July 22, 2019
As more disappointing results emerge from anti-amyloid drug trials in Alzheimer's disease (AD), there is growing interest in novel treatment approaches for this condition.
One such approach is based on the hypothesis that Porphyromonas gingivalis (Pg), the bacteria involved in periodontal disease, may cause AD. The biopharmaceutical company Cortexyme Inc is testing this theory with an investigational agent COR388, which targets gingipains, the toxic proteases released by Pg.
Early results show the drug is well tolerated and promising in terms of biomarker findings. Organizers hope that a phase 2/3 trial of the treatment now under way will provide definitive efficacy results.
"Our findings suggest that there's hope for a new and very different approach to Alzheimer's disease", Michael Detke, MD, PhD, chief medical officer for Cortexyme, told Medscape Medical News.
The findings were presented here at the Alzheimer's Association International Conference (AAIC) 2019.
The gingipain hypothesis assumes that Pg is a key etiologic agent in AD. The thinking is that the infection causes downstream inflammation and signs of AD such as tau tangles, and amyloid-beta.
Research has already found that people with periodontal disease are at higher risk for AD. Some studies have shown that gum disease precedes AD, which dispels the idea that the bacteria is an effect of AD rather than a cause, said Detke.
The most important causal data come from a mouse model of AD, he said. "Several studies show that if you put Pg in the mouth of a mouse, it causes all the signs of Alzheimer's disease."
Studies have also shown that most patients with AD have Pg present in the brain. In collaboration with other groups, Detke's team compared brain bank samples of patients with AD and age-matched controls who did not have AD. They found that the AD patients were statistically significantly more likely to have been infected with Pg.
"Almost 100% of the samples from Alzheimer's patients had Pg, and about one third or maybe 40% of the control brains had Pg," Detke said.
It makes sense that some individuals without AD would have the infection, as the hallmark pathological signs of the disease appear about 20 years before clinical symptoms, said Detke.
Several factors likely determine why one individual with Pg gets AD while another with the pathogen does not. For instance, it may be that some individuals have better blood–brain barrier function or a better immune response.
"This is pretty common in the infectious disease area," said Detke. He noted that while the Helicobacter pylori (H pylori) bacteria is closely linked to ulcers, only about 10% of those with the bacteria actually get an ulcer.
For the study, the investigators administered cognitive tests to a small sample of subjects with mild to moderate AD — 6 taking COR388 (50 mg of the oral drug twice daily) and 3 taking placebo.
One of these tests examined change from baseline to 28 days on the mini-mental state exam (MMSE). Here, subjects taking COR388 had about a 1-point improvement while those on placebo worsened by about half a point. The improvement in the treatment group almost reached statistical significance (P = .052).
In addition, the study results showed beneficial trends in other cognitive tests.
The researchers also examined blood levels of RANTES, a key inflammatory biomarker. Levels of RANTES dropped by about 30% in the treated group (P <.01 vs placebo). They also investigated cerebrospinal fluid (CSF) levels of ApoE fragments, a marker of gingipain activity in the brain.
"We know that ApoE is chopped up into fragments in the brains of people with Alzheimer's, and we know that those fragments cause neuro toxicity," said Detke.
If COR388 is affecting Pg, this should reduce the number of these fragments. Again, investigators found a 30% reduction in CSF ApoE fragments in the AD group (P <.05).
The drug was well tolerated in phase 1 a/b single ascending dose (SAD), and multiple ascending dose (MAD) studies. There were only a few side effects such as headache and dizziness, and these were mild. Detke noted that there seemed to be a higher rate of side effects in the placebo group.
"This drug targets a bacterial protein. It's not supposed to, and doesn't seem to, be hitting anything in the human body, so you would expect it to be safe and have minimal side effects."
Detke's team is reaching out to possible collaborators to try to determine if Pg is linked to dementias other than AD. "That's definitely possible and we're exploring it," he said.
The company has launched the phase 2/3 GingipAIN inhibitor (GAIN) for treatment of Alzheimer's disease trial. The study includes 570 patients with mild to moderate AD (MMSE score 12-22) at 100 sites in the US and Europe.
Subjects have been randomly assigned to placebo or to a low (40 mg twice daily) or high dose (80 mg twice daily) of COR388, for 48 weeks, after which there will be a 6-week safety follow-up.
"The 40 mg dose produces levels in humans that are comparable to, or a little higher than, blood levels in mice that worked," said Detke.
"We have chosen 40 mg to be sort of minimally effective and then 80 mg should produce blood levels that are at least twice that level, but both doses are still well within the safety range."
Although the study is not enrolling based on the presence or absence of gum disease, about half the sites will include dental assessments of participants, said Detke.
Results of the GAIN study are expected in late 2021.
Asked to comment on this new research for Medscape Medical News, Rebecca Edelmayer, PhD, director of scientific engagement at the Alzheimer's Association, said "it's a new way of thinking" about treating AD.
"It's highlighting some of the new avenues, new ways, that we are continuing to try to build and foster a pipeline of innovation for developing new treatments for Alzheimer's disease," she said.
However, much of the data related to a bacterial link to AD "is still very preliminary" and "we have to learn more", said Edelmayer.
"We still need a lot more evidence to prove any kind of causality."
The role of infectious agents in AD — including bacteria and viruses — is a hot topic in neuroscience. The AAIC 2019 hosted what organizers billed as a "robust discussion on this emerging provocative topic" as well as a related press briefing. The panel included noted experts in the field.
While some speakers agreed there's evidence suggesting microbes and/or viruses in the brain may trigger immune reactions related to the buildup of amyloid plaques and tau tangles, others sounded a cautionary note.
One of these was Michael Heneka, MD, PhD, professor of Clinical Neuroscience, Department of Neurodegenerative Diseases and Geriatric Psychiatry/Neurology, University of Bonn Medical Center, and the German Center for Neurodegenerative Disease. Heneka said several questions need to be answered before antibacterial therapies should be tested in patients.
"For example, the precise time point in the pathogenesis of Alzheimer's at which bacteria may enter the brain needs to be determined, since these phenomena may present just late-stage events," Heneka said in a statement released by the Alzheimer's Association.
"Furthermore, it remains elusive how bacteria would overcome the intrinsic innate immune defense of microglia that usually shield the brain from such invasion."
The research related to the role of infectious agents in AD "reinforces the complexity of Alzheimer's Disease" and highlights the importance of sharing data across the research community, Maria C. Carrillo, PhD, chief science officer at the Alzheimer's Association, said in the press release.
"No stone should be left unturned," Carrillo said, "in the vigorous search for better treatments, prevention, and a cure for Alzheimer's Disease."
The study was funded by Cortexyme. Detke is an employee of Cortexyme.