You are here

Linagliptin for patients aged 70 years or older with type 2 diabetes inadequately controlled with common antidiabetes treatments: a randomised, double-blind, placebo-controlled trial

Lancet. 2013;382:1413–23

Summary

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 HbA1clevel [<8·5%vs≥8·5%] and insulin use [yesvsno], 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 HbA1cfrom baseline to week 24. This trial is registered with ClinicalTrials.gov , number NCT01084005 .

Findings

241 community-living outpatients were randomised (162 linagliptin, 79 placebo). Mean age was 74·9 years (SD 4·3). Mean HbA1cwas 7·8% (SD 0·8). At week 24, placebo-adjusted mean change in HbA1cwith 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% (five) 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.

Introduction

Elderly people (≥65 years) are the main users of medications but have been substantially under-represented in clinical trials.1 and 2About a quarter of people aged 65 years or older have diabetes mellitus, 3 nearly all of whom have type 2 disease. 4 However, an analysis 5 showed that only 0·6% of interventional trials in diabetes specifically targeted this age group, 31% excluded patients older than 65 years, and almost all excluded those older than 75 years. Glucose-lowering treatment of elderly patients with type 2 diabetes is generally deemed necessary to alleviate symptoms associated with hyperglycaemia, improve general wellbeing, and, in some cases, to reduce the risk of long-term complications; 4 untreated hyperglycaemia is also frequently associated with cognitive impairment. 6 Randomised clinical trials of glucose-lowering treatments in elderly patients with type 2 diabetes are particularly rare in those whose hyperglycaemia is uncontrolled by common antidiabetes treatments. 4 This evidence gap hinders clinical decision making for elderly patients, because the risks and benefits of treatment are unclear.4, 7, and 8

Treatment of elderly patients with type 2 diabetes is challenging because of the high prevalence of comorbidities, use of polypharmacy, frailty, and age-related reduction in pancreatic islet function. 9 Safety is therefore an important consideration for treatment, especially avoidance of iatrogenic hypoglycaemia, which occurs frequently in elderly patients and can have severe consequences.4 and 7Renal impairment is also very common in elderly patients with type 2 diabetes, increasing their risk for hypoglycaemia and complicating treatment strategies. 10 Consequently, although the general goal for patients with type 2 diabetes of a glycated haemoglobin A1c(HbA1c) level of less than 7% might still be reasonable for some elderly patients, it is judged important to individualise this target to balance potential benefits and risks of treatment.4, 9, and 10For some patients, a more appropriate treatment goal might be, for example, HbA1cof 7·5% or higher, depending on individual circumstances.

Many glucose-lowering drugs have potential disadvantages for elderly patients, particularly a risk for hypoglycaemia (sulfonylureas, insulin) or fractures (thiazolidinediones, of which pioglitazone is the only agent remaining widely available), or contraindication or dose adjustments for renal impairment (most oral and injectable agents). 4 Additionally, efficacy of glucose-lowering drugs can be affected by disease duration, because of deterioration over time in pancreatic β-cell function. Dipeptidyl peptidase-4 (DPP4) inhibitors are oral agents with little risk for hypoglycaemia and are generally well tolerated. 11 Unlike other DPP4 inhibitors, linagliptin, given in a once-daily, single-dose 5 mg regimen, is excreted mainly by non-renal pathways and dose adjustment is not needed for renal impairment or any other factor. 12 Linagliptin also has minimal drug interactions. 12 In phase 3 studies in patients with type 2 diabetes aged up to 80 years, linagliptin elicited meaningful glucose-lowering effects and was well tolerated with little intrinsic risk for hypoglycaemia.13, 14, 15, 16, and 17This pharmacological profile suggests that linagliptin might be particularly useful for treatment of elderly patients with type 2 diabetes.

The aim of this study was to investigate the glucose-lowering efficacy, safety, and tolerability of linagliptin in patients aged 70 years and older with type 2 diabetes with inadequate glycaemic control on other common glucose-lowering drugs.

Methods

Study design and patients

This randomised, double-blind, placebo-controlled, parallel-group, phase 3 clinical trial was done at 33 clinics in five countries (Australia, Canada, Denmark, the Netherlands, and Sweden). The first patient was enrolled on March 10, 2010, and the last patient visit occurred on June 22, 2011.

Men and women with type 2 diabetes were eligible to participate if they were aged 70 years or older, had insufficient glycaemic control (HbA1c≥7·0%), and had been receiving stable doses of metformin, sulfonylureas, or basal insulin, or combinations of these drugs, for at least 8 weeks. The main exclusion criteria were: fasting plasma glucose (FPG) greater than 13·3 mmol/L; impaired hepatic function (serum concentrations of alanine transaminase [ALT], aspartate transaminase [AST], or alkaline phosphatase [ALP] more than three times the upper limit of normal); myocardial infarction, stroke, or transient ischaemic attack within 3 months before the study; previous bariatric surgery; present treatment with rapid-acting or premixed insulin or systemic steroids; treatment within the previous 3 months with a thiazolidinedione, α-glucosidase inhibitor, meglitinide, GLP1 analogue, DPP4 inhibitor, or anti-obesity drug.

The study protocol was approved by the independent ethics committee or institutional review board of participating centres, and the study was done in accordance with the principles of the Declaration of Helsinki and the International Conference of Harmonisation Harmonised Tripartite Guideline for Good Clinical Practice. All patients provided written informed consent before study initiation.

Randomisation and masking

Randomisation was done in a 2:1 ratio (linagliptin:placebo) for ethical reasons (so that only a third of patients received inert treatment); this allocation also provided additional safety information for the active treatment. Randomisation was stratified by HbA1clevel (<8·5%vs≥8·5%) and insulin use (yesvsno) with a block size of four. Computer-generated randomisation lists were produced by the sponsor, with allocation concealed using a central interactive voice–web response system. Linagliptin and placebo tablets were identical in appearance, and investigators and patients were masked to treatment assignment throughout the study.

Procedures

After screening, eligible patients underwent a 2 week, open-label, placebo run-in period. Patients were then randomised to receive once-daily oral treatment with linagliptin 5 mg or placebo for 24 weeks, in addition to their existing glucose-lowering treatment. Doses of background treatments were maintained for the first 12 weeks of randomised treatment, after which dose adjustments were permitted. Rescue medication for hyperglycaemia (confirmed glucose level: fasting >13·3 mmol/L in weeks 1–12, >11·1 mmol/L in weeks 13–24; or random test >22·2 mmol/L; two or more measurements on different days, one done after an overnight fast) was permitted during randomised treatment.

The primary endpoint was change in HbA1cfrom baseline to week 24. Secondary endpoints were the proportion of patients achieving HbA1cless than 7·0% after 24 weeks, the proportion of patients with a 0·5% or greater reduction in HbA1cafter 24 weeks, change in HbA1cfrom baseline over time, change in FPG from baseline at week 24, change in FPG from baseline over time, and use of rescue therapy. The proportion of patients achieving other levels of HbA1c(<7·5%, <8·0%, <8·5%) was analysed post hoc.

