Safety and tolerability of linagliptin in Asians with type 2 diabetes: a pooled analysis of 4457 patients from 21 randomized, double-blind, placebo-controlled clinical trials

ABSTRACT

Background

Safety and tolerability of glucose-lowering drugs is a key consideration for use in type 2 diabetes (T2D). We evaluated the safety and tolerability of the dipeptidyl peptidase-4 inhibitor linagliptin in Asian patients with T2D.

Research design and methods

This was a post-hoc, descriptive pooled analysis of 21 randomized, double-blind, placebo-controlled clinical trials of linagliptin in T2D patients lasting ≤52 weeks. We evaluated adverse events (AEs) and laboratory parameters in Asian participants living in Asia, both overall and in the East Asian subgroup.

Results

This analysis included 4457 Asian patients overall (2712 receiving linagliptin; 1745 receiving placebo) and 3057 (68.6%) East Asians. AEs were reported in 1510 (55.7%) Asian patients receiving linagliptin and 1032 (59.1%) receiving placebo but were considered drug-related in only 13.0% of each group. Serious AEs occurred in 109 (4.0%) linagliptin patients and 90 (5.2%) placebo patients. The most common AEs were nasopharyngitis (6.4% linagliptin, 7.3% placebo), upper respiratory tract infection (5.7% linagliptin, 6.5% placebo), and hypoglycemia (7.3% linagliptin, 6.3% placebo). One linagliptin patient had pancreatitis; none had bullous pemphigoid. No clinically relevant mean changes in laboratory parameters occurred. These findings were consistent in East Asians.

Conclusions

Linagliptin is well tolerated in Asian T2D patients, including East Asians, with low risk for AEs.

KEYWORDS (MeSH terms):

• Asia
• Asian continental ancestry group
• diabetes mellitus
• type 2
• dipeptidyl-peptidase iv inhibitors
• drug-related side effects and adverse reactions
• linagliptin

1. Introduction

Diabetes mellitus has become highly prevalent in Asia where – like other regions – the vast majority of patients (~90%) have the type 2 form of the disease [1]. Type 2 diabetes (T2D) is characterized by pancreatic β-cell dysfunction and insulin resistance rather than the overt loss of insulin secretion seen in type 1 diabetes. Together, the Western Pacific region (including Japan and China), with an estimated 163 million cases in 2019, and South-East Asia (including India), with 88 million cases, account for over half of the 463 million diabetes cases worldwide [1].

Asian and Western guidelines on managing hyperglycemia in people with T2D indicate that most individuals require pharmacotherapy to achieve glycemic control, and many glucose-lowering drugs from novel molecular classes have been added to the therapeutic armamentarium over the past 15 years [2–6]. Linagliptin is an oral drug from the dipeptidyl peptidase-4 (DPP-4) inhibitor class that is indicated in many countries as an adjunct to healthy diet and exercise to improve glycemic control in adults with T2D.

Linagliptin was launched between 2011 and 2013 in several Asian countries, where it is widely used as monotherapy or in combination with other glucose-lowering drugs. A pooled analysis in 2014 of 22 placebo-controlled, multinational clinical trials of the glycemic efficacy and tolerability of linagliptin treatment found that it had an acceptable overall safety and tolerability profile, with few adverse events occurring at a higher rate than with placebo treatment [7]. Patients whose ethnicity was identified as Asian comprised approximately 37% of the cohort in this analysis. In 2018 and 2019, the results from the CARMELINA and CAROLINA cardiovascular outcome trials (CVOTs), respectively, demonstrated the cardiovascular and kidney safety profile of linagliptin added to standard of care in patients with high cardiovascular risk with [8] or without high kidney risk [9] (ClinicalTrials.gov, NCT01897532 and NCT01243424). Analyses of the Asian participants in these trials showed similar findings to the overall trial cohorts [10,11].

