英格兰医药杂志》
Testosterone treatment for elderly men made the news again recently, when the Institute of Medicine Committee on Assessing the Need for Clinical Trials of Testosterone Replacement Therapy issued its long-awaited report,1 which concluded that there is insufficient evidence that testosterone treatment benefits elderly men. What is the background behind this report, and what are its implications?
Many studies document that serum testosterone concentrations in men decrease as they age. In contrast to the precipitous and profound decrease in estradiol concentrations that occur when women enter menopause, the decrease in testosterone in men occurs moderately and gradually over a period of several decades. The serum total testosterone concentration decreases from a mean of about 600 ng per deciliter (20.8 nmol per liter) at 30 years of age to a mean of about 400 ng per deciliter (13.9 nmol per liter) at 80 years, although the range is wide at all ages. In one study, approximately 20 percent of men older than 60 years of age had total serum testosterone concentrations that were below the normal range for young men.
An essential but still unanswered question is whether this decrease in the testosterone concentration is physiologic, perhaps conveying a benefit, or pathologic, causing harm. Testosterone normally has many different effects on many different tissues, which can be explained in part by its ability to act directly through the androgen receptor, through conversion to dihydrotestosterone, which also acts through the androgen receptor, or through conversion to estradiol, which acts through the estrogen receptor (see Figure). The decrease in testosterone with aging, therefore, could affect many different tissues.
Figure. Mechanisms of Action of Testosterone, Tissues Affected by Testosterone, and Potential Effects of Testosterone-Replacement Therapy.
Testosterone has many effects in many different tissues, at least in part because it can act as three different hormones. In many tissues, it binds to the androgen receptor and acts directly as an androgen. In tissues that express the enzyme 5-reductase, it is converted to dihydrotestosterone, which binds to the androgen receptor with even greater affinity than testosterone itself and thereby has effects that a physiologic concentration of testosterone would not have. In tissues that express the enzyme aromatase, it is converted to estradiol, which acts through the estrogen receptor.
One reason for thinking that this decrease might be pathologic is that there are parallels between the consequences of frank hypogonadism due to known pituitary or testicular disease and the consequences of aging, such as decreases in bone density, muscle mass and strength, energy, and libido. Studies, usually uncontrolled, in which replacement doses of testosterone have been administered to men with frank hypogonadism and known disease have generally demonstrated a correction of these consequences of testosterone deficiency. Studies, a few of which were placebo-controlled, in which testosterone was administered to elderly men with low-normal serum testosterone concentrations have generally shown an increase in lean body mass, a decrease in fat mass, and a trend toward an increase in bone density, but no clear improvement in muscle strength, energy, or libido.
Another essential, but also unanswered, question is whether reversing this decrease in testosterone concentrations will exacerbate the testosterone-dependent diseases to which elderly men are prone, including prostate cancer, benign prostatic hyperplasia, erythrocytosis, and perhaps sleep apnea. No data, unfortunately, are available with which to answer this question. Although testosterone treatment did not result in a significant increase in the risk of these conditions in a few placebo-controlled studies, the numbers of subjects in these studies were far too small to permit the conclusion that testosterone does no harm. According to one estimate, a study would need to include 6000 elderly hypogonadal men randomly assigned to receive testosterone or placebo for six years in order to determine whether testosterone treatment increases the risk of prostate cancer by 30 percent.
The Institute of Medicine committee evaluated the existing data concerning the efficacy of treating elderly men who have low-normal testosterone concentrations (300 to 400 ng per deciliter ) and concluded that efficacy has not been demonstrated sufficiently well to justify embarking on a long-term study to determine the risks associated with testosterone replacement. Instead, the committee recommended first performing short-term, randomized, placebo-controlled studies of the effect of testosterone on several outcomes in elderly men whose testosterone concentrations are below 300 ng per deciliter and then, only if the short-term studies demonstrate efficacy, performing a long-term study to evaluate the risks.
