Gynaecomastia on androgen receptor blockade did not track with testosterone

Design

prospective analysis within a trial of intermittent androgen receptor pathway inhibitor monotherapy, no comparator arm

Population

38 men with biochemically recurrent prostate cancer; median age 64.6 years, median prostate-specific antigen 4.38 ng/mL

Primary outcome

relationship between serum testosterone kinetics and gynaecomastia

Effect

testosterone median peak 832.5 ng/dL (median rise 63%); gynaecomastia in 16/38 (42%) with no correlation to peak or duration

Androgen receptor pathway inhibitor monotherapy, without androgen deprivation, raises testosterone through loss of negative feedback, and gynaecomastia has been assumed to follow from that rise. Thirty-eight men with biochemically recurrent prostate cancer were given three months of enzalutamide intermittently with no androgen deprivation, and testosterone and adverse events were tracked prospectively.

Testosterone rose as expected, from a median 316 ng/dL at baseline to a median peak of 832.5 ng/dL — a median increase of 63%. Gynaecomastia symptoms were reported by 16 of 38 men (42%). But neither the peak testosterone nor the duration for which it was raised correlated with whether a man developed gynaecomastia. Nor did the first course predict the second: men who had symptoms in course one did not reliably have them again.

That unpredictability is the practical finding. If the symptom cannot be predicted from the hormone level, monitoring testosterone is not a way of anticipating it, and a man who tolerated one course cannot be told he will tolerate the next. Prophylactic measures — where they are considered at all — have to be offered on the basis of the regimen rather than the individual's response so far.

Thirty-eight men is small, 16 events smaller still, and a correlation analysis on that base has limited power to detect a relationship that exists. The honest summary is that the simple hormonal explanation was not supported here, not that it has been excluded. Intermittent androgen receptor pathway inhibitor monotherapy remains an approach used in selected men with biochemical recurrence rather than a standard of care.

• Do not use testosterone level to predict or monitor for gynaecomastia on androgen receptor pathway inhibitor monotherapy
• Warn men before each course, not only the first — tolerance of one course did not predict the next
• Quote a symptom rate of around 40% on this regimen when consenting
• Assess breast symptoms directly at review; men do not always volunteer them
• Treat this as a small prospective series in a selected group, not evidence about androgen receptor blockade generally

Why it matters

The hormonal explanation for this side effect is what clinicians use to reassure patients, and it did not hold up.

Don't overread it

Thirty-eight men with 16 events cannot exclude a real relationship between testosterone and gynaecomastia — this is an underpowered null.

The statistics, in plain English

With 38 men and 16 events, an analysis looking for a correlation has very limited power: a moderate relationship between testosterone and gynaecomastia could easily be present and undetected. The absence of a correlation here is therefore weak evidence of absence rather than a demonstration that testosterone is irrelevant. The wide range of baseline testosterone (167 to 723 ng/dL) means the men started from very different places, which further complicates relating a peak value to a symptom.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.