TBD Meeting (3102-19)

LOS ANGELES – A single look at the gut microbiome of patients with Alzheimer’s disease (AD) suggests an interaction between anti-inflammatory gut bacteria and long-term exposure to an investigational sigma 1 receptor agonist.

After up to 148 weeks treatment with Anavex 2-73, patients with stable or improved functional scores showed significantly higher levels of both Ruminococcaceae and Porphyromonadaceae, compared with patients who had declining function. Both bacterial families produce butyrate, an anti-inflammatory short-chain fatty acid.

Conversely, poor response was associated with a low level of Verrucomicrobia, a mucin-degrading phylum thought to be important in gut homeostasis. These bacteria live mainly in the intestinal mucosa – the physical interface between the microbiome and the rest of the body.

The data, presented at the Alzheimer’s Association International Conference, represent the first microbiome measurements reported in a clinical trial of an investigational Alzheimer’s therapy. Because they come from a single sample taken from a small group in an extension study, without a baseline comparator, it’s impossible to know what these associations mean. But the findings are enough to nudge Anavex Life Sciences into adding microbiome changes to its new study of Anavex 2-73, according to Christopher Missling, PhD, president and chief executive officer of the company.

The study, ramping up now, aims to recruit 450 patients with mild AD. They will be randomized to high-dose or mid-dose Anavex 2-73 for 48 weeks. The primary outcomes are measures of cognition and function. Stool sampling at baseline and at the end of the study will be included as well, Dr. Missling said in an interview.

Anavex 2-73 is a sigma-1 receptor agonist. A chaperone protein, sigma-1 is activated in response to acute and chronic cellular stressors, several which are important in neurodegeneration. The sigma-1 receptor is found on neurons and glia in many areas of the central nervous system. It modulates several processes implicated in neurodegenerative diseases, including glutamate and calcium activity, reaction to oxidative stress, and mitochondrial function. There is some evidence that sigma-1 receptor activation can induce neuronal regrowth and functional recovery after stroke. It also appears to play a role in helping cells clear misfolded proteins – a pathway that makes it an attractive drug target in Alzheimer’s disease, as well as other neurodegenerative diseases with aberrant proteins, such as Parkinson’s and Huntington’s diseases.

Anavex 2-73’s phase 2 development started with a 5-week crossover trial of 32 patients. This was followed by a 52-week, open-label extension trial of 10, 20, 30, and 50 mg/day orally, in which each patient was titrated to the maximum tolerated dose. The main endpoints were change on the Mini Mental State Exam and change on the Alzheimer’s Disease Cooperative Study-activities of daily living (ADCS-ADL) scale.

At 57 weeks, six patients had improved on the Mini Mental State Exam score: four with high plasma levels and two with low plasma levels, correlating to the dosage obtained. On the functional measure of activities of daily living, nine patients had improved, including five with high plasma levels, three with moderate levels, and one with a low level. One patient, with a moderate level, remained stable. The remaining 14 patients declined.

The company then enrolled 21 of the cohort in a 208-week extension trial, primarily because of patient request, Dr. Missling said. “They know they are doing better. Their families know they’re doing better. They did not want to give this up.”

Last fall, the company released 148-week functional and cognitive data confirming the initial findings: Patients with higher plasma levels (correlating with higher doses) declined about 2 points on the ADCS-ADL scale, compared with a mean decline of about 25 points among those with lower blood levels – an 88% difference in favor of treatment. Cognition scores showed a similar pattern, with the high-concentration group declining 64% less than the low-concentration group.

Sixteen patients consented to stool sampling. A sophisticated computer algorithm characterized the microbiome of each, measuring the relative abundance of phyla. Microbiome analysis wasn’t included as an endpoint in the original study design because, at that time, the idea of a connection between AD and the gut microbiome was barely on the research radar.

Things shifted dramatically in 2017, with a seminal paper finding that germ-free mice inoculated with stool from Parkinson’s patients developed Parkinson’s symptoms. This study was widely heralded as a breakthrough in the field – the first time any neurodegenerative disease had been conclusively linked to dysregulations in the human microbiome.

Last year, Vo Van Giau, PhD, of Gachon University, South Korea, and his colleagues published an extensive review of the data suggesting a similar link with Alzheimer’s disease.

Dr. Giau and his coauthors laid out a potential pathogenic pathway for this interaction.

“The microbiota is closely related to neurological dysfunction and plays a significant role in neuroinflammation through the secretion of proinflammatory cytokines. Changes in the homeostatic state of the microbiota lead to increased intestinal permeability, which may promote the translocation of bacteria and endotoxins across the epithelial barrier, inducing an immunological response associated with the production of proinflammatory cytokines. The activation of both enteric neurons and glial cells may result in various neurological disorders,” including Alzheimer’s, he wrote.

Dr. Missling said this paper, and smaller studies appearing at Alzheimer’s meetings, prompted the company to add the stool sampling as a follow-up measure.

“It’s something of great interest, we think, and deserves to be investigated.”

[email protected]

SOURCE: Missling C et al. AAIC 2019, Abstract 32260.

LOS ANGELES – A single look at the gut microbiome of patients with Alzheimer’s disease (AD) suggests an interaction between anti-inflammatory gut bacteria and long-term exposure to an investigational sigma 1 receptor agonist.

After up to 148 weeks treatment with Anavex 2-73, patients with stable or improved functional scores showed significantly higher levels of both Ruminococcaceae and Porphyromonadaceae, compared with patients who had declining function. Both bacterial families produce butyrate, an anti-inflammatory short-chain fatty acid.

Conversely, poor response was associated with a low level of Verrucomicrobia, a mucin-degrading phylum thought to be important in gut homeostasis. These bacteria live mainly in the intestinal mucosa – the physical interface between the microbiome and the rest of the body.

The data, presented at the Alzheimer’s Association International Conference, represent the first microbiome measurements reported in a clinical trial of an investigational Alzheimer’s therapy. Because they come from a single sample taken from a small group in an extension study, without a baseline comparator, it’s impossible to know what these associations mean. But the findings are enough to nudge Anavex Life Sciences into adding microbiome changes to its new study of Anavex 2-73, according to Christopher Missling, PhD, president and chief executive officer of the company.

The study, ramping up now, aims to recruit 450 patients with mild AD. They will be randomized to high-dose or mid-dose Anavex 2-73 for 48 weeks. The primary outcomes are measures of cognition and function. Stool sampling at baseline and at the end of the study will be included as well, Dr. Missling said in an interview.

Anavex 2-73 is a sigma-1 receptor agonist. A chaperone protein, sigma-1 is activated in response to acute and chronic cellular stressors, several which are important in neurodegeneration. The sigma-1 receptor is found on neurons and glia in many areas of the central nervous system. It modulates several processes implicated in neurodegenerative diseases, including glutamate and calcium activity, reaction to oxidative stress, and mitochondrial function. There is some evidence that sigma-1 receptor activation can induce neuronal regrowth and functional recovery after stroke. It also appears to play a role in helping cells clear misfolded proteins – a pathway that makes it an attractive drug target in Alzheimer’s disease, as well as other neurodegenerative diseases with aberrant proteins, such as Parkinson’s and Huntington’s diseases.

Anavex 2-73’s phase 2 development started with a 5-week crossover trial of 32 patients. This was followed by a 52-week, open-label extension trial of 10, 20, 30, and 50 mg/day orally, in which each patient was titrated to the maximum tolerated dose. The main endpoints were change on the Mini Mental State Exam and change on the Alzheimer’s Disease Cooperative Study-activities of daily living (ADCS-ADL) scale.

At 57 weeks, six patients had improved on the Mini Mental State Exam score: four with high plasma levels and two with low plasma levels, correlating to the dosage obtained. On the functional measure of activities of daily living, nine patients had improved, including five with high plasma levels, three with moderate levels, and one with a low level. One patient, with a moderate level, remained stable. The remaining 14 patients declined.

The company then enrolled 21 of the cohort in a 208-week extension trial, primarily because of patient request, Dr. Missling said. “They know they are doing better. Their families know they’re doing better. They did not want to give this up.”

Last fall, the company released 148-week functional and cognitive data confirming the initial findings: Patients with higher plasma levels (correlating with higher doses) declined about 2 points on the ADCS-ADL scale, compared with a mean decline of about 25 points among those with lower blood levels – an 88% difference in favor of treatment. Cognition scores showed a similar pattern, with the high-concentration group declining 64% less than the low-concentration group.

Sixteen patients consented to stool sampling. A sophisticated computer algorithm characterized the microbiome of each, measuring the relative abundance of phyla. Microbiome analysis wasn’t included as an endpoint in the original study design because, at that time, the idea of a connection between AD and the gut microbiome was barely on the research radar.

Things shifted dramatically in 2017, with a seminal paper finding that germ-free mice inoculated with stool from Parkinson’s patients developed Parkinson’s symptoms. This study was widely heralded as a breakthrough in the field – the first time any neurodegenerative disease had been conclusively linked to dysregulations in the human microbiome.

Last year, Vo Van Giau, PhD, of Gachon University, South Korea, and his colleagues published an extensive review of the data suggesting a similar link with Alzheimer’s disease.

Dr. Giau and his coauthors laid out a potential pathogenic pathway for this interaction.

“The microbiota is closely related to neurological dysfunction and plays a significant role in neuroinflammation through the secretion of proinflammatory cytokines. Changes in the homeostatic state of the microbiota lead to increased intestinal permeability, which may promote the translocation of bacteria and endotoxins across the epithelial barrier, inducing an immunological response associated with the production of proinflammatory cytokines. The activation of both enteric neurons and glial cells may result in various neurological disorders,” including Alzheimer’s, he wrote.

Dr. Missling said this paper, and smaller studies appearing at Alzheimer’s meetings, prompted the company to add the stool sampling as a follow-up measure.

“It’s something of great interest, we think, and deserves to be investigated.”

[email protected]

SOURCE: Missling C et al. AAIC 2019, Abstract 32260.

LOS ANGELES – A single look at the gut microbiome of patients with Alzheimer’s disease (AD) suggests an interaction between anti-inflammatory gut bacteria and long-term exposure to an investigational sigma 1 receptor agonist.

After up to 148 weeks treatment with Anavex 2-73, patients with stable or improved functional scores showed significantly higher levels of both Ruminococcaceae and Porphyromonadaceae, compared with patients who had declining function. Both bacterial families produce butyrate, an anti-inflammatory short-chain fatty acid.

Conversely, poor response was associated with a low level of Verrucomicrobia, a mucin-degrading phylum thought to be important in gut homeostasis. These bacteria live mainly in the intestinal mucosa – the physical interface between the microbiome and the rest of the body.

The data, presented at the Alzheimer’s Association International Conference, represent the first microbiome measurements reported in a clinical trial of an investigational Alzheimer’s therapy. Because they come from a single sample taken from a small group in an extension study, without a baseline comparator, it’s impossible to know what these associations mean. But the findings are enough to nudge Anavex Life Sciences into adding microbiome changes to its new study of Anavex 2-73, according to Christopher Missling, PhD, president and chief executive officer of the company.

The study, ramping up now, aims to recruit 450 patients with mild AD. They will be randomized to high-dose or mid-dose Anavex 2-73 for 48 weeks. The primary outcomes are measures of cognition and function. Stool sampling at baseline and at the end of the study will be included as well, Dr. Missling said in an interview.

Anavex 2-73 is a sigma-1 receptor agonist. A chaperone protein, sigma-1 is activated in response to acute and chronic cellular stressors, several which are important in neurodegeneration. The sigma-1 receptor is found on neurons and glia in many areas of the central nervous system. It modulates several processes implicated in neurodegenerative diseases, including glutamate and calcium activity, reaction to oxidative stress, and mitochondrial function. There is some evidence that sigma-1 receptor activation can induce neuronal regrowth and functional recovery after stroke. It also appears to play a role in helping cells clear misfolded proteins – a pathway that makes it an attractive drug target in Alzheimer’s disease, as well as other neurodegenerative diseases with aberrant proteins, such as Parkinson’s and Huntington’s diseases.

Anavex 2-73’s phase 2 development started with a 5-week crossover trial of 32 patients. This was followed by a 52-week, open-label extension trial of 10, 20, 30, and 50 mg/day orally, in which each patient was titrated to the maximum tolerated dose. The main endpoints were change on the Mini Mental State Exam and change on the Alzheimer’s Disease Cooperative Study-activities of daily living (ADCS-ADL) scale.

At 57 weeks, six patients had improved on the Mini Mental State Exam score: four with high plasma levels and two with low plasma levels, correlating to the dosage obtained. On the functional measure of activities of daily living, nine patients had improved, including five with high plasma levels, three with moderate levels, and one with a low level. One patient, with a moderate level, remained stable. The remaining 14 patients declined.

The company then enrolled 21 of the cohort in a 208-week extension trial, primarily because of patient request, Dr. Missling said. “They know they are doing better. Their families know they’re doing better. They did not want to give this up.”

Last fall, the company released 148-week functional and cognitive data confirming the initial findings: Patients with higher plasma levels (correlating with higher doses) declined about 2 points on the ADCS-ADL scale, compared with a mean decline of about 25 points among those with lower blood levels – an 88% difference in favor of treatment. Cognition scores showed a similar pattern, with the high-concentration group declining 64% less than the low-concentration group.

Sixteen patients consented to stool sampling. A sophisticated computer algorithm characterized the microbiome of each, measuring the relative abundance of phyla. Microbiome analysis wasn’t included as an endpoint in the original study design because, at that time, the idea of a connection between AD and the gut microbiome was barely on the research radar.

Things shifted dramatically in 2017, with a seminal paper finding that germ-free mice inoculated with stool from Parkinson’s patients developed Parkinson’s symptoms. This study was widely heralded as a breakthrough in the field – the first time any neurodegenerative disease had been conclusively linked to dysregulations in the human microbiome.

Last year, Vo Van Giau, PhD, of Gachon University, South Korea, and his colleagues published an extensive review of the data suggesting a similar link with Alzheimer’s disease.

Dr. Giau and his coauthors laid out a potential pathogenic pathway for this interaction.