Practicing physicians, meanwhile, remain in a quandary about the best approach to the care of elderly men with symptoms that might be caused by hypogonadism. A few basic principles can guide the approach until definitive studies are available. The first principle is that the criteria for the diagnosis of testosterone deficiency should be more stringent in the absence than in the presence of a disease that is known to cause hypogonadism, such as a pituitary macroadenoma, and should be more stringent in an elderly man — say, older than 65 years of age — than in a younger man. According to this principle, an elderly man can be considered to be hypogonadal if his early morning serum total testosterone concentration on three occasions is consistently and unequivocally subnormal — perhaps below 200 ng per deciliter (6.9 nmol per liter). A measurement of total testosterone should generally be used, because reliable assays for it are widely available, whereas the most widely available assays for free testosterone are not reliable. In a patient in whom the level of sex hormone–binding globulin is low because of obesity, so that the total testosterone concentration does not accurately reflect the free testosterone concentration, the free testosterone concentration should be measured, but only by means of equilibrium dialysis. When the serum testosterone concentration is unequivocally low, the serum luteinizing hormone concentration should be measured. An elevated value would indicate primary hypogonadism, and a nonelevated value would indicate secondary hypogonadism. If secondary hypogonadism appears to be likely, the serum concentrations of thyroxine, cortisol, and prolactin should be measured, and magnetic resonance imaging of the sellar region should be performed.
The second principle is that the replacement of testosterone should be considered only if testosterone deficiency can be demonstrated on the basis of the criteria outlined above. It is entirely possible that testosterone treatment would also benefit elderly men whose testosterone values are not quite so low — say, 200 to 300 ng per deciliter — but until the efficacy and safety of treating this group of men have been established, the prudent course is to limit treatment to men who are more severely hypogonadal, on the premise that those men are more likely to benefit.
The third principle is that elderly hypogonadal men who are treated with testosterone should be monitored to ensure that the goal of replacing testosterone is achieved. The most appropriate variable to monitor is the response of the serum testosterone concentration to treatment, especially given its relative stability when a transdermal preparation is used. The response of presenting symptoms is much harder to interpret, because the symptoms that usually lead to the diagnosis of hypogonadism are nonspecific and might well have resulted from another cause. In monitoring the serum testosterone concentration, however, it is uncertain whether the goal should be to maintain the concentration within the midnormal range for young men (i.e., 500 to 700 ng per deciliter ) or within the midnormal range for elderly men (i.e., 300 to 450 ng per deciliter ). Until more information is available, the latter seems more prudent.
The fourth principle is that elderly hypogonadal men who are treated with testosterone should be monitored for possible exacerbation of testosterone-dependent diseases, given the concern that testosterone might increase the risk of these diseases, since the reduction of even physiologic levels of testosterone can sometimes ameliorate testosterone-dependent conditions. Current treatments for metastatic prostate cancer and benign prostatic hyperplasia, in fact, are based on the reduction of testosterone secretion or action. In addition, approximately 50 percent of men older than 50 years of age harbor an occult prostate cancer, and there is concern that testosterone treatment might convert occult lesions into clinically significant ones. Until the actual risk is known, it seems prudent for physicians who prescribe testosterone to elderly hypogonadal men — even if they are hypogonadal according to the relatively strict criteria discussed above — to screen for testosterone-dependent diseases before initiating treatment and to monitor patients for their development during treatment. Rhoden and Morgentaler discuss approaches to screening in this issue of the Journal (pages 482–492).
These principles constitute a "what to do until the evidence comes" approach to testosterone treatment in elderly hypogonadal men. We can only hope that the studies recommended by the Institute of Medicine committee will be performed and will provide better evidence for future guidelines.
Source Information
From the Division of Endocrinology, Diabetes, and Metabolism, Department of Medicine, University of Pennsylvania, Philadelphia.
References
Liverman CT, Blazer DG, eds. Testosterone and aging: clinical research directions. Washington, D.C.: National Academies Press (in press).
