Journal Club
Heneka MT, Fink A, Doblhammer G
Effect of pioglitazone medication on the incidence of dementia.
Ann Neurol. 2015 May 14. doi: 10.1002/ana.24439.
Marklund M, Leander K, Vikström M, Laguzzi F, Gigante B, Sjögren P,
Cederholm T, de Faire U, Hellénius ML, Risérus U.
Polyunsaturated Fat Intake Estimated by Circulating Biomarkers and Risk of
Cardiovascular Disease and All-Cause Mortality in a Population-Based Cohort
of 60-Year-Old Men and Women.
Circulation. 2015 Jun 17. pii: CIRCULATIONAHA.115.015607.
2015年7月9日 8:30-8:55
８階 医局
埼玉医科大学 総合医療センター 内分泌・糖尿病内科
Department of Endocrinology and Diabetes,
Saitama Medical Center, Saitama Medical University
松田 昌文
Matsuda, Masafumi
http://www.beasleyallen.com/news/diabetes-drug-actos-pulled-from-french-german-market-due-to-cancer-link/
A large German study is the latest clinical trial to suggest that a cheap generic treatment for diabetes can stave off symptoms of Alzheimer's
disease, although conclusive proof from a more formal trial could be about five years away.
Earlier studies have suggested that people and animals given the widely used pill for type-2 diabetes, called pioglitazone, were less likely to
develop Alzheimer's or other forms of dementia. The medicine is sold under the brand name Actos by Japanese drugmaker Takeda
Pharmaceutical Co Ltd.
Researchers in the new trial used routine data from German healthcare plans for the years 2004 until 2010. They tracked a database of about
146,000 patients age 60 and older who initially did not have evidence of dementia.
The analysis showed that 13,841 subjects eventually developed dementia, and that for those taking pioglitazone the risk of dementia was
significantly reduced with each additional three months the drug was prescribed.
"The long-term use of pioglitazone reduces the risk of dementia incidence," based on examination of health claims data, concluded Anne Fink, a
researcher for the German Center for Neurodegenerative Diseases who helped lead the trial. Her data was presented Monday at the annual
meeting of the Alzheimer's Association International Conference in Copenhagen.
Fink speculated that pioglitazone helped prevent Alzheimer's by reducing inflammation in the brain and nervous system, although other effects of
the drug might also be at play.
Separate earlier studies of patients with type 2 diabetes have found that those with poor blood sugar control are much more likely to develop
dementia. Moreover, those taking medicines like Actos - called thiazolidinediones (TZDs) - have been shown to be at almost 20 percent less risk
of Alzheimer's than those who took insulin.
Takeda last year began a five-year study, in collaboration with privately held Zinfandel Pharmaceuticals Inc, to assess whether low
doses of pioglitazone can delay the onset of mild cognitive impairment due to Alzheimer's disease. Using a special diagnostic test, the
trial will enroll cognitively normal people who have genetic variations known to increase the risk of early onset of Alzheimer's symptoms.
Stephen Brannan, in charge of central nervous system drug development at Takeda, speculated that pioglitazone may help arrest Alzheimer's by
improving the function of mitochondria: energy-producing compartments in every cell of the body except red blood cells.
"The brain requires a lot of energy," he said in an interview, adding that more efficient mitochondria could improve brain function and thereby help
stave off Alzheimer's.
Some 18 million Americans suffer from Alzheimer's, according to Takeda, with the rate of occurrence doubling every five years for those between
65 and 85 years of age.
http://www.reuters.com/article/2014/07/14/us-alzheimers-prevention-actos-idUSKBN0FJ0VJ20140714
•Article first published online: 30 June 2015
•Accepted manuscript online: 14 May 2015
•Manuscript Revised: 11 May 2015
•Manuscript Accepted: 11 May 2015
•Manuscript Received: 28 January 2015
Funded by
•Deutsche Forschungsgemeinschaft. Grant Numbers: KFO177, TP8
Ann Neurol. 2015 May 14. doi: 10.1002/ana.24439.
