Testosterone Study Is Disappointing For More Than The Obvious Reasons

Testosterone

Testosterone replacement has been a controversial issue, despite a large number of physicians who prescribe it, FDA approval of products for mass marketing, and a number of advocates that see it as an anti-aging and performance enhancing drug for aging men.  The science behind is has been surprisingly slow.  In an era where inequities in health care are often argued by race or sex, this seems to be one where the science at least lags far behind what is known for women.  Expert opinion about testosterone replacement has changed very little over the years, despite much popular literature about how this has occurred for years in Europe and there is widespread benefit for physical and emotional well being of men as they age.  From a clinical perspective, psychiatrists are seeing more and more middle-aged men on testosterone replacement.  This treatment modality has significant implications for any medical treatment provided by psychiatrists as well as associated comorbidity.  From a psychiatric perspective it is also important to determine if the testosterone use is more than indicated, is associated with increasing muscle mass and body image changes, and if it occurring in the context of unreasonable expectations like preserving youth , reversing the effects of aging, or athletic performance enhancement.

In the most recent study of men 65 years of age and older the authors examined the effects of treatment with testosterone gel for a period of one year (1).  The total number of subjects was 790 out of a total of 51,085 men who were screened.  In order to qualify for the trial testosterone level had to be sufficiently low (less than 275 ng/dl) on one of two samples and that eliminated 85% of all of the screened men.  The trial was designed to look at physical function, sexual function, and vitality.  The treatment resulted in testosterone levels that were in the mid-normal range for 19 to 40 year old men.  Despite considerable medical comorbidity in the recruited patients there were clear cutoffs for medical comorbidity.  Those cutoffs included a history of prostate cancer or risk of prostate cancer on the Prostate Cancer Risk Calculator, other conditions known to cause hypogonadism, medications known to effect the concentration of testosterone, high cardiovascular and severe depression (PHQ-9 score >20).   The overall results showed that there was moderate benefit on sexual functioning and some benefit with mood and depressive symptoms but no effect on physical functioning or vitality.  A look at these measurements is instructive.

Physical Function was assessed using the 6 minute walk.  Recent normative data has been established for various age cohorts using subjects from several countries.   That data establishes normative curves for age cohort stratified by sex.  The six-minute walk distance (6MWD) ranged from 559 +/- 80 meters in 60-69 year old men to 514 +/- 71 meters in 70-80 year old men.  On two trials subjects usually walked on average 12 meter farther on the second trial.  In the present study, the criterion for a positive effects from testosterone was a walking distance increase of 50 meters.  When compared with controls, the testosterone treated men had significantly more men increasing the distance walked by 50 meters, the percentage of men exceeding this measure, the change from baseline, and the percentage of men with higher scores on the physical function domain test (PF-10) from the Medical Outcomes Study 36-Item Short Form Instrument (SF-36).  One of the concerns I had reading this methodology is the use of testosterone for physical performance enhancement.  Is it effective without using a training program?  Were the subjects in this study advised to extend their walking distances or speed?  Without it I am skeptical that it would have much of an effect.

Vitality was determined by measures on the FACIT-Fatigue score as the primary outcome measure.  Interestingly FACIT stands for Functional Assessment of Cancer Treatment and their web page links to an extensive list of modified rating scales.  I could not tell if the rating scale had been validated in normal 65 year old men without a cancer diagnosis.  No significant differences were noted between testosterone treated men and controls.  On secondary measures testosterone treated men were more likely to report higher energy at the end of the trial, but that measure was unspecified and may have been a line item in the SF-36.

Sexual Function was measured primarily by question 4 of the Psychosexual Daily Questionnaire.  This item is a 12 item yes-no checklist of male sexual behaviors, thoughts, and fantasies.  Greater testosterone levels achieved in treatment correlated with higher testosterone levels.  A secondary measure - the Derogatis Interview for Sexual Functioning in Men-II (DISF-M-II) also documented increased sexual desire and it was also used as a screen for participants in the Sexual Function arm of the trial.  The sexual function measures do appear to be more straightforward, but they are all self report measures over considerable time frames.

