Celecoxib and aspirin
Anti-inflammatory painkillers tested as add-ons to antipsychotics. Small trials of celecoxib were positive; the largest aspirin trials found no benefit.
At a glance
- Type
- Medicine: non-steroidal anti-inflammatory drugs (NSAIDs)
- Used for
- Tested as add-ons to antipsychotics
- Status
- Experimental
- Evidence
- EarlyHow evidence is rated
- Administration
- Tablets, added to an antipsychotic. In the trials, for 5 weeks to 4 months
- Common side effects
- Stomach upset, about as often as on placebo in the trials
- In Europe
- Tested in Germany and the Netherlands. A German trial limited to people with raised inflammation has finished but not yet reported
Clinical summaryFor health professionals
- Celecoxib (COX-2 inhibitor) 400 mg a day, added to antipsychotics: 7 trials, 438 patients; PANSS total 8.5 points better than placebo (Farag 2025). The largest single trial had 90 patients.
- TargetFlame (Germany): celecoxib only in patients with an inflammatory cytokine profile; 30 enrolled of 109 planned; completed January 2025; no results yet.
- Aspirin 1,000 mg a day: Laan 2010 (Netherlands), 70 patients, PANSS 4.9 points better at 3 months. Weiser 2021, 200 and 160 patients, the second limited to CRP above 1 mg/L: no significant benefit.
- Naproxen: no effect in a 232-patient trial (Weiser 2023).
What they are
Celecoxib and aspirin belong to the anti-inflammatory painkillers. They block cyclo-oxygenase (COX) enzymes, which make prostaglandins, messengers of inflammation. Celecoxib blocks one form of the enzyme, COX-2; aspirin blocks both forms.1 They were tested in schizophrenia on the idea that inflammation contributes to the illness in some people, as explained in Beyond dopamine (opens in a new tab).
How well they work
The first trial, in Munich, added celecoxib or a placebo to risperidone in 50 people during an acute episode; after five weeks, symptoms had improved more with celecoxib.1 A second Munich trial, in 49 people early in their illness, found the same with amisulpride.2 A US trial in 38 people with long-standing symptoms found no difference.3 In the largest trial, in China, 90 people with a first episode who had never taken antipsychotics improved more on celecoxib, though their thinking skills did not.4 Pooled across seven trials with 438 people, celecoxib improved symptoms by 8.5 points on the 30-to-210-point PANSS scale.5
A German trial, TargetFlame, gave celecoxib only to people whose blood showed raised inflammatory signals. It aimed to include 109 people and stopped at 30; it finished in January 2025 and has not reported its results.6,7
In the Netherlands, 70 people took aspirin or a placebo for three months. Symptoms improved by 4.9 points more with aspirin, and more in people whose immune system showed the greatest changes.8 Two later trials did not confirm this. In the first, 200 people took aspirin or a placebo for 16 weeks; in the second, 160 people, all with raised C-reactive protein, did the same. Neither found a clear benefit.9 Another painkiller, naproxen, had no effect in a trial of 232 people.10
Side effects
In the trials, stomach upset was about as common as on placebo, and the trials reviewed found no clear difference in side effects.11,5
Alternatives
Other medicines tested against inflammation in schizophrenia include minocycline (opens in a new tab), statins (opens in a new tab) and N-acetylcysteine (opens in a new tab). None is an established treatment.
What we don’t know yet
- Whether celecoxib helps people whose blood shows raised inflammation, as TargetFlame set out to test.
- Whether the benefit seen in small trials holds in larger ones.