Safety endpoints were incidence and intensity of adverse events (including adverse changes noted during physical examinations or 12-lead electrocardiographs [ECGs]); withdrawals because of adverse events; hypoglycaemia; cardiovascular events; and changes in vital signs, laboratory variables, and background treatment. Adverse events were categorised using the Medical Dictionary for Regulatory Activities (version 14.1). Events suspected to be stroke or cardiac ischaemia (including myocardial infarction) were reviewed by an independent clinical endpoint committee (CEC), consisting of three academic cardiologists and three academic neurologists who were masked to assignment. Hypoglycaemia was defined as plasma glucose of 3·9 mmol/L or less, with or without symptoms; severe hypoglycaemia was defined as needing assistance from another person to administer resuscitative action, irrespective of plasma glucose concentration. On the basis of clinical experience with other DPP4 inhibitors, significant adverse events prespecified for investigation were hypersensitivity reactions (angio-oedema, angio-oedema-like event, anaphylaxis), renal events (kidney failure, two times or greater increase in creatinine), increases in hepatic enzymes (ALT, AST, or ALP above more than three times the upper limit of normal), severe cutaneous adverse reactions, and pancreatitis.

During assessments at randomisation and every 6 weeks thereafter, patients reported to their local study centre for blood sampling for HbA1cand fasting glucose, examination of vital signs, and documentation of any adverse events and use of concomitant therapies. Bodyweight and 12-lead ECGs were recorded at randomisation and week 24; urine and blood samples for laboratory safety assessments were taken at randomisation and weeks 12 and 24. Physical examinations were done at the start of the placebo run-in period and at week 24. Health-related quality-of-life data were collected using the short-form health survey 36 (SF-36) at randomisation, and the European quality of life-5 dimensions (EQ-5D) at randomisation and weeks 6, 12, and 24. Charlson combined age-comorbidity scores at baseline were calculated post hoc, using previously described methods. 18

Laboratory measurements were done by a central laboratory (Quintiles Laboratories: Livingston, UK; Singapore; Atlanta, GA, USA).

Statistical analysis

231 patients (linagliptin n=154, placebo n=77) were needed to achieve 90% power to detect a 0·5% difference in HbA1cchange from baseline to week 24, with a two-sided test with a significance level (α) of 0·05, assuming a common SD of 1·1% for change in HbA1cand a baseline HbA1cof 8%. Hence, randomisation of about 243 patients was needed, assuming 5% dropout.

The primary efficacy analysis, change in HbA1cfrom baseline to week 24, was done with the full analysis set (all randomised patients who received at least one dose of study medication, and who had a baseline and at least one on-treatment HbA1cmeasurement), using a last-observation-carried-forward approach to impute missing data. ANCOVA was used, with treatment and previous insulin use as fixed classification effects, and baseline HbA1cas a linear covariate. Superiority of linagliptin versus placebo for change in HbA1cfrom baseline to week 24 was tested using the two-sided p value and the 95% CI of the treatment difference (linagliptin minus placebo). The effect of the study centre on the primary endpoint was explored with an ANCOVA model that included a treatment by centre interaction.

Change in FPG from baseline to week 24 was assessed for the full analysis set with the same ANCOVA model with baseline FPG as an additional linear covariate. Change in HbA1cover time was analysed with a mixed model for repeated measurements with treatment, previous use of insulin, visit, and visit by treatment interaction as fixed classification effects, and baseline HbA1cas a linear covariate. Available measurements were used (observed cases) with no imputation of missing data. Change in FPG over time was analysed using a similar mixed model for repeated measurements with baseline FPG as an additional linear covariate. For analysis of patients achieving HbA1cof less than 7·0% or a reduction in HbA1cof 0·5% or more, missing values were imputed using the non-completers-considered-failure approach. Logistic regression models were used to calculate odds ratios (OR) for achievement of HbA1cof less than 7·0% and use of rescue therapy.

Safety data were generally analysed descriptively for the treated set, which consisted of all randomised patients who received at least one dose of study drug. The effect of treatment on occurrence of hypoglycaemia was explored with logistic regression.

This study is registered with ClinicalTrials.gov , number NCT01084005 , and the European Clinical Trials database (EudractCT), number 2009–015255–25.

Role of the funding source

The sponsor was involved in study design, and data collection, review, and analysis. Individual authors, some of whom are employees of the sponsor, interpreted the data. All authors had full access to the data and had final responsibility for the content of the manuscript. AHB had the final decision to submit for publication.

Results

241 community-living outpatients (the treated set) were randomised to linagliptin 5 mg (n=162) or placebo (n=79), of whom 220 (91·3%) completed 24 weeks of treatment ( figure 1 ). The full analysis set consisted of 238 patients (160 linagliptin, 78 placebo). The mean age of patients at baseline was 74·9 years (range 70–91), with 44·4% aged 75 years or older, and nearly all were white ( table 1 ). Overall, patients had mean baseline HbA1cof 7·8% (SD 0·8), FPG of 8·3 mmol/L (1·6), bodyweight of 85·7 kg (16·0), and body-mass index of 29·7 kg/m2(4·7). Baseline demographic and clinical characteristics were generally much the same between treatment groups. Overall, there was a very high prevalence of long-standing diabetes, cardiovascular disease, renal impairment, use of insulin or sulfonylureas, and use of concomitant drugs and polypharmacy ( figure 2A ); mean Charlson age-comorbidity score was 5·1 (SD 0·7; figure 2B ). One patient (in the linagliptin group) had Alzheimer's disease. About one in five patients were taking basal insulin ( table 1 ), the mean daily dose for different types of insulin ranged from 30 to 60 IU.

gr1

Figure 1 Trial profile

Table 1 Baseline demographics and clinical characteristics in the treated set of patients