“The microbiota is closely related to neurological dysfunction and plays a significant role in neuroinflammation through the secretion of proinflammatory cytokines. Changes in the homeostatic state of the microbiota lead to increased intestinal permeability, which may promote the translocation of bacteria and endotoxins across the epithelial barrier, inducing an immunological response associated with the production of proinflammatory cytokines. The activation of both enteric neurons and glial cells may result in various neurological disorders,” including Alzheimer’s, he wrote.

Dr. Missling said this paper, and smaller studies appearing at Alzheimer’s meetings, prompted the company to add the stool sampling as a follow-up measure.

“It’s something of great interest, we think, and deserves to be investigated.”

[email protected]

SOURCE: Missling C et al. AAIC 2019, Abstract 32260.

LOS ANGELES – While a large number of older adults take prescription and nonprescription medications to help them sleep, the effect of these medications on dementia risk is unclear, with most researchers advocating a cautious and conservative approach to prescribing.

Nastasic/iStockphoto.com

Research is increasingly revealing a bidirectional relationship between sleep and dementia. Poor sleep – especially from insomnia, sleep deprivation, or obstructive sleep apnea – is known to increase dementia risk. Dementias, meanwhile, are associated with serious circadian rhythm disturbances, leading to nighttime sleep loss and increasing the likelihood of institutionalization.

At the Alzheimer’s Association International Conference, researchers presented findings assessing the links between sleep medication use and dementia and also what agents or approaches might safely improve sleep in people with sleep disorders who are at risk for dementia or who have been diagnosed with dementia.
 

Sex- and race-based differences in risk

Yue Leng, PhD, of the University of California, San Francisco, reported a link between frequent sleep medication use and later dementia – but only in white adults. Dr. Leng presented findings from the National Institutes of Health–funded Health, Aging, and Body Composition Study, which recruited 3,068 subjects aged 70-79 and followed them for 15 years. At baseline, 2.7% of African Americans and 7.7% of whites in the study reported taking sleep medications “often” or “almost always.”

Dr. Leng and her colleagues found that white subjects who reported taking sleep aids five or more times a month at baseline had a nearly 80% higher risk of developing dementia during the course of the study (hazard ratio, 1.79; 95% confidence interval, 1.21-2.66), compared with people who reported never taking sleep aids or taking them less frequently.

The researchers saw no between-sex differences for this finding, and adjusted for a variety of genetic and lifestyle confounders. Importantly, no significant increase in dementia risk was seen for black subjects, who made up more than one-third of the cohort.

Dr. Leng told the conference that the researchers could not explain why black participants did not see similarly increased dementia risk. Also, she noted, researchers did not have information on the specific sleep medications people used: benzodiazepines, antihistamines, antidepressants, or other types of drugs. Nonetheless, she told the conference, the findings ratified the cautious approach many dementia experts are already stressing.


“Do we really need to prescribe so many sleep meds to older adults who are already at risk for cognitive impairment?” Dr. Leng said, adding: “I am a big advocate of behavioral sleep interventions.” People with clinical sleep problems “should be referred to sleep centers” for a fuller assessment before medication is prescribed, she said.

Findings from another cohort study, meanwhile, suggest that there could be sex-related differences in how sleep aids affect dementia risk. Investigators at Utah State University in Logan used data from some 3,656 older adults in the Cache County Study on Memory and Aging, an NIH-backed cohort study of white adults in Utah without dementia at baseline who were followed for 12 years.

The investigators, led by doctoral student Elizabeth Vernon, found that men reporting use of sleep medication saw more than threefold higher risk of developing Alzheimer’s disease than did men who did not use sleep aids (HR, 3.604; P = .0001).

Women who did not report having sleep disturbance but used sleep-inducing medications were at nearly fourfold greater risk for developing Alzheimer’s disease (HR, 3.916; P = .0001). Women who self-reported sleep disturbances at baseline, meanwhile, saw a reduction in Alzheimer’s risk of about one-third associated with the use of sleep medications.

Ms. Vernon told the conference that, despite the finding of risk reduction for this particular group of women, caution in prescribing sleep aids was warranted.

Common sleep drugs linked to cognitive aging

Chris Fox, MD, a researcher at the University of East Anglia in Norwich, England, and his colleagues demonstrated in 2018 that long-term exposure to anticholinergic drugs, a class that includes some antidepressants and antihistamines used to promote sleep, was associated with a higher risk of dementia, while use of benzodiazepines, a class of sedatives used commonly in older people as sleep aids, was not. (Whether benzodiazepine exposure relates to dementia remains controversial.)

At AAIC 2019, Dr. Fox presented findings from a study of 337 brains in a U.K. brain bank, of which 17% and 21% came from users of benzodiazepines and anticholinergic drugs, whose usage history was well documented. Dr. Fox and his colleagues found that, while neither anticholinergic nor benzodiazepine exposure was associated with brain pathology specific to that seen in Alzheimer’s disease, both classes of drugs were associated with “slight signals in neuronal loss” in one brain region, the nucleus basalis of Meynert. Dr. Fox described the drugs as causing “an increase in cognitive aging” which could bear on Alzheimer’s risk without being directly causative.
 

Newer sleep drugs may help Alzheimer’s patients

Scientists working for drug manufacturers presented findings on agents to counter the circadian rhythm disturbances seen in people with Alzheimer’s disease. Margaret Moline, PhD, of Eisai in Woodcliff Lake, N.J., showed some results from a phase 2, dose-ranging, placebo-controlled study of the experimental agent lemborexant in 62 subjects aged 60-90 with mild to moderate Alzheimer’s disease and sleep disturbances. (Lemborexant, an orexin receptor agonist that acts to regulate wakefulness, is being investigated in a broad range of sleep disorders.) Patients were randomized to one of four doses of lemborexant or placebo and wore a device for sleep monitoring. Nighttime activity indicating arousal was significantly lower for people in two dosage arms, 5 mg and 10 mg, compared with placebo, and treatment groups saw trends toward less sleep fragmentation and higher total sleep time, Dr. Moline told the conference.

Suvorexant (Belsomra), the only orexin receptor antagonist currently licensed as a sleep aid, is also being tested in people with Alzheimer’s disease. At AAIC 2019, Joseph Herring, MD, PhD, of Merck in Kenilworth, N.J., presented results from a placebo-controlled trial of 277 patients with Alzheimer’s disease and insomnia, and reported that treatment with 10 or 20 mg of suvorexant over 4 weeks was associated with about an extra half hour of total nightly sleep, with a 73-minute mean increase from baseline, compared with 45 minutes for patients receiving placebo (95% CI, 11-45; P less than .005).
 

Trazodone linked to slower cognitive decline

An inexpensive antidepressant used in low doses as a sleep aid, including in people with Alzheimer’s disease, was associated with a delay in cognitive decline in older adults, according to results from a retrospective study. Elissaios Karageorgiou, MD, PhD, of the University of California, San Francisco, and the Neurological Institute of Athens presented results derived from two cohorts: patients enrolled at the UCSF Memory and Aging Center and women enrolled in the Study for Osteoporotic Fractures (SOF) in Women. The investigators were able to identify trazodone users in the studies (with two or more contiguous study visits reporting trazodone use) and match them with control patients from the same cohorts who did not use trazodone.

Trazodone was studied because previous research suggests it increases total sleep time in patients with Alzheimer’s disease without affecting next-day cognitive performance.

Trazodone-using patients in the UCSF cohort (n = 25) saw significantly less decline in Mini Mental State Exam (MMSE) scores over 4 years, compared with nonusers (0.27 vs. 0.70 points per year; P = .023), an effect that remained statistically significant even after adjusting for sedative and stimulant use and the expected progression of Alzheimer’s disease pathology. Importantly, the slower decline was seen only among subjects with sleep complaints at baseline and especially those whose sleep improved over time, suggesting that the cognitive benefit was mediated by improved sleep.

In the SOF cohort of 46 trazodone users matched with 148 nonusers, no significant protective or negative effect related to long-term trazodone use was found using the MMSE or the Trails Making Test. In this analysis, however, baseline and longitudinal sleep quality was not captured in the group-matching process, neither was the use of other medications. The patient group was slightly older, and all patients were women.

Dr. Karageorgiou said in an interview that the link between improved sleep, trazodone, and cognition needs to be validated in prospective intervention studies. Trazodone, he said, appears to work best in people with a specific type of insomnia characterized by cortical and behavioral hyperarousal, and its cognitive effect appears limited to people whose sleep improves with treatment. “You’re not going to see long-term cognitive benefits if it’s not improving your sleep,” Dr. Karageorgiou said. “So, whether trazodone improves sleep or not in a patient after a few months can be an early indicator for the clinician to continue using it or suspend it, because it is unlikely to help their cognition otherwise.”

He stressed that physicians need to be broadly focused on improving sleep to help patients with, or at risk for, dementia by consolidating their sleep rhythms.

“Trazodone is not the magic bullet, and I don’t think we will ever have a magic bullet,” Dr. Karageorgiou said. “Because when our brain degenerates, it’s not just one chemical, or one system, it’s many. And our body changes as well. The important thing is to help the patient consolidate their rhythms, whether through light therapy, daily exercise, cognitive behavioral therapy for insomnia, or other evidence-based interventions and their combination. The same applies for a person with dementia as for the rest of us.”

None of the investigators outside of the industry-sponsored studies had relevant disclosures.

LOS ANGELES – While a large number of older adults take prescription and nonprescription medications to help them sleep, the effect of these medications on dementia risk is unclear, with most researchers advocating a cautious and conservative approach to prescribing.

Nastasic/iStockphoto.com

Research is increasingly revealing a bidirectional relationship between sleep and dementia. Poor sleep – especially from insomnia, sleep deprivation, or obstructive sleep apnea – is known to increase dementia risk. Dementias, meanwhile, are associated with serious circadian rhythm disturbances, leading to nighttime sleep loss and increasing the likelihood of institutionalization.

At the Alzheimer’s Association International Conference, researchers presented findings assessing the links between sleep medication use and dementia and also what agents or approaches might safely improve sleep in people with sleep disorders who are at risk for dementia or who have been diagnosed with dementia.
 

Sex- and race-based differences in risk

Yue Leng, PhD, of the University of California, San Francisco, reported a link between frequent sleep medication use and later dementia – but only in white adults. Dr. Leng presented findings from the National Institutes of Health–funded Health, Aging, and Body Composition Study, which recruited 3,068 subjects aged 70-79 and followed them for 15 years. At baseline, 2.7% of African Americans and 7.7% of whites in the study reported taking sleep medications “often” or “almost always.”

Dr. Leng and her colleagues found that white subjects who reported taking sleep aids five or more times a month at baseline had a nearly 80% higher risk of developing dementia during the course of the study (hazard ratio, 1.79; 95% confidence interval, 1.21-2.66), compared with people who reported never taking sleep aids or taking them less frequently.

The researchers saw no between-sex differences for this finding, and adjusted for a variety of genetic and lifestyle confounders. Importantly, no significant increase in dementia risk was seen for black subjects, who made up more than one-third of the cohort.

Dr. Leng told the conference that the researchers could not explain why black participants did not see similarly increased dementia risk. Also, she noted, researchers did not have information on the specific sleep medications people used: benzodiazepines, antihistamines, antidepressants, or other types of drugs. Nonetheless, she told the conference, the findings ratified the cautious approach many dementia experts are already stressing.


“Do we really need to prescribe so many sleep meds to older adults who are already at risk for cognitive impairment?” Dr. Leng said, adding: “I am a big advocate of behavioral sleep interventions.” People with clinical sleep problems “should be referred to sleep centers” for a fuller assessment before medication is prescribed, she said.

Findings from another cohort study, meanwhile, suggest that there could be sex-related differences in how sleep aids affect dementia risk. Investigators at Utah State University in Logan used data from some 3,656 older adults in the Cache County Study on Memory and Aging, an NIH-backed cohort study of white adults in Utah without dementia at baseline who were followed for 12 years.

The investigators, led by doctoral student Elizabeth Vernon, found that men reporting use of sleep medication saw more than threefold higher risk of developing Alzheimer’s disease than did men who did not use sleep aids (HR, 3.604; P = .0001).

Women who did not report having sleep disturbance but used sleep-inducing medications were at nearly fourfold greater risk for developing Alzheimer’s disease (HR, 3.916; P = .0001). Women who self-reported sleep disturbances at baseline, meanwhile, saw a reduction in Alzheimer’s risk of about one-third associated with the use of sleep medications.

Ms. Vernon told the conference that, despite the finding of risk reduction for this particular group of women, caution in prescribing sleep aids was warranted.

Common sleep drugs linked to cognitive aging

Chris Fox, MD, a researcher at the University of East Anglia in Norwich, England, and his colleagues demonstrated in 2018 that long-term exposure to anticholinergic drugs, a class that includes some antidepressants and antihistamines used to promote sleep, was associated with a higher risk of dementia, while use of benzodiazepines, a class of sedatives used commonly in older people as sleep aids, was not. (Whether benzodiazepine exposure relates to dementia remains controversial.)

At AAIC 2019, Dr. Fox presented findings from a study of 337 brains in a U.K. brain bank, of which 17% and 21% came from users of benzodiazepines and anticholinergic drugs, whose usage history was well documented. Dr. Fox and his colleagues found that, while neither anticholinergic nor benzodiazepine exposure was associated with brain pathology specific to that seen in Alzheimer’s disease, both classes of drugs were associated with “slight signals in neuronal loss” in one brain region, the nucleus basalis of Meynert. Dr. Fox described the drugs as causing “an increase in cognitive aging” which could bear on Alzheimer’s risk without being directly causative.
 

Newer sleep drugs may help Alzheimer’s patients

Scientists working for drug manufacturers presented findings on agents to counter the circadian rhythm disturbances seen in people with Alzheimer’s disease. Margaret Moline, PhD, of Eisai in Woodcliff Lake, N.J., showed some results from a phase 2, dose-ranging, placebo-controlled study of the experimental agent lemborexant in 62 subjects aged 60-90 with mild to moderate Alzheimer’s disease and sleep disturbances. (Lemborexant, an orexin receptor agonist that acts to regulate wakefulness, is being investigated in a broad range of sleep disorders.) Patients were randomized to one of four doses of lemborexant or placebo and wore a device for sleep monitoring. Nighttime activity indicating arousal was significantly lower for people in two dosage arms, 5 mg and 10 mg, compared with placebo, and treatment groups saw trends toward less sleep fragmentation and higher total sleep time, Dr. Moline told the conference.