Objective
Peroxisome proliferator activated receptor γ–
activating drugs show various salutary effects in
preclinical models of neurodegenerative disease.
The decade-long clinical usage of these drugs as
antidiabetics now allows for evaluation of patientoriented data sources.
Methods
Using observational data from 2004–2010, we
analyzed the association of pioglitazone and incidence
of dementia in a prospective cohort study of 145,928
subjects aged ≥60 years who, at baseline, were free of
dementia and insulin-dependent diabetes mellitus. We
distinguished between nondiabetics, diabetics without
pioglitazone, diabetics with prescriptions of <8
calendar quarters of pioglitazone, and diabetics with
≥8 quarters. Cox proportional hazard models explored
the relative risk (RR) of dementia incidence dependent
on pioglitazone use adjusted for sex, age, use of
rosiglitazone or metformin, and cardiovascular
comorbidities.
Sample and study design
Analyses were based on a longitudinal 2.18 % -sample of the largest
German mandatory public health insurance, Allgemeine
Ortskrankenkassen (AOK), from the year 2004 to 2010. The sample
included 250,000 persons born in or prior to 1954 with at least one day
of insurance in the first quarter of 2004.
Dementia is defined as having been given one of the ICD-10 codes
G30, G31.0, G31.82, G23.1, F00, F01, F02, F03, and F05.1.
Validation of diagnoses
Since routine data of public sickness funds are created for the purpose of cost calculation and reimbursement and
are subject to legal changes and to changes in the data-handling procedures of the health insurers a two-stage
validation procedure was applied in order to internally validate the diagnosis of dementia. For more details see11.
This procedure excludes false positive diagnoses of dementia which otherwise would lead to an overestimation of
the true dementia incidence10.
First, diagnoses from the outpatient sector were taken into account only if the physician had indicated them as
verified. Diagnoses from the inpatient sector had to be either discharge or secondary diagnoses.
Second, dementia diagnoses had to be confirmed by co-occurrence. Diagnoses were considered valid if they
occurred simultaneously in the inpatient and outpatient sectors, or if at least two physicians made a diagnosis of
dementia in the same quarter. Furthermore, dementia diagnoses were considered valid by a co-occurrence over
time, with all five years of study being used as the validation period. If the patient died within the quarter with the
first dementia diagnosis, the case was considered valid even though the initial diagnosis could not be confirmed
by a second diagnosis.
Table 1: Characteristic s of the study
population and dementia incidence rate
per 1000 person-years, 95 % confidence
interval
Source: AOK Observational Data 2004-2010
Table 2: Relative risks of dementia
RR: Rate ratios CI: Confidence intervals PIO: pioglitazone
Source: AOK Observational Data 2004-2010
Figure 2: Dementia incidence rates from AOK and previous studies.
Table 3: Relative risks of dementia, rosiglitazone nonusers split up into non-diabetics and
diabetics without rosiglitazone
This lower rate of rosiglitazone
prescription follows several
studies, which revealed that
rosiglitazone therapy is
associated with an increased risk
of myocardial infarction24 and a
subsequent black box warning by
the FDA in 200725. Prescription
of rosiglitazone has been halted
in Germany since 2010 for this
reason.
Table 4: Relative risks of dementia, metformin nonusers split up into non-diabetics and diabetics
without metformin
Figure 1: Dementia incidence rate by number of quarters with pioglitazone (A) and extended
Kaplan-Meier estimators of time to the first dementia diagnosis dependent on the use of
pioglitazone (B),rosiglitazone (C) and metformin (D).
