TRAVERSE and Anemia: A Real Benefit, at Real Numbers

Testosterone deficiency causes mild anemia, and the TRAVERSE anemia sub-study asked whether replacing the hormone corrects it. Among 815 men who were anemic at baseline, anemia remitted in 41.0% of testosterone-treated men versus 27.5% on placebo at six months, with the advantage holding at 12, 24, 36 and 48 months (P = .002). Among the 4,379 men who were not anemic, significantly fewer developed anemia on testosterone than on placebo.[1]

This is one of the more straightforwardly positive results in the TRAVERSE programme — and it is worth quoting at its actual size, because inflated versions of these numbers circulate.

The mechanism is not incidental

Testosterone stimulates erythropoiesis. That is the same physiology behind the best-known side effect of therapy — rising hematocrit and polycythemia — read from the other end.

In a man whose red cell mass is already adequate, pushing it higher is a monitoring problem. In a man who is anemic because he is hypogonadal, the same effect is the treatment. One mechanism, two clinical meanings, decided entirely by where the patient started.

What the sub-study did

Nested inside the main TRAVERSE trial, at 316 US sites, enrolling between May 2018 and February 2022.

• Population: 5,204 men analyzed — 815 with anemia (mean age 64.8) and 4,379 without (mean age 63.0).
• Anemia defined as: hemoglobin below 12.7 g/dL. Remission: hemoglobin at or above 12.7 g/dL.
• Intervention: daily 1.62% testosterone gel or placebo gel, randomized with stratification for pre-existing cardiovascular disease.
• Primary endpoint: proportion of anemic participants whose anemia remitted over the study.
• Secondary endpoint: incidence of new anemia among men who were not anemic.

What it found

Correction of anemia, testosterone versus placebo, at each timepoint:[1]

Timepoint	Testosterone	Placebo
6 months	143/349 — 41.0%	103/375 — 27.5%
12 months	152/338 — 45.0%	122/360 — 33.9%
24 months	124/290 — 42.8%	95/307 — 30.9%
36 months	94/216 — 43.5%	76/229 — 33.2%
48 months	41/92 — 44.6%	38/97 — 39.2%

Overall P = .002.

Three things stand out. The placebo correction rate is substantial — between a quarter and two-fifths of untreated men corrected on their own, which is why an uncontrolled series would badly overstate the drug effect. The absolute advantage is roughly 10–13 percentage points across most timepoints, narrowing at 48 months where the sample is smallest. And prevention worked too: among men who started without anemia, fewer on testosterone developed it.

The paper also reports that changes in hemoglobin were associated with changes in energy level — a link to symptom rather than to number alone, though an association within the trial rather than a demonstrated causal chain.

On the numbers that circulate

Figures of "54.4% versus 26.3%" appear in secondary write-ups of this trial. They do not appear in the paper. No timepoint in the published results matches that pair. The correct six-month figures are 41.0% and 27.5%.

The direction of the claim is right; the magnitude is inflated by more than 13 percentage points on the treatment arm. Anyone citing this study should take the numbers from the table above.

What it does not establish

1. "Unexplained" anemia was the target. This is not evidence for testosterone in anemia from iron deficiency, blood loss, chronic kidney disease, or marrow disorders — those have their own causes and their own treatments.
2. The men were hypogonadal. Everyone in TRAVERSE had two testosterone concentrations below 300 ng/dL plus symptoms. This says nothing about testosterone as an anemia treatment in men with normal levels.
3. The same mechanism cuts both ways. A trial showing testosterone raises hemoglobin in anemic men is the same trial reminding you that hematocrit needs watching in everyone else. That monitoring requirement does not go away because the effect is beneficial for one subgroup.
4. Correction is not the same as feeling better. The endpoint was a laboratory threshold. The energy association is suggestive, not an outcome trial.

Bottom line

In hypogonadal men who are also anemic, testosterone corrects the anemia meaningfully more often than placebo — about 41% versus 28% at six months, sustained through four years — and reduces the chance of becoming anemic in men who are not. It is a real benefit at a moderate effect size, sitting on the same physiology that makes hematocrit monitoring non-negotiable for everyone on therapy.

References

1 source

1. Pencina KM, Travison TG, Artz AS, et al. Efficacy of Testosterone Replacement Therapy in Correcting Anemia in Men With Hypogonadism: A Randomized Clinical Trial. JAMA Netw Open. 2023;6(10):e2340030. doi:10.1001/jamanetworkopen.2023.40030 · PMID 37889486