Suvorexant (Belsomra), the only orexin receptor antagonist currently licensed as a sleep aid, is also being tested in people with Alzheimer’s disease. At AAIC 2019, Joseph Herring, MD, PhD, of Merck in Kenilworth, N.J., presented results from a placebo-controlled trial of 277 patients with Alzheimer’s disease and insomnia, and reported that treatment with 10 or 20 mg of suvorexant over 4 weeks was associated with about an extra half hour of total nightly sleep, with a 73-minute mean increase from baseline, compared with 45 minutes for patients receiving placebo (95% CI, 11-45; P less than .005).
 

Trazodone linked to slower cognitive decline

An inexpensive antidepressant used in low doses as a sleep aid, including in people with Alzheimer’s disease, was associated with a delay in cognitive decline in older adults, according to results from a retrospective study. Elissaios Karageorgiou, MD, PhD, of the University of California, San Francisco, and the Neurological Institute of Athens presented results derived from two cohorts: patients enrolled at the UCSF Memory and Aging Center and women enrolled in the Study for Osteoporotic Fractures (SOF) in Women. The investigators were able to identify trazodone users in the studies (with two or more contiguous study visits reporting trazodone use) and match them with control patients from the same cohorts who did not use trazodone.

Trazodone was studied because previous research suggests it increases total sleep time in patients with Alzheimer’s disease without affecting next-day cognitive performance.

Trazodone-using patients in the UCSF cohort (n = 25) saw significantly less decline in Mini Mental State Exam (MMSE) scores over 4 years, compared with nonusers (0.27 vs. 0.70 points per year; P = .023), an effect that remained statistically significant even after adjusting for sedative and stimulant use and the expected progression of Alzheimer’s disease pathology. Importantly, the slower decline was seen only among subjects with sleep complaints at baseline and especially those whose sleep improved over time, suggesting that the cognitive benefit was mediated by improved sleep.

In the SOF cohort of 46 trazodone users matched with 148 nonusers, no significant protective or negative effect related to long-term trazodone use was found using the MMSE or the Trails Making Test. In this analysis, however, baseline and longitudinal sleep quality was not captured in the group-matching process, neither was the use of other medications. The patient group was slightly older, and all patients were women.

Dr. Karageorgiou said in an interview that the link between improved sleep, trazodone, and cognition needs to be validated in prospective intervention studies. Trazodone, he said, appears to work best in people with a specific type of insomnia characterized by cortical and behavioral hyperarousal, and its cognitive effect appears limited to people whose sleep improves with treatment. “You’re not going to see long-term cognitive benefits if it’s not improving your sleep,” Dr. Karageorgiou said. “So, whether trazodone improves sleep or not in a patient after a few months can be an early indicator for the clinician to continue using it or suspend it, because it is unlikely to help their cognition otherwise.”

He stressed that physicians need to be broadly focused on improving sleep to help patients with, or at risk for, dementia by consolidating their sleep rhythms.

“Trazodone is not the magic bullet, and I don’t think we will ever have a magic bullet,” Dr. Karageorgiou said. “Because when our brain degenerates, it’s not just one chemical, or one system, it’s many. And our body changes as well. The important thing is to help the patient consolidate their rhythms, whether through light therapy, daily exercise, cognitive behavioral therapy for insomnia, or other evidence-based interventions and their combination. The same applies for a person with dementia as for the rest of us.”

None of the investigators outside of the industry-sponsored studies had relevant disclosures.

LOS ANGELES – While a large number of older adults take prescription and nonprescription medications to help them sleep, the effect of these medications on dementia risk is unclear, with most researchers advocating a cautious and conservative approach to prescribing.

Nastasic/iStockphoto.com

Research is increasingly revealing a bidirectional relationship between sleep and dementia. Poor sleep – especially from insomnia, sleep deprivation, or obstructive sleep apnea – is known to increase dementia risk. Dementias, meanwhile, are associated with serious circadian rhythm disturbances, leading to nighttime sleep loss and increasing the likelihood of institutionalization.

At the Alzheimer’s Association International Conference, researchers presented findings assessing the links between sleep medication use and dementia and also what agents or approaches might safely improve sleep in people with sleep disorders who are at risk for dementia or who have been diagnosed with dementia.
 

Sex- and race-based differences in risk

Yue Leng, PhD, of the University of California, San Francisco, reported a link between frequent sleep medication use and later dementia – but only in white adults. Dr. Leng presented findings from the National Institutes of Health–funded Health, Aging, and Body Composition Study, which recruited 3,068 subjects aged 70-79 and followed them for 15 years. At baseline, 2.7% of African Americans and 7.7% of whites in the study reported taking sleep medications “often” or “almost always.”

Dr. Leng and her colleagues found that white subjects who reported taking sleep aids five or more times a month at baseline had a nearly 80% higher risk of developing dementia during the course of the study (hazard ratio, 1.79; 95% confidence interval, 1.21-2.66), compared with people who reported never taking sleep aids or taking them less frequently.

The researchers saw no between-sex differences for this finding, and adjusted for a variety of genetic and lifestyle confounders. Importantly, no significant increase in dementia risk was seen for black subjects, who made up more than one-third of the cohort.

Dr. Leng told the conference that the researchers could not explain why black participants did not see similarly increased dementia risk. Also, she noted, researchers did not have information on the specific sleep medications people used: benzodiazepines, antihistamines, antidepressants, or other types of drugs. Nonetheless, she told the conference, the findings ratified the cautious approach many dementia experts are already stressing.


“Do we really need to prescribe so many sleep meds to older adults who are already at risk for cognitive impairment?” Dr. Leng said, adding: “I am a big advocate of behavioral sleep interventions.” People with clinical sleep problems “should be referred to sleep centers” for a fuller assessment before medication is prescribed, she said.

Findings from another cohort study, meanwhile, suggest that there could be sex-related differences in how sleep aids affect dementia risk. Investigators at Utah State University in Logan used data from some 3,656 older adults in the Cache County Study on Memory and Aging, an NIH-backed cohort study of white adults in Utah without dementia at baseline who were followed for 12 years.

The investigators, led by doctoral student Elizabeth Vernon, found that men reporting use of sleep medication saw more than threefold higher risk of developing Alzheimer’s disease than did men who did not use sleep aids (HR, 3.604; P = .0001).

Women who did not report having sleep disturbance but used sleep-inducing medications were at nearly fourfold greater risk for developing Alzheimer’s disease (HR, 3.916; P = .0001). Women who self-reported sleep disturbances at baseline, meanwhile, saw a reduction in Alzheimer’s risk of about one-third associated with the use of sleep medications.

Ms. Vernon told the conference that, despite the finding of risk reduction for this particular group of women, caution in prescribing sleep aids was warranted.

Common sleep drugs linked to cognitive aging

Chris Fox, MD, a researcher at the University of East Anglia in Norwich, England, and his colleagues demonstrated in 2018 that long-term exposure to anticholinergic drugs, a class that includes some antidepressants and antihistamines used to promote sleep, was associated with a higher risk of dementia, while use of benzodiazepines, a class of sedatives used commonly in older people as sleep aids, was not. (Whether benzodiazepine exposure relates to dementia remains controversial.)

At AAIC 2019, Dr. Fox presented findings from a study of 337 brains in a U.K. brain bank, of which 17% and 21% came from users of benzodiazepines and anticholinergic drugs, whose usage history was well documented. Dr. Fox and his colleagues found that, while neither anticholinergic nor benzodiazepine exposure was associated with brain pathology specific to that seen in Alzheimer’s disease, both classes of drugs were associated with “slight signals in neuronal loss” in one brain region, the nucleus basalis of Meynert. Dr. Fox described the drugs as causing “an increase in cognitive aging” which could bear on Alzheimer’s risk without being directly causative.
 

Newer sleep drugs may help Alzheimer’s patients

Scientists working for drug manufacturers presented findings on agents to counter the circadian rhythm disturbances seen in people with Alzheimer’s disease. Margaret Moline, PhD, of Eisai in Woodcliff Lake, N.J., showed some results from a phase 2, dose-ranging, placebo-controlled study of the experimental agent lemborexant in 62 subjects aged 60-90 with mild to moderate Alzheimer’s disease and sleep disturbances. (Lemborexant, an orexin receptor agonist that acts to regulate wakefulness, is being investigated in a broad range of sleep disorders.) Patients were randomized to one of four doses of lemborexant or placebo and wore a device for sleep monitoring. Nighttime activity indicating arousal was significantly lower for people in two dosage arms, 5 mg and 10 mg, compared with placebo, and treatment groups saw trends toward less sleep fragmentation and higher total sleep time, Dr. Moline told the conference.

Suvorexant (Belsomra), the only orexin receptor antagonist currently licensed as a sleep aid, is also being tested in people with Alzheimer’s disease. At AAIC 2019, Joseph Herring, MD, PhD, of Merck in Kenilworth, N.J., presented results from a placebo-controlled trial of 277 patients with Alzheimer’s disease and insomnia, and reported that treatment with 10 or 20 mg of suvorexant over 4 weeks was associated with about an extra half hour of total nightly sleep, with a 73-minute mean increase from baseline, compared with 45 minutes for patients receiving placebo (95% CI, 11-45; P less than .005).
 

Trazodone linked to slower cognitive decline

An inexpensive antidepressant used in low doses as a sleep aid, including in people with Alzheimer’s disease, was associated with a delay in cognitive decline in older adults, according to results from a retrospective study. Elissaios Karageorgiou, MD, PhD, of the University of California, San Francisco, and the Neurological Institute of Athens presented results derived from two cohorts: patients enrolled at the UCSF Memory and Aging Center and women enrolled in the Study for Osteoporotic Fractures (SOF) in Women. The investigators were able to identify trazodone users in the studies (with two or more contiguous study visits reporting trazodone use) and match them with control patients from the same cohorts who did not use trazodone.

Trazodone was studied because previous research suggests it increases total sleep time in patients with Alzheimer’s disease without affecting next-day cognitive performance.

Trazodone-using patients in the UCSF cohort (n = 25) saw significantly less decline in Mini Mental State Exam (MMSE) scores over 4 years, compared with nonusers (0.27 vs. 0.70 points per year; P = .023), an effect that remained statistically significant even after adjusting for sedative and stimulant use and the expected progression of Alzheimer’s disease pathology. Importantly, the slower decline was seen only among subjects with sleep complaints at baseline and especially those whose sleep improved over time, suggesting that the cognitive benefit was mediated by improved sleep.

In the SOF cohort of 46 trazodone users matched with 148 nonusers, no significant protective or negative effect related to long-term trazodone use was found using the MMSE or the Trails Making Test. In this analysis, however, baseline and longitudinal sleep quality was not captured in the group-matching process, neither was the use of other medications. The patient group was slightly older, and all patients were women.

Dr. Karageorgiou said in an interview that the link between improved sleep, trazodone, and cognition needs to be validated in prospective intervention studies. Trazodone, he said, appears to work best in people with a specific type of insomnia characterized by cortical and behavioral hyperarousal, and its cognitive effect appears limited to people whose sleep improves with treatment. “You’re not going to see long-term cognitive benefits if it’s not improving your sleep,” Dr. Karageorgiou said. “So, whether trazodone improves sleep or not in a patient after a few months can be an early indicator for the clinician to continue using it or suspend it, because it is unlikely to help their cognition otherwise.”

He stressed that physicians need to be broadly focused on improving sleep to help patients with, or at risk for, dementia by consolidating their sleep rhythms.

“Trazodone is not the magic bullet, and I don’t think we will ever have a magic bullet,” Dr. Karageorgiou said. “Because when our brain degenerates, it’s not just one chemical, or one system, it’s many. And our body changes as well. The important thing is to help the patient consolidate their rhythms, whether through light therapy, daily exercise, cognitive behavioral therapy for insomnia, or other evidence-based interventions and their combination. The same applies for a person with dementia as for the rest of us.”

None of the investigators outside of the industry-sponsored studies had relevant disclosures.

LOS ANGELES – Benzodiazepines and hypnotics, including the so-called “Z drugs,” don’t significantly increase the pathological features typical of Alzheimer’s disease but long-term users may experience some neuronal loss in the nucleus basalis, Chris Fox, MD, reported at the Alzheimer’s Association International Conference.

Michele G. Sullivan/MDedge News

The nucleus basalis is rich in cholinergic neurons and associated with arousing stimuli, including positive and aversive appetite, sustained attention, and the interplay of reality and visual perception.

“Neuronal loss in the nucleus basalis offers mechanisms for the impact of benzodiazepine and anticholinergic drug use on the aging brain and highlights important areas for future research,” said Dr. Fox, professor of clinical psychiatry at the University of East Anglia, Norwich, England.

“The risk [for taking a Z drug] in the United Kingdom is high, with about 7.5 million older adults using potentially inappropriately prescribed anticholinergic and/or Z-drug medications. Despite well-documented cognitive impairment associated with these medicines, hypnotics are still used for long durations and exceed the recommended limits,” Dr. Fox said. “There’s no association with better cognition, quality of life, or improved behavior when they are given to people with dementia. In fact, we’ve seen a 60% increased risk of hip fractures – an increase from a 3% to a 15% yearly risk.”

Dr. Fox and colleagues studied the brains of 337 subjects who were included in the U.K. Medical Research Council’s Cognitive Function and Ageing Studies (CFAS). The study was intended to explore the incidence of dementia in the United Kingdom, examine incidence variation among regions, and explore factors increasing dementia risk and rate of progression.

The first study, which began in 1989 and lasted until 2015, followed subjects older than 65 years for up to 12 years. Each subject was regularly interviewed and underwent cognitive testing about every 1.5 years. Benzodiazepine use was considered an especially important aspect, because the medications are frequently used in the elderly and seem linked to injuries and cognitive status at last follow-up.

In CFAS, 21% of subjects reported at least one incidence of anticholinergic use, and 12% reported recurrent use. Another 17% reported any hypnotic use, and 11% reported recurrent use. The main indications were as an antidepressant (13%), for urological issues (4%), as antiparkinsonism drugs (1%), as antipsychotics (3%), and as antihistamines (3%). Overall, 18% reported concurrent use of benzodiazepines and hypnotics. At time of death, 46% had a diagnosis of dementia.