Related Letters:
Risks of Testosterone Replacement
Shames D. A., Schaeffer E. M., Walsh P. C., Redmond G., Harris R. M., Morgentaler A., Rhoden E. L.(Peter J. Snyder, M.D.)
Many studies document that serum testosterone concentrations in men decrease as they age. In contrast to the precipitous and profound decrease in estradiol concentrations that occur when women enter menopause, the decrease in testosterone in men occurs moderately and gradually over a period of several decades. The serum total testosterone concentration decreases from a mean of about 600 ng per deciliter (20.8 nmol per liter) at 30 years of age to a mean of about 400 ng per deciliter (13.9 nmol per liter) at 80 years, although the range is wide at all ages. In one study, approximately 20 percent of men older than 60 years of age had total serum testosterone concentrations that were below the normal range for young men.
An essential but still unanswered question is whether this decrease in the testosterone concentration is physiologic, perhaps conveying a benefit, or pathologic, causing harm. Testosterone normally has many different effects on many different tissues, which can be explained in part by its ability to act directly through the androgen receptor, through conversion to dihydrotestosterone, which also acts through the androgen receptor, or through conversion to estradiol, which acts through the estrogen receptor (see Figure). The decrease in testosterone with aging, therefore, could affect many different tissues.
Figure. Mechanisms of Action of Testosterone, Tissues Affected by Testosterone, and Potential Effects of Testosterone-Replacement Therapy.
Testosterone has many effects in many different tissues, at least in part because it can act as three different hormones. In many tissues, it binds to the androgen receptor and acts directly as an androgen. In tissues that express the enzyme 5-reductase, it is converted to dihydrotestosterone, which binds to the androgen receptor with even greater affinity than testosterone itself and thereby has effects that a physiologic concentration of testosterone would not have. In tissues that express the enzyme aromatase, it is converted to estradiol, which acts through the estrogen receptor.
One reason for thinking that this decrease might be pathologic is that there are parallels between the consequences of frank hypogonadism due to known pituitary or testicular disease and the consequences of aging, such as decreases in bone density, muscle mass and strength, energy, and libido. Studies, usually uncontrolled, in which replacement doses of testosterone have been administered to men with frank hypogonadism and known disease have generally demonstrated a correction of these consequences of testosterone deficiency. Studies, a few of which were placebo-controlled, in which testosterone was administered to elderly men with low-normal serum testosterone concentrations have generally shown an increase in lean body mass, a decrease in fat mass, and a trend toward an increase in bone density, but no clear improvement in muscle strength, energy, or libido.
Another essential, but also unanswered, question is whether reversing this decrease in testosterone concentrations will exacerbate the testosterone-dependent diseases to which elderly men are prone, including prostate cancer, benign prostatic hyperplasia, erythrocytosis, and perhaps sleep apnea. No data, unfortunately, are available with which to answer this question. Although testosterone treatment did not result in a significant increase in the risk of these conditions in a few placebo-controlled studies, the numbers of subjects in these studies were far too small to permit the conclusion that testosterone does no harm. According to one estimate, a study would need to include 6000 elderly hypogonadal men randomly assigned to receive testosterone or placebo for six years in order to determine whether testosterone treatment increases the risk of prostate cancer by 30 percent.
The Institute of Medicine committee evaluated the existing data concerning the efficacy of treating elderly men who have low-normal testosterone concentrations (300 to 400 ng per deciliter ) and concluded that efficacy has not been demonstrated sufficiently well to justify embarking on a long-term study to determine the risks associated with testosterone replacement. Instead, the committee recommended first performing short-term, randomized, placebo-controlled studies of the effect of testosterone on several outcomes in elderly men whose testosterone concentrations are below 300 ng per deciliter and then, only if the short-term studies demonstrate efficacy, performing a long-term study to evaluate the risks.