    Linagliptin (n=162) Placebo (n=79)
Age (years)
All patients 74·9 (4·4) 74·9 (4·2)
<75 years 91 (56·2%) 43 (54·4%)
≥75 years 71 (43·8%) 36 (45·6%)
Sex
Men 116 (71·6%) 49 (62·0%)
Women 46 (28·4%) 30 (38·0%)
Race
White 157 (96·9%) 76 (96·2%)
Asian 3 (1·9%) 2 (2·5%)
Black 2 (1·2%) 1 (1·3%)
Weight and BMI
Bodyweight (kg) 86·3 (16·4) 84·4 (15·3)
BMI (kg/m2) 29·6 (4·7) 29·8 (4·5)
Renal function (eGFR, mL/min per 1·73 m2, according to MDRD)
Normal (≥90) 36 (22·2%) 15 (19·0%)
Mild impairment (60 to <90) 83 (51·2%) 42 (53·2%)
Moderate impairment (30 to <60) 41 (25·3%) 21 (26·6%)
Severe impairment (<30) 2 (1·2%) 1 (1·3%)
HbA1c * (%) and FPG * (mmol/L)
All patients 7·8 (0·8) 7·7 (0·7)
HbA1c <7% 11 (6·9%) 6 (7·7%)
HbA1c ≥7% and <8% 91 (56·9%) 52 (66·7%)
HbA1c ≥8% and <9% 45 (28·1%) 15 (19·2%)
HbA1c ≥9% 13 (8·1%) 5 (6·4%)
FPG (mmol/L) 8·5 (1·6) 8·0 (1·6)
Time since diagnosis of type 2 diabetes *
≤1 year 3 (1·9%) 0
>1 to ≤5 years 20 (12·5%) 7 (9·0%)
>5 to ≤10 years 48 (30·0%) 29 (37·2%)
>10 years 89 (55·6%) 42 (53·8%)
Antidiabetes drugs at screening *
Drug class
  Metformin 133 (83·1%) 69 (88·5%)
  Sulfonylurea 94 (58·8%) 43 (55·1%)
  Insulin 35 (21·9%) 15 (19·2%)
  Meglitinide 1 (0·6%) 0
  α-glucosidase inhibitor 1 (0·6%) 0
Specific regimen
  Metformin 43 (26·9%) 21 (26·9%)
  Metformin plus sulfonylurea 64 (40·0%) 36 (46·2%)
  Metformin plus sulfonylurea plus α-glucosidase inhibitor 1 (0·6%) 0
  Metformin plus sulfonylurea plus insulin 9 (5·6%) 0
  Metformin plus meglitinide 1 (0·6%) 0
  Metformin plus insulin 15 (9·4%) 12 (15·4%)
  Sulfonylurea 16 (10·0%) 6 (7·7%)
  Sulfonylurea plus insulin 4 (2·5%) 1 (1·3%)
  Insulin 7 (4·4%) 2 (2·6%)
Concomitant drugs (non-diabetes)
≥1 159 (98·1%) 77 (97·5%)
≥3 141 (87·0%) 67 (84·8%)
≥5 96 (59·3%) 54 (68·4%)
Total drugs (diabetes and non-diabetes)
1 1 (0·6%) 0
2–3 13 (8·0%) 7 (8·9%)
4–5 38 (23·5%) 11 (13·9%)
>5 110 (67·9%) 61 (77·2%)

* Full analysis set (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; linagliptin n=160, placebo n=78).

These patients had an HbA1c level of 7·0% or more at screening and therefore were eligible for inclusion.

These patients were enrolled by investigators despite being on drugs that were not listed as permitted in this study (meglitinides, α-glucosidase inhibitors).

Data are n (%) or mean (SD). Treated set consisted of all patients who received at least one dose of study drug. BMI=body-mass index. FPG=fasting plasma glucose. eGFR=estimated glomerular filtration rate. HbA1c=glycated haemoglobin A1c. MDRD=Modification of Diet in Renal Disease equation.

gr2

Figure 2 Baseline disease-related characteristics and comorbidities (A) and Charlson age-comorbidity score (B) in the treated set of patients Treated set consisted of all patients who received at least one dose of study drug. FPG=fasting plasma glucose. GFR=glomerular filtration rate. HbA1c=glycated haemoglobin A1c. *Renal function according to GFR estimated by the Modification of Diet in Renal Disease equation. † Microvascular disease consisted of diabetic retinopathy, nephropathy, and neuropathy. ‡Macrovascular disease consisted of coronary artery 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.

After 24 weeks, the placebo-adjusted mean change from baseline in HbA1cwith linagliptin was −0·64% (95% CI −0·81 to −0·48, p<0·0001; table 2 ). Change in HbA1cover time is shown in figure 3 . HbA1cof less than 7·0%, when baseline level was 7·0% or greater, was achieved by more patients in the linagliptin group (38·9%, 58 of 149) than in the placebo group (8·3%, six of 72; OR 8·32, 95% CI 3·32 to 20·84, p<0·0001), as was an HbA1creduction of 0·5% or more (54·4% [87 of 160]vs12·8% [ten of 78]). In patients with baseline HbA1cabove these levels, in the linagliptin group, HbA1cof less than 7·5% was achieved by 49·5% (51 of 103) of patients, of less than 8·0% by 63·8% (37 of 58), and of less than 8·5% by 71·4% (20 of 28); in the placebo group the corresponding values were 27·3% (12 of 44), 30·0% (six of 20), and 45·5% (five of 11). The placebo-adjusted mean reduction in FPG from baseline with linagliptin was −1·15 mmol/L (95% CI −1·68 to −0·62, p<0·0001; table 2 ). Mean FPG decreased from baseline at every timepoint with linagliptin but increased with placebo (data not shown). Rescue treatment was needed by fewer patients in the linagliptin group than in the placebo group (4·4% [seven of 160]vs14·1% [11 of 78]; OR 0·21, 95% CI 0·07–0·63, p=0·0048). From weeks 12 to 24, insulin dosage changes (defined as a change of ≥20%) occurred in one of 35 patients in the linagliptin group and one of 15 patients in the placebo group; in those taking sulfonylureas, dosage changes occurred in four of 94 linagliptin patients (one increase, three decrease) and in seven of 43 placebo patients (six increase, one decrease).

Table 2 Change from baseline in HbA1cand FPG after 24 weeks in the full analysis set of patients, last observation carried forward