One of the most interesting aspects of this trial was an analysis of the side effects.  One of the main concerns in treating men with testosterone is the risk of prostate cancer that is sensitive to testosterone levels.   I have had personal communication with physicians who were treating patients with testosterone who developed widely metastatic prostate cancer.  The significant adverse effects in this trial were all summarized in a table.  There were 394 placebo treated men compared with 394 testosterone treated men.  The events were summarized according to prostate-specific antigen (PSA) level increases, diagnoses of prostate cancer, elevations of hemoglobin, myocardial infarction, stroke, all cause cardiovascular death, death and hospitalization.  Although there were a greater number of PSA elevations in the testosterone treated men, there was only 1 case of prostate cancer (compared to 0 in the controls) during the year of study and 2 cases of prostate cancer in the testosterone group and one in the placebo group in the subsequent year of the study.

The rationale for this study was try to to add to the results of previous trials that showed inconsistent results of testosterone replacement on muscle mass, body fat, sexual function, energy level, and physical function.  It is disappointing to read through all of these details and find out that there was not much of an effect and that the methodology in the 21st century is so weak.  The only good news that I can find in the article is that it is relatively safe to replace testosterone in 65 year old men who are rigorously screened for hypogonadism and prostate cancer risk.  Although selection bias for a project like this is certainly a consideration, I asked myself whether men self-selected for replacement by low testosterone ads would be rejected at the 85% level by their physicians.  Of course the researchers here point out that the risk over the long run can only be determined by larger trials or (my opinion) post marketing surveillance.  But post marketing surveillance as suggested by the FDA is really a blunt instrument compared with any type of organized pharmacosurveillance or screening all new cases of prostate cancer on a particular electronic health record for testosterone prescriptions. At some level,  I wonder if the research questions would have been answered more completely by trials focused solely on each groups, but there were a very large number of investigators.  Many of the investigators have the appearance of conflict of interest through affiliations with pharmaceutical companies and many do not.

The results of this study are consistent with what clinical psychiatrists see in their practice.  It is not uncommon to see men taking supraphysiological doses of testosterone and not wanting to reduce those doses.  It is common to see men on testosterone who stop taking it because there is no clear efficacy and/or because they are concerned that the women and girls in their life will accidentally come into contact with testosterone.  But let's face it - this is really not about hypoandrogenism.  That is not what is fueling the debate.  I am not an advocate of aging gracefully and accepting what happens.  I think if you have been dieting and exercising through the first 65 years of life it would be useful if medical science could provide something to help you maintain an active lifestyle and prevent the kind of morbidity that slows you down, leads to deconditioning and the host of metabolic factors that hasten aging and death.  Preventing that pathway of physical deterioration depends on maintaining attitude, muscle, joints, and bone.  This is another paper documenting that there is no clear solution to that array of problems.

It is unfortunate that medicine does not have a scientific solution at this point - even after hearing all of the unscientific hype about testosterone and growth hormone for the past 30 years.    

                  

George Dawson, MD, DLFAPA



References:

1:  Snyder PJ, Bhasin S, Cunningham GR, Matsumoto AM, Stephens-Shields AJ, Cauley JA, Gill TM, Barrett-Connor E, Swerdloff RS, Wang C, Ensrud KE, Lewis CE, Farrar JT, Cella D, Rosen RC, Pahor M, Crandall JP, Molitch ME, Cifelli D, Dougar D, Fluharty L, Resnick SM, Storer TW, Anton S, Basaria S, Diem SJ, Hou X, Mohler ER 3rd, Parsons JK, Wenger NK, Zeldow B, Landis JR, Ellenberg SS; Testosterone Trials Investigators. Effects of Testosterone Treatment in Older Men. N Engl J Med. 2016 Feb 18;374(7):611-24. doi: 10.1056/NEJMoa1506119. PubMed PMID: 26886521.

2:  Casanova C, Celli BR, Barria P, Casas A, Cote C, de Torres JP, Jardim J, LopezMV, Marin JM, Montes de Oca M, Pinto-Plata V, Aguirre-Jaime A; Six Minute Walk Distance Project (ALAT). The 6-min walk distance in healthy subjects: reference standards from seven countries. Eur Respir J. 2011 Jan;37(1):150-6. PubMed PMID: 20525717.

