Journal Club
Barnett AH, Huisman H, Jones R, von Eynatten M, Patel S, Woerle HJ.
Linagliptin for patients aged 70 years or older with type 2 diabetes inadequately
controlled with common antidiabetes treatments: a randomised, double-blind, placebocontrolled trial.
Lancet. 2013 Aug 12. doi:pii: S0140-6736(13)61500-7. 10.1016/S0140-6736(13)61500-7.
Cefalu WT, Leiter LA, Yoon KH, Arias P, Niskanen L, Xie J, Balis DA, Canovatchel W,
Meininger G.
Efficacy and safety of canagliflozin versus glimepiride in patients with type 2 diabetes
inadequately controlled with metformin (CANTATA-SU): 52 week results from a
randomised, double-blind, phase 3 non-inferiority trial.
Lancet. 2013 Jul 11. pii: S0140-6736(13)60683-2. doi: 10.1016/S0140-6736(13)60683-2.
2013年9月12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
吉永 玲恵 松田 昌文
Yoshinaga, Ree, Matsuda, Masafumi
Diabetes Centre, Heart of England NHS Foundation Trust, Birmingham, UK (Prof A H
Barnett MD); University of Birmingham, Birmingham, UK (Prof A H Barnett); Boehringer
Ingelheim, Alkmaar, Netherlands (H Huisman MSc); Boehringer Ingelheim, Bracknell,
UK (R Jones MSc, S Patel MB ChB); and Boehringer Ingelheim, Ingelheim, Germany
(M von Eynatten MD, H-J Woerle MD)
Lancet. 2013 Aug 12. doi:pii: S0140-6736(13)61500-7. 10.1016/S0140-6736(13)61500-
Background
A substantial proportion of patients with type
2 diabetes are elderly (≥65 years) but this
group has been largely excluded from
clinical studies of glucose-lowering drugs.
We aimed to assess the effectiveness of
linagliptin, a dipeptidyl peptidase-4 inhibitor,
in elderly patients with type 2 diabetes.
Methods
In this randomised, double-blind, parallel-group,
multinational phase 3 study, patients aged 70 years or
older with type 2 diabetes, glycated haemoglobin A1c
(HbA1c) of 7・0% or more, receiving metformin,
sulfonylureas, or basal insulin, or combinations of these
drugs, were randomised (by computer-generated
randomisation sequence, concealed with a voice–
response system, stratified by HbA1c level [<8・5% vs
≥8・5%] and insulin use [yes vs no], block size four) in a
2:1 ratio to once-daily oral treatment with linagliptin 5 mg
or matching placebo for 24 weeks. Investigators and
participants were masked to assignment throughout the
study. The primary endpoint was change in HbA1c from
baseline to week 24. This trial is registered with
ClinicalTrials.gov, number NCT01084005.
Figure 1: Trial profile
Table 1: Baseline demographics and
clinical characteristics in the treated set of
patients
FPG=fasting plasma
glucose.
GFR=glomerular
filtration rate.
HbA1c=glycated
haemoglobin A1c.
Figure 2: Baseline disease-related characteristics and comorbidities in the treated set of patients Treated set
consisted of all patients who received at least one dose of study drug. *Renal function according to GFR estimated
by the Modifi cation of Diet in Renal Disease equation. †Microvascular disease consisted of diabetic retinopathy,
nephropathy, and neuropathy. ‡Macrovascular disease consisted of coronary art ery disease, peripheral artery
disease, cerebrovascular disease, and hypertension. §Sulfonylureas, or meglitinides, or insulin, or combination
thereof. ¶Drug in addition to glucose-lowering drugs. IIGlucose-lowering drugs and other drugs.
Figure 2: Charlson age-comorbidity score in the treated set of patients Treated set consisted of all patients who
received at least one dose of study drug.
Charlson M, Szatrowski TP, Peterso n J, Gold J. Validation of a
combined comorbidity index. J Clin Epidemiol 1994; 47: 1245–51.