Source: AOK Observational Data 2004-2010
Results
Long-term use of pioglitazone was associated
with a lower dementia incidence. Relative to
nondiabetics, the cumulative long-term use of
pioglitazone reduced the dementia risk by 47%
(RR = 0.53, p = 0.029). If diabetes patients used
pioglitazone <8 quarters, the dementia risk was
comparable to those of nondiabetics (RR = 1.16,
p = 0.317), and diabetes patients without a
pioglitazone treatment had a 23% increase in
dementia risk (RR = 1.23, p < 0.001). We did not
find evidence for age effects, nor for selection into
pioglitazone treatment due to obesity.
Interpretation
These findings indicate that pioglitazone
treatment is associated with a reduced dementia
risk in initially non–insulin-dependent diabetes
mellitus patients. Prospective clinical trials are
needed to evaluate a possible neuroprotective
effect in these patients in an ageing population.
Message
Pioglitazoneが認知症によさそうな話はあった
が、ヒトのデータで示された！しかもACOTSご禁
制のドイツから！
前向き研究でないので、解釈は難しいが、高齢
者にはmetforminよりもpioglitazoneを使用する
とよいかもしれない。ただし２年以上使用がよ
いらしい。
そのうちTakedaからZinfandel Pharmaceuticals
Incとの2013から５年計画の研究データが出るの
であろう。認知症予防で糖尿病でない人も服用
するようになるのであろうか？
油脂の加工・精製でできるもの
魚介類供給量と平均寿命の関係
（歳）
イタリア
84
豪州
スイス
82
日本
ニュージーランド
スウェーデン
フランス
ドイツ
カナダ
80
ギリシア
英国
78
平
均 76
寿
命
スペイン
アイルランド
オーストリア
オランダ
チェコ
アイスランド
ノルウェー
フィンランド
ポルトガル
韓国
デンマーク
米国
メキシコ
ポーランド
74
スロバキア
トルコ
72
70
0
10
20
30
40
50
60
70
80
90 （kg/人/年）
FAO「Food balance sheets」（日本以外の国）、農林水産省「食料需給表」、WHO「Statistical Information System
（WHOSIS）」に基づき水産庁で作成。
平成22年水産白書
In our study, the average daily
consumption of ω-3
PUFAs(polyunsaturated fatty acid)
from seal oil was ~8 g. We based
this estimate on a 30% ω-3 PUFA
content of seal oil.
Compared with less-than-daily consumption, both daily seal oil (odds ratio [OR] 0.2, 95%
confidence interval [CI] 0.1-0.8) and daily salmon consumption (OR 0.5, CI 0.2-1.1) were
associated with a lower prevalence of glucose intolerance, controlling for age, ethnicity, body
mass index, and sex. The effects were similar when limited to newly discovered cases: OR 0.3, CI
0.1-1.3 for seal oil and OR 0.4, CI 0.1-1.3 for salmon. Consumption of seal oil at least five times
per week was required to reduce risk.
Adler AI, Boyko EJ, Schraer CD, Murphy NJ.: Lower prevalence of impaired glucose tolerance and diabetes associated with
daily seal oil or salmon consumption among Alaska Natives. Diabetes Care. 1994 Dec;17(12):1498-501.
eicosapentaenoic acid (EPA).
1800mg/day
Kromhout D, Giltay EJ, Geleijnse JM; Alpha Omega Trial Group.: n-3 fatty acids and cardiovascular events after
myocardial infarction. N Engl J Med. 2010 Nov 18;363(21):2015-26.
Low-dose supplementation with EPA–DHA
or ALA did not significantly reduce the rate
of major cardiovascular events
Figure 2. Kaplan–Meier Curves for Primary and Secondary End Points. Kaplan–Meier curves are shown for the cumulative
incidence of major cardiovascular events (the primary end point) and fatal coronary heart disease (a secondary end point) among
4837 patients who had had a myocardial infarction and were assigned to receive a study margarine containing supplemental
eicosapentaenoic acid (EPA) combined with docosahexaenoic acid (DHA), a margarine containing alpha-linolenic acid (ALA), a
margarine containing both EPA–DHA and ALA, or a placebo margarine.
