Ginkgolide B

Pharmacology & Research

Ginkgolide B (GB) is a diterpene trilactone of Ginkgo biloba and the prototypical selective antagonist of the platelet-activating-factor (PAF) receptor — its research drug code is BN52021 ^{1Reference 1DeFeudis FV · 2000Ginkgo biloba extract (EGb 761) and CNS functions: basic studies and clinical applicationsView study →}. The single most important framing point is a naming distinction that the page keeps sharp throughout: nearly all the encouraging human data attributed to “ginkgolide B” are actually for ginkgo diterpene lactone meglumine (GDLM), an injectable mixture of ginkgolides A/B/C (plus bilobalide) used clinically in China for ischemic stroke ^{3Reference 32023RCTEfficacy and safety of ginkgo diterpene lactone meglumine in acute ischemic stroke: a randomised clinical trialView study →} — not the purified oral molecule, and not the whole leaf extract EGb 761. Isolated GB’s own human evidence is mechanistic (antiplatelet, historical anti-asthma) ^{2,13Reference 2Liu Y et al. · 2014Ginkgolide B inhibits platelet release by blocking Syk and p38 MAPK phosphorylation in thrombin-stimulated plateletsView study →Reference 131991Effects of the PAF antagonist BN52021 on PAF-, methacholine- and allergen-induced bronchoconstriction in asthmatic childrenView study →}. Its defining mechanism — PAF antagonism — underlies both its circulatory effects and its (contested) bleeding-risk signal.

1. Ischemic stroke (injection mixture)

The largest evidence — but for the mixture. In a multicentre double-blind RCT of ~3450 acute-ischemic-stroke patients, IV GDLM within 48 hours of onset increased the proportion reaching a favourable 90-day modified Rankin score versus placebo, with comparable safety ^{3Reference 32023RCTEfficacy and safety of ginkgo diterpene lactone meglumine in acute ischemic stroke: a randomised clinical trialView study →}; a smaller RCT adding GDLM to aspirin showed an antiplatelet/functional signal ^{4Reference 42023RCTAntiplatelet effect of ginkgo diterpene lactone meglumine injection in acute ischemic stroke: a randomised controlled trialView study →}; and meta-analyses are cautiously positive ^{6,7Reference 62025Meta-analysisGinkgo diterpene lactone meglumine for functional recovery in acute ischemic stroke: a systematic review and meta-analysisView study →Reference 72022Meta-analysisThe efficacy and safety of ginkgo terpene-lactone preparations in the treatment of ischemic stroke: a systematic review and meta-analysis of RCTsView study →}.

Gap: the human efficacy is for the GDLM meglumine mixture given intravenously, not oral isolated ginkgolide B, and the trials are overwhelmingly single-country with heterogeneity and bias risk flagged even in the positive meta-analyses — no RCT of pure GB for stroke exists ^{3,6Reference 32023RCTEfficacy and safety of ginkgo diterpene lactone meglumine in acute ischemic stroke: a randomised clinical trialView study →Reference 62025Meta-analysisGinkgo diterpene lactone meglumine for functional recovery in acute ischemic stroke: a systematic review and meta-analysisView study →}.

2. PAF antagonism (antiplatelet / asthma)

GB/BN52021 is the prototypical selective PAF antagonist ^{1Reference 1DeFeudis FV · 2000Ginkgo biloba extract (EGb 761) and CNS functions: basic studies and clinical applicationsView study →}. In human platelets it blocks PAF-driven Syk/p38 signalling and granule release ^{2Reference 2Liu Y et al. · 2014Ginkgolide B inhibits platelet release by blocking Syk and p38 MAPK phosphorylation in thrombin-stimulated plateletsView study →}, and inhaled BN52021 blunted PAF- and allergen-induced (but not methacholine-induced) bronchoconstriction in asthmatic children ^{13Reference 131991Effects of the PAF antagonist BN52021 on PAF-, methacholine- and allergen-induced bronchoconstriction in asthmatic childrenView study →}.