The mean Charlson
age-comorbidity score
of more than 5
suggests a highrisk
population.
The Charlson comorbidity index predicts the ten-year mortality for a patient who may have a range of comorbid conditions, such as
heart disease, AIDS, or cancer (a total of 22 conditions). Each condition is assigned a score of 1, 2, 3, or 6, depending on the risk of
dying associated with each one. Scores are summed to provide a total score to predict mortality. Many variations of the Charlson
comorbidity index have been presented, including the Charlson/Deyo, Charlson/Romano, Charlson/Manitoba, and
Charlson/D'Hoores comorbidity indices.
Clinical conditions and associated scores are as follows:
1 each: Myocardial infarct, congestive heart failure, peripheral vascular disease, dementia, cerebrovascular disease, chronic lung
disease, connective tissue disease, ulcer, chronic liver disease, diabetes.
2 each: Hemiplegia, moderate or severe kidney disease, diabetes with end organ damage, tumor, leukemia, lymphoma.
3 each: Moderate or severe liver disease.
6 each: Malignant tumor, metastasis, AIDS.
For a physician, this score is helpful in deciding how aggressively to treat a condition. For example, a patient may have cancer with
comorbid heart disease and diabetes. These comorbidities may be so severe that the costs and risks of cancer treatment would
outweigh its short-term benefit.
Since patients often do not know how severe their conditions are, nurses were originally supposed to review a patient's chart and
determine whether a particular condition was present in order to calculate the index. Subsequent studies have adapted the
comorbidity index into a questionnaire for patients.
Table 2: Change from baseline in
HbA1c and FPG after 24 weeks in the
full analysis set of patients, last
observation carried forward
Figure 3: Change from baseline in HbA1c over 24 weeks in the full analysis set
Adjusted mean changes in HbA1c. Error bars are SE. Full analysis set consisted of all
randomised patients who received at least one dose of study drug, and who had a
baseline and at least one on-treatment HbA1c measurement. Data are from a mixed
model for repeated measurements, using observed cases with treatment, visit, previous
use of insulin, and visit by treatment interaction as fixed classification effects, and
baseline HbA1c as a linear covariate. HbA1c=glycated haemoglobin A1c.
Table 4: Summary of
adverse events during
24 weeks in the treated
set of patients not
receiving concomitant
sulfonylurea treatment
Data are n (%).
Treated set consisted of all
patients who received at
least one dose of study
drug.
CEC=clinical endpoint
committee.
*Deemed to be related to
study drug by the
investigator.
†Hypersensitivity reaction,
renal adverse event, hepatic
adverse event.
‡Preferred terms from the
Medical Dictionary for
Regulatory Activities
Findings
241 community-living outpatients were randomised (162
linagliptin, 79 placebo). Mean age was 74・9 years (SD 4・3).
Mean HbA1c was 7・8% (SD 0・8). At week 24, placeboadjusted mean change in HbA1c with linagliptin was −0・64%
(95% CI −0・81 to −0・48, p<0・0001). Overall safety and
tolerability were much the same between the linagliptin and
placebo groups; 75・9% of patients in both groups had an
adverse event (linagliptin n=123, placebo n=60). No deaths
occurred. Serious adverse events occurred in 8・6% (14) of
patients in the linagliptin group and 6・3% (fi ve) patients in the
placebo group; none were deemed related to study drug.
Hypoglycaemia was the most common adverse event in both
groups, but did not differ between groups (24・1% [39] in the
linagliptin group, 16・5% [13] in the placebo group; odds ratio 1・
58, 95% CI 0・78–3・78, p=0・2083).
Interpretation
In elderly patients with type 2 diabetes
linagliptin was efficacious in lowering
glucose with a safety profile similar to
placebo. These findings could inform
treatment decisions for achieving
individualised glycaemic goals with minimal
risk in this important population of patients.
Funding Boehringer Ingelheim.