“Those reporting benzodiazepine use were more likely to be women and to have depression or sleep problems,” Dr. Fox noted, although he didn’t give specific hazard ratios. After adjustment for numerous factors, including age, sex, stroke, hypertension, depression, anxiety, asthma, Parkinson’s disease, duration of sleep problems, education, and smoking, he found no statistically increased risk of amyloid brain plaques or tau tangles, the pathologic hallmarks of Alzheimer’s disease.

Anticholinergic use was associated with a significant 60% reduction in cortical atrophy (odds ratio, 0.40) and recurrent use with a 61% reduction in amyloid angiopathy (OR, 0.39).

However, both medication classes were associated with greater neuronal loss in the nucleus basalis. Recurrent use of anticholinergic drugs increased neuronal loss by 300% (OR, 4.12), while any use nearly tripled it (OR, 2.87). Recurrent use of benzodiazepines was associated with increased neuronal loss in the region (OR, 3.76) as well. However, these associations did not reach statistical significance. But there was a statistically significant association with any use of benzodiazepines and neuronal loss in the nucleus basalis (OR, 6.84).

“We did find greater neuronal loss in the nucleus basalis associated with benzodiazepine and anticholinergic drugs use,” Dr. Fox said. “The nucleus basalis is rich in neurons that stimulate the cholinergic system of the neocortex. Neuronal loss in this region is thought to occur in the early stages of Alzheimer’s. Other studies have suggested that volume loss in the basal forebrain cholinergic site leads to widespread cortical atrophy in patients with mild cognitive impairment. We did not observe the widespread cortical atrophy, however.

“Given that the strongest associations were observed for benzodiazepines and neuronal loss in the nucleus basalis, it may be that the drugs were prescribed to treat the symptoms of ‘cholinergic deficiency syndrome,’ Our findings suggest that the symptoms of dementia lead to an increase of benzodiazepines as opposed to the medications actually causing Alzheimer’s disease,” he said.

Dr. Fox reported no financial disclosures.

[email protected]

SOURCE: Fox C et al. AAIC 2019, Abstract 34017.

LOS ANGELES – Benzodiazepines and hypnotics, including the so-called “Z drugs,” don’t significantly increase the pathological features typical of Alzheimer’s disease but long-term users may experience some neuronal loss in the nucleus basalis, Chris Fox, MD, reported at the Alzheimer’s Association International Conference.

Michele G. Sullivan/MDedge News

The nucleus basalis is rich in cholinergic neurons and associated with arousing stimuli, including positive and aversive appetite, sustained attention, and the interplay of reality and visual perception.

“Neuronal loss in the nucleus basalis offers mechanisms for the impact of benzodiazepine and anticholinergic drug use on the aging brain and highlights important areas for future research,” said Dr. Fox, professor of clinical psychiatry at the University of East Anglia, Norwich, England.

“The risk [for taking a Z drug] in the United Kingdom is high, with about 7.5 million older adults using potentially inappropriately prescribed anticholinergic and/or Z-drug medications. Despite well-documented cognitive impairment associated with these medicines, hypnotics are still used for long durations and exceed the recommended limits,” Dr. Fox said. “There’s no association with better cognition, quality of life, or improved behavior when they are given to people with dementia. In fact, we’ve seen a 60% increased risk of hip fractures – an increase from a 3% to a 15% yearly risk.”

Dr. Fox and colleagues studied the brains of 337 subjects who were included in the U.K. Medical Research Council’s Cognitive Function and Ageing Studies (CFAS). The study was intended to explore the incidence of dementia in the United Kingdom, examine incidence variation among regions, and explore factors increasing dementia risk and rate of progression.

The first study, which began in 1989 and lasted until 2015, followed subjects older than 65 years for up to 12 years. Each subject was regularly interviewed and underwent cognitive testing about every 1.5 years. Benzodiazepine use was considered an especially important aspect, because the medications are frequently used in the elderly and seem linked to injuries and cognitive status at last follow-up.

In CFAS, 21% of subjects reported at least one incidence of anticholinergic use, and 12% reported recurrent use. Another 17% reported any hypnotic use, and 11% reported recurrent use. The main indications were as an antidepressant (13%), for urological issues (4%), as antiparkinsonism drugs (1%), as antipsychotics (3%), and as antihistamines (3%). Overall, 18% reported concurrent use of benzodiazepines and hypnotics. At time of death, 46% had a diagnosis of dementia.

“Those reporting benzodiazepine use were more likely to be women and to have depression or sleep problems,” Dr. Fox noted, although he didn’t give specific hazard ratios. After adjustment for numerous factors, including age, sex, stroke, hypertension, depression, anxiety, asthma, Parkinson’s disease, duration of sleep problems, education, and smoking, he found no statistically increased risk of amyloid brain plaques or tau tangles, the pathologic hallmarks of Alzheimer’s disease.

Anticholinergic use was associated with a significant 60% reduction in cortical atrophy (odds ratio, 0.40) and recurrent use with a 61% reduction in amyloid angiopathy (OR, 0.39).

However, both medication classes were associated with greater neuronal loss in the nucleus basalis. Recurrent use of anticholinergic drugs increased neuronal loss by 300% (OR, 4.12), while any use nearly tripled it (OR, 2.87). Recurrent use of benzodiazepines was associated with increased neuronal loss in the region (OR, 3.76) as well. However, these associations did not reach statistical significance. But there was a statistically significant association with any use of benzodiazepines and neuronal loss in the nucleus basalis (OR, 6.84).

“We did find greater neuronal loss in the nucleus basalis associated with benzodiazepine and anticholinergic drugs use,” Dr. Fox said. “The nucleus basalis is rich in neurons that stimulate the cholinergic system of the neocortex. Neuronal loss in this region is thought to occur in the early stages of Alzheimer’s. Other studies have suggested that volume loss in the basal forebrain cholinergic site leads to widespread cortical atrophy in patients with mild cognitive impairment. We did not observe the widespread cortical atrophy, however.

“Given that the strongest associations were observed for benzodiazepines and neuronal loss in the nucleus basalis, it may be that the drugs were prescribed to treat the symptoms of ‘cholinergic deficiency syndrome,’ Our findings suggest that the symptoms of dementia lead to an increase of benzodiazepines as opposed to the medications actually causing Alzheimer’s disease,” he said.

Dr. Fox reported no financial disclosures.

[email protected]

SOURCE: Fox C et al. AAIC 2019, Abstract 34017.

LOS ANGELES – Benzodiazepines and hypnotics, including the so-called “Z drugs,” don’t significantly increase the pathological features typical of Alzheimer’s disease but long-term users may experience some neuronal loss in the nucleus basalis, Chris Fox, MD, reported at the Alzheimer’s Association International Conference.

Michele G. Sullivan/MDedge News

The nucleus basalis is rich in cholinergic neurons and associated with arousing stimuli, including positive and aversive appetite, sustained attention, and the interplay of reality and visual perception.

“Neuronal loss in the nucleus basalis offers mechanisms for the impact of benzodiazepine and anticholinergic drug use on the aging brain and highlights important areas for future research,” said Dr. Fox, professor of clinical psychiatry at the University of East Anglia, Norwich, England.

“The risk [for taking a Z drug] in the United Kingdom is high, with about 7.5 million older adults using potentially inappropriately prescribed anticholinergic and/or Z-drug medications. Despite well-documented cognitive impairment associated with these medicines, hypnotics are still used for long durations and exceed the recommended limits,” Dr. Fox said. “There’s no association with better cognition, quality of life, or improved behavior when they are given to people with dementia. In fact, we’ve seen a 60% increased risk of hip fractures – an increase from a 3% to a 15% yearly risk.”

Dr. Fox and colleagues studied the brains of 337 subjects who were included in the U.K. Medical Research Council’s Cognitive Function and Ageing Studies (CFAS). The study was intended to explore the incidence of dementia in the United Kingdom, examine incidence variation among regions, and explore factors increasing dementia risk and rate of progression.

The first study, which began in 1989 and lasted until 2015, followed subjects older than 65 years for up to 12 years. Each subject was regularly interviewed and underwent cognitive testing about every 1.5 years. Benzodiazepine use was considered an especially important aspect, because the medications are frequently used in the elderly and seem linked to injuries and cognitive status at last follow-up.

In CFAS, 21% of subjects reported at least one incidence of anticholinergic use, and 12% reported recurrent use. Another 17% reported any hypnotic use, and 11% reported recurrent use. The main indications were as an antidepressant (13%), for urological issues (4%), as antiparkinsonism drugs (1%), as antipsychotics (3%), and as antihistamines (3%). Overall, 18% reported concurrent use of benzodiazepines and hypnotics. At time of death, 46% had a diagnosis of dementia.

“Those reporting benzodiazepine use were more likely to be women and to have depression or sleep problems,” Dr. Fox noted, although he didn’t give specific hazard ratios. After adjustment for numerous factors, including age, sex, stroke, hypertension, depression, anxiety, asthma, Parkinson’s disease, duration of sleep problems, education, and smoking, he found no statistically increased risk of amyloid brain plaques or tau tangles, the pathologic hallmarks of Alzheimer’s disease.

Anticholinergic use was associated with a significant 60% reduction in cortical atrophy (odds ratio, 0.40) and recurrent use with a 61% reduction in amyloid angiopathy (OR, 0.39).

However, both medication classes were associated with greater neuronal loss in the nucleus basalis. Recurrent use of anticholinergic drugs increased neuronal loss by 300% (OR, 4.12), while any use nearly tripled it (OR, 2.87). Recurrent use of benzodiazepines was associated with increased neuronal loss in the region (OR, 3.76) as well. However, these associations did not reach statistical significance. But there was a statistically significant association with any use of benzodiazepines and neuronal loss in the nucleus basalis (OR, 6.84).

“We did find greater neuronal loss in the nucleus basalis associated with benzodiazepine and anticholinergic drugs use,” Dr. Fox said. “The nucleus basalis is rich in neurons that stimulate the cholinergic system of the neocortex. Neuronal loss in this region is thought to occur in the early stages of Alzheimer’s. Other studies have suggested that volume loss in the basal forebrain cholinergic site leads to widespread cortical atrophy in patients with mild cognitive impairment. We did not observe the widespread cortical atrophy, however.

“Given that the strongest associations were observed for benzodiazepines and neuronal loss in the nucleus basalis, it may be that the drugs were prescribed to treat the symptoms of ‘cholinergic deficiency syndrome,’ Our findings suggest that the symptoms of dementia lead to an increase of benzodiazepines as opposed to the medications actually causing Alzheimer’s disease,” he said.

Dr. Fox reported no financial disclosures.

[email protected]

SOURCE: Fox C et al. AAIC 2019, Abstract 34017.

LOS ANGELES – Neurofilament light, or NfL, is an increasingly studied biomarker of axonal damage across a range of neurodegenerative diseases, including Alzheimer’s disease. And because it is a biomarker that can be measured in blood, it is a less invasive measure of disease progression in Alzheimer’s than cerebrospinal fluid markers.

At the Alzheimer’s Association International Conference, scientists studying the world’s largest cohort of early-onset Alzheimer’s families presented results from a cross-sectional and longitudinal study of more than 2,000 carriers and noncarriers of a single Alzheimer’s-causing mutation (Presenilin 1 E280A) that occurs in an extended Colombian family.

While previous studies have also looked at NfL in cohorts of autosomal dominant mutation carriers, this study strengthens evidence for NfL as an Alzheimer’s biomarker in the largest single-mutation cohort to date.

Yakeel Quiroz, PhD, of Harvard University in Boston and colleagues reported that, in a cross-sectional study of 1,070 mutation carriers and 1,074 noncarriers aged 8-75 years (mean age, 29-30 years; 46% male), mean plasma NfL levels were elevated in cognitively unimpaired carriers (18.08 pg/mL), compared with noncarriers (9.09 pg/mL; P less than .0001).

Longitudinal data from 504 of those carriers and noncarriers showed that NfL levels begin to diverge significantly between the groups at age 22, more than 2 decades before the mean onset of mild cognitive impairment for this cohort (44 years). The between-group differences in NfL continued to widen with advancing age.

“At approximately age 22, the axons, the neurons are already changing, and this measure serves as an early sign of degeneration,” Dr. Quiroz said in an interview. “This is really telling us about neurodegeneration related to Alzheimer’s disease because these are people destined to develop Alzheimer’s dementia later in life and have no age-related comorbidities that could cause elevation in NfL.”

A study published early this year in a different cohort of about 400 autosomal dominant Alzheimer’s disease mutation carriers and noncarriers found that the longitudinal rate of change of serum NfL could discriminate carriers from noncarriers almost a decade earlier than cross-sectional absolute NfL levels – at 16 years and 7 years, respectively, before expected onset of symptoms (Nat Med. 2019 Feb;25[2]:277–83).

In Dr. Quiroz and colleagues’ study, both cross-sectional and longitudinal findings showed carriers to significantly differ from noncarriers by age 22 years. “We’re seeing differences between groups that reach statistical significance earlier” – decades, in this case, before onset of symptoms, Dr. Quiroz said. The current study is distinguished by its exceptional power, she said: “No one has done this with such a large number of carriers with a single genetic mutation.”

Eric Reiman, MD, of Banner Alzheimer’s Institute in Phoenix, the coauthor on the study who presented the findings to the conference, commented in an interview that they illustrate “the opportunity for fluid biomarkers to be used in trials.”

Dr. Reiman cautioned, however, that the NfL measurements are likely a more useful measure of preclinical neurodegeneration in genetic early-onset Alzheimer’s than in late-onset or sporadic disease, which represents the lion’s share of Alzheimer’s cases.

“In autosomal dominant Alzheimer’s disease, these [NfL] changes really go up – probably more so than in late onset,” he said.

Dr. Reiman said these cohort findings add to growing interest in less-invasive biomarkers for Alzheimer’s, both in research and clinical practice. “If NfL is already elevated as a marker of active neurodegeneration, in early phase trials you might think about looking to it as a proof of concept – so where in 6-12 months you can see reductions [in NfL].”

Dr. Reiman added that “there will be other fluid biomarkers coming down the pike that will be exciting as well, and which people will learn a lot more about in the next few months.”

Dr. Quiroz had no disclosures related to her findings. Other authors on the study, including Dr. Reiman, have received research support and/or consulting fees from pharmaceutical manufacturers.