3:  Lee KK, Berman N, Alexander GM, Hull L, Swerdloff RS, Wang C. A simple self-report diary for assessing psychosexual function in hypogonadal men. J Androl. 2003 Sep-Oct;24(5):688-98. PubMed PMID: 12954659.


Brief Video:

The two minute NEJM video summarizing this trial can be found at this link.  Interestingly the video attributes sexual function, vitality, and physical function to testosterone decrease with age and yet full replacement to the point where it is equivalent to levels in much younger men does nothing for physical function or vitality.

"Low T Syndrome" and the Fountain of Youth

Psychiatrists are seeing increasing numbers of male patients who are being treated for low testosterone.  The symptoms of "Low T" are being promoted as a reason to get assessed and treated with testosterone.  Not surprisingly, "Low T" is a highly successful pharmaceutical company promotion.  So successful that testosterone seems to have disappeared from the vernacular, replaced by "T".  I have had the opportunity to follow this controversy for the last 30 years.  Back in the days when there were a subgroup of psychiatrists who considered themselves to be "microendocrinologists" testosterone, LH, FSH, and GnRH were studied along with the components of the hypothalamic-pituitary-adrenal and hypothalamic-pituitary-thyroid axes,  particularly in depressives.  Despite a lengthy but low intensity research effort on gonadal steroids in both men and women the data on baseline neuroendocrine correlates and results of supplemental treatments are equivocal.  That said, most psychiatrists have encountered women who have often had a significant response to treatment with gonadal steroids.  Seeing a consistent treatment effect is difficult.

As a clinical psychiatrist, I have found that the clearest information on drugs is available in the FDA approved package insert.  In the case of the product being marketed by the "Low T" ads is a form of testosterone gel that is available in two different strengths.  The only indication for the the product according to that package insert is primary hypogonadism (congenital or acquired) or hypogonadotropic hypogonadism (congential or acquired).  Any use for treating depression or augmenting antidepressants is off label use.  If testosterone was effective as an augmenting agent it would join the ranks of most antidepressant augmenting agents as being an off label prescription.  There are three related issues.

The first is the diagnosis of primary or hypogonadotropic hypogonadism.  In most cases, my speculation would be that a middle aged man sees or hears about the "Low T" ad and goes in to see their primary care physician and a testosterone level is ordered.  Most authoritative sources like UpToDate state that testosterone replacement should only occur in men who are hypogonadal.  Making that determination generally requires two low testosterone levels or in the indeterminant cases some expertise in the hypothalamic-pituitary-gonadal axis (HPG), as well as access to a laboratory with some specialized endocrine capabilities.  This is the level where most assessments seem to break down.  The range I am used to seeing is total testosterone levels ranging from about 300-1,000 ng/ml.  Scattergrams of testosterone versus behavioral parameters of interest like libido, aggression, and energy usually show low levels of correlation.  Many men are getting treated for (like the commercial suggests) low testosterone, or a level in the low normal range but not in the deficient range.  That takes treatment into the realm of off label indications because they do not have a diagnosis of hypogonadism.  A definitive algorithm (1) from endocrinologists is available including when to refer to an endocrinologist.

The second issue is whether there is any evidence testosterone either treats depression or is an effective agent to augment the effects of antidepressants.  That would conceivably move testosterone to the level of the augmenting agents used in the STAR*D study of depression.  The best guidance in the literature comes from Pope, et al (2) article on testosterone replacement in men 65 years of age or younger taking serotonergic antidepressants, a total testosterone level of less than or equal to 350 ng/ml, a PSA less than 4.0 ng/ml and an incomplete response to the antidepressant.  The author's conclude that there were no significant differences in response to testosterone or placebo gel.  Their conclusion is that the current practice of testosterone supplementation of antidepressants is not supported, but that there may be identifiable subgroups in larger studies.