Practicing physicians, meanwhile, remain in a quandary about the best approach to the care of elderly men with symptoms that might be caused by hypogonadism. A few basic principles can guide the approach until definitive studies are available. The first principle is that the criteria for the diagnosis of testosterone deficiency should be more stringent in the absence than in the presence of a disease that is known to cause hypogonadism, such as a pituitary macroadenoma, and should be more stringent in an elderly man — say, older than 65 years of age — than in a younger man. According to this principle, an elderly man can be considered to be hypogonadal if his early morning serum total testosterone concentration on three occasions is consistently and unequivocally subnormal — perhaps below 200 ng per deciliter (6.9 nmol per liter). A measurement of total testosterone should generally be used, because reliable assays for it are widely available, whereas the most widely available assays for free testosterone are not reliable. In a patient in whom the level of sex hormone–binding globulin is low because of obesity, so that the total testosterone concentration does not accurately reflect the free testosterone concentration, the free testosterone concentration should be measured, but only by means of equilibrium dialysis. When the serum testosterone concentration is unequivocally low, the serum luteinizing hormone concentration should be measured. An elevated value would indicate primary hypogonadism, and a nonelevated value would indicate secondary hypogonadism. If secondary hypogonadism appears to be likely, the serum concentrations of thyroxine, cortisol, and prolactin should be measured, and magnetic resonance imaging of the sellar region should be performed.
The second principle is that the replacement of testosterone should be considered only if testosterone deficiency can be demonstrated on the basis of the criteria outlined above. It is entirely possible that testosterone treatment would also benefit elderly men whose testosterone values are not quite so low — say, 200 to 300 ng per deciliter — but until the efficacy and safety of treating this group of men have been established, the prudent course is to limit treatment to men who are more severely hypogonadal, on the premise that those men are more likely to benefit.
The third principle is that elderly hypogonadal men who are treated with testosterone should be monitored to ensure that the goal of replacing testosterone is achieved. The most appropriate variable to monitor is the response of the serum testosterone concentration to treatment, especially given its relative stability when a transdermal preparation is used. The response of presenting symptoms is much harder to interpret, because the symptoms that usually lead to the diagnosis of hypogonadism are nonspecific and might well have resulted from another cause. In monitoring the serum testosterone concentration, however, it is uncertain whether the goal should be to maintain the concentration within the midnormal range for young men (i.e., 500 to 700 ng per deciliter ) or within the midnormal range for elderly men (i.e., 300 to 450 ng per deciliter ). Until more information is available, the latter seems more prudent.
The fourth principle is that elderly hypogonadal men who are treated with testosterone should be monitored for possible exacerbation of testosterone-dependent diseases, given the concern that testosterone might increase the risk of these diseases, since the reduction of even physiologic levels of testosterone can sometimes ameliorate testosterone-dependent conditions. Current treatments for metastatic prostate cancer and benign prostatic hyperplasia, in fact, are based on the reduction of testosterone secretion or action. In addition, approximately 50 percent of men older than 50 years of age harbor an occult prostate cancer, and there is concern that testosterone treatment might convert occult lesions into clinically significant ones. Until the actual risk is known, it seems prudent for physicians who prescribe testosterone to elderly hypogonadal men — even if they are hypogonadal according to the relatively strict criteria discussed above — to screen for testosterone-dependent diseases before initiating treatment and to monitor patients for their development during treatment. Rhoden and Morgentaler discuss approaches to screening in this issue of the Journal (pages 482–492).
These principles constitute a "what to do until the evidence comes" approach to testosterone treatment in elderly hypogonadal men. We can only hope that the studies recommended by the Institute of Medicine committee will be performed and will provide better evidence for future guidelines.
Source Information
From the Division of Endocrinology, Diabetes, and Metabolism, Department of Medicine, University of Pennsylvania, Philadelphia.
References
Liverman CT, Blazer DG, eds. Testosterone and aging: clinical research directions. Washington, D.C.: National Academies Press (in press).
Related Letters:
Risks of Testosterone Replacement
Shames D. A., Schaeffer E. M., Walsh P. C., Redmond G., Harris R. M., Morgentaler A., Rhoden E. L.(Peter J. Snyder, M.D.)