Ginkgo biloba Extract and Preventing Alzheimer Disease

Ginkgo biloba leaves and extracts are widely used over-the-counter preparations marketed in the United States as food supplements or nutraceuticals, and as such, explicit health claims are not listed in their labeling. In other countries and in the popular press, G biloba is advocated for the treatment of a broad and seemingly ever-increasing range of medical conditions.

The extract used in the Ginkgo Evaluation of Memory (GEM) study reported in this issue of JAMA by DeKosky and colleagues¹ and in many other clinical and preclinical studies is Ginkgo extract EGb 761 (Schwabe Pharmaceuticals, Karlsruhe, Germany). The extract is standardized to contain 2 major constituents: 22% to 27% flavonoids and 5% to 7% terpene lactones (ginkgolides and bilobalide).² Unique to G biloba trees, the terpene lactones consist of several ginkgolides and bilobalide.^{2 - 3} The flavonoids and ginkgolides have protean biological activity in preclinical research. The flavonoids are active as antioxidants and appear neuroprotective.^{2 ,4} Ginkgolide B is a potent antagonist of the platelet activating factor receptor.⁵ Ginkgolides A and J variously inhibit hippocampal neuron dysfunction and neuronal cell death caused by amyloid beta protein-42 (Aβ42).^{6 - 7} Ginkgolides A and J decrease Aβ42-induced pathological behaviors,⁸ enhance neurogenesis in animal models of Alzheimer disease,⁹ and inhibit Aβ aggregation,¹⁰ providing considerable rationale for G biloba extracts as potential treatments for Alzheimer disease. Some of the components of G biloba extract are as active in preclinical models of neurodegeneration and Alzheimer disease as new drug candidates being developed by pharmaceutical companies and some research universities.

The GEM trial involved 3069 participants who were not cognitively impaired or had mild cognitive impairment and were randomized to receive EGb 761 (120 mg twice daily) or identically appearing and tasting placebo and were followed up for the onset of dementia.¹ The main results were that use of EGb 761 over the median follow-up period of 6.1 years did not delay the onset of dementia due to any cause including Alzheimer disease specifically. The hazard ratios for all-cause dementia (1.12 [95% confidence interval {CI}, 0.94-1.33]) and for Alzheimer disease (1.16 [95% CI, 0.97-1.39]) numerically favored placebo and the lower limits of the 95% CIs were not compatible with any potentially meaningful level of dementia risk reduction.

The investigators also found no favorable effects of EGb 761 for reducing cardiovascular serious adverse events or total mortality, 2 of 4 secondary outcomes. The relationship of ginkgo extract use and cardiovascular disease is potentially more complex, however, in that the 25% of participants with cardiovascular disease prior to randomized treatment showed increased risk for developing dementia with EGb 761 treatment (hazard ratio, 1.56 [95% CI, 1.14-2.15], P = .006). Moreover, the infrequently occurring serious adverse events of hemorrhagic stroke favored placebo—8 events with placebo vs 16 events with EGb 761—although the difference was not statistically significant. Furthermore, this difference appeared inconsistent with the similarly infrequent incidence of vascular dementia (by definition dementia subsequent to stroke) that favored EGb 761, 17 placebo cases vs 7 with EGb 761 (hazard ratio, 0.41 [95% CI, 0.17-0.98], P = .05). These 2 latter observations may be the result of chance, often seen in post hoc analyses of very few events. The former observation, however, suggests that at least in patients aged 75 years or older with cardiovascular disease, G biloba may have risks and the decision to use this agent should be carefully considered.

Although DeKosky and colleagues do not report results of their 2 other secondary outcomes—overall cognitive decline and functional disability—it is unlikely that a trial with no difference in dementia outcomes would yield significant benefits in the cognitive and functional impairments that define the dementia outcomes. Nevertheless, the effects of EGb 761 on actual cognitive test scores and daily function ratings are important because individuals without cognitive impairment who use G biloba may expect it to noticeably improve their intellectual function over the short term but not necessarily to prevent Alzheimer disease or other dementia over the long term.