  Linagliptin Placebo Difference (linagliptin–placebo) p value
HbA1c in overall population
n 160 78 .. ..
Adjusted * mean percentage change from baseline (SE) −0·61% (0·06) 0·04% (0·07) −0·64% (0·08) <0·0001
95% CI .. .. −0·81 to −0·48 ..
FPG in overall population
n 160 78 .. ..
Adjusted mean change from baseline, mmol/L (SE) −0·59 (0·18) 0·56 (0·24) −1·15 (0·27) <0·0001
95% CI .. .. −1·68 to −0·62 ..
HbA1c in patients aged <75 years
n 90 42 .. ..
Adjusted mean percentage change from baseline (SE) −0·59% (0·07) −0·01% (0·10) −0·58% (0·11) <0·0001
HbA1c in patients aged ≥75 years
n 70 36 .. ..
Adjusted mean percentage change from baseline (SE) −0·62% (0·08) 0·10% (0·11) −0·73% (0·12) <0·0001
HbA1c in patients with type 2 diabetes for >1 year and ≤5 years
n 20 7 .. ..
Adjusted mean percentage change from baseline (SE) −0·60% (0·14) 0·17% (0·23) −0·77% (0·27) 0·0042
HbA1c in patients with type 2 diabetes for >5 years and ≤10 years
n 48 29 .. ..
Adjusted mean percentage change from baseline (SE) −0·61% (0·10) −0·11% (0·12) −0·50% (0·14) 0·0005
HbA1c in patients with type 2 diabetes for >10 years
n 89 42 .. ..
Adjusted mean percentage change from baseline (SE) −0·62% (0·07) 0·11% (0·10) −0·73% (0·11) <0·0001
HbA1c in patients with normal renal function
n 34 15 .. ..
Adjusted mean percentage change from baseline (SE) −0·44% (0·11) −0·06% (0·16) −0·39% (0·19) 0·0397
HbA1c in patients with mild renal impairment
n 83 42 .. ..
Adjusted mean percentage change from baseline (SE) −0·62% (0·07) 0·12% (0·10) −0·74% (0·11) <0·0001
HbA1c in patients with moderate renal impairment
n 41 20 .. ..
Adjusted mean percentage change from baseline (SE) −0·70% (0·10) −0·05% (0·14) −0·64% (0·17) 0·0001
HbA1c in patients with severe renal impairment
n 2 1 .. ..
Adjusted mean percentage change from baseline (SE) −0·51% (0·43) 0·23% (0·61) −0·73% (0·74) 0·3262
HbA1c in patients on metformin
n 44 21 .. ..
Adjusted mean percentage change from baseline (SE) −0·61% (0·09) −0·12% (0·13) −0·49% (0·16) 0·0024
HbA1c in patients on sulfonylurea, with or without metformin
n 81 42 .. ..
Adjusted mean percentage change from baseline (SE) −0·73% (0·07) −0·01% (0·09) −0·72% (0·12) <0·0001
HbA1c in patients on insulin (alone or in combination)
n 35 15 .. ..
Adjusted mean percentage change from baseline (SE) −0·53% (0·10) 0·14% (0·16) −0·66% (0·19) 0·0005

* Model includes treatment, continuous baseline HbA1c, and previous insulin use.

Model includes treatment, continuous baseline FPG, continuous baseline HbA1c, and previous insulin use.

FPG was analysed in units of mg/dL and converted to mmol/L values.

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 HbA1cmeasurement. FPG=fasting plasma glucose. HbA1c=glycated haemoglobin A1c.

gr3

Figure 3 Change from baseline in HbA1cover 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 HbA1cmeasurement. 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 HbA1cas a linear covariate. HbA1c=glycated haemoglobin A1c.

Placebo-adjusted mean changes in HbA1cwith linagliptin were generally similar between subgroups of age, duration of diabetes, renal function, and concomitant therapy ( table 2 ). Notably, efficacy was much the same across subgroups for disease duration, with a placebo-adjusted HbA1creduction of −0·73% in patients with diabetes for more than 10 years ( table 2 ).

The proportion of patients reporting adverse events was the same in both treatment groups (75·9%) ( table 3 ). Most events were mild or moderate in intensity with severe events reported for nine (5·6%) patients in the linagliptin group and three (3·8%) in the placebo group. 21·0% of patients in the linagliptin group and 13·9% of those in the placebo group had adverse events deemed to be related to study drug ( table 3 ). No deaths occurred. Serious adverse events occurred in 14 (8·6%) patients in the linagliptin group and five (6·3%) patients in the placebo group; none were deemed related to study drug. Drug-related adverse events leading to discontinuation of study drug occurred in one patient in the linagliptin group (insomnia) and one patient in the placebo group (depressed mood). CEC-confirmed cardiovascular events occurred in no placebo patients and in two linagliptin patients (non-fatal ischaemic stroke, admission to hospital because of unstable angina), but were judged to be unrelated to study drug. Neoplasms occurred in one (0·6%) patient in the linagliptin group (melanocytic naevus) and three (3·8%) patients in the placebo group (lung adenocarcinoma, malignant melanoma, and prostate cancer), and were all judged to be unrelated to study drug. No cases of pancreatitis were recorded. There were no clinically relevant changes in bodyweight in patients with available data (mean change at week 24 of −0·2 kg [SD 2·3] with linagliptin [n=146] and −0·6 kg [2·2] with placebo [n=63]; data were not adjusted for background sulfonylureas or insulin, which were used by slightly more patients in the linagliptin group than in the placebo group). Nor were there clinically relevant changes in renal function ( figure 4 ), laboratory variables, physical examination, ECGs, or vital signs, including heart rate and blood pressure (data not shown). Fewer of the patients not receiving a concomitant sulfonylurea had an adverse event with linagliptin (73%) than with placebo (83%), although slightly more patients in the linagliptin group reported drug-related, serious, severe, or treatment-ending events ( table 4 ).

Table 3 Summary of adverse events during 24 weeks in the treated set of patients

    Linagliptin (n=162) Placebo (n=79)
Overall
Any adverse event 123 (75·9%) 60 (75·9%)
Drug-related adverse event * 34 (21·0%) 11 (13·9%)
Serious adverse event 14 (8·6%) 5 (6·3%)
  Death 0 0
  Needing admission to hospital 14 (8·6%) 4 (5·1%)
  Other 0 2 (2·5%)
Individual serious adverse event
  Atrial fibrillation 1 (0·6%) 0
  Atrioventricular block complete 1 (0·6%) 0
  Bradycardia 1 (0·6%) 0
  Cerebrovascular accident 1 (0·6%) 0
  Coronary artery disease 1 (0·6%) 0
  Dehydration 1 (0·6%) 0
  Fall 2 (1·2%) 0
  Limb traumatic amputation 1 (0·6%) 0
  Lower limb fracture 1 (0·6%) 0
  Lower respiratory tract infection 0 1 (1·3%)
  Lung adenocarcinoma 0 1 (1·3%)
  Malignant melanoma 0 1 (1·3%)
  Pneumonia 3 (1·9%) 0
  Presyncope 1 (0·6%) 0
  Prostate cancer 0 1 (1·3%)
  Rectal haemorrhage 1 (0·6%) 0
  Thoracic vertebral fracture 1 (0·6%) 0
  Urinary bladder polyp 0 1 (1·3%)
  Vestibular neuronitis 1 (0·6%) 0
Adverse event leading to discontinuation 8 (4·9%) 1 (1·3%)
Severe adverse event 9 (5·6%) 3 (3·8%)
Significant adverse event § 4 (2·5%) 0
CEC-confirmed cardiovascular event 2 (1·2%) 0
Adverse events with an incidence ≥3·0% in either group
Hypoglycaemia 37 (22·8%) 13 (16·5%)
Nasopharyngitis 17 (10·5%) 7 (8·9%)
Diarrhoea 9 (5·6%) 2 (2·5%)
Hyperglycaemia 9 (5·6%) 8 (10·1%)
Back pain 7 (4·3%) 0
Fall 7 (4·3%) 2 (2·5%)
Urinary tract infection 7 (4·3%) 5 (6·3%)
Dizziness 6 (3·7%) 4 (5·1%)
Headache 6 (3·7%) 3 (3·8%)
Upper respiratory tract infection 6 (3·7%) 5 (6·3%)
Arthralgia 5 (3·1%) 0
Pain in extremity 5 (3·1%) 2 (2·5%)
Vertigo 5 (3·1%) 0
Influenza 3 (1·9%) 3 (3·8%)
Fatigue 2 (1·2%) 3 (3·8%)
Lower respiratory tract infection 1 (0·6%) 3 (3·8%)
Musculoskeletal pain 1 (0·6%) 3 (3·8%)

* Judged to be related to study drug by the investigator.

