The 12-Year Registry: The Longest Follow-Up We Have, and Why It Isn't a Trial

Almost every randomized trial of testosterone runs for months. This registry ran for up to twelve years. Saad and colleagues followed 805 hypogonadal men with varying degrees of erectile dysfunction — 412 who elected testosterone undecanoate and 393 who did not — and reported sustained improvement in erectile function, cardiometabolic risk factors and urinary function in the treated group, with significantly lower incidence of prostate cancer, major adverse cardiovascular events and mortality.[1]

The duration is genuinely valuable and there is nothing else like it. The design means it answers a different question than it appears to.

What it did

• Design: observational, prospective, cumulative registry.
• Population: 805 hypogonadal men with different degrees of erectile dysfunction, assessed by the International Index of Erectile Function — Erectile Function Domain.
• Groups: 412 underwent testosterone therapy; 393 served as controls. Assignment was by patient choice, not randomization.
• Observation: up to 12 years.

What it found

Improvement in erectile function was significant for each successive year until year 9, with benefits stronger in men with moderate or severe ED than in men with no or minor ED. Cardiometabolic risk factors and urinary function improved across the full twelve-year window. Incidence of prostate cancer, major adverse cardiovascular events and mortality were all significantly lower in the treated men.[1]

The authors' own framing of the practical implication is worth quoting for what it concedes: patients must stay on therapy consistently for a long time to achieve maximum benefit.

The design problem, stated directly

The men chose their own group. The control arm is composed of men who were offered testosterone and declined it.

That single fact runs through every result. Men who elect a long-term injectable therapy, attend appointments quarterly for a decade, and remain in a registry differ systematically from men who decline — in health engagement, in socioeconomic position, in comorbidity, in the very motivation that also predicts taking other medications and following other advice. None of that is fixed by statistical adjustment, because the relevant variables are mostly unmeasured.

This is the same healthy adherer problem that inflates the 83,010-veteran cohort, and here it is stronger, not weaker: that study at least compared men who all received prescriptions, differing in whether levels normalised. This one compares acceptance against refusal.

How to reconcile it with the randomized evidence

Where this registry and the trials cover the same ground, the trials win — and the gaps are instructive.

Finding	Registry	Randomized evidence
Erectile function	Sustained improvement over years	TRAVERSE: desire and activity improved, erectile function did not
Prostate cancer	Significantly lower on treatment	TRAVERSE: no significant difference
Mortality / MACE	Significantly lower on treatment	TRAVERSE: MACE statistically equivalent; Hudson: OR 1.07

On erectile function the divergence is direct: an observational registry reports years of improvement where a randomized trial of 1,161 men found none. The likeliest explanation is not that the trial was insensitive.

A treatment that genuinely lowered prostate cancer, cardiovascular events and all-cause mortality would be one of the more remarkable drugs in medicine. That no randomized trial has reproduced any of those three is the strongest argument that what the registry captured is largely the difference between the men, not the effect of the molecule.

What it does contribute

Discounting the comparative claims does not make the study worthless.

1. Duration nothing else matches. Twelve years of prospective follow-up on men receiving testosterone undecanoate is a real observational safety record, and no alarming signal emerged over that span.
2. A dose-of-time observation. The finding that benefit accrued year over year, and that stopping forfeits it, is consistent with how a replacement therapy should behave.
3. A severity gradient. Larger effects in men with worse baseline ED is an internal pattern that random noise would not typically produce.

Bottom line

The longest prospective follow-up available on testosterone therapy reports broad, sustained benefit — and compares men who chose treatment against men who refused it, which is the weakest comparison in observational research. Read it as long-horizon safety observation and as a description of what happens to men who stay on therapy for a decade. Do not read its mortality, prostate or erectile findings as evidence of causal benefit; on all three, larger randomized trials say otherwise.

References

1 source

1. Saad F, Caliber M, Doros G, et al. Long-term treatment with testosterone undecanoate injections in men with hypogonadism alleviates erectile dysfunction and reduces risk of major adverse cardiovascular events, prostate cancer, and mortality. Aging Male. 2020;23(1):81–92. doi:10.1080/13685538.2019.1575354 · PMID 30782054