The pathophysiology of T2D in Asians may have clinically important differences compared with other ethnicities, and there may also be considerable heterogeneity between Asians from different geographical regions [12–14]. Compared with white patients, Asians generally have greater abdominal obesity, but lower body mass index (BMI), and higher postprandial glucose excursions [12,13]. East Asian patients also tend to have lower pancreatic β-cell function while South Asians may have greater insulin resistance [12–15]. However, amplification of prandial insulin secretion by gastrointestinal peptide hormones (i.e. the incretin effect) may not be reduced in East Asians with T2D as it is in white patients [16], potentially explaining the greater glycemic response to incretin therapies such as DPP-4 inhibitors seen in Asians compared with non-Asians in some studies [17–20]. Genetic factors in East Asians potentially contributing to T2D were identified in a recent meta-analysis of genome-wide association studies, with some differences compared with European populations [21].

Therefore, we conducted a comprehensive analysis of the safety and tolerability of linagliptin as monotherapy or combination treatment in Asian patients, including a subgroup analysis of East Asians alone. Asian data from the CARMELINA and CAROLINA CVOTs were not included in this analysis, due to the substantial differences in study design compared with the glycemic efficacy trials; these included much greater study duration (as event-driven trials evaluating clinical outcomes), trial cohorts with increased cardiorenal risk, an active comparator in CAROLINA, and differences in the concomitant use of other glucose-lowering drugs. Safety data for the Asian patients in CARMELINA and CAROLINA have previously been published [10,11].

2. Patients and methods

2.1. Study design and patient population

This was a post hoc, descriptive analysis of patient-level safety and tolerability data from the 21 phase II, III or IV clinical trials in the linagliptin global development program that included Asian patients [22–42]. In these trials, which were conducted in more than 40 countries, patients were randomized to daily oral treatment with linagliptin (generally 5 mg) or placebo administered double-blind. These trials were primarily designed to evaluate the glycemic efficacy of linagliptin as either monotherapy or in combination with other glucose-lowering drugs. The duration of the trials ranged from 12 weeks to 52 weeks but was most commonly 24 weeks (Table 1). Trials with only an active (non-placebo) comparator were excluded in order to focus on the inherent safety profile of linagliptin rather than its relative safety versus other glucose-lowering drugs.

Table 1. Randomized, double-blind, placebo-controlled phase II, III, and Ⅳ trials of linagliptin that included Asian patients

ClinicalTrials.gov identifier [reference]	Phase	Patient population	Number of Asian patients		Duration (weeks)	Treatment regimen for linagliptin 5 mg or placebo
			Linagliptin	Placebo
NCT00641043 [22]	III	Multinational^a, 18–80 years	65	32	24	Initial combination with pioglitazone
NCT00621140 [23]	III	Multinational, 18–80 years	155	76	24	Monotherapy
NCT00601250 [24]	III	Multinational, 18–80 years	110	31	24	Add-on to metformin
NCT00602472 [25]	III	Multinational^a, 18–80 years	389	139	24	Add-on to metformin + sulfonylurea
NCT00654381 [26]	IIb/III	Japanese^a, 20–80 years	159	80	12^b	Monotherapy
NCT00819091 [27]	III	Multinational^a, 18–80 years	75	42	18	Add-on to sulfonylurea
NCT00954447 [31]	III	Multinational^a, ≥18 years	70	69	52	Add-on to basal insulin ± metformin and/or pioglitazone
NCT00800683 [32]	III	Multinational^a, 18–80 years, renal impairment	8	9	52	Add-on to existing glucose-lowering treatment
NCT00798161 [28]	III	Multinational, 18–80 years	125	124	24	Initial combination with metformin
NCT00740051 [29]	III	Multinational, 18–80 years, metformin-ineligible	40	20	18	Monotherapy
NCT00996658 [33]	III	Multinational, 18–80 years	124	61	24	Add-on to metformin + pioglitazone
NCT01012037 [30]	IIb	Multinational^a, 18–80 years	135	11	12	Add-on to metformin
NCT01087502 [35]	III	Multinational^a, ≥18 years, renal impairment	20	22	52	Monotherapy or add-on to existing glucose-lowering treatment
NCT01215097 [37]	III	Asian^a, 18–80 years	205	100	24	Add-on to metformin
NCT01214239 [36]	III	Asian^a, 18–80 years	200	99	24	Monotherapy
NCT01792518 [39]	IIIb	Multinational^a, 18–80 years, albuminuria	112	116	24	Monotherapy or add-on to existing glucose-lowering treatment
NCT02897349 [42]	III	Chinese^a, ≥18 years	104	102	24	Add-on to insulin ± metformin
NCT02240680 [40]	IV	Japanese^a, ≥60 years	52	50	52	Add-on to basal insulin
NCT01422876 [34]	III	Multinational^a, ≥18 years	47	49	24	Combined with empagliflozin
NCT02489968 [41]	III	Japanese^a, ≥20 years	223	224	52	Add-on to empagliflozin
NCT01708902 [38]	III	Asian^a, 18–80 years	294	289	24	Combined with metformin