Gap: the asthma work is old and small and never became a therapy — PAF antagonists broadly failed as anti-asthma drugs — and the antiplatelet effect, while real in vitro, is likely weak at oral exposures (see Pharmacokinetics) ^{13,15Reference 131991Effects of the PAF antagonist BN52021 on PAF-, methacholine- and allergen-induced bronchoconstriction in asthmatic childrenView study →Reference 15Koch E · 2005Inhibition of PAF-induced aggregation of human thrombocytes by ginkgolides: considerations on possible bleeding complications after oral Ginkgo biloba extractsView study →}.

3. Anti-inflammatory

In mouse cerebral ischemia/reperfusion, isolated GB inhibited NF-κB activation with anti-inflammatory and anti-apoptotic effects ^{14Reference 142012AnimalInhibition of NF-κB activation is associated with anti-inflammatory and anti-apoptotic effects of ginkgolide B in a mouse model of cerebral ischemia/reperfusion injuryView study →}, as reviewed alongside microglial and cytokine data ^{16Reference 162019ReviewThe neuroprotective mechanisms of ginkgolides and bilobalide in cerebral ischemic injury: a literature reviewView study →}.

Gap: entirely animal and in-vitro for the isolate, downstream of the PAF/NF-κB mechanism rather than an independent claim, with no human anti-inflammatory endpoint ^{14Reference 142012AnimalInhibition of NF-κB activation is associated with anti-inflammatory and anti-apoptotic effects of ginkgolide B in a mouse model of cerebral ischemia/reperfusion injuryView study →}.

4. Cognitive

A pre-specified cognitive analysis within a GDLM stroke trial suggested benefit ^{5Reference 52024Clinical trialEffectiveness of ginkgo diterpene lactone meglumine on cognitive function in patients with acute ischemic strokeView study →}, and ginkgolides are framed within the CNS effects of the whole extract ^{1Reference 1DeFeudis FV · 2000Ginkgo biloba extract (EGb 761) and CNS functions: basic studies and clinical applicationsView study →}.

Gap: again the mixture, in stroke patients, as a secondary/exploratory endpoint — not isolated GB and not healthy-cognition use; the weakest tier, and it should not be generalised to nootropic use ^{5Reference 52024Clinical trialEffectiveness of ginkgo diterpene lactone meglumine on cognitive function in patients with acute ischemic strokeView study →}.

Mechanisms

Target / pathway	Effect	Relevant to
PAF receptor (defining) — competitive antagonist (GB = BN52021)	↓ platelet aggregation and granule release (Syk/p38), ↓ PAF-driven bronchoconstriction and vascular permeability	antiplatelet, bleeding signal, anti-asthma, neuroprotection
Glycine receptor — non-competitive pore blocker	antagonises inhibitory glycine-gated Cl⁻ channels (weak GABA-A activity too)	CNS excitability; convulsant ceiling at high exposure
NF-κB pathway	↓ activation → ↓ cytokines, ↓ apoptosis (preclinical)	ischemic neuroprotection, anti-inflammatory

Pharmacokinetics

Load-bearing, with a nuance. Ginkgolide B is a small, relatively polar trilactone with — for a terpene lactone — reasonable fractional oral absorption, yet only tiny absolute plasma levels: human single-dose studies of oral ginkgo preparations reported GB Cmax of roughly 1–10 ng/mL despite high estimated fractional bioavailability ^{8,9Reference 82010Pharmacokinetics of bilobalide, ginkgolide A and ginkgolide B after three different Ginkgo biloba preparations in humansView study →Reference 91995Pharmacokinetic properties of bilobalide and ginkgolides A and B after IV and oral EGb 761 in healthy subjectsView study →}. Elimination is rapid (half-life on the order of a few hours) and predominantly renal, largely as unchanged drug. The GDLM injection deliberately bypasses oral absorption, delivering far higher plasma ginkgolide exposure than any oral dose ^{10Reference 102016AnimalPharmacokinetics and tissue distribution of ginkgolide A, B and K after IV infusion of ginkgo diterpene lactones in a rat modelView study →} — which is why the clinical efficacy data come from the injectable, not from capsules. The key implication: oral exposures to isolated GB sit roughly two orders of magnitude below its in-vitro antiplatelet IC50, so oral GB’s systemic PAF effect is modest.