Pope's observations are also critical in that he is an expert in anabolic androgenic steroid (AAS) abuse and has observed a euphorigenic hypomanic response to both AAS and prescription testosterone.  In his article he cites this response occurring in about 4.8% of 105 volunteers taking the equivalent of 500 mg/week of testosterone or an equivalent.  These observations are critical because they factor in addictive behaviors associated with substance use and Pope's group has proposed criteria for anabolic-androgenic steroid dependence.  The criteria highlight that fact that there are a number of associated mood symptoms including depression during the withdrawal phase.  Screening for an AAS use disorder and associated comorbidity like muscle dysmorphia

The third issue is the risk benefit analysis and that makes testosterone as an augmenting agent more unique than the other STAR*D agents.  Testosterone has unique medical risks beyond any risk of an additive effect with an antidepressant.  The main risk with other augmenting agents is usually rare cases of serotonin syndrome or side effects specific to the agent.  They are essentially being prescribed for the same indication.  With testosterone, there is no professional body advocating for supplementation in men with a eugonadal state and the risks may be significant starting with the contraindications (breast cancer, known or suspected prostate cancer) and warnings (benign prostatic hypertrophy, exposure to women and children, edema, sleep apnea, and the need to monitor a number of biochemical parameters).  There have also been recent articles showing a possible correlation with a number concerns about increased myocardial infarction, ischemic stroke, and other mortality (3,4,5).

So for all of you psychiatrists out there who are being referred men who are being treated for "Low T" or being sent to you with a question about testosterone there are a couple of necessary steps at this point.  Make sure that a diagnosis of hypogonadism has been established.  Let your patient know that testosterone supplementation of antidepressants in eugonadal men at this time is experimental and carries risks.  I would also inquire about a past history of anabolic androgenic steroid use, their conscious experience of that use ranging from mood changes to body image concerns and any prior psychiatric history including a history of addictions or using performance enhancing drugs.  For men considering an evaluation and treatment for "Low T" it is much more complex than filling out an online questionnaire on pharmaceutical company website.  Have realistic expectations about what you can expect, especially if someone is suggesting that testosterone supplementation is a treatment for depression.  Take a good look at the risk and consider that there will probably never be a major study that takes a good look at this issue with a large population of men.  The prospective studies will probably be similar to Pope, et al of about 100 men followed for a short period of time and the retrospective studies will have some innovative designs but they will also be limited by selection factors and significant stratification factors.  That generally means that additional information about risk may only be available in the form of FDA warnings from post marketing surveillance or their own analysis of data that may not be publicly available.

George Dawson, MD, DFAPA

References:

1.  Dandona P, Rosenberg MT. A practical guide to male hypogonadism in the primary care setting. Int J Clin Pract. 2010 May;64(6):682-96. doi: 10.1111/j.1742-1241.2010.02355.x. Review. PubMed PMID: 20518947; PubMed Central PMCID: PMC2948422

2.  Pope HG Jr, Amiaz R, Brennan BP, Orr G, Weiser M, Kelly JF, Kanayama G, Siegel A, Hudson JI, Seidman SN. Parallel-group placebo-controlled trial of testosterone gel in men with major depressive disorder displaying an incomplete response to standard antidepressant treatment. J Clin Psychopharmacol. 2010 Apr;30(2):126-34. doi: 10.1097/JCP.0b013e3181d207ca. PubMed PMID: 20520285

3.  Vigen R, O’Donnell CI, Barón AE, et al. (2013) Association of testosterone therapy with mortality, myocardial infarction, and stroke in men with low testosterone levels. JAMA 310:1829-1836.

4.  Basaria S, Coviello AD, Travison TG, et al.  Adverse events associated with testosterone administration. N Engl J Med. 2010 Jul 8;363(2):109-22. doi: 10.1056/NEJMoa1000485. Epub 2010 Jun 30. PubMed PMID: 20592293; PubMed Central PMCID: PMC3440621.

5. Cappola AR. Testosterone therapy and risk of cardiovascular disease in men.  JAMA. 2013 Nov 6;310(17):1805-6. doi: 10.1001/jama.2013.280387. PubMed PMID: 24193077.

Tip:  Follow the lead authors of these references on Medline and you will have a comprehensive look at the literature in this area.