The outcomes of the GEM study should be considered within the context of the numerous clinical trials of G biloba extract assessing its cognitive effects that have been conducted over the past 2 decades. Trials in older and younger adults who do not have cognitive impairment show mixed results at best. One meta-analysis of 8 trials did not find evidence for cognitive benefits with G biloba in non–cognitively impaired participants younger than 60 years treated for up to 13 weeks.¹¹ Two placebo-controlled trials reported contradictory effects in non–cognitively impaired older adults, and the magnitude of the cognitive effects were small in the positive trial.^{12 - 13}

A 2007 Cochrane systematic review that included 35 clinical trials and 4247 participants found “inconsistent and unconvincing” evidence that G biloba had clinically significant benefits for individuals with dementia or cognitive impairment.¹⁴ One 6-month regulatory trial of 513 patients with mild to moderate Alzheimer disease sponsored by Schwabe Pharmaceuticals, conducted with the hope of gaining US Food and Drug Administration marketing approval for EGb 761, failed to demonstrate efficacy,⁴ as did a recent trial performed at British primary care sites with 120-mg/d doses of EGb 761.¹⁵

A recent trial involving 118 participants without mild cognitive impairment or dementia, all older than 85 years, who were randomized to receive G biloba extract (240 mg/d) or placebo and were followed up for 42 months showed a nonsignificant effect for G biloba to delay progression to mild cognitive impairment.¹⁶ Of potential concern, however, was that more ischemic strokes and transient ischemic episodes occurred in the G biloba group.

The GEM trial is the largest and longest randomized controlled trial of G biloba to be performed and comprises the substantial bulk of the non–industry-funded, placebo-controlled clinical database for G biloba extract. The 3069 participants in the GEM trial were continuously exposed to G biloba or placebo for a median duration of more than 6 years and can be compared with the 4247 included in the Cochrane review who were mainly in short-term trials of 6 to 26 weeks. Future analyses of treatment-related adverse events from GEM will provide vital information for those who are still considering using the extract.

There is an acute recognition that the effective pharmacological treatment of Alzheimer disease needs to begin early in the course of the neuropathogenesis, before symptoms of dementia manifest. The GEM trial also serves to demonstrate the feasibility of primary prevention trials for Alzheimer disease and could serve as a model for other Alzheimer disease prevention trials.¹⁷ Notably, participants were efficiently recruited, not from clinic patients or by advertising, but rather proactively from the community using voter registration records and mailing lists, and are more representative of the population. Systematic and centralized methods were used to maintain participation and medication adherence.

One consideration with respect to the generalizability of results is that the trial, by design, excluded the substantial proportion of persons who had developed Alzheimer disease earlier in life than age 75 years and included many participants, mainly those with mild cognitive impairment, who almost certainly already had the pathology of Alzheimer disease but did not yet have dementia. This is particularly likely given that the incident dementia rate was about 42% in the participants with mild cognitive impairment at baseline. Although the GEM trial is mainly a primary prevention trial of late-age-onset Alzheimer disease, it is partly a secondary prevention trial of individuals with incipient Alzheimer disease. The outcomes in the mild cognitive impairment group can be interpreted as outcomes in early Alzheimer disease as well. Regardless, there were no clinical effects for ginkgo extract in either group.

Despite 2 decades of research with standardized extracts of G biloba, considerable uncertainty about its pharmacology and clinical effects remains. Preclinical scientific reports exude promise but generally have not identified the relevant active molecules of this biochemically complex extract, and the preclinical promise has not translated to clinical research benefits. The clinical research, in turn, has not adequately defined potential cognitive indications, potentially effective dosing ranges, pharmacodynamic markers, or convincing evidence for efficacy for any one cognitive condition.

The GEM study adds to the substantial body of evidence that G biloba extract as it is generally used does not prevent dementia in individuals with or without cognitive impairment and is not effective for Alzheimer disease. With respect to preventing stroke or minimizing the damaging sequelae from stroke, the slight evidence is vague and contradictory. Users of this extract should not expect it to be helpful. Moreover, the potential adverse effects of G biloba extract illustrate why it is untenable to recommend a drug or nutraceutical in the absence of efficacy evidence simply because it could possibly help and initially appears harmless.