An ACE signal in early psychosis, and what it is not

Design

Cross-sectional case-control study with polygenic scoring and enzymatic assays

Population

200 participants aged 13 to 35: 122 with early-stage psychosis (onset within 2 years) and 78 healthy controls

Primary outcome

ACE protein levels in cerebrospinal fluid and serum

Effect

Lower in patients: CSF Cohen d −1.16 (P=.005), serum d −0.92 (P<.001); treatment-resistant vs not, serum d −0.50 (P=.03)

Two hundred people aged 13 to 35 — 122 with psychosis of less than two years' duration and 78 controls — had angiotensin-converting enzyme measured in cerebrospinal fluid and serum. Protein levels were lower in patients in both compartments, with large effect sizes, and the two fluids correlated with each other. Within the patient group, those with treatment-resistant illness had lower serum ACE than those without.

The mechanistic point is the dissociation. Higher ACE genetic burden for schizophrenia tracked lower protein levels but not the enzyme's canonical renin-angiotensin activity, and the treatment-resistant difference likewise appeared in protein but not activity. Whatever ACE is doing here, it is not doing it through blood pressure biology.

This is a biomarker in the first stage of its life. It is cross-sectional, from one centre, in medicated patients, and there is no threshold, no assay standard and no prospective validation. It is worth knowing about because a serum measure that separates treatment-resistant illness early would be valuable — not because anyone should order it.

• Do not send ACE levels to investigate psychosis; there is no validated cut-off.
• An ACE result sent for sarcoidosis in a patient with psychosis still means what it always meant.
• Nothing here implies ACE inhibitors help or harm in psychosis; canonical activity was unchanged.
• Treatment resistance remains a clinical definition — two adequate antipsychotic trials of adequate duration.
• Watch for prospective cohorts; cross-sectional separation is the easy part.

Why it matters

It raises the possibility of separating treatment-resistant psychosis early, by a blood test rather than by two failed drug trials.

Don't overread it

This is cross-sectional and in medicated patients; it cannot show that low ACE precedes or causes the illness.

The statistics, in plain English

Cohen d of −1.16 in cerebrospinal fluid is a large group separation, but group separation is not diagnostic accuracy: with 122 patients and 78 controls the distributions still overlap, and no sensitivity or specificity is reported. The treatment-resistant contrast (d −0.50) rests on a subgroup split made after the fact, which is the weakest part of the design and the part most likely to shrink on replication.

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.