ORIGIN Trial Investigators, Bosch J, Gerstein HC, Dagenais GR, Díaz R, Dyal L, Jung H, Maggiono AP, Probstfield J, Ramachandran A, Riddle MC, Rydén LE, Yusuf S.: n-3 fatty acids and cardiovascular outcomes in patients with dysglycemia. N Engl J Med. 2012 Jul 26;367(4):309-18.
n−3 fatty acids (1 g daily) vs placebo (olive oil).
N Engl J Med 2013;368:1800-8.
n−3 fatty acids (1 g daily) vs placebo (olive oil).
N Engl J Med 2013;368:1800-8.
Background— High intake of polyunsaturated
fatty acids (PUFA) may reduce the risk of
cardiovascular disease (CVD) and mortality.
Large prospective studies including both sexes
and circulating PUFA as dietary biomarkers are
needed. We investigated sex-specific
associations of major dietary PUFA;
eicosapentaenoic acid (EPA), docohexaenoic
acid (DHA), linoleic acid (LA), and α-linolenic acid
(ALA), with incident CVD and all-cause mortality
in a population based cohort.
Methods— PUFA in serum cholesterol esters
were measured at baseline in 2193 Swedish 60year old women and 2039 men. Using national
registers, 484 incident CVD events (294 men and
190 women) and 456 all-cause deaths (265 men
and 191 women) were identified during follow-up
(median 14.5y), in individuals without prior CVD
at baseline. Associations of PUFA with CVD and
mortality were evaluated using Cox proportional
hazard models.
EPA
Figure 1. Multivariate-adjusted associations of serum eicosapentaenoic acid [A, B],
docosahexaenoic acid [C, D], and α-linolenic acid [E, F] with incident CVD among 1938
women [A, C, E] and 1733 men [B, D, F].
Hazard ratios (circles) and 95% CI (error bars) of quartiles of serum fatty acid
concentrations (% of total fatty acids) were estimated using Cox models adjusted for
BMI, smoking, physical activity, education, alcohol intake, diabetes, drug-treated
hypertension, and drug-treated hypercholesterolemia, and plotted against the median
fatty acid concentration of each quartile, with the lowest quartile as reference. Linear
trends across quartiles were evaluated by assigning participants the median FA
concentration of each quartile and assessing this as a continuous variable. Non-linearity
was evaluated using restricted cubic splines with five knots located at percentiles 5, 27.5,
50, 72.5, and 95.
DHA
Women
Men
Figure 1. Multivariate-adjusted associations of serum eicosapentaenoic acid [A, B],
docosahexaenoic acid [C, D], and α-linolenic acid [E, F] with incident CVD among 1938
women [A, C, E] and 1733 men [B, D, F].
Hazard ratios (circles) and 95% CI (error bars) of quartiles of serum fatty acid
concentrations (% of total fatty acids) were estimated using Cox models adjusted for
BMI, smoking, physical activity, education, alcohol intake, diabetes, drug-treated
hypertension, and drug-treated hypercholesterolemia, and plotted against the median
fatty acid concentration of each quartile, with the lowest quartile as reference. Linear
trends across quartiles were evaluated by assigning participants the median FA
concentration of each quartile and assessing this as a continuous variable. Non-linearity
was evaluated using restricted cubic splines with five knots located at percentiles 5, 27.5,
50, 72.5, and 95.
ALA
Women
Men
Figure 1. Multivariate-adjusted associations of serum eicosapentaenoic acid [A, B],
docosahexaenoic acid [C, D], and α-linolenic acid [E, F] with incident CVD among 1938
women [A, C, E] and 1733 men [B, D, F].