LOS ANGELES – Neurofilament light, or NfL, is an increasingly studied biomarker of axonal damage across a range of neurodegenerative diseases, including Alzheimer’s disease. And because it is a biomarker that can be measured in blood, it is a less invasive measure of disease progression in Alzheimer’s than cerebrospinal fluid markers.

At the Alzheimer’s Association International Conference, scientists studying the world’s largest cohort of early-onset Alzheimer’s families presented results from a cross-sectional and longitudinal study of more than 2,000 carriers and noncarriers of a single Alzheimer’s-causing mutation (Presenilin 1 E280A) that occurs in an extended Colombian family.

While previous studies have also looked at NfL in cohorts of autosomal dominant mutation carriers, this study strengthens evidence for NfL as an Alzheimer’s biomarker in the largest single-mutation cohort to date.

Yakeel Quiroz, PhD, of Harvard University in Boston and colleagues reported that, in a cross-sectional study of 1,070 mutation carriers and 1,074 noncarriers aged 8-75 years (mean age, 29-30 years; 46% male), mean plasma NfL levels were elevated in cognitively unimpaired carriers (18.08 pg/mL), compared with noncarriers (9.09 pg/mL; P less than .0001).

Longitudinal data from 504 of those carriers and noncarriers showed that NfL levels begin to diverge significantly between the groups at age 22, more than 2 decades before the mean onset of mild cognitive impairment for this cohort (44 years). The between-group differences in NfL continued to widen with advancing age.

“At approximately age 22, the axons, the neurons are already changing, and this measure serves as an early sign of degeneration,” Dr. Quiroz said in an interview. “This is really telling us about neurodegeneration related to Alzheimer’s disease because these are people destined to develop Alzheimer’s dementia later in life and have no age-related comorbidities that could cause elevation in NfL.”

A study published early this year in a different cohort of about 400 autosomal dominant Alzheimer’s disease mutation carriers and noncarriers found that the longitudinal rate of change of serum NfL could discriminate carriers from noncarriers almost a decade earlier than cross-sectional absolute NfL levels – at 16 years and 7 years, respectively, before expected onset of symptoms (Nat Med. 2019 Feb;25[2]:277–83).

In Dr. Quiroz and colleagues’ study, both cross-sectional and longitudinal findings showed carriers to significantly differ from noncarriers by age 22 years. “We’re seeing differences between groups that reach statistical significance earlier” – decades, in this case, before onset of symptoms, Dr. Quiroz said. The current study is distinguished by its exceptional power, she said: “No one has done this with such a large number of carriers with a single genetic mutation.”

Eric Reiman, MD, of Banner Alzheimer’s Institute in Phoenix, the coauthor on the study who presented the findings to the conference, commented in an interview that they illustrate “the opportunity for fluid biomarkers to be used in trials.”

Dr. Reiman cautioned, however, that the NfL measurements are likely a more useful measure of preclinical neurodegeneration in genetic early-onset Alzheimer’s than in late-onset or sporadic disease, which represents the lion’s share of Alzheimer’s cases.

“In autosomal dominant Alzheimer’s disease, these [NfL] changes really go up – probably more so than in late onset,” he said.

Dr. Reiman said these cohort findings add to growing interest in less-invasive biomarkers for Alzheimer’s, both in research and clinical practice. “If NfL is already elevated as a marker of active neurodegeneration, in early phase trials you might think about looking to it as a proof of concept – so where in 6-12 months you can see reductions [in NfL].”

Dr. Reiman added that “there will be other fluid biomarkers coming down the pike that will be exciting as well, and which people will learn a lot more about in the next few months.”

Dr. Quiroz had no disclosures related to her findings. Other authors on the study, including Dr. Reiman, have received research support and/or consulting fees from pharmaceutical manufacturers.

LOS ANGELES – Neurofilament light, or NfL, is an increasingly studied biomarker of axonal damage across a range of neurodegenerative diseases, including Alzheimer’s disease. And because it is a biomarker that can be measured in blood, it is a less invasive measure of disease progression in Alzheimer’s than cerebrospinal fluid markers.

At the Alzheimer’s Association International Conference, scientists studying the world’s largest cohort of early-onset Alzheimer’s families presented results from a cross-sectional and longitudinal study of more than 2,000 carriers and noncarriers of a single Alzheimer’s-causing mutation (Presenilin 1 E280A) that occurs in an extended Colombian family.

While previous studies have also looked at NfL in cohorts of autosomal dominant mutation carriers, this study strengthens evidence for NfL as an Alzheimer’s biomarker in the largest single-mutation cohort to date.

Yakeel Quiroz, PhD, of Harvard University in Boston and colleagues reported that, in a cross-sectional study of 1,070 mutation carriers and 1,074 noncarriers aged 8-75 years (mean age, 29-30 years; 46% male), mean plasma NfL levels were elevated in cognitively unimpaired carriers (18.08 pg/mL), compared with noncarriers (9.09 pg/mL; P less than .0001).

Longitudinal data from 504 of those carriers and noncarriers showed that NfL levels begin to diverge significantly between the groups at age 22, more than 2 decades before the mean onset of mild cognitive impairment for this cohort (44 years). The between-group differences in NfL continued to widen with advancing age.

“At approximately age 22, the axons, the neurons are already changing, and this measure serves as an early sign of degeneration,” Dr. Quiroz said in an interview. “This is really telling us about neurodegeneration related to Alzheimer’s disease because these are people destined to develop Alzheimer’s dementia later in life and have no age-related comorbidities that could cause elevation in NfL.”

A study published early this year in a different cohort of about 400 autosomal dominant Alzheimer’s disease mutation carriers and noncarriers found that the longitudinal rate of change of serum NfL could discriminate carriers from noncarriers almost a decade earlier than cross-sectional absolute NfL levels – at 16 years and 7 years, respectively, before expected onset of symptoms (Nat Med. 2019 Feb;25[2]:277–83).

In Dr. Quiroz and colleagues’ study, both cross-sectional and longitudinal findings showed carriers to significantly differ from noncarriers by age 22 years. “We’re seeing differences between groups that reach statistical significance earlier” – decades, in this case, before onset of symptoms, Dr. Quiroz said. The current study is distinguished by its exceptional power, she said: “No one has done this with such a large number of carriers with a single genetic mutation.”

Eric Reiman, MD, of Banner Alzheimer’s Institute in Phoenix, the coauthor on the study who presented the findings to the conference, commented in an interview that they illustrate “the opportunity for fluid biomarkers to be used in trials.”

Dr. Reiman cautioned, however, that the NfL measurements are likely a more useful measure of preclinical neurodegeneration in genetic early-onset Alzheimer’s than in late-onset or sporadic disease, which represents the lion’s share of Alzheimer’s cases.

“In autosomal dominant Alzheimer’s disease, these [NfL] changes really go up – probably more so than in late onset,” he said.

Dr. Reiman said these cohort findings add to growing interest in less-invasive biomarkers for Alzheimer’s, both in research and clinical practice. “If NfL is already elevated as a marker of active neurodegeneration, in early phase trials you might think about looking to it as a proof of concept – so where in 6-12 months you can see reductions [in NfL].”

Dr. Reiman added that “there will be other fluid biomarkers coming down the pike that will be exciting as well, and which people will learn a lot more about in the next few months.”

Dr. Quiroz had no disclosures related to her findings. Other authors on the study, including Dr. Reiman, have received research support and/or consulting fees from pharmaceutical manufacturers.

LOS ANGELES – Estrogen therapy may have scored another goal in its comeback game, as a 7-year prospective study shows that a transdermal formulation preserves some measures of cognitive function and brain architecture in postmenopausal women.

JackF/thinkstockphotos.com

In addition to performing better on subjective tests of memory, women using the estrogen patch experienced less cortical atrophy and were less likely to show amyloid on brain imaging. The observations were moderately associated with the improved sleep these women reported, Burcu Zeydan, MD, said at the Alzheimer’s Association International Conference.

“By 7 years, among the cognitive domains studied ... [less brain and cognitive change] correlated with lower global sleep score, meaning better sleep quality in the estradiol group,” said Dr. Zeydan, assistant professor of radiology at the Mayo Clinic in Rochester, Minn. “We previously found that preservation of dorsolateral prefrontal cortex over 7 years was associated with lower cortical beta-amyloid deposition on PET only in the estradiol group, pointing out the potential role of estrogen receptors in modulating this relationship.”

Dysregulated sleep is more common among women than men, particularly as menopause approaches and estrogen levels fluctuate, then decline, Dr. Zeydan said.

Dr. Zeydan reported the sleep substudy of KEEPS (the Kronos Early Estrogen Prevention Study), a randomized, double-blind, placebo-controlled, multisite trial that compared oral conjugated equine estrogen with transdermal estradiol. A control group received oral placebo and a placebo patch.*

Brain architecture was similar between the placebo and transdermal groups, but it was actually worse in some measures in the oral-estrogen group, compared with the placebo group. Women taking oral estrogen had more white matter hyperintensities, greater ventricle enlargement, and more cortical thinning. Those differences resolved after they stopped taking the oral formulation, bringing them into line with the transdermal and placebo groups.

The investigation also found that the transdermal group showed lower cerebral amyloid binding on PET scans relative to both placebo and oral estrogen.

“The relative preservation of dorsolateral prefrontal cortical volume in the [transdermal estradiol] group over 7 years indicates that hormone therapy may have long-term effects on the brain,” the team concluded. They noted that the original KEEPS study didn’t find any cognitive correlation with these changes.

The subanalysis looked at 69 women of the KEEPS cohort who had been followed for the full 7 years (4 years on treatment and 3 years off treatment). They were randomized to oral placebo and a placebo patch,* oral conjugated equine estrogen (0.45 mg/day), or transdermal estradiol (50 mcg/day). Participants in the active treatment groups received oral micronized progesterone 12 days each month. All had complete data on cognitive testing and brain imaging. Sleep quality was measured by the Pittsburgh Sleep Quality Index. Dr. Zeydan compared cognition and brain architecture findings in relation to the sleep score; lower scores mean better sleep.

The women were aged 42-58 years at baseline, and within 36 months from menopause. They had no history of menopausal hormone therapy or cardiovascular disease.

The investigators were particularly interested in how estrogen might have modulated the disturbed sleep patterns that often accompany perimenopause and early menopause, and whether the observed brain and cognitive changes tracked with sleep quality.

“During this time, 40% to 60% of women report problems sleeping, and estrogen decline seems to play an important role in sleep disturbances during this phase,” Dr. Zeydan said. “Although poor sleep quality is common in recently menopausal women, sleep quality improves with hormone therapy, as was previously demonstrated in KEEPS hormone therapy trial in recently menopausal women.”

By year 7, the cohort’s mean age was 61 years. The majority had at least some college education. The percentage who carried an apolipoprotein E epsilon-4 allele varied by group, with 15% positivity in the oral group, 48% in the transdermal group, and 16% in the placebo group.


Cognitive function was estimated with a global cognitive measure and four cognitive domain scores: verbal learning and memory, auditory attention and working memory, visual attention and executive function, and mental flexibility.

Higher attention and executive function scores were moderately correlated with a lower sleep score in the transdermal group (r = –0.54, a significant difference compared with the oral formulation). Lower sleep scores also showed a moderate correlation with preserved cortical volume of the dorsolateral prefrontal region (r = –0.47, also significantly different from the oral group).

Lower brain amyloid also positively correlated with better sleep. The correlation between sleep and global amyloid burden in the transdermal group was also moderate (r = 0.45), while the correlation in the oral group was significantly weaker (r = 0.18).

“We can say that sleep quality and transdermal estradiol during early postmenopausal years somehow interact to influence beta-amyloid deposition, preservation of dorsolateral prefrontal cortex volume, and attention and executive function,” Dr. Zeydan said.

Dr. Zeydan had no financial disclosures.

[email protected]

*Correction, 8/7/2019: An earlier version of this story did not make clear that participants in the control group received oral placebo and a placebo patch.

LOS ANGELES – Estrogen therapy may have scored another goal in its comeback game, as a 7-year prospective study shows that a transdermal formulation preserves some measures of cognitive function and brain architecture in postmenopausal women.

JackF/thinkstockphotos.com

In addition to performing better on subjective tests of memory, women using the estrogen patch experienced less cortical atrophy and were less likely to show amyloid on brain imaging. The observations were moderately associated with the improved sleep these women reported, Burcu Zeydan, MD, said at the Alzheimer’s Association International Conference.

“By 7 years, among the cognitive domains studied ... [less brain and cognitive change] correlated with lower global sleep score, meaning better sleep quality in the estradiol group,” said Dr. Zeydan, assistant professor of radiology at the Mayo Clinic in Rochester, Minn. “We previously found that preservation of dorsolateral prefrontal cortex over 7 years was associated with lower cortical beta-amyloid deposition on PET only in the estradiol group, pointing out the potential role of estrogen receptors in modulating this relationship.”

Dysregulated sleep is more common among women than men, particularly as menopause approaches and estrogen levels fluctuate, then decline, Dr. Zeydan said.

Dr. Zeydan reported the sleep substudy of KEEPS (the Kronos Early Estrogen Prevention Study), a randomized, double-blind, placebo-controlled, multisite trial that compared oral conjugated equine estrogen with transdermal estradiol. A control group received oral placebo and a placebo patch.*

Brain architecture was similar between the placebo and transdermal groups, but it was actually worse in some measures in the oral-estrogen group, compared with the placebo group. Women taking oral estrogen had more white matter hyperintensities, greater ventricle enlargement, and more cortical thinning. Those differences resolved after they stopped taking the oral formulation, bringing them into line with the transdermal and placebo groups.

The investigation also found that the transdermal group showed lower cerebral amyloid binding on PET scans relative to both placebo and oral estrogen.

“The relative preservation of dorsolateral prefrontal cortical volume in the [transdermal estradiol] group over 7 years indicates that hormone therapy may have long-term effects on the brain,” the team concluded. They noted that the original KEEPS study didn’t find any cognitive correlation with these changes.

The subanalysis looked at 69 women of the KEEPS cohort who had been followed for the full 7 years (4 years on treatment and 3 years off treatment). They were randomized to oral placebo and a placebo patch,* oral conjugated equine estrogen (0.45 mg/day), or transdermal estradiol (50 mcg/day). Participants in the active treatment groups received oral micronized progesterone 12 days each month. All had complete data on cognitive testing and brain imaging. Sleep quality was measured by the Pittsburgh Sleep Quality Index. Dr. Zeydan compared cognition and brain architecture findings in relation to the sleep score; lower scores mean better sleep.