^aTrial included East Asian patients.

^bStudy duration was 52 weeks: an initial 12-week placebo-controlled phase was followed by a 14-week active-controlled phase and then a final 26-week open-label extension phase; data shown are from patients receiving linagliptin 5 mg or placebo in the initial 12-week phase.

In general, these trials enrolled T2D patients aged ≥18 years (≥20 years in Japan) either naive to glucose-lowering pharmacotherapy or already receiving glucose-lowering drugs, with glycated hemoglobin (HbA1c) levels in the range of 6.5% to 11.0% (most commonly 7.0% to 10.0%), and BMI <40 kg/m². Patients with impaired hepatic function, kidney failure, or recent cardiovascular event (myocardial infarction, stroke, transient ischemic attack) were generally excluded.

The trial protocols were approved by local institutional review boards or independent ethics committees at each trial site or country before commencement, and all participants provided written informed consent prior to enrollment. All trials were registered with ClinicalTrials.gov prior to commencement (Table 1), and were subsequently conducted in accordance with Good Clinical Practice guidelines and the principles of the Declaration of Helsinki.

The current pooled analysis was restricted to participants of Asian race living in a country in Asia, which included 11 territories in East Asia (Japan, China, Hong Kong, South Korea, Taiwan) and elsewhere in the region (India, Israel, Malaysia, Philippines, Thailand, Vietnam).

2.2. Assessments

In this pooled analysis, we evaluated the incidence of adverse events reported by patients and recorded by trial investigators, which were coded using the Medical Dictionary for Regulatory Activities (MedDRA) version 23.1. Based on previous studies and theoretical considerations, we defined the following adverse events as being of special interest: angioedema, bullous pemphigoid, heart failure, hepatic injury, hypersensitivity, hypoglycemia, interstitial lung disease, intestinal obstruction, pancreatic cancer, pancreatitis, rhabdomyolysis, skin lesions, and worsening of kidney function. We also evaluated clinical laboratory test results, which were performed by central laboratories in each trial.

2.3. Statistical analysis

We pooled and analyzed individual patient data for all participants who received at least one dose of linagliptin 5 mg or placebo, i.e. the treated set. All analyses were descriptive in nature. We analyzed the number and percentage of patients with reported adverse events; we also calculated incidence rates (number of patients reporting adverse event per 100 patient-years) for adverse events of special interest. For analysis of laboratory parameters, we calculated means with standard deviations (SD) for baseline and change from baseline values; for amylase and lipase, we also calculated the number and percentage of patients with last value on treatment above or below the upper limit of normal.

3. Results

3.1. Patient disposition, baseline characteristics, and drug exposure

In the 21 eligible trials, 4457 Asian patients received at least one dose of study drug and were included in the treated set for this pooled analysis (2712 receiving linagliptin; 1745 receiving placebo). These 4457 Asian patients comprised 44% of the total population of the 21 phase II, III or IV clinical trials in the linagliptin global development program that included Asian patients.