14 patients in the linagliptin group had 17 serious adverse events (none judged related to study drug). Five patients in the placebo group had five serious adverse events (none judged related to study drug).

In the linagliptin group, one patient had two severe adverse events (fall, lower limb fracture), and eight patients had one severe adverse event (atrial fibrillation, atrioventricular block complete, chest pain, pneumonia, back pain, cerebrovascular accident, dizziness, urinary retention). Three patients in the placebo group had one severe adverse event each (lower respiratory tract infection, lung adenocarcinoma, urinary bladder polyp).

§ Hypersensitivity reaction, renal adverse event, hepatic adverse event.

Preferred terms from the Medical Dictionary for Regulatory Activities (version 14.1).

Data are n (%). Treated set consisted of all patients who received at least one dose of study drug. CEC=clinical endpoint committee.

gr4

Figure 4 Renal function over 24 weeks in the treated set of patients Treated set consisted of all patients who received at least one dose of study drug. Data are from observed cases, including those from patients who received glycaemic rescue drugs. Error bars are SD. GFR=glomerular filtration rate. *GFR was estimated by the Modification of Diet in Renal Disease equation.

Table 4 Summary of adverse events during 24 weeks in the treated set of patients not receiving concomitant sulfonylurea treatment

    Linagliptin (n=67) Placebo (n=36)
Overall
Any adverse event 49 (73·1%) 30 (83·3%)
Drug-related adverse event * 11 (16·4%) 5 (13·9%)
Serious adverse event 5 (7·5%) 2 (5·6%)
  Death 0 0
  Needing admission to hospital 5 (7·5%) 1 (2·8%)
  Other 0 1 (2·8%)
Adverse event leading to discontinuation 4 (6·0%) 1 (2·8%)
Severe adverse event 3 (4·5%) 1 (2·8%)
Significant adverse event 2 (3·0%) 0
CEC-confirmed cardiovascular event 1 (1·5%) 0
Adverse events with an incidence ≥3·0% in either group
Hypoglycaemia 8 (11·9%) 6 (16·7%)
Nasopharyngitis 7 (10·4%) 4 (11·1%)
Upper respiratory tract infection 4 (6·0%) 2 (5·6%)
Vertigo 4 (6·0%) 0
Cystitis 3 (4·5%) 0
Fall 3 (4·5%) 1 (2·8%)
Gastroenteritis 3 (4·5%) 0
Peripheral oedema 3 (4·5%) 0
Rhinorrhoea 3 (4·5%) 0
Urinary tract infection 3 (4·5%) 2 (5·6%)
Headache 2 (3·0%) 1 (2·8%)
Dry mouth 1 (1·5%) 2 (5·6%)
Hyperglycaemia 1 (1·5%) 2 (5·6%)
Lower respiratory tract infection 0 2 (5·6%)
Fatigue 0 2 (5·6%)

* 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 (version 14.1).

Data are n (%). Treated set consisted of all patients who received at least one dose of study drug. CEC=clinical endpoint committee.

Investigator-defined hypoglycaemia occurred in 39 (24·1%) patients in the linagliptin group and 13 (16·5%) in the placebo group ( figure 5 ; OR for linagliptinvsplacebo 1·58, 95% CI 0·78–3·78, p=0·2083). Severe hypoglycaemia occurred in one patient, who was receiving linagliptin with metformin and sulfonylurea. The slightly higher overall incidence of hypoglycaemia with linagliptin, although not statistically significant, was mainly because of increased incidence in patients receiving sulfonylureas (30·5% linagliptin, 16·3% placebo); hypoglycaemia was much the same between groups in patients not receiving sulfonylureas (14·9% linagliptin, 16·7% placebo), and in those receiving metformin alone or insulin ( figure 5 ).

gr5

Figure 5 Hypoglycaemia in the treated set of patients Treated set consisted of all patients who received at least one dose of study drug. INS=insulin (with or without metformin). MET=metformin only. SU=sulfonylurea (with or without other antidiabetes drugs). *Confirmed plasma glucose concentration of 3·9 mmol/L or less, or symptoms attributed to hypoglycaemia, or both.

Discussion

In this randomised clinical trial, linagliptin improved glycaemia and was well tolerated in elderly patients with type 2 diabetes with inadequate glycaemic control from other glucose-lowering drugs, a group generally excluded from clinical trials despite being a large and growing proportion of the population of patients with type 2 diabetes.

There was a high prevalence of long-standing diabetes (>10 years in more than half the patients) and comorbidities—particularly cardiovascular disease (87%) and renal impairment (79%)—and a generally high use of concomitant drugs. The mean Charlson age-comorbidity score of more than 5 suggests a high-risk population. 18 These characteristics are typical of elderly patients with type 2 diabetes who are encountered in clinical practice and show why safety is a key consideration for glucose-lowering treatment. 10 Linagliptin was well tolerated with no new safety signals compared with previous phase 3 trials in younger patients.13, 14, 15, 16, and 17More than 75% of patients in the linagliptin group did not have an episode of hypoglycaemia. Although hypoglycaemia was reported by more linagliptin (24·1%) than placebo patients (16·5%), this difference seems to be because more linagliptin than placebo patients received sulfonylureas; in patients not receiving sulfonylureas, hypoglycaemia was not increased with linagliptin (14·9%, placebo 16·7%), consistent with previous studies.13, 14, 15, and 16Hypoglycaemia in elderly patients is of particular concern because it is more common than in younger patients because of age-related impairment of the counter-regulatory response to low plasma glucose and of symptom awareness. Furthermore, hypoglycaemia in elderly individuals frequently has severe consequences (falls, fractures, admissions to hospital, and possibly an increased risk for cardiovascular events and mortality).10 and 19Elderly populations have a high prevalence of renal impairment, which itself increases the risk of hypoglycaemia, 10 particularly when patients are receiving sulfonylureas. 20 About 80% of cases of sulfonylurea-induced severe hypoglycaemia occur in elderly patients.21 and 22Position statements by expert groups of diabetologists and gerontologists underscore the growing awareness that hypoglycaemia in the elderly is highly prevalent, under-recognised, and has severe consequences.4 and 7

