Background

Benfotiamine is a synthetic, fat-soluble precursor of Vitamin B1 with high bioavailability. Taken orally, benfotiamine produces supra-physiological blood levels of the active coenzyme form of thiamine. This supplement is available by prescription in some countries, and over the counter in the U.S. and other places. It is used to treat diabetic neuropathy at doses of 150-300 mg per day.

Thiamine-dependent enzymes function in glucose metabolism. Thiamine deficiency, and reduced activity of thiamine-dependent enzymes in the brain, have been documented in Alzeimer's disease patents (e.g. Gibson et al., 1998). This deficiency has been related to changes in brain glucose metabolism and cognitive function (Sang et al., 2018). Earlier trials of high-dose conventional thiamine demonstrated little or no benefit on cognition (e.g. Meador et al., 1993). A pilot trial of benfotiamine in five people with Alzheimer’s reported an increase of 3.2 points on the MMSE after 18 months of 300 mg daily, but it lacked a placebo group for comparison (Pan et al., 2016).

Preclinical work from two different groups showed that increasing thiamine to very high levels using benfotiamine supplementation diminished pathology, was neuroprotective, and improved behavior deficits in mouse models of amyloidosis and tauopathy (Pan et al., 2010; Tapias et al., 2018).

Findings

From February 2015 to July 2020, a Phase 2 pilot study in New York tested the ability of benfotiamine to improve cognition in people with mild cognitive impairment or mild AD dementia. Seventy-one participants with PET-confirmed brain amyloid took 600 mg benfotiamine or placebo daily for one year, against a primary outcome of change from baseline in the ADAS-Cog. Secondary outcomes were change in brain glucose metabolism in the posterior cingulate by FDG-PET, as well ADCS-Activities of Daily Living, Neuropsychiatric Inventory, Clinical Dementia Rating, and Buschke Selective Reminding test. According to published results, the trial missed its primary endpoint. The benfotiamine group had a 43 percent slowing in decline on the ADAS-Cog, which fell short of statistical significance. Decline on the CDR was significantly delayed by 77 percent; no differences were noted in other secondary outcomes (Gibson et al., 2020). An exploratory analysis suggested improvement in brain glucose utilization on FDG-PET, with a better response in ApoE4 noncarriers than carriers. Benfotiamine increased blood thiamine levels 161-fold and significantly decreased the levels of abnormally glycated fats and proteins in blood, indicating normalization of glucose metabolism. No treatment-related adverse events were reported. A separate serum metabolomic and lipidomic profiling analysis was published (Bhawal et al., 2021).

According to a poster presentation at the 2022 CTAD conference in San Diego, a follow-on Phase 2 trial is planned to test 18 months of benfotiamine in 406 people with mild AD (Feldman et al. 2022 CTAD abstract, Gibson et al., 2022). The adaptive design will initially compare safety of 600 or 1,200 mg to placebo, then randomize participants to the highest safe dose or placebo. The primary outcome will be within-person change from baseline in ADAS-cog and CDR-SB. Secondary outcomes include other measures of cognition, pharmacokinetics, safety and tolerability, MRI, plasma biomarkers of amyloid, tau and neurodegeneration, and age-related glycation end products are exploratory endpoints. This multicenter study, coordinated by the ADCS in San Diego, is supported by National Institute on Aging (press release). It has not appeared in a trial registry yet.

Also at CTAD, another group reported on a trial in China that compared one year of 300 or 600 mg benfotiamine to placebo, as an add-on to donepezil in 302 clinically diagnosed AD patients. The study found no significant change in the primary outcome of ADAS-Cog in the group as a whole (Pan et al. 2022 CTAD abstract).  The authors did report a dose-dependent slowing of decline on ADAS-Cog in the subgroup of people with moderate AD (MMSE between 11 and 19), but not in those with mild AD (MMSE 20-24). Adverse events were similar between treatment and placebo. This group had previously published Phase 1 results, where benfotiamine had similar safety findings at these doses (Sheng et al., 2021).

For details on benfotiamine trials, see clinicaltrials.gov.