Message
メトホルミンやスルホニル尿素薬などで血糖コ
ントロール不良の70歳以上の2型糖尿病（DM）患
者241人を対象に、リナグリプチンの有効性を無
作為化プラセボ対照試験で検討。リナグリプチ
ン群におけるHbA1c値のベースラインからのプラ
セボ調整後平均変化量は－0.64％だった（P＜
0.0001）。安全性と忍容性は両群で同等だっ
た。
Diamant M, Morsink LM.: SGLT2 inhibitors for diabetes: turning symptoms into therapy.
Lancet. 2013 Jul 11. doi:pii: S0140-6736(13)60902-2. 10.1016/S0140-6736(13)60902-2.
[Epub ahead of print]
Pennington Biomedical Research Center, Baton Rouge, LA, USA (Prof W T Cefalu MD); Louisiana State
University Health Sciences Centre School of Medicine, New Orleans, LA, USA (Prof W T Cefalu); Keenan
Research Centre, Li Ka Shing Knowledge Institute, St Michael’s Hospital, University of Toronto, Toronto, ON,
Canada (Prof L A Leiter MD); The Catholic University of Korea, Seoul St Mary’s Hospital, Seoul, South Korea
(Prof K-H Yoon MD); University of Rosario Medical School, Rosario, Argentina (Prof P Arias MD); Litoral
University Medical School, Santa Fe, Argentina (Prof P Arias); University of Eastern Finland, Kuopio, Finland (Prof
L Niskanen MD); and Janssen Research & Development, LLC, Raritan, NJ, USA (J Xie PhD, D A Balis PharmD,
W Canovatchel MD, G Meininger MD)
Lancet. 2013 Jul 11. pii: S0140-6736(13)60683-2. doi: 10.1016/S0140-6736(13)60683-2.
Background
Sodium–glucose cotransporter 2
(SGLT2) inhibitors improve glycaemia
in patients with type 2 diabetes by
enhancing urinary glucose excretion.
We compared the efficacy and safety of
canagliflozin, an SGLT2 inhibitor, with
glimepiride in patients with type 2
diabetes inadequately controlled with
metformin.
Methods
We undertook this 52 week, randomised, double-blind, activecontrolled, phase 3 non-inferiority trial at 157 centres in 19 countries
between Aug 28, 2009, and Dec 21, 2011. Patients aged 18–80
years with type 2 diabetes and glycated haemoglobin A1c (HbA1c) of
7・0–9・5% on stable metformin were randomly assigned (1:1:1) by
computergenerated random sequence via an interactive voice or web
response system to receive canagliflozin 100 mg or 300 mg, or
glimepiride (up-titrated to 6 mg or 8 mg per day) orally once daily.
Patients, study investigators, and local sponsor personnel were
masked to treatment. The primary endpoint was change in HbA1c
from baseline to week 52, with a non-inferiority margin of 0・3% for
the comparison of each canagliflozin dose with glimepiride. If
noninferiority was shown, we assessed superiority on the basis of an
upper bound of the 95% CI for the difference of each canagliflozin
dose versus glimepiride of less than 0・0%. Analysis was done in a
modified intention-to-treat population, including all randomised
patients who received at least one dose of study drug.
This study is registered with ClinicalTrials.gov, number NCT00968812.
Figure 1: Trial profile
eGFR=estimated glomerular filtration rate. *484 patients randomly assigned.
Figure 2: Change in HbA1c
(A), and mean HbA1c over
time (B)
Last observation carried
forward analyses.
Mean baseline HbA1c of 7・
8% for each treatment
group.
LS=least squares.
HbA1c=glycated haemoglobin
A1c.
Table 4:
Changes from
baseline in
blood pressure,
pulse rate,
fasting plasma
lipids, and
fasting insulin
at week 52
Conversion factor
Cholesterol 0.0259
Triglyceride 0.0113
168(106)
104(35)
186(133)
100 (35)
186 (186) mg/dl
108(35) mg/dl
Increased!
46(12)
46 (12)
46(12) mg/dl
not different!
Last observation
carried forward
analyses.