Despite the possible risks of glucose-lowering treatment, the potential for hyperglycaemic crises and vascular complications in elderly patients means that undertreatment also has risks. 4 Although there is a scarcity of data for glucose-lowering treatment in elderly patients, it is thought to be beneficial because of improved clinical outcomes identified in younger patients with type 2 diabetes,23 and 24possible improvements in cognitive function, 6 and the benefit of modification of other cardiometabolic risk factors in elderly individuals (dyslipidaemia, hypertension).4 and 10In the present study, linagliptin treatment was associated with a clinically meaningful improvement in glycaemic control on average (placebo-corrected change in HbA1cof −0·64% from a baseline of 7·8%) and in most individuals (HbA1creduction of ≥0·5%). The efficacy of linagliptin was unaffected by age, renal impairment, or diabetes duration—an important finding in view of the advanced disease and many comorbidities prevalent in both the study population and clinical practice. More patients in the linagliptin than in the placebo group achieved HbA1cin the ranges recommended by European clinical guidelines 8 for elderly patients with type 2 diabetes (7–7·5% for those without major comorbidities; 7·6–8·5% for frail individuals). The notable efficacy identified in patients with diabetes for more than 10 years might be indicative of suppression of inappropriate glucagon secretion, which could be an important aspect of DPP4 inhibition in this age group. Consistent efficacy irrespective of diabetes duration has been previously identified with linagliptin and other DPP4 inhibitors,25, 26, 27, and 28although not in elderly patients with such long-standing disease.

This study was done in a clinically relevant population of patients with type 2 diabetes (aged ≥70 years) with long-standing disease inadequately controlled by common glucose-lowering drugs, including basal insulin ( panel ). Few drugs from other classes have been rigorously investigated in elderly patients,4 and 9but many have noteworthy limitations for elderly patients: for example, metformin, the usual first-line drug, is contraindicated for some degrees of renal impairment and can cause intolerable gastrointestinal side-effects; sulfonylureas are associated with hypoglycaemia and weight gain; pioglitazone can cause fluid retention, fractures, and weight gain, and has risk for congestive heart failure and possibly bladder cancer; and injectables (insulin and GLP1 receptor agonists) might not be convenient for elderly individuals. The higher direct cost of DPP4 inhibitors compared with generic oral drugs is an issue, but this difference could be balanced or even outweighed by reduced indirect costs—eg, lower rates of admissions to hospital for hypoglycaemia.33 and 34Health-related quality of life is an important consideration for treatment of elderly patients, and data collected with the SF-36 and EQ-5D standardised instruments in the present study are under analysis.

Panel Research in context

Systematic review

We searched PubMed and ISI Web of Science for papers published before June 3, 2013, reporting randomised, double-blind clinical trials of dipeptidyl peptidase-4 (DPP4) inhibitors done exclusively in elderly patients with type 2 diabetes, using the terms “type 2 diabetes”, “dipeptidyl peptidase-4”, “elderly”, and various synonyms. We identified three manuscripts reporting studies in patients aged 65 years or older in which monotherapy with sitagliptin, 26 vildagliptin, 29 or alogliptin 30 elicited reductions in glycated haemoglobin A1c(HbA1c) in patients not receiving other glucose-lowering drugs. Whereas sitagliptin was tested versus placebo, metformin was the comparator in the vildagliptin study and had much the same glycaemic efficacy and tolerability. In the alogliptin study, the sulfonylurea comparator, glipizide, was associated with substantially more hypoglycaemia (5%vs26% of patients). 30 Additionally, in another vildagliptin study that enrolled elderly patients who were treatment-naive or receiving oral glucose-lowering drugs, more vildagliptin than placebo patients achieved individualised glycaemic targets set by the study investigators, most of whom had a target HbA1cof less than 7·0%, 31 the lowest level recommended for elderly patients. 7 By contrast with the vildagliptin study, the present linagliptin study had a more vulnerable patient population (including those receiving insulin and those with severe renal impairment) with high use of polypharmacy, used a conventional definition of hypoglycaemia (blood glucose <3·9 mmol/L compared with <3·1 mmol/L), and provided data showing efficacy irrespective of diabetes duration. 31 We additionally searched ClinicalTrials.gov and identified completed studies in elderly patients of sitagliptin ( NCT01189890 ) and saxagliptin ( NCT01006603 ), for which results are not yet published.

Interpretation

Clinical decision making for glucose-lowering treatment in elderly patients with type 2 diabetes is challenging because of their comorbidities, potential frailty, and use of polypharmacy, as well as the scarcity of data from clinical trials to guide treatment selection. The importance of balancing efficacy with safety might unfavourably shift the risk–benefit profile of many glucose-lowering drugs in elderly individuals, particularly those that carry a substantial risk for hypoglycaemia, such as sulfonylureas, or have restrictions for renal impairment. Oral DPP4 inhibitors improve glycaemia in elderly patients with minimal risk for hypoglycaemia or other side-effects. The previous studies assessed DPP4 inhibitors as monotherapy, an uncommon treatment regimen in elderly patients, or in those not receiving insulin; by contrast, linagliptin showed these effects in patients whose glycaemia was not controlled by up to three other glucose-lowering drugs, including basal insulin. The present linagliptin study, therefore, closely represents a real-world situation, particularly in view of the patients' age (mean 75 years), polypharmacy burden, and clinical characteristics, since most had long-standing diabetes and comorbidities such as cardiovascular disease or renal impairment. Further efforts are needed to address the evidence gap in glucose-lowering treatment for elderly patients, including clinical-outcome studies to establish whether glycaemic efficacy and tolerability translate into reduced morbidity; the CAROLINA cardiovascular outcomes study of linagliptin will provide useful information in this regard. 32

This study has certain strengths and limitations. Like most randomised trials, the robust design confers high internal validity for comparison of the study drugs. Equally importantly, the patients' age and characteristics render the results highly relevant to clinical practice, although most participants were white. Clinical trials are frequently done in younger and healthier patients than are encountered in clinical practice; 2 although this is usually done for safety or logistical reasons (difficulty enrolling patients who are elderly or have comorbidities), it limits their external validity. The paucity of data on glucose-lowering in elderly patients—either from studies of glycaemic efficacy and tolerability or clinical-outcome studies—poses a challenge for clinical practitioners because many of the patients with type 2 diabetes who they encounter will be elderly.4 and 10A consensus report 4 on diabetes in elderly people recommended that clinical studies broaden inclusion criteria to address this evidence gap. This trial, like most of glucose-lowering drugs, was short (24 weeks) and consequently did not assess long-term clinical outcomes. However, the CAROLINA cardiovascular outcomes study of linagliptin 32 is enrolling patients up to 80 years old; a predefined requirement for enrolment is high risk for cardiovascular events, for which age 70 years or older is sufficient to qualify. Other limitations include possible under-representation of patients with cognitive impairment (although they were not specifically excluded), absence of frailty assessment, and use of an inert control (placebo) rather than an active comparator. Of relevance to the last point, further analysis of a 2 year comparative study in patients inadequately controlled by metformin—about a third of whom were older than 65 years 17 —showed that significantly more linagliptin (54%) than glimepiride patients (23%) achieved a composite endpoint of HbA1cof less than 7·0% without hypoglycaemia or weight gain. 35

In conclusion, the present study showed that linagliptin added to existing glucose-lowering drugs was well tolerated, weight neutral, and improved glycaemic control in patients with type 2 diabetes aged 70 years or older with characteristics typical of those encountered in clinical practice (eg, long-standing diabetes, renal impairment, use of combination therapies, including basal insulin). Hypoglycaemia was increased only with concomitant sulfonylureas, providing further evidence that sulfonylureas should be used with caution in elderly patients. Linagliptin might be a useful glucose-lowering drug for elderly patients with type 2 diabetes, a prevalent population for which other treatment options are understudied and have important limitations.