The women were aged 42-58 years at baseline, and within 36 months from menopause. They had no history of menopausal hormone therapy or cardiovascular disease.

The investigators were particularly interested in how estrogen might have modulated the disturbed sleep patterns that often accompany perimenopause and early menopause, and whether the observed brain and cognitive changes tracked with sleep quality.

“During this time, 40% to 60% of women report problems sleeping, and estrogen decline seems to play an important role in sleep disturbances during this phase,” Dr. Zeydan said. “Although poor sleep quality is common in recently menopausal women, sleep quality improves with hormone therapy, as was previously demonstrated in KEEPS hormone therapy trial in recently menopausal women.”

By year 7, the cohort’s mean age was 61 years. The majority had at least some college education. The percentage who carried an apolipoprotein E epsilon-4 allele varied by group, with 15% positivity in the oral group, 48% in the transdermal group, and 16% in the placebo group.


Cognitive function was estimated with a global cognitive measure and four cognitive domain scores: verbal learning and memory, auditory attention and working memory, visual attention and executive function, and mental flexibility.

Higher attention and executive function scores were moderately correlated with a lower sleep score in the transdermal group (r = –0.54, a significant difference compared with the oral formulation). Lower sleep scores also showed a moderate correlation with preserved cortical volume of the dorsolateral prefrontal region (r = –0.47, also significantly different from the oral group).

Lower brain amyloid also positively correlated with better sleep. The correlation between sleep and global amyloid burden in the transdermal group was also moderate (r = 0.45), while the correlation in the oral group was significantly weaker (r = 0.18).

“We can say that sleep quality and transdermal estradiol during early postmenopausal years somehow interact to influence beta-amyloid deposition, preservation of dorsolateral prefrontal cortex volume, and attention and executive function,” Dr. Zeydan said.

Dr. Zeydan had no financial disclosures.

[email protected]

*Correction, 8/7/2019: An earlier version of this story did not make clear that participants in the control group received oral placebo and a placebo patch.

LOS ANGELES – Estrogen therapy may have scored another goal in its comeback game, as a 7-year prospective study shows that a transdermal formulation preserves some measures of cognitive function and brain architecture in postmenopausal women.

JackF/thinkstockphotos.com

In addition to performing better on subjective tests of memory, women using the estrogen patch experienced less cortical atrophy and were less likely to show amyloid on brain imaging. The observations were moderately associated with the improved sleep these women reported, Burcu Zeydan, MD, said at the Alzheimer’s Association International Conference.

“By 7 years, among the cognitive domains studied ... [less brain and cognitive change] correlated with lower global sleep score, meaning better sleep quality in the estradiol group,” said Dr. Zeydan, assistant professor of radiology at the Mayo Clinic in Rochester, Minn. “We previously found that preservation of dorsolateral prefrontal cortex over 7 years was associated with lower cortical beta-amyloid deposition on PET only in the estradiol group, pointing out the potential role of estrogen receptors in modulating this relationship.”

Dysregulated sleep is more common among women than men, particularly as menopause approaches and estrogen levels fluctuate, then decline, Dr. Zeydan said.

Dr. Zeydan reported the sleep substudy of KEEPS (the Kronos Early Estrogen Prevention Study), a randomized, double-blind, placebo-controlled, multisite trial that compared oral conjugated equine estrogen with transdermal estradiol. A control group received oral placebo and a placebo patch.*

Brain architecture was similar between the placebo and transdermal groups, but it was actually worse in some measures in the oral-estrogen group, compared with the placebo group. Women taking oral estrogen had more white matter hyperintensities, greater ventricle enlargement, and more cortical thinning. Those differences resolved after they stopped taking the oral formulation, bringing them into line with the transdermal and placebo groups.

The investigation also found that the transdermal group showed lower cerebral amyloid binding on PET scans relative to both placebo and oral estrogen.

“The relative preservation of dorsolateral prefrontal cortical volume in the [transdermal estradiol] group over 7 years indicates that hormone therapy may have long-term effects on the brain,” the team concluded. They noted that the original KEEPS study didn’t find any cognitive correlation with these changes.

The subanalysis looked at 69 women of the KEEPS cohort who had been followed for the full 7 years (4 years on treatment and 3 years off treatment). They were randomized to oral placebo and a placebo patch,* oral conjugated equine estrogen (0.45 mg/day), or transdermal estradiol (50 mcg/day). Participants in the active treatment groups received oral micronized progesterone 12 days each month. All had complete data on cognitive testing and brain imaging. Sleep quality was measured by the Pittsburgh Sleep Quality Index. Dr. Zeydan compared cognition and brain architecture findings in relation to the sleep score; lower scores mean better sleep.

The women were aged 42-58 years at baseline, and within 36 months from menopause. They had no history of menopausal hormone therapy or cardiovascular disease.

The investigators were particularly interested in how estrogen might have modulated the disturbed sleep patterns that often accompany perimenopause and early menopause, and whether the observed brain and cognitive changes tracked with sleep quality.

“During this time, 40% to 60% of women report problems sleeping, and estrogen decline seems to play an important role in sleep disturbances during this phase,” Dr. Zeydan said. “Although poor sleep quality is common in recently menopausal women, sleep quality improves with hormone therapy, as was previously demonstrated in KEEPS hormone therapy trial in recently menopausal women.”

By year 7, the cohort’s mean age was 61 years. The majority had at least some college education. The percentage who carried an apolipoprotein E epsilon-4 allele varied by group, with 15% positivity in the oral group, 48% in the transdermal group, and 16% in the placebo group.


Cognitive function was estimated with a global cognitive measure and four cognitive domain scores: verbal learning and memory, auditory attention and working memory, visual attention and executive function, and mental flexibility.

Higher attention and executive function scores were moderately correlated with a lower sleep score in the transdermal group (r = –0.54, a significant difference compared with the oral formulation). Lower sleep scores also showed a moderate correlation with preserved cortical volume of the dorsolateral prefrontal region (r = –0.47, also significantly different from the oral group).

Lower brain amyloid also positively correlated with better sleep. The correlation between sleep and global amyloid burden in the transdermal group was also moderate (r = 0.45), while the correlation in the oral group was significantly weaker (r = 0.18).

“We can say that sleep quality and transdermal estradiol during early postmenopausal years somehow interact to influence beta-amyloid deposition, preservation of dorsolateral prefrontal cortex volume, and attention and executive function,” Dr. Zeydan said.

Dr. Zeydan had no financial disclosures.

[email protected]

*Correction, 8/7/2019: An earlier version of this story did not make clear that participants in the control group received oral placebo and a placebo patch.

LOS ANGELES – Scientists testing an unusual hypothesis about the pathogenesis of Alzheimer’s disease say that use of an experimental antimicrobial drug to target a common oral infection was associated with biomarker improvements in people with the disease.

At the Alzheimer’s Association International Conference, Michael Detke, MD, PhD, of Cortexyme in South San Francisco, Calif., presented findings from a small cohort (n = 9) of people with mild to moderate Alzheimer’s disease (AD).

Six patients, mean age 72, were randomized to 4 weeks’ treatment with an agent called COR388, which inhibits toxic proteases produced by Porphyromonas gingivalis, a bacterium that colonizes the mouth and gums and has been found in the brains and cerebrospinal fluid of people with AD more than in non-AD patients. Another three subjects were randomized to placebo.

At 28 days, the CSF levels of an Alzheimer’s-associated apolipoprotein E (ApoE) fragment and serum levels of RANTES, a chemokine associated with chronic inflammation, were reduced from baseline by about one-third in treated subjects, compared with placebo subjects (P less than .05).

“In AD, ApoE is known to be fragmented and the fragments are known to be neurotoxic,” Dr. Detke, the lead study author, said in an interview. “We hypothesized that these P. gingivalis proteases may be acting on ApoE cleavage – so if you bind the proteases you should reduce the fragments. We saw in this study that ApoE fragment was reduced by 30%, or back to about normal levels.”

It is difficult to eradicate P. gingivalis infection using conventional antibiotics. The experimental agent COR388 “does not kill the bacteria but rather neutralizes it by binding to the proteases it produces, making the bacteria benign,” Dr. Detke said.


While hypotheses involving infectious causes of Alzheimer’s remain on the periphery of dementia research, Dr. Detke and his colleagues will soon be able to test theirs in a more meaningful way. At the conference, Dr. Detke presented detailed plans for a phase 2/3, placebo-controlled, clinical trial of COR388 in 570 patients aged 55-80 with mild to moderate AD.

Patients in the study are currently being randomized to 48 weeks of treatment with one of two doses, or placebo, with cognitive and biomarker endpoints planned, including for amyloid-beta, tau, and serum, plasma, and CSF markers of neuroinflammation.

Dr. Detke is an employee of Cortexyme, as are another 8 of the study’s 11 authors.

LOS ANGELES – Scientists testing an unusual hypothesis about the pathogenesis of Alzheimer’s disease say that use of an experimental antimicrobial drug to target a common oral infection was associated with biomarker improvements in people with the disease.

At the Alzheimer’s Association International Conference, Michael Detke, MD, PhD, of Cortexyme in South San Francisco, Calif., presented findings from a small cohort (n = 9) of people with mild to moderate Alzheimer’s disease (AD).

Six patients, mean age 72, were randomized to 4 weeks’ treatment with an agent called COR388, which inhibits toxic proteases produced by Porphyromonas gingivalis, a bacterium that colonizes the mouth and gums and has been found in the brains and cerebrospinal fluid of people with AD more than in non-AD patients. Another three subjects were randomized to placebo.

At 28 days, the CSF levels of an Alzheimer’s-associated apolipoprotein E (ApoE) fragment and serum levels of RANTES, a chemokine associated with chronic inflammation, were reduced from baseline by about one-third in treated subjects, compared with placebo subjects (P less than .05).

“In AD, ApoE is known to be fragmented and the fragments are known to be neurotoxic,” Dr. Detke, the lead study author, said in an interview. “We hypothesized that these P. gingivalis proteases may be acting on ApoE cleavage – so if you bind the proteases you should reduce the fragments. We saw in this study that ApoE fragment was reduced by 30%, or back to about normal levels.”

It is difficult to eradicate P. gingivalis infection using conventional antibiotics. The experimental agent COR388 “does not kill the bacteria but rather neutralizes it by binding to the proteases it produces, making the bacteria benign,” Dr. Detke said.


While hypotheses involving infectious causes of Alzheimer’s remain on the periphery of dementia research, Dr. Detke and his colleagues will soon be able to test theirs in a more meaningful way. At the conference, Dr. Detke presented detailed plans for a phase 2/3, placebo-controlled, clinical trial of COR388 in 570 patients aged 55-80 with mild to moderate AD.

Patients in the study are currently being randomized to 48 weeks of treatment with one of two doses, or placebo, with cognitive and biomarker endpoints planned, including for amyloid-beta, tau, and serum, plasma, and CSF markers of neuroinflammation.

Dr. Detke is an employee of Cortexyme, as are another 8 of the study’s 11 authors.

LOS ANGELES – Scientists testing an unusual hypothesis about the pathogenesis of Alzheimer’s disease say that use of an experimental antimicrobial drug to target a common oral infection was associated with biomarker improvements in people with the disease.

At the Alzheimer’s Association International Conference, Michael Detke, MD, PhD, of Cortexyme in South San Francisco, Calif., presented findings from a small cohort (n = 9) of people with mild to moderate Alzheimer’s disease (AD).

Six patients, mean age 72, were randomized to 4 weeks’ treatment with an agent called COR388, which inhibits toxic proteases produced by Porphyromonas gingivalis, a bacterium that colonizes the mouth and gums and has been found in the brains and cerebrospinal fluid of people with AD more than in non-AD patients. Another three subjects were randomized to placebo.

At 28 days, the CSF levels of an Alzheimer’s-associated apolipoprotein E (ApoE) fragment and serum levels of RANTES, a chemokine associated with chronic inflammation, were reduced from baseline by about one-third in treated subjects, compared with placebo subjects (P less than .05).

“In AD, ApoE is known to be fragmented and the fragments are known to be neurotoxic,” Dr. Detke, the lead study author, said in an interview. “We hypothesized that these P. gingivalis proteases may be acting on ApoE cleavage – so if you bind the proteases you should reduce the fragments. We saw in this study that ApoE fragment was reduced by 30%, or back to about normal levels.”

It is difficult to eradicate P. gingivalis infection using conventional antibiotics. The experimental agent COR388 “does not kill the bacteria but rather neutralizes it by binding to the proteases it produces, making the bacteria benign,” Dr. Detke said.


While hypotheses involving infectious causes of Alzheimer’s remain on the periphery of dementia research, Dr. Detke and his colleagues will soon be able to test theirs in a more meaningful way. At the conference, Dr. Detke presented detailed plans for a phase 2/3, placebo-controlled, clinical trial of COR388 in 570 patients aged 55-80 with mild to moderate AD.

Patients in the study are currently being randomized to 48 weeks of treatment with one of two doses, or placebo, with cognitive and biomarker endpoints planned, including for amyloid-beta, tau, and serum, plasma, and CSF markers of neuroinflammation.

Dr. Detke is an employee of Cortexyme, as are another 8 of the study’s 11 authors.

LOS ANGELES – While uncontrolled hypertension is an established risk factor for dementias including Alzheimer’s disease, the pathways by which it might lead to dementia remain poorly understood.

GlobalStock/Getty Images

In research presented at the Alzheimer’s Association International Conference, Jérémie Lespinasse, PharmD, an investigator with the University of Bordeaux and INSERM U1219 in Bordeaux, France, presented data showing that hypertension-linked increases in the brain’s load of white matter hyperintensities – an imaging biomarker linked to small-vessel disease – was associated with cognitive decline independent of amyloid cascade biomarkers.

“We know that hypertension is related to cognitive decline and dementia, including Alzheimer’s dementia,” Carole Dufouil, PhD, the study’s last author and the research director at INSERM U1219, said in an interview. “But we didn’t know the pathway. It could go through what’s more typical of vascular pathology or it could go through what’s more typical of AD – and that’s what we wanted to test.”