At baseline, the study cohort was relatively young (mean age: 55.4 years) and slightly overweight, especially by Asian standards (mean BMI: 25.7 kg/m²) (Table 2). On average, patients had moderate hyperglycemia (mean HbA1c: 8.3%) and normal kidney function (mean estimated glomerular filtration rate [eGFR]: 89 ml/min/1.73 m²). Almost half the patients (45.6%) had been diagnosed with T2D over five years previously. Most patients (75.7%) were taking at least one oral glucose-lowering drug, but few were taking insulin (12.5%) (Table 2, Supplemental Table S1). More of the linagliptin group were taking sulfonylurea-containing regimens compared with the placebo group: metformin and sulfonylurea in 14.3% and 8.0% of patients, respectively, sulfonylurea monotherapy in 2.8% and 2.4%, and insulin and sulfonylurea (with or without other oral glucose-lowering drugs) in 0.5% and 0.5% (Supplemental Table S1).

Table 2. Demographic and baseline characteristics of the Asian patients included in this pooled analysis

	Linagliptin	Placebo
	(n = 2712)	(n = 1745)
Asian region
East	1817 (67.0)	1240 (71.1)
South	533 (19.7)	276 (15.8)
Southeast	361 (13.3)	229 (13.1)
West	1 (0.0)	0
Sex
Male	1520 (56.0)	1033 (59.2)
Female	1192 (44.0)	712 (40.8)
Age (years)	55.4 ± 10.4	55.4 ± 10.5
<65	2147 (79.2)	1377 (78.9)
65 to <75	503 (18.5)	327 (18.7)
≥75	62 (2.3)	41 (2.4)
HbA1c (%)	8.3 ± 0.9	8.4 ± 0.9
Fasting plasma glucose (mg/dl)	157.8 ± 39.3	160.1 ± 40.8
Body weight (kg)	67.7 ± 12.6	67.9 ± 12.1
Body mass index (kg/m^2)	25.7 ± 3.9	25.6 ± 3.7
<25	1279 (47.2)	849 (48.7)
25 to <30	1089 (40.2)	704 (40.3)
≥30	344 (12.7)	192 (11.0)
Time since T2D diagnosis (years)
≤1	672 (24.8)	469 (26.9)
>1 to 5	792 (29.2)	491 (28.1)
>5	1248 (46.0)	785 (45.0)
eGFR (MDRD) (ml/min/1.73 m^2)	89.2 ± 22.4	88.2 ± 22.3
≥90	1315 (48.5)	814 (46.6)
60 to <90	1169 (43.1)	772 (44.2)
30 to <60	216 (8.0)	138 (7.9)
<30	12 (0.4)	21 (1.2)
Systolic blood pressure (mmHg)	126.8 ± 14.4	126.8 ± 14.6
Diastolic blood pressure (mmHg)	77.9 ± 9.1	77.9 ± 9.3
Concomitant OAD
0	682 (25.1)	399 (22.9)
1	1427 (52.6)	1057 (60.6)
≥2	603 (22.2)	289 (16.6)
Insulin
Yes	281 (10.4)	274 (15.7)
No	2431 (89.6)	1471 (84.3)

Data are n (%) or mean ± SD for participants treated with ≥1 dose of study drug. eGFR: estimated glomerular filtration rate; HbA1c: glycated hemoglobin; MDRD: Modification of Diet in Renal Disease study equation; OAD: oral anti-diabetes drug; T2D: type 2 diabetes.

Over two-thirds of the study cohort (n = 3057; 68.6%) was from East Asia and the baseline characteristics for these patients were similar to the overall cohort (Supplemental Table S2).

The median duration of exposure to study drug was 169 days in patients receiving linagliptin and 170 days in those receiving placebo, in both the overall cohort and the East Asian subgroup.