Contributors

AHB participated in design of the study, collection and interpretation of data, and drafting and revision of the manuscript. HH participated in design of the study, interpretation of data, and drafting and revision of the manuscript. RJ participated in design of the study, performed the statistical analysis, and participated in interpretation of data and drafting and revision of the manuscript. MvE participated in design of the study, interpretation of data, and drafting and revision of the manuscript. SP participated in design of the study, interpretation of data, and drafting and revision of the manuscript. H-JW participated in design of the study, interpretation of data, and drafting and revision of the manuscript. All authors have approved the final version of the manuscript.

Conflicts of interest

AHB has received honoraria for lectures and advisory work from Boehringer Ingelheim (the manufacturer of linagliptin), MSD, Novartis, Bristol-Myers Squibb/AstraZeneca, Takeda, Eli Lilly, Novo Nordisk and Sanofi-Aventis. HH, RJ, MvE, SP, and H-JW are employees of Boehringer Ingelheim.

Acknowledgments

This study was sponsored by Boehringer Ingelheim. We thank the patients and staff who participated in the study. Medical writing assistance, supported financially by Boehringer Ingelheim, was provided by Giles Brooke (Envision Scientific Solutions, Horsham, UK) during the preparation of this manuscript. Data in this manuscript have been presented in a poster at the 72nd Scientific Sessions of the American Diabetes Association, June 8–12, 2012, Philadelphia, PA, USA, and orally at the 9th International Diabetes Federation Western Pacific Region congress, November 24–27, 2012, Kyoto, Japan.

References

  • 1 F Cerreta, HG Eichler, G Rasi. Drug policy for an aging population—the European Medicines Agency's geriatric medicines strategy. N Engl J Med. 2012;367:1972-1974 Crossref
  • 2 AP Herrera, SA Snipes, DW King, I Torres-Vigil, DS Goldberg, AD Weinberg. Disparate inclusion of older adults in clinical trials: priorities and opportunities for policy and practice change. Am J Public Health. 2010;100(suppl 1):S105-S112
  • 3 Centers for Disease Control and Prevention. National diabetes fact sheet: national estimates and general information on diabetes and prediabetes in the United States, 2011 (US Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, GA, 2011)
  • 4 MS Kirkman, VJ Briscoe, N Clark, et al. Diabetes in older adults. Diabetes Care. 2012;35:2650-2664 Crossref
  • 5 WC Lakey, K Barnard, BC Batch, K Chiswell, A Tasneem, JB Green. Are current clinical trials in diabetes addressing important issues in diabetes care?. Diabetologia. 2013;56:1226-1235 Crossref
  • 6 JA Luchsinger. Type 2 diabetes and cognitive impairment: linking mechanisms. J Alzheimers Dis. 2012;30(suppl 2):S185-S198
  • 7 A Sinclair, JE Morley, L Rodriguez-Mañas, et al. Diabetes mellitus in older people: position statement on behalf of the International Association of Gerontology and Geriatrics (IAGG), the European Diabetes Working Party for Older People (EDWPOP), and the International Task Force of Experts in Diabetes. J Am Med Dir Assoc. 2012;13:497-502 Crossref
  • 8 AJ Sinclair, G Paolisso, M Castro, I Bourdel-Marchasson, R Gadsby, L Rodriguez Mañas, the European Diabetes Working Party for Older People. European Diabetes Working Party for Older People 2011 clinical guidelines for type 2 diabetes mellitus. Executive summary. Diabetes Metab. 2011;37(suppl 3):S27-S38 Crossref
  • 9 MA Fravel, DL McDanel, MB Ross, KG Moores, MJ Starry. Special considerations for treatment of type 2 diabetes mellitus in the elderly. Am J Health Syst Pharm. 2011;68:500-509 Crossref
  • 10 SE Inzucchi, RM Bergenstal, JB Buse, et al., the American Diabetes Association (ADA), the European Association for the Study of Diabetes (EASD). Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2012;35:1364-1379 Crossref
  • 11 T Karagiannis, P Paschos, K Paletas, DR Matthews, A Tsapas. Dipeptidyl peptidase-4 inhibitors for treatment of type 2 diabetes mellitus in the clinical setting: systematic review and meta-analysis. BMJ. 2012;344:e1369 Crossref
  • 12 U Graefe-Mody, S Retlich, C Friedrich. Clinical pharmacokinetics and pharmacodynamics of linagliptin. Clin Pharmacokinet. 2012;51:411-427
  • 13 S Del Prato, AH Barnett, H Huisman, D Neubacher, HJ Woerle, KA Dugi. Effect of linagliptin monotherapy on glycaemic control and markers of β-cell function in patients with inadequately controlled type 2 diabetes: a randomized controlled trial. Diabetes Obes Metab. 2011;13:258-267 Crossref
  • 14 R Gomis, RM Espadero, R Jones, HJ Woerle, KA Dugi. Efficacy and safety of initial combination therapy with linagliptin and pioglitazone in patients with inadequately controlled type 2 diabetes: a randomized, double-blind, placebo-controlled study. Diabetes Obes Metab. 2011;13:653-661 Crossref
  • 15 MR Taskinen, J Rosenstock, I Tamminen, et al. Safety and efficacy of linagliptin as add-on therapy to metformin in patients with type 2 diabetes: a randomized, double-blind, placebo-controlled study. Diabetes Obes Metab. 2011;13:65-74 Crossref
  • 16 DR Owens, R Swallow, KA Dugi, HJ Woerle. Efficacy and safety of linagliptin in persons with type 2 diabetes inadequately controlled by a combination of metformin and sulphonylurea: a 24-week randomized study. Diabet Med. 2011;28:1352-1361 Crossref
  • 17 B Gallwitz, J Rosenstock, T Rauch, et al. 2-year efficacy and safety of linagliptin compared with glimepiride in patients with type 2 diabetes inadequately controlled on metformin: a randomised, double-blind, non-inferiority trial. Lancet. 2012;380:475-483 Crossref
  • 18 M Charlson, TP Szatrowski, J Peterson, J Gold. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47:1245-1251 Crossref
  • 19 GS Meneilly, E Cheung, H Tuokko. Altered responses to hypoglycemia of healthy elderly people. J Clin Endocrinol Metab. 1994;78:1341-1348
  • 20 Y Handelsman, JI Mechanick, L Blonde, et al., the AACE Task Force for Developing Diabetes Comprehensive Care Plan. American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for developing a diabetes mellitus comprehensive care plan. Endocr Pract. 2011;17(suppl 2):1-53 Crossref
  • 21 YD Schejter, E Turvall, Z Ackerman. Characteristics of patients with sulphonurea-induced hypoglycemia. J Am Med Dir Assoc. 2012;13:234-238 Crossref
  • 22 A Holstein, C Hammer, M Hahn, NS Kulamadayil, P Kovacs. Severe sulfonylurea-induced hypoglycemia: a problem of uncritical prescription and deficiencies of diabetes care in geriatric patients. Expert Opin Drug Saf. 2010;9:675-681 Crossref
  • 23 UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-853
  • 24 RR Holman, SK Paul, MA Bethel, DR Matthews, HA Neil. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359:1577-1589 Crossref
  • 25 S Patel, S-A Weber, A Emser, M von Eynatten, H-J Woerle. Linagliptin improves glycemic control independently of diabetes duration and insulin resistance in patients with type 2 diabetes. Diabetologia. 2011;54(suppl 1):S339
  • 26 N Barzilai, H Guo, EM Mahoney, et al. Efficacy and tolerability of sitagliptin monotherapy in elderly patients with type 2 diabetes: a randomized, double-blind, placebo-controlled trial. Curr Med Res Opin. 2011;27:1049-1058 Crossref
  • 27 R Kawamori, N Inagaki, E Araki, et al. Linagliptin monotherapy provides superior glycaemic control versus placebo or voglibose with comparable safety in Japanese patients with type 2 diabetes: a randomized, placebo and active comparator-controlled, double-blind study. Diabetes Obes Metab. 2012;14:348-357 Crossref
  • 28 A Schweizer, S Dejager, JE Foley. Impact of insulin resistance, body mass index, disease duration, and duration of metformin use on the efficacy of vildagliptin. Diabetes Ther. 2012;3:8 Crossref
  • 29 A Schweizer, S Dejager, E Bosi. Comparison of vildagliptin and metformin monotherapy in elderly patients with type 2 diabetes: a 24-week, double-blind, randomized trial. Diabetes Obes Metab. 2009;11:804-812 Crossref
  • 30 J Rosenstock, C Wilson, P Fleck. Alogliptin versus glipizide monotherapy in elderly type 2 diabetes patients with mild hyperglycemia: a prospective, double-blind, randomized, 1-year study. Diabetes Obes Metab. 2013;10.1111/dom.12102 published online March 26.
  • 31 WD Strain, V Lukashevich, W Kothny, MJ Hoellinger, PM Paldánius. Individualised treatment targets for elderly patients with type 2 diabetes using vildagliptin add-on or lone therapy (INTERVAL): a 24 week, randomised, double-blind, placebo-controlled study. Lancet. 2013;382:409-416 Crossref
  • 32 J Rosenstock, N Marx, SE Kahn, et al. Cardiovascular outcome trials in type 2 diabetes and the sulphonylurea controversy: Rationale for the active-comparator CAROLINA trial. Diab Vasc Dis Res. 2013;10:289-301 Crossref
  • 33 M Bron, M Marynchenko, H Yang, AP Yu, EQ Wu. Hypoglycemia, treatment discontinuation, and costs in patients with type 2 diabetes mellitus on oral antidiabetic drugs. Postgrad Med. 2012;124:124-132 Crossref
  • 34 JC Simeone, BJ Quilliam. Predictors of emergency department and outpatient visits for hypoglycemia in type 2 diabetes: an analysis of a large US administrative claims database. Ann Pharmacother. 2012;46:157-168 Crossref
  • 35 B Gallwitz, J Rosenstock, A Emser, M von Eynatten, HJ Woerle. Linagliptin is more effective than glimepiride at achieving a composite outcome of target HbA1c <7% with no hypoglycaemia and no weight gain over 2 years. Int J Clin Pract. 2013;67:317-321 Crossref