Hazard ratios (circles) and 95% CI (error bars) of quartiles of serum fatty acid
concentrations (% of total fatty acids) were estimated using Cox models adjusted for
BMI, smoking, physical activity, education, alcohol intake, diabetes, drug-treated
hypertension, and drug-treated hypercholesterolemia, and plotted against the median
fatty acid concentration of each quartile, with the lowest quartile as reference. Linear
trends across quartiles were evaluated by assigning participants the median FA
concentration of each quartile and assessing this as a continuous variable. Non-linearity
was evaluated using restricted cubic splines with five knots located at percentiles 5, 27.5,
50, 72.5, and 95.
EPA
Figure 2. Multivariate-adjusted associations of serum eicosapentaenoic acid [A, B],
docosahexaenoic acid [C, D], and linoleic acid [E, F] with all-cause mortality among
1938 women [A, C, E] and 1733 men [B, D, F].
Hazard ratios (circles) and 95% CI (error bars) of quartiles of serum fatty acid
concentrations (% of total fatty acids) were estimated using Cox models adjusted for
BMI, smoking, physical activity, education, alcohol intake, diabetes, drugtreated
hypertension, and drug-treated hypercholesterolemia, and plotted against the median
fatty acid concentration of each quartile, with the lowest quartile as reference. Linear
trends across quartiles were evaluated by assigning participants the median FA
concentration of each quartile and assessing this as a continuous variable. Non-linearity
was evaluated using restricted cubic splines with five knots located at percentiles 5, 27.5,
50, 72.5, and 95.
DHA
Women
Men
Figure 2. Multivariate-adjusted associations of serum eicosapentaenoic acid [A, B],
docosahexaenoic acid [C, D], and linoleic acid [E, F] with all-cause mortality among
1938 women [A, C, E] and 1733 men [B, D, F].
Hazard ratios (circles) and 95% CI (error bars) of quartiles of serum fatty acid
concentrations (% of total fatty acids) were estimated using Cox models adjusted for
BMI, smoking, physical activity, education, alcohol intake, diabetes, drugtreated
hypertension, and drug-treated hypercholesterolemia, and plotted against the median
fatty acid concentration of each quartile, with the lowest quartile as reference. Linear
trends across quartiles were evaluated by assigning participants the median FA
concentration of each quartile and assessing this as a continuous variable. Non-linearity
was evaluated using restricted cubic splines with five knots located at percentiles 5, 27.5,
50, 72.5, and 95.
ALA
Women
Men
Figure 2. Multivariate-adjusted associations of serum eicosapentaenoic acid [A, B],
docosahexaenoic acid [C, D], and linoleic acid [E, F] with all-cause mortality among
1938 women [A, C, E] and 1733 men [B, D, F].
Hazard ratios (circles) and 95% CI (error bars) of quartiles of serum fatty acid
concentrations (% of total fatty acids) were estimated using Cox models adjusted for
BMI, smoking, physical activity, education, alcohol intake, diabetes, drugtreated
hypertension, and drug-treated hypercholesterolemia, and plotted against the median
fatty acid concentration of each quartile, with the lowest quartile as reference. Linear
trends across quartiles were evaluated by assigning participants the median FA
concentration of each quartile and assessing this as a continuous variable. Non-linearity
was evaluated using restricted cubic splines with five knots located at percentiles 5, 27.5,
50, 72.5, and 95.
Results— In multivariable adjusted models, 1-SD
increments of EPA and DHA were associated with
lower risk of incident CVD among women (hazard
ratios 0.79 [95% CI 0.64-0.97] and 0.74 [0.610.89], respectively). ALA was associated with
moderately increased CVD risk in women (1.16
[1.02- 1.32]). Inverse associations with all-cause
mortality was observed for EPA and DHA among
all participants (0.81 [0.72-0.91] and 0.80 [0.720.89], respectively) and for LA in men (0.73 [0.640.83]).
Conclusions— Serum LA and very
long-chain n-3 PUFA, partly reflecting
vegetable oil and fish intake,
respectively, were inversely associated
with all-cause mortality. Inverse
associations of EPA and DHA with
incident CVD were only observed in
women.
Message
EPAやDHAはよさそうなのだが．．．