The researchers found that the impact of hypertension on cognition doesn’t “go through amyloid,” Dr. Dufouil said, but rather a typically vascular pathway of white matter hyperintensities and neurodegeneration. “This is a big difference from other brain markers for which you know they exist, but you don’t know how to treat them,” she said. “This one is treatable.”

For their research, Dr. Lespinasse, Dr. Dufouil, and colleagues used a cross-sectional sample of data from the MEMENTO study, a 5-year observational cohort of 2,323 patients recruited at 26 memory centers in France between 2011 and 2014. Of the patients in MEMENTO, 62% were women, and the mean age was 71. All patients were deemed free of dementia and had isolated cognitive complaints or mild cognitive impairment at baseline. A total of 60% had hypertension, and 17% had uncontrolled hypertension defined as above 140/90 mm Hg despite treatment. Cognitive testing and MRI was conducted on all patients, while 60% also had ¹⁸F-fluorodeoxyglucose PET scanning and a minority, 18%, had cerebrospinal fluid samples.


The investigators found in using a structural equation model that the uncontrolled hypertension subjects had significantly lower cognition when compared against those without (P = .001). About half of the harmful effect of uncontrolled hypertension on brain functions was mediated by white matter hyperintensities load (P = .021) and neurodegeneration (P = .024) but not by cerebrospinal fluid biomarkers for amyloid-beta 42/40 ratio or tau.

The study’s main limitation was its use of cross-sectional data, the investigators said, while its strength was in a multifactorial model that allowed for a more integrative look at the relationships among hypertension, Alzheimer’s biomarkers, white matter hyperintensities, and cognition.

The investigators stressed the importance of controlling hypertension generally – and for all clinicians to be more aware of its cognitive impacts. Dr. Dufouil said that memory clinics should make blood pressure monitoring a key part of their workups, and should ensure that people are well controlled. “Up until recently it hasn’t been obvious” that control of hypertension has a key role in dementia prevention.

“It’s now obvious,” she said.

Dr. Lespinasse and Dr. Dufouil disclosed no industry relationships.

LOS ANGELES – While uncontrolled hypertension is an established risk factor for dementias including Alzheimer’s disease, the pathways by which it might lead to dementia remain poorly understood.

GlobalStock/Getty Images

In research presented at the Alzheimer’s Association International Conference, Jérémie Lespinasse, PharmD, an investigator with the University of Bordeaux and INSERM U1219 in Bordeaux, France, presented data showing that hypertension-linked increases in the brain’s load of white matter hyperintensities – an imaging biomarker linked to small-vessel disease – was associated with cognitive decline independent of amyloid cascade biomarkers.

“We know that hypertension is related to cognitive decline and dementia, including Alzheimer’s dementia,” Carole Dufouil, PhD, the study’s last author and the research director at INSERM U1219, said in an interview. “But we didn’t know the pathway. It could go through what’s more typical of vascular pathology or it could go through what’s more typical of AD – and that’s what we wanted to test.”

The researchers found that the impact of hypertension on cognition doesn’t “go through amyloid,” Dr. Dufouil said, but rather a typically vascular pathway of white matter hyperintensities and neurodegeneration. “This is a big difference from other brain markers for which you know they exist, but you don’t know how to treat them,” she said. “This one is treatable.”

For their research, Dr. Lespinasse, Dr. Dufouil, and colleagues used a cross-sectional sample of data from the MEMENTO study, a 5-year observational cohort of 2,323 patients recruited at 26 memory centers in France between 2011 and 2014. Of the patients in MEMENTO, 62% were women, and the mean age was 71. All patients were deemed free of dementia and had isolated cognitive complaints or mild cognitive impairment at baseline. A total of 60% had hypertension, and 17% had uncontrolled hypertension defined as above 140/90 mm Hg despite treatment. Cognitive testing and MRI was conducted on all patients, while 60% also had ¹⁸F-fluorodeoxyglucose PET scanning and a minority, 18%, had cerebrospinal fluid samples.


The investigators found in using a structural equation model that the uncontrolled hypertension subjects had significantly lower cognition when compared against those without (P = .001). About half of the harmful effect of uncontrolled hypertension on brain functions was mediated by white matter hyperintensities load (P = .021) and neurodegeneration (P = .024) but not by cerebrospinal fluid biomarkers for amyloid-beta 42/40 ratio or tau.

The study’s main limitation was its use of cross-sectional data, the investigators said, while its strength was in a multifactorial model that allowed for a more integrative look at the relationships among hypertension, Alzheimer’s biomarkers, white matter hyperintensities, and cognition.

The investigators stressed the importance of controlling hypertension generally – and for all clinicians to be more aware of its cognitive impacts. Dr. Dufouil said that memory clinics should make blood pressure monitoring a key part of their workups, and should ensure that people are well controlled. “Up until recently it hasn’t been obvious” that control of hypertension has a key role in dementia prevention.

“It’s now obvious,” she said.

Dr. Lespinasse and Dr. Dufouil disclosed no industry relationships.

LOS ANGELES – While uncontrolled hypertension is an established risk factor for dementias including Alzheimer’s disease, the pathways by which it might lead to dementia remain poorly understood.

GlobalStock/Getty Images

In research presented at the Alzheimer’s Association International Conference, Jérémie Lespinasse, PharmD, an investigator with the University of Bordeaux and INSERM U1219 in Bordeaux, France, presented data showing that hypertension-linked increases in the brain’s load of white matter hyperintensities – an imaging biomarker linked to small-vessel disease – was associated with cognitive decline independent of amyloid cascade biomarkers.

“We know that hypertension is related to cognitive decline and dementia, including Alzheimer’s dementia,” Carole Dufouil, PhD, the study’s last author and the research director at INSERM U1219, said in an interview. “But we didn’t know the pathway. It could go through what’s more typical of vascular pathology or it could go through what’s more typical of AD – and that’s what we wanted to test.”

The researchers found that the impact of hypertension on cognition doesn’t “go through amyloid,” Dr. Dufouil said, but rather a typically vascular pathway of white matter hyperintensities and neurodegeneration. “This is a big difference from other brain markers for which you know they exist, but you don’t know how to treat them,” she said. “This one is treatable.”

For their research, Dr. Lespinasse, Dr. Dufouil, and colleagues used a cross-sectional sample of data from the MEMENTO study, a 5-year observational cohort of 2,323 patients recruited at 26 memory centers in France between 2011 and 2014. Of the patients in MEMENTO, 62% were women, and the mean age was 71. All patients were deemed free of dementia and had isolated cognitive complaints or mild cognitive impairment at baseline. A total of 60% had hypertension, and 17% had uncontrolled hypertension defined as above 140/90 mm Hg despite treatment. Cognitive testing and MRI was conducted on all patients, while 60% also had ¹⁸F-fluorodeoxyglucose PET scanning and a minority, 18%, had cerebrospinal fluid samples.


The investigators found in using a structural equation model that the uncontrolled hypertension subjects had significantly lower cognition when compared against those without (P = .001). About half of the harmful effect of uncontrolled hypertension on brain functions was mediated by white matter hyperintensities load (P = .021) and neurodegeneration (P = .024) but not by cerebrospinal fluid biomarkers for amyloid-beta 42/40 ratio or tau.

The study’s main limitation was its use of cross-sectional data, the investigators said, while its strength was in a multifactorial model that allowed for a more integrative look at the relationships among hypertension, Alzheimer’s biomarkers, white matter hyperintensities, and cognition.

The investigators stressed the importance of controlling hypertension generally – and for all clinicians to be more aware of its cognitive impacts. Dr. Dufouil said that memory clinics should make blood pressure monitoring a key part of their workups, and should ensure that people are well controlled. “Up until recently it hasn’t been obvious” that control of hypertension has a key role in dementia prevention.

“It’s now obvious,” she said.

Dr. Lespinasse and Dr. Dufouil disclosed no industry relationships.

LOS ANGELES – Seizures are not uncommon among patients with Alzheimer’s disease – particularly as patients live longer with the disease – and are often associated with worse cognitive and functional performance, according to research findings presented at the Alzheimer’s Association International Conference.

Jonathan Vöglein, MD, of the German Center for Neurodegenerative Diseases and Ludwig-Maximilian University in Munich presented results from a cohort of 9,127 patients with Alzheimer’s disease (AD), of whom 287 had experienced a seizure, and more than 10,000 non-AD control subjects recruited at clinics during 2005-2016.

Dr. Vöglein and colleagues found that seizure risk increased with duration of disease, from 1.5% of patients at 4.8 years with the disease to 5.4% at 11 years, with likelihood of a seizure increasing steadily over time.

Moreover, 70% of AD patients who experienced a seizure had a second one within 7.5 months. People who had seizures fared worse on cognitive and functional tests: a mean 16.6 on the Mini Mental State Examination, compared with 19.6 for patients without seizures. On a severity rating scale, the Clinical Dementia Rating Sum of Boxes, patients with seizures also fared worse, with scores of 9.3, compared with 6.8 for patients without seizures (P less than .0001 for all, with results adjusted for age and disease duration).

“The data of our study show that there’s an association of seizures with worse cognitive and functional performance,” Dr. Vöglein said in an interview.

“It’s important for clinicians to know that Alzheimer’s patients are at an increased risk for seizures,” Dr. Vöglein said. “In my clinical care experience, seizures are rarely the main complaint of patients with Alzheimer’s disease.” Detailed interviews with the patient and a proxy are important, he added, because patients with Alzheimer’s disease may not always remember events that could be a seizure.

Dr. Vöglein noted that, to his knowledge, there are no reliable data showing that treating seizures with antiepileptic drugs slows cognitive decline. “The results of our study suggest that an antiepileptic treatment after a first seizure in patients with Alzheimer’s dementia may be considered,” he said.

Also at the conference, researcher Ruby Castilla-Puentes, MD, DrPH, of Janssen Pharmaceuticals in Hopewell, N.J., along with Miguel Habeych, MD, MPH, of the University of Cincinnati presented findings on dementia and seizure risk from a large U.S. national managed care database of nearly 3 million people aged 60 years and older, of whom 56% were women.

The researchers analyzed this cohort during 2005-2014 and identified 80,000 people (2.8% of the cohort) as having any dementia diagnosis. The overall incidence of new-onset seizures in patients with dementia was 12.3% per year. In general, all subtypes of seizures and epileptic disorders (partial, generalized, or undifferentiated) occurred more frequently in patients with dementia, compared against patients without dementia (P less than .0001).

People with dementia had more than six times greater risk for experiencing recurring epileptic seizures than did people without dementia (95% confidence interval, 4.4-9.5). They were at six times higher risk for partial seizures (95% CI, 5.5-6.6); fivefold higher risk for generalized (95% CI, 4.9-5.5) and undifferentiated epilepsy (95% CI, 4.8-5.2); and 4.75 times higher risk for generalized seizures (95% CI, 4.5-5.0) and partial epilepsy (95% CI, 4.4-5.1).

“Although there are limitations with the use of administrative claims databases to calculate incidence rates, this analysis suggests that patients of 60 years of age or older have higher risks of new-onset seizures associated with a dementia diagnosis,” Dr. Castilla-Puentes commented.

The findings, she said, reinforce the need for clinicians to monitor for seizures to ensure that patients with dementia receive appropriate treatment.

Dr. Vöglein disclosed no financial conflicts of interest. Dr. Castilla-Puentes disclosed being an employee of Janssen, which funded her study.

LOS ANGELES – Seizures are not uncommon among patients with Alzheimer’s disease – particularly as patients live longer with the disease – and are often associated with worse cognitive and functional performance, according to research findings presented at the Alzheimer’s Association International Conference.

Jonathan Vöglein, MD, of the German Center for Neurodegenerative Diseases and Ludwig-Maximilian University in Munich presented results from a cohort of 9,127 patients with Alzheimer’s disease (AD), of whom 287 had experienced a seizure, and more than 10,000 non-AD control subjects recruited at clinics during 2005-2016.

Dr. Vöglein and colleagues found that seizure risk increased with duration of disease, from 1.5% of patients at 4.8 years with the disease to 5.4% at 11 years, with likelihood of a seizure increasing steadily over time.

Moreover, 70% of AD patients who experienced a seizure had a second one within 7.5 months. People who had seizures fared worse on cognitive and functional tests: a mean 16.6 on the Mini Mental State Examination, compared with 19.6 for patients without seizures. On a severity rating scale, the Clinical Dementia Rating Sum of Boxes, patients with seizures also fared worse, with scores of 9.3, compared with 6.8 for patients without seizures (P less than .0001 for all, with results adjusted for age and disease duration).

“The data of our study show that there’s an association of seizures with worse cognitive and functional performance,” Dr. Vöglein said in an interview.

“It’s important for clinicians to know that Alzheimer’s patients are at an increased risk for seizures,” Dr. Vöglein said. “In my clinical care experience, seizures are rarely the main complaint of patients with Alzheimer’s disease.” Detailed interviews with the patient and a proxy are important, he added, because patients with Alzheimer’s disease may not always remember events that could be a seizure.

Dr. Vöglein noted that, to his knowledge, there are no reliable data showing that treating seizures with antiepileptic drugs slows cognitive decline. “The results of our study suggest that an antiepileptic treatment after a first seizure in patients with Alzheimer’s dementia may be considered,” he said.

Also at the conference, researcher Ruby Castilla-Puentes, MD, DrPH, of Janssen Pharmaceuticals in Hopewell, N.J., along with Miguel Habeych, MD, MPH, of the University of Cincinnati presented findings on dementia and seizure risk from a large U.S. national managed care database of nearly 3 million people aged 60 years and older, of whom 56% were women.

The researchers analyzed this cohort during 2005-2014 and identified 80,000 people (2.8% of the cohort) as having any dementia diagnosis. The overall incidence of new-onset seizures in patients with dementia was 12.3% per year. In general, all subtypes of seizures and epileptic disorders (partial, generalized, or undifferentiated) occurred more frequently in patients with dementia, compared against patients without dementia (P less than .0001).

People with dementia had more than six times greater risk for experiencing recurring epileptic seizures than did people without dementia (95% confidence interval, 4.4-9.5). They were at six times higher risk for partial seizures (95% CI, 5.5-6.6); fivefold higher risk for generalized (95% CI, 4.9-5.5) and undifferentiated epilepsy (95% CI, 4.8-5.2); and 4.75 times higher risk for generalized seizures (95% CI, 4.5-5.0) and partial epilepsy (95% CI, 4.4-5.1).