3.2. Summary of adverse events

Fewer Asian patients in the linagliptin group reported an adverse event compared with the placebo group (55.7% and 59.1%, respectively) (Table 3). Similarly, fewer linagliptin-treated patients had a serious adverse event compared with placebo-treated patients (4.0% and 5.2%, respectively) or an adverse event leading to discontinuation (2.8% and 3.9%) (Table 3). Drug-related adverse events were reported in 13.0% of patients in each group. Of the individual adverse events (i.e. MedDRA preferred terms) that occurred in more than 5% of either treatment group, only hypoglycemia was more common with linagliptin (7.3% of patients) than placebo (6.0%). This pattern of adverse events was similar in East Asians (Supplemental Table S3).

Table 3. Summary of adverse events in Asian patients

	Linagliptin	Placebo
	(n = 2712)	(n = 1745)
Any AE	1510 (55.7)	1032 (59.1)
Drug-related AE^a	352 (13.0)	226 (13.0)
Severe AE^b	54 (2.0)	35 (2.0)
AE leading to discontinuation	76 (2.8)	68 (3.9)
Serious AE^c	109 (4.0)	90 (5.2)
Fatal AE	2 (0.1)	4 (0.2)
Most frequent individual AEs^d
Hypoglycemia	197 (7.3)	104 (6.0)
Nasopharyngitis	174 (6.4)	127 (7.3)
Upper respiratory tract infection	155 (5.7)	113 (6.5)
Hyperglycemia	103 (3.8)	93 (5.3)

Data are n (%) of participants treated with ≥1 dose of study drug. AE: adverse event; MedDRA: Medical Dictionary for Regulatory Activities version 23.1.

^aAE that was related to study drug in the opinion of the investigator.

^bAE that was incapacitating or caused inability to work or to perform usual activities.

^cAE that resulted in death, was immediately life threatening, resulted in persistent or significant disability/incapacity, required or prolonged patient hospitalization, was a congenital anomaly/birth defect, or was deemed serious for any other reason.

^dMedDRA preferred term that occurred in ≥5% of either group.

3.3. Adverse events of special interest

All adverse events of special interest, except hypoglycemia, occurred in <3% of Asian patients (Table 4). Angioedema, hypersensitivity events, and hypoglycemia were slightly more common in the linagliptin group than the placebo group. Very few events of heart failure were identified in this analysis (three patients in the linagliptin group and one patient in the placebo group). With regards to intestinal obstruction, two cases were identified in the linagliptin group and one in the placebo group.

Table 4. Adverse events of special interest in Asian patients

	Linagliptin		Placebo
	(n = 2712)		(n = 1745)
	n (%)	Rate/100 patient-years		n (%)	Rate/100 patient-years
Hypoglycemia^a	197 (7.26)	15.21		104 (5.96)	11.22
Angioedema^b	23 (0.85)	1.67		8 (0.46)	0.82
Hypersensitivity^c	81 (2.99)	5.95		43 (2.46)	4.47
Heart failure^d	3 (0.11)	0.22		1 (0.06)	0.10
Worsening of renal function^e	10 (0.37)	0.72		5 (0.29)	0.51
Pancreatitis^f	1 (0.04)	0.07		0	0
Skin lesions^g	0	0		0	0
Bullous pemphigoid^h	0	0		0	0
Hepatic injury^i	54 (1.99)	3.93		51 (2.92)	5.31
Pancreatic cancer^j	0	0		0	0
Interstitial lung disease^k	0	0		1 (0.06)	0.10
Intestinal obstruction^l	2 (0.07)	0.14		1 (0.06)	0.10
Rhabdomyolysis^m	0	0		0	0

Data are n (%) of participants treated with ≥1 dose of study drug. MedDRA: Medical Dictionary for Regulatory Activities version 23.1; SMQ: Standardized MedDRA Query.