Statistical
comparison for
canagliflozin 100 mg
and 300 mg versus
glimepiride not
undertaken (not
prespecified).
LS=least squares.
135(39)
135 (39)
143(42) mg/dl
Conversion factor
Insulin pmol/L to
microunits/mL 7.18
9.9(10.0)
9.6 (7.0)
9.0(6.1) mU/ml
Table 5: Overall safety and selected adverse events
Table 6: Summary of laboratory parameters at baseline and week 52
We recorded no notable
differences in serum
electrolytes, sodium and
potassium, with canagliflozin
compared with glimepiride
(data not shown).
Statistical comparison for
canagliflozin 100 mg and 300
mg versus glimepiride not
undertaken (not prespecified).
ALT=alanine aminotransferase.
AST=aspartate aminotransferase.
BUN=blood urea nitrogen.
GGT=gamma-
Findings
1450 of 1452 randomised patients received at least one dose of
glimepiride (n=482), canagliflozin 100 mg (n=483), or canagliflozin
300 mg (n=485). For lowering of HbA1c at 52 weeks, canagliflozin
100 mg was non-inferior to glimepiride (least-squares mean
difference –0・01% [95% CI –0・11 to 0・09]), and canaglifl ozin
300 mg was superior to glimepiride (–0・12% [–0・22 to –0・02]).
39 (8%) patients had serious adverse events in the glimepiride
group versus 24 (5%) in the canagliflozin 100 mg group and 26
(5%) in the 300 mg group. In the canagliflozin 100 mg and 300 mg
groups versus the glimepiride group, we recorded a greater
number of genital mycotic infections (women: 26 [11%] and 34
[14%] vs five [2%]; men: 17 [7%] and 20 [8%] vs three [1%]),
urinary tract infections (31 [6%] for both canagliflozin doses vs 22
[5%]), and osmotic diuresis-related events (pollakiuria: 12 [3%] for
both doses vs one [<1%]; polyuria: four [<1%] for both doses vs
two [<1%]).
Interpretation
Canaglifl ozin provides greater HbA1c
reduction than does glimepiride, and is well
tolerated in patients with type 2 diabetes
receiving metformin. These fi ndings
support the use of canaglifl ozin as a viable
treatment option for patients who do not
achieve suffi cient glycaemic control with
metformin therapy.
Funding Janssen Research &
Development, LLC.
Message
CANTATA-SU試験は、メトホルミン単独では血糖値の管理が十分で
ない2型糖尿病患者において、カナグリフロジンの上乗せ効果
の、グリメピリド（商品名：アマリールほか）に対する非劣性を
検証する二重盲検無作為化第III相試験。カナグリフロジン