Clinical trials

Human trials are for the GDLM injection — a ginkgolide mixture — not purified ginkgolide B, and not the oral isolate: the JAMA Network Open 2023 RCT (~3450 patients, IV GDLM within 48 hours) ^{3Reference 32023RCTEfficacy and safety of ginkgo diterpene lactone meglumine in acute ischemic stroke: a randomised clinical trialView study →}, a smaller GDLM-plus-aspirin RCT ^{4Reference 42023RCTAntiplatelet effect of ginkgo diterpene lactone meglumine injection in acute ischemic stroke: a randomised controlled trialView study →}, an exploratory cognitive analysis ^{5Reference 52024Clinical trialEffectiveness of ginkgo diterpene lactone meglumine on cognitive function in patients with acute ischemic strokeView study →}, and cautiously positive meta-analyses that flag heterogeneity and bias ^{6,7Reference 62025Meta-analysisGinkgo diterpene lactone meglumine for functional recovery in acute ischemic stroke: a systematic review and meta-analysisView study →Reference 72022Meta-analysisThe efficacy and safety of ginkgo terpene-lactone preparations in the treatment of ischemic stroke: a systematic review and meta-analysis of RCTsView study →}. The only human study of the pure molecule is the small 1991 asthma-challenge study of BN52021 ^{13Reference 131991Effects of the PAF antagonist BN52021 on PAF-, methacholine- and allergen-induced bronchoconstriction in asthmatic childrenView study →} — there are zero RCTs of oral or purified ginkgolide B for any other indication.

Last checked: July 2026.

Toxicity & Safety

Ginkgolide B is one of the better-tolerated ginkgo constituents — distinct from the allergenic ginkgolic-acid contaminants and the seed neurotoxin ginkgotoxin (both separate entities, not GB) — and GDLM injection RCTs reported adverse-event rates comparable to placebo. The theoretically important flag is bleeding: GB is a PAF antagonist and inhibits PAF-induced platelet aggregation in vitro, but the inhibitory concentrations are roughly 100-fold above peak plasma levels after normal oral dosing, and PAF is a weak platelet agonist unlikely to drive primary hemostasis — so a meaningful hemorrhage risk from oral ginkgolides is contested, with the real-world signal coming chiefly from whole-leaf extract case reports ^{15,2Reference 15Koch E · 2005Inhibition of PAF-induced aggregation of human thrombocytes by ginkgolides: considerations on possible bleeding complications after oral Ginkgo biloba extractsView study →Reference 2Liu Y et al. · 2014Ginkgolide B inhibits platelet release by blocking Syk and p38 MAPK phosphorylation in thrombin-stimulated plateletsView study →}. Even so, theoretical additive risk with anticoagulants, antiplatelets and NSAIDs is prudent to observe, especially perioperatively and with the IV injection. At the CNS level, GB also acts as a non-competitive blocker of inhibitory glycine (and weakly GABA) receptors ^{11,12Reference 112002BN52021, a platelet-activating-factor antagonist, is a selective blocker of the glycine-gated chloride channelView study →Reference 122007Probing the pharmacophore of ginkgolides as glycine-receptor antagonistsView study →}, giving a theoretical pro-convulsant ceiling at very high exposure that is not seen at therapeutic doses.

Pregnancy & lactation

Avoid. There is no adequate human pregnancy or lactation data for isolated GB or the GDLM injection, and a PAF antagonist with an antiplatelet mechanism (PAF also has roles in implantation and parturition), combined with general ginkgo bleeding concerns, makes avoidance the prudent default.

Dosage

There is no established oral human dose for purified ginkgolide B, and nothing here is a recommendation. The stroke RCTs used the marketed GDLM injection — a hospital IV drug (roughly 25 mg once daily over about two weeks, per product-defined ginkgolide content), not a supplement and not the isolate ^{3Reference 32023RCTEfficacy and safety of ginkgo diterpene lactone meglumine in acute ischemic stroke: a randomised clinical trialView study →}; the asthma work used administered BN52021 in research settings only ^{13Reference 131991Effects of the PAF antagonist BN52021 on PAF-, methacholine- and allergen-induced bronchoconstriction in asthmatic childrenView study →}. For exposure reference (not a GB dose), oral EGb 761 at 120–240 mg/day delivers only a milligram-fraction of ginkgolide B, reaching ng/mL plasma levels ^{8Reference 82010Pharmacokinetics of bilobalide, ginkgolide A and ginkgolide B after three different Ginkgo biloba preparations in humansView study →}.

References