Footnotes

a Diabetes Centre, Heart of England NHS Foundation Trust, Birmingham, UK

b University of Birmingham, Birmingham, UK

c Boehringer Ingelheim, Alkmaar, Netherlands

d Boehringer Ingelheim, Bracknell, UK

e Boehringer Ingelheim, Ingelheim, Germany

* Correspondence to: Prof Anthony H Barnett, Diabetes Centre, Heart of England NHS Foundation Trust, Birmingham, B9 5SS, UK

Reprint Sales Representative
Derrick Imasa
T: +1 212 633 3877
F: + 1212 633 3820
d.imasa@elsevier.com
The Lancet is a weekly
subscription journal. For further
information on how to subscribe
please contact our
Subscription Department
T: +44 (0) 1865 843077
F: +44 (0) 1865 843970
custserv@lancet.com
(North America)
T: +1 (800) 462 6198
F: +1 (800) 327 9021
USLancetCS@elsevier.com

Printed by

Elsevier España, S.L.U.
(A member of Elsevier)
Av. Josep Tarradellas, 20-30
08029 Barcelona
Tel.: 932 000 711
Fax: 932 091 136

This eprint is distributed with the support of Boehringer Ingelheim.

© 2015 Elsevier Ltd. All rights reserved. This journal and the individual contributions contained in it are protected under copyright by Elsevier Ltd, and the following terms and conditions apply to their use. The Lancet® is a registered trademark of Elsevier Properties S.A., used under licence. 
Publication information The Lancet (ISSN 0140-6736) is published weekly by Elsevier (The Boulevard, Langford Lane, Kidlington, Oxford, OX5 1GB). Periodicals Postage Paid at Rahway, NJ, USA. POSTMASTER: send address corrections to The Lancet, c/o Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. 
Photocopying Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery.Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier Global Rights Department, The boulevard, Langford Lane, Kidlington, Oxford, OX5 1GB, UK; tel: +1 215 239 3804 or +44 (0)1865 843 830, fax: +44 (0)1865 853 333, email healthpermissions@elsevier.com. In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc, 222 Rosewood Drive, Danvers, MA 01923, USA; tel: +1 (978) 7508400, fax: +1 (978) 646 8600. Other countries may have a local reprographic rights agency for payments. 
Derivative works Subscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions. Permission of the Publisher is required for resale or distribution outside the institution. Permission of the Publisher is required for all other derivative works, including compilations and translations.
Electronic storage or usage Permission of  the Publisher is required to store or use electronically any material contained in this journal, including any article or part of an article. Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier Health Sciences Rights Department, at the mail, fax, and e-mail addresses noted above.
Notice No responsibility is assumed by Elsevier for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.