Does testosterone make you age faster? The evidence points the other way

Nothing in this research shows testosterone making men age faster, and most of it points the other way. Researchers compared men's testosterone against their biological age, an estimate of how worn a body looks in its blood chemistry rather than how many birthdays it has had, across four groups: 6,976 men in the United States, 2,080 and 2,133 in China, and 6,058 in the UK Biobank, a long-running British health study [2].

In all three of the groups measured at a single point in time, men with more testosterone had a smaller gap between their biological age and their real age [2]. The British group was tested twice, and the men whose testosterone had risen between the two readings showed less of that gap than the men whose testosterone had fallen [2]. The pattern held in every subgroup the researchers checked [2].

The clearest test was taking the hormone away. 38 men being treated for prostate cancer with androgen deprivation therapy, drugs that shut testosterone production down to slow the tumour, were measured before and after, and their biological ageing sped up once the hormone was gone [2].

If testosterone did accelerate ageing, the suspect would be oxidative stress, the chemical damage left behind when reactive oxygen molecules attack cells faster than the body repairs them. One trial tested that directly. 38 men aged 60 to 78 with low testosterone and a waist of 94 cm or more were randomly given either testosterone gel or an identical gel containing no hormone for 24 weeks, with neither the men nor their doctors told which they were using [7].

Oxidative stress fell rather than rose. A damage marker the body flushes into urine when reactive oxygen attacks its RNA, the working copies a cell makes of its genetic instructions, dropped by about 30% over the 24 weeks, a bigger fall than in the men on the dummy gel, while a second marker for damaged DNA, the instructions themselves, did not change [7]. The men on testosterone also gained lean tissue, the weight of everything in the body that is not fat, and lost fat [7]. The men who gained the most lean tissue showed the biggest drops in the RNA marker, which suggests the change in body composition, rather than the hormone itself, did the work [7].

The question is usually asked backwards: ageing damages the machinery that makes testosterone, which is why the two fall together. In mice, the Leydig cells, the cells inside the testicles that manufacture testosterone, are unusually vulnerable to age [1]. Young testicles hold far more ketone bodies, a fuel the body makes out of fat, than the bloodstream does, and old ones lose both that fuel and the enzyme needed to make it [1].

Switching that enzyme off in young cells pushed them into senescence, a state where a cell stops dividing but keeps leaking inflammatory signals, and aged the testicles early; restoring it in old mice partly reversed the damage [1].

A second mouse study found a different route to the same place. Ageing and obesity both drove fat droplets to pile up inside Sertoli cells, the testicular cells that nurse developing sperm, which flooded them with reactive oxygen and tipped them into senescence [6]. Blocking that pathway in aged mice restored their testosterone levels [6]. In both studies the hormone fell because the gland aged, which makes low testosterone easier to read as a symptom of ageing than a cause of it.

Testosterone also acts on blood vessels, and the human evidence there points the same way. In men, low testosterone is associated with stiffer large arteries and with vessel linings that fail to relax properly, two of the features that define a vascular system getting old [4]. In naturally aged mice, testosterone treatment reduced cell senescence and the structural changes of ageing vessels, and mice bred without the receptor that signal works through got no benefit from it [5].

The core human finding is a snapshot. In three of the four groups, testosterone and biological age were measured at the same moment, which shows the two move together but not which one moves first [2]. The twice-tested British group is stronger evidence, and the prostate cancer comparison stronger still, but that was 38 men on a cancer treatment that does a good deal more to the body than lower one hormone [2].

The trial is small and short. It ran 24 weeks and enrolled 38 men, all aged 60 to 78, all with low testosterone and a waist of at least 94 cm [7]. It says nothing about younger men, men whose testosterone is already normal, or anything past six months.

None of this shows that testosterone slows ageing either, and none of it was measured in a mirror. Biological age here is a calculation from blood markers [2]; no study in this set looked at skin, hair, or whether testosterone makes you look older.

The evidence does not cover everyone:

• Women are left out. The effect of androgens, the hormone family testosterone belongs to, on vascular ageing in women is still unclear [4], and raising testosterone in female mice increased oxidative stress while the same change lowered it in males [12].
• Men whose testosterone is already normal are left out too. The trial and the review both deal with men carrying a deficiency confirmed by blood test [7] [15].
• Nor do large doses appear anywhere. These studies measured replacement for a diagnosed deficiency, not the much higher amounts taken to build muscle.
• And the mechanism has not been shown in people. The Leydig cell, Sertoli cell, vascular and oxidative stress pathways were all worked out in mice and rats [1] [5] [6] [9] [12].

Testosterone's effect on oxidative wear also depends on the tissue it is acting on. In mice it weakened the antioxidant defences of the testicles while looking protective in the prostate and nervous system, and part of the confusion comes from the body converting testosterone into oestrogen, which has antioxidant effects of its own [12].

One source in this set, published in 2016, calculated from existing life tables that replacement raised five-year survival by 9 to 10% in men with late-onset testosterone deficiency [13]. That is arithmetic on published tables rather than a trial result, and no trial here measured survival at all. Starting or stopping testosterone is a prescription decision made with a doctor, and the review that lists its benefits also lists worsening symptoms of an enlarged prostate, thickened blood, a rise in red cell count, liver toxicity and untreated sleep apnoea among its risks [15].

Take a man of 58 who feels older than he is, reads that testosterone declines with age, and comes back from a blood test at the low end of the range. On this evidence his low reading is more likely to be a signal of something else going on than the cause of his ageing. In rats, the groups made diabetic had lower testosterone and higher oxidant readings than healthy controls, while the rats made merely obese differed from controls on neither [9]. In mice, ageing and obesity were the two things that drove fat droplets into the testicular cells that support developing sperm [6]. What the research cannot tell him is whether to treat it: that calls for a deficiency confirmed by blood test alongside symptoms, and the decision, risks included, sits with a doctor [15].