Comparison
Alpha-GPC vs. Citicoline (CDP-Choline)
Alpha-GPC or citicoline – which choline supplement has the better data?
The short answer
Both supply choline, both are sold as nootropics, and both have a problem that is missing from the advertising. With citicoline, it is study size: the ICTUS trial with 2,298 patients found no effect on recovery after stroke and thus refuted what smaller studies had previously suggested. With alpha-GPC, it is a safety signal: a Korean cohort study of twelve million people with nine years of follow-up found a 46 percent higher stroke risk among users, with a dose-response trend; a second, smaller cohort found no increased risk. For use in healthy people – the advertised target group – there is no robust evidence for either.
The numbers side by side
| Axis | Alpha-GPC | Citicoline (CDP-Choline) | Difference |
|---|---|---|---|
| Human evidenceHow much and how good the data in humans are | 5 | 7 | Citicoline better studied |
| MechanismHow well the chain of action is understood | 6 | 6 | even |
| Safety dataHow well safety has been studied – not how safe it is | 7 | 8 | Citicoline better studied |
| Hype gap10 means: the advertising stays level with the data | 2 | 3 | Citicoline closer to the data |
| Track record of useHow much documented use in humans exists – does not measure whether it works | 6 | 7 | Citicoline in use longer |
| Rating | Thin human evidence · Supported, with caveats | ||
“Track record of use” measures how long and how widely something has been used in humans – not whether it works. The BK-Score is a subjective assessment by Biohacking Kompakt based on published scoring rules (German). It is not a scientific rating, not the consensus of a professional society and not a medical recommendation.
Why these numbers
Alpha-GPC
The most striking contradiction in the supplement field: an RCT with 100 patients with mild cognitive impairment (BMC Geriatrics 2024) shows an advantage on the ADAS-Cog over twelve weeks – at the same time, a Korean cohort study with more than twelve million people and nine years of follow-up (JAMA Network Open 2021) found a 46 percent higher stroke risk among alpha-GPC users, with a dose-response trend. This is an observational study, confounding is not ruled out, and TMAO is discussed as the mechanism. This signal practically never appears in the advertising. A second Korean cohort (J Prev Alzheimers Dis 2025, 508,107 people with mild cognitive impairment) found fewer transitions to dementia and no increased stroke risk – in people without a transition to dementia, even a lower one. According to press reports, the placebo-controlled trials ordered by the authorities in Korea (48 weeks, mild cognitive impairment) missed the primary endpoint in 2026; the per-protocol analysis was positive (67.8 versus 60.1 percent); a publication is lacking. In May 2026, EFSA saw no safety concerns for alpha-GPC as a novel food at a limited daily intake.
Citicoline (CDP-Choline)
A textbook case of what happens when a large trial arrives: an earlier pooled analysis of smaller RCTs saw benefits in stroke; the ICTUS trial (Lancet 2012) then randomized 2,298 patients and found no difference in the global recovery endpoint after 90 days (OR 1.03; 95% CI 0.86–1.25). For nootropic use in healthy people, there are individual small, manufacturer-funded RCTs with positive secondary findings, for example on episodic memory in 100 older adults over 12 weeks, whose primary outcome, however, was missed; in 2024, EFSA saw no sufficient evidence for an effect on memory. Hence mixed rather than negative: the data clearly speak against a benefit in stroke; on use in healthy people they are thin, but not negative. Tolerability is well known from decades of clinical trials.
Frequently asked questions
How robust is the stroke signal for alpha-GPC?
It comes from an observational study, not from a randomized trial – confounding is not ruled out, for example because users more often had high blood pressure. With twelve million people analyzed and a discernible dose-response trend, however, it is not something that should be left out. TMAO, a metabolite of the gut microbiome, is discussed as an explanation. A second Korean cohort with 508,107 people with mild cognitive impairment, by contrast, found no increased stroke risk.
Why is citicoline considered refuted in stroke, even though a pooled analysis saw a benefit?
The positive pooled analysis combines older, smaller studies. The largest and methodologically most rigorous single trial came later and found no difference from placebo. When a large, clean trial contradicts a collection of small ones, the large one counts. This does not apply to use in healthy people: there, the data are thin, but not negative.
Is there any well-supported nootropic at all?
For healthy adults, the data are thin for almost all nootropics. The most reproduced are small effects of Bacopa Monnieri on memory after about twelve weeks and of L-theanine on attention after a single dose.
See all 255 ratings in the ranking (German app)
Information only, not medical advice and not a usage or dosage recommendation. Prescription-only and unapproved substances belong in a doctor's hands.