“Although there are limitations with the use of administrative claims databases to calculate incidence rates, this analysis suggests that patients of 60 years of age or older have higher risks of new-onset seizures associated with a dementia diagnosis,” Dr. Castilla-Puentes commented.

The findings, she said, reinforce the need for clinicians to monitor for seizures to ensure that patients with dementia receive appropriate treatment.

Dr. Vöglein disclosed no financial conflicts of interest. Dr. Castilla-Puentes disclosed being an employee of Janssen, which funded her study.

LOS ANGELES – Seizures are not uncommon among patients with Alzheimer’s disease – particularly as patients live longer with the disease – and are often associated with worse cognitive and functional performance, according to research findings presented at the Alzheimer’s Association International Conference.

Jonathan Vöglein, MD, of the German Center for Neurodegenerative Diseases and Ludwig-Maximilian University in Munich presented results from a cohort of 9,127 patients with Alzheimer’s disease (AD), of whom 287 had experienced a seizure, and more than 10,000 non-AD control subjects recruited at clinics during 2005-2016.

Dr. Vöglein and colleagues found that seizure risk increased with duration of disease, from 1.5% of patients at 4.8 years with the disease to 5.4% at 11 years, with likelihood of a seizure increasing steadily over time.

Moreover, 70% of AD patients who experienced a seizure had a second one within 7.5 months. People who had seizures fared worse on cognitive and functional tests: a mean 16.6 on the Mini Mental State Examination, compared with 19.6 for patients without seizures. On a severity rating scale, the Clinical Dementia Rating Sum of Boxes, patients with seizures also fared worse, with scores of 9.3, compared with 6.8 for patients without seizures (P less than .0001 for all, with results adjusted for age and disease duration).

“The data of our study show that there’s an association of seizures with worse cognitive and functional performance,” Dr. Vöglein said in an interview.

“It’s important for clinicians to know that Alzheimer’s patients are at an increased risk for seizures,” Dr. Vöglein said. “In my clinical care experience, seizures are rarely the main complaint of patients with Alzheimer’s disease.” Detailed interviews with the patient and a proxy are important, he added, because patients with Alzheimer’s disease may not always remember events that could be a seizure.

Dr. Vöglein noted that, to his knowledge, there are no reliable data showing that treating seizures with antiepileptic drugs slows cognitive decline. “The results of our study suggest that an antiepileptic treatment after a first seizure in patients with Alzheimer’s dementia may be considered,” he said.

Also at the conference, researcher Ruby Castilla-Puentes, MD, DrPH, of Janssen Pharmaceuticals in Hopewell, N.J., along with Miguel Habeych, MD, MPH, of the University of Cincinnati presented findings on dementia and seizure risk from a large U.S. national managed care database of nearly 3 million people aged 60 years and older, of whom 56% were women.

The researchers analyzed this cohort during 2005-2014 and identified 80,000 people (2.8% of the cohort) as having any dementia diagnosis. The overall incidence of new-onset seizures in patients with dementia was 12.3% per year. In general, all subtypes of seizures and epileptic disorders (partial, generalized, or undifferentiated) occurred more frequently in patients with dementia, compared against patients without dementia (P less than .0001).

People with dementia had more than six times greater risk for experiencing recurring epileptic seizures than did people without dementia (95% confidence interval, 4.4-9.5). They were at six times higher risk for partial seizures (95% CI, 5.5-6.6); fivefold higher risk for generalized (95% CI, 4.9-5.5) and undifferentiated epilepsy (95% CI, 4.8-5.2); and 4.75 times higher risk for generalized seizures (95% CI, 4.5-5.0) and partial epilepsy (95% CI, 4.4-5.1).

“Although there are limitations with the use of administrative claims databases to calculate incidence rates, this analysis suggests that patients of 60 years of age or older have higher risks of new-onset seizures associated with a dementia diagnosis,” Dr. Castilla-Puentes commented.

The findings, she said, reinforce the need for clinicians to monitor for seizures to ensure that patients with dementia receive appropriate treatment.

Dr. Vöglein disclosed no financial conflicts of interest. Dr. Castilla-Puentes disclosed being an employee of Janssen, which funded her study.

LOS ANGELES – Dysregulated sleep-wake cycles may be linked to cerebral glucose hypometabolism and subtle cognitive changes, both of which are early signs of Alzheimer’s disease–like neurodegeneration, according to a 2-year study of older Korean adults.

Michele G. Sullivan/MDedge News

The association was particularly strong in subjects who experienced delayed acrophase, the peak of the normal sleep-wake cycle, So-Yeon Jeon, MD, said at the Alzheimer’s Association International Conference. It’s not yet clear whether the changes are a risk factor for dementia or a prodromal sign of neurodegeneration, but even without full elucidation, the findings could have value as a signal of impending neurodegeneration, said Dr. Jeon of Seoul (South Korea) National University.

“Our findings suggest that delayed acrophase may be used as a predictor for the progression of Alzheimer’s-type neurodegeneration and cognitive decline in the near future in old individuals with diverse cognitive status,” she said. “But the relationship between circadian phases and neurodegeneration is complex and not yet well understood.”

The 24-month study comprised 215 elderly adults enrolled in the Korean Brain Aging Study for the Early Diagnosis and Prediction of Alzheimer’s Disease (KBASE). They were a mean of 70 years old at baseline; 143 were cognitively normal, 40 had mild cognitive impairment, and 32 had Alzheimer’s dementia. Both at baseline and 2 years, everyone underwent a comprehensive neuropsychological assessment, amyloid PET brain imaging with Pittsburgh compound B, and an [18F]-fluorodeoxyglucose PET scan to determine brain glucose metabolic rate.

Before each assessment, the investigators measured sleep and circadian rhythms with 8 days of actigraphy. This assessed sleep variables (total sleep time, sleep latency, sleep efficiency, and wakefulness after sleep); rest-activity rhythm variables (midline estimated statistic of rhythm, amplitude, and acrophase), and some nonparametric values including interdaily stability, intradaily variability, and relative amplitude of sleep cycles. Subjects also completed sleep diaries during these periods.

The study’s main outcomes were 2-year changes in the Mini Mental State Exam (MMSE) score and in Alzheimer’s imaging biomarkers, including glucose metabolism and amyloid deposition. All analyses controlled for age, sex, Clinical Dementia Rating score, apolipoprotein E allele status, and baseline cognition.

At baseline, lower total sleep time was significantly associated with hypometabolism in areas associated with Alzheimer’s pathology as well as lower mean MMSE scores. Circadian variables showed no significant associations with these characteristics. However, the relative amplitude of circadian rhythm was significantly associated with hypometabolism and with lower MMSE score. There were no associations with brain amyloid load.

At 2 years, acrophase was associated with declines in cerebral glucose metabolism and further changes in the MMSE, even after the researchers controlled for the potential confounders. Delayed acrophase, although not associated with either metabolic rate or cognition at baseline, did significantly influence both at 2 years, suggesting a rapidly eroding clinical picture.

“Neurodegeneration over 2 years means the disease is progressing rapidly and subjects are likely to have tauopathies or other proteinopathy,” Dr. Jeon said. “These pathologies may either be resulting in delayed acrophase followed by neurodegeneration, or they may be prodromal symptoms of impending neurodegeneration. Whether they are early symptoms or early risk factors is not currently known, however. Two years is too short of a follow-up to determine these questions.”

Dr. Jeon had no financial declarations.

[email protected]

SOURCE: Jeon SY et al. AAIC 2019, abstract 33543.

LOS ANGELES – Dysregulated sleep-wake cycles may be linked to cerebral glucose hypometabolism and subtle cognitive changes, both of which are early signs of Alzheimer’s disease–like neurodegeneration, according to a 2-year study of older Korean adults.

Michele G. Sullivan/MDedge News

The association was particularly strong in subjects who experienced delayed acrophase, the peak of the normal sleep-wake cycle, So-Yeon Jeon, MD, said at the Alzheimer’s Association International Conference. It’s not yet clear whether the changes are a risk factor for dementia or a prodromal sign of neurodegeneration, but even without full elucidation, the findings could have value as a signal of impending neurodegeneration, said Dr. Jeon of Seoul (South Korea) National University.

“Our findings suggest that delayed acrophase may be used as a predictor for the progression of Alzheimer’s-type neurodegeneration and cognitive decline in the near future in old individuals with diverse cognitive status,” she said. “But the relationship between circadian phases and neurodegeneration is complex and not yet well understood.”

The 24-month study comprised 215 elderly adults enrolled in the Korean Brain Aging Study for the Early Diagnosis and Prediction of Alzheimer’s Disease (KBASE). They were a mean of 70 years old at baseline; 143 were cognitively normal, 40 had mild cognitive impairment, and 32 had Alzheimer’s dementia. Both at baseline and 2 years, everyone underwent a comprehensive neuropsychological assessment, amyloid PET brain imaging with Pittsburgh compound B, and an [18F]-fluorodeoxyglucose PET scan to determine brain glucose metabolic rate.

Before each assessment, the investigators measured sleep and circadian rhythms with 8 days of actigraphy. This assessed sleep variables (total sleep time, sleep latency, sleep efficiency, and wakefulness after sleep); rest-activity rhythm variables (midline estimated statistic of rhythm, amplitude, and acrophase), and some nonparametric values including interdaily stability, intradaily variability, and relative amplitude of sleep cycles. Subjects also completed sleep diaries during these periods.

The study’s main outcomes were 2-year changes in the Mini Mental State Exam (MMSE) score and in Alzheimer’s imaging biomarkers, including glucose metabolism and amyloid deposition. All analyses controlled for age, sex, Clinical Dementia Rating score, apolipoprotein E allele status, and baseline cognition.

At baseline, lower total sleep time was significantly associated with hypometabolism in areas associated with Alzheimer’s pathology as well as lower mean MMSE scores. Circadian variables showed no significant associations with these characteristics. However, the relative amplitude of circadian rhythm was significantly associated with hypometabolism and with lower MMSE score. There were no associations with brain amyloid load.

At 2 years, acrophase was associated with declines in cerebral glucose metabolism and further changes in the MMSE, even after the researchers controlled for the potential confounders. Delayed acrophase, although not associated with either metabolic rate or cognition at baseline, did significantly influence both at 2 years, suggesting a rapidly eroding clinical picture.

“Neurodegeneration over 2 years means the disease is progressing rapidly and subjects are likely to have tauopathies or other proteinopathy,” Dr. Jeon said. “These pathologies may either be resulting in delayed acrophase followed by neurodegeneration, or they may be prodromal symptoms of impending neurodegeneration. Whether they are early symptoms or early risk factors is not currently known, however. Two years is too short of a follow-up to determine these questions.”

Dr. Jeon had no financial declarations.

[email protected]

SOURCE: Jeon SY et al. AAIC 2019, abstract 33543.

LOS ANGELES – Dysregulated sleep-wake cycles may be linked to cerebral glucose hypometabolism and subtle cognitive changes, both of which are early signs of Alzheimer’s disease–like neurodegeneration, according to a 2-year study of older Korean adults.

Michele G. Sullivan/MDedge News

The association was particularly strong in subjects who experienced delayed acrophase, the peak of the normal sleep-wake cycle, So-Yeon Jeon, MD, said at the Alzheimer’s Association International Conference. It’s not yet clear whether the changes are a risk factor for dementia or a prodromal sign of neurodegeneration, but even without full elucidation, the findings could have value as a signal of impending neurodegeneration, said Dr. Jeon of Seoul (South Korea) National University.

“Our findings suggest that delayed acrophase may be used as a predictor for the progression of Alzheimer’s-type neurodegeneration and cognitive decline in the near future in old individuals with diverse cognitive status,” she said. “But the relationship between circadian phases and neurodegeneration is complex and not yet well understood.”

The 24-month study comprised 215 elderly adults enrolled in the Korean Brain Aging Study for the Early Diagnosis and Prediction of Alzheimer’s Disease (KBASE). They were a mean of 70 years old at baseline; 143 were cognitively normal, 40 had mild cognitive impairment, and 32 had Alzheimer’s dementia. Both at baseline and 2 years, everyone underwent a comprehensive neuropsychological assessment, amyloid PET brain imaging with Pittsburgh compound B, and an [18F]-fluorodeoxyglucose PET scan to determine brain glucose metabolic rate.

Before each assessment, the investigators measured sleep and circadian rhythms with 8 days of actigraphy. This assessed sleep variables (total sleep time, sleep latency, sleep efficiency, and wakefulness after sleep); rest-activity rhythm variables (midline estimated statistic of rhythm, amplitude, and acrophase), and some nonparametric values including interdaily stability, intradaily variability, and relative amplitude of sleep cycles. Subjects also completed sleep diaries during these periods.

The study’s main outcomes were 2-year changes in the Mini Mental State Exam (MMSE) score and in Alzheimer’s imaging biomarkers, including glucose metabolism and amyloid deposition. All analyses controlled for age, sex, Clinical Dementia Rating score, apolipoprotein E allele status, and baseline cognition.

At baseline, lower total sleep time was significantly associated with hypometabolism in areas associated with Alzheimer’s pathology as well as lower mean MMSE scores. Circadian variables showed no significant associations with these characteristics. However, the relative amplitude of circadian rhythm was significantly associated with hypometabolism and with lower MMSE score. There were no associations with brain amyloid load.

At 2 years, acrophase was associated with declines in cerebral glucose metabolism and further changes in the MMSE, even after the researchers controlled for the potential confounders. Delayed acrophase, although not associated with either metabolic rate or cognition at baseline, did significantly influence both at 2 years, suggesting a rapidly eroding clinical picture.

“Neurodegeneration over 2 years means the disease is progressing rapidly and subjects are likely to have tauopathies or other proteinopathy,” Dr. Jeon said. “These pathologies may either be resulting in delayed acrophase followed by neurodegeneration, or they may be prodromal symptoms of impending neurodegeneration. Whether they are early symptoms or early risk factors is not currently known, however. Two years is too short of a follow-up to determine these questions.”

Dr. Jeon had no financial declarations.

[email protected]

SOURCE: Jeon SY et al. AAIC 2019, abstract 33543.