100mgまたは300mg、あるいはグリメピリド（6から8mgへ漸増）を
1日1回経口投与する群に無作為化に割り付けられた。グリメピリ
ド群に482例、カナグリフロジン 100mg群に483例、300mg群には
485例が割り付けられた。治療52週時のHbA1cは3群ともにベース
ラインよりも低下し、最小二乗平均値の変化率はグリメピリド群
が－0.81、カナグリフロジン 100mg群は－0.82％、300mg群は－
0.93％であった。体重は、グリメピリド群がわずかに増加したの
に対し、2つのカナグリフロジン群は有意に低下した。尿路感染
症も100mg群が31例（6％）、300mg群が31例（6％）で、グリメピ
リド群の22例（5％）より多い傾向がみられた。
糖質食べても尿から出すからOK！というアメリカで大人気のお薬
SGLT2阻害薬追加、血糖管理不良な2型糖尿病に有用
ナトリウム/グルコース共輸送体（SGLT）2阻害薬カナグリフロジン（canagliflozin）は、メトホルミン単独では血糖管理が十分でない2型糖尿
病患者に対する追加治療として良好な血糖改善効果を発揮し、忍容性も良好なことが、米国・ペニントン生物医学研究所のWilliam T Cefalu
氏らが行ったCANTATA-SU試験で示された。メトホルミンへの上乗せが可能な既存薬の多くは体重増加や低血糖の懸念があるが、SGLT2
の阻害という作用機序はこれらの問題を回避し、また尿糖排泄の促進作用により総カロリーの消費が進むため体重が減少する可能性もある
という。一方、本薬剤には軽度の浸透圧利尿がみられるため、頻尿や多尿の懸念があるという。
標準治療に対する非劣性を無作為化試験で検証
CANTATA-SU試験は、メトホルミン単独では血糖値の管理が十分でない2型糖尿病患者において、カナグリフロジンの上乗せ効果の、グリ
メピリド（商品名：アマリールほか）に対する非劣性を検証する二重盲検無作為化第III相試験。対象は、年齢18～80歳、HbA1c 7.0～9.5％、
10週以上のメトホルミン投与を受けている2型糖尿病患者であった。
これらの患者が、カナグリフロジン 100mgまたは300mg、あるいはグリメピリド（6から8mgへ漸増）を1日1回経口投与する群に無作為化に
割り付けられた。主要評価項目はベースラインから治療52週までのHbA1cの変化で、非劣性マージンは0.3％とした。
100mg群の非劣性、300mg群の優位性を確認
2009年8月28日～2011年12月21日までに19ヵ国157施設から1,450例が登録され、グリメピリド群に482例、カナグリフロジン 100mg群に
483例、300mg群には485例が割り付けられた。
全体の平均年齢は56.2歳（9.2 SD）、男性52％、白人67％、アジア人20％、平均HbA1c 7.8％、平均空腹時血糖9.2mmol/L
（≒165.6mg/dL）、平均体重86.6kg、平均BMI 31.0、平均罹病期間6.6年（中央値5.0年）であった。
治療52週時のHbA1cは3群ともにベースラインよりも低下し、最小二乗平均値の変化率はグリメピリド群が－0.81、カナグリフロジン 100mg
群は－0.82％、300mg群は－0.93％であった。
カナグリフロジン 100mg群とグリメピリド群の最小二乗平均値の差は－0.01％（95％信頼区間[CI]：－0.11～0.09）であり、カナグリフロジン
100 mg群はグリメピリド群に対し非劣性であった。また、300mg群とグリメピリド群の最小二乗平均値の差は－0.12％（－0.22～－0.02）であ
り、300mg群のグリメピリド群に対する優位性が示された。
体重は、グリメピリド群がわずかに増加したのに対し、2つのカナグリフロジン群は有意に低下した［最小二乗平均値の変化率：0.7％、－
3.7％、－4.0％、100mg群、300mg群とグリメピリド群の差：－4.4（－4.8～－3.9）、－4.7（－5.2～－4.3）］。
良好な安全性プロフィール、浸透圧利尿関連イベントは多い傾向
重篤な有害事象は、カナグリフロジン 100 mg群が24例（5％）、300mg群が26例（5％）、グリメピリド群は39例（8％）に認められた。
カナグリフロジン群で多い有害事象として性器真菌感染症が挙げられ、女性では100mg群が26例（11％）、300mg群が34例（14％）で、グリ
メピリド群は5例（2％）であり、男性では100mg群が17例（7％）、300mg群が20例（8％）で、グリメピリド群は3例（1％）だった。
尿路感染症も100mg群が31例（6％）、300mg群が31例（6％）で、グリメピリド群の22例（5％）より多い傾向がみられた。浸透圧利尿関連イ
ベントもカナグリフロジン群で多い傾向にあり、頻尿が100mg群、300mg群ともに12例（3％）ずつ、グリメピリド群は1例（＜1％）にみられ、多
尿はカナグリフロジン群が4例（＜1％）ずつ、グリメピリド群は2例（＜1％）に認められた。
著者は、「これらの結果は、メトホルミンで血糖管理が不十分な2型糖尿病患者において、カナグリフロジンは実行可能な治療選択肢である
ことを示すもの」と指摘している。