Anxious distress marks out who gained most from adjunctive lumateperone

Design

Exploratory post hoc subgroup analysis of a phase 3, randomised, double-blind, placebo-controlled trial

Population

481 outpatients aged 18-65 with MDD and inadequate response to 1-2 antidepressants; 43% with DSM-5 anxious distress

Primary outcome

Change in MADRS total score from baseline to day 43

Effect

With anxious distress: LSMD -6.8 (95% CI -9.00 to -4.51), ES 0.85. Without: LSMD -3.5 (95% CI -5.47 to -1.58), ES 0.44

This is a post hoc analysis of a phase 3, randomised, double-blind trial of adjunctive lumateperone 42 mg in outpatients aged 18 to 65 with DSM-5 major depressive disorder and an inadequate response to one or two antidepressants in the current episode. Entry needed a MADRS total of 24 or more and a CGI-Severity of at least 4. Patients were randomised to six weeks of lumateperone plus antidepressant (n=239) or placebo plus antidepressant (n=242), and split for this analysis by whether they met DSM-5 criteria for anxious distress. Of 481 patients, 43% did.

The split was large. In the anxious distress group the least squares mean difference from placebo on MADRS total at day 43 was -6.8 points (95% CI -9.00 to -4.51), effect size 0.85. Without anxious distress it was -3.5 (95% CI -5.47 to -1.58), effect size 0.44. Response and remission rates were significantly greater with lumateperone in both groups, and CGI-S, the MADRS inner tension item and QIDS-SR-16 all improved in both. GAD-7 improved significantly only in the anxious distress group (P<.0001 versus P=.379).

The specifier costs nothing to apply — it is a DSM-5 specifier you can record in the consultation. What it may buy you is a better prior about who is likely to gain from augmenting rather than switching. That is a hypothesis this analysis generates, not one it tests.

• Record the anxious distress specifier at the point you are deciding between switching and augmenting
• Inner tension on MADRS improved in both groups — the anxiety signal was on GAD-7, not on that item
• Lumateperone is not widely available in India; check before you build a plan around it
• Six weeks is the horizon this trial measured, not a long-term maintenance result
• Both subgroups improved — this is about magnitude, not about who responds at all

Why it matters

It offers a way to choose between augmenting and switching that does not need a new test or a new scale.

Don't overread it

This is an exploratory post hoc subgroup analysis; it was not designed to compare the two subgroups and cannot establish that anxious distress predicts response.

The statistics, in plain English

An effect size of 0.85 versus 0.44 is a genuinely large difference in magnitude, but this was a subgroup split decided after the trial finished. The trial was powered for the whole population, not for the interaction between anxious distress and treatment, so the honest reading is that both groups benefited and the anxious group appears to have benefited more. The GAD-7 result in the non-anxious group (P=.379) is unsurprising: those patients had little anxiety to improve.

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.