^aBased on narrow SMQ Hypoglycemia.

^bBased on narrow SMQ Angioedema.

^cBased on narrow SMQ Hypersensitivity.

^dBased on narrow SMQ Cardiac failure.

^eBased on narrow SMQ Acute renal failure.

^fBased on narrow SMQ Pancreatitis plus MedDRA preferred term Pancreatitis chronic.

^gBased on narrow SMQ Severe cutaneous adverse reactions.

^hBased on MedDRA high-level term Bullous conditions primary path

ⁱBased on the following narrow sub-SMQs: Liver related investigations, signs and symptoms; Cholestasis and jaundice of hepatic origin; Hepatitis, noninfectious; Hepatic failure, fibrosis and cirrhosis and other liver damage-related conditions.

^jBased on the following MedDRA preferred terms: Intraductal papillary-mucinous carcinoma of pancreas, Malignant neoplasm of islets of Langerhans, Metastatic glucagonoma, Neurotensinoma, Pancreatic carcinoma, Pancreatic carcinoma metastatic, Pancreatic carcinoma recurrent, Pancreatic carcinoma stage 0–IV, Pancreatic neoplasm, Pancreatic neuroendocrine tumor, Pancreatic neuroendocrine tumor metastatic, Pancreatic sarcoma, Pancreatoblastoma, Solid pseudopapillary tumor of the pancreas, Somatostatinoma, Vipoma.

^kBased on narrow SMQ Interstitial lung disease plus MedDRA preferred term Organizing pneumonia.

^lBased on narrow SMQ Gastrointestinal obstruction plus the following MedDRA preferred terms: Gastric atony, Gastric hypomotility, Gastrointestinal hypomotility, Intestinal atony, Intestinal pseudo-obstruction, Intestinal transit time abnormal, Intestinal transit time increased, Megacolon, Esophageal achalasia, Esophageal atony, Esophageal hypomotility, Toxic dilatation of intestine.

^mBased on narrow SMQ Rhabdomyolysis/myopathy.

Worsening of renal function was also slightly more common in the linagliptin group (Table 4). Mean eGFR decreased slightly from baseline to study end in both the linagliptin group (–1.27 ml/min/1.73 m² from a baseline of 89.19) and the placebo group (–0.13 ml/min/1.73 m² from baseline 87.97). Mean uric acid increased slightly in both groups (by 0.38 and 0.17 μmol/l with linagliptin and placebo, respectively), as did serum creatinine (0.02 and 0.01 mg/dl, respectively) (Table 5).

Table 5. Changes in laboratory parameters in Asian patients

	Linagliptin			Placebo
	n	Baseline	Change from baseline	n	Baseline	Change from baseline
Serum creatinine (mg/dl)	2649	0.85 ± 0.24	0.02 ± 0.17	1696	0.87 ± 0.30	0.01 ± 0.14
eGFR (ml/min/1.73 m^2)	2649	89.19 ± 22.44	–1.27 ± 12.00	1696	87.97 ± 22.36	–0.13 ± 11.87
Uric acid (µmol/l)	2648	5.60 ± 2.27	0.38 ± 1.52	1696	5.67 ± 2.41	0.17 ± 1.59
AST/GOT (U/l)	2643	26.01 ± 18.37	0.46 ± 50.42	1695	28.13 ± 22.29	0.13 ± 40.16
ALT/GPT (U/l)	2643	31.95 ± 22.97	–0.42 ± 49.33	1695	33.57 ± 26.84	–0.43 ± 53.48
GGT (U/l)	2383	45.67 ± 63.28	0.64 ± 42.29	1428	44.69 ± 58.67	–2.21 ± 43.75
Total bilirubin (μmol/l)	2626	0.47 ± 0.27	–0.03 ± 0.20	1674	0.47 ± 0.28	–0.02 ± 0.21
Amylase (U/l)	2549	70.92 ± 31.21	5.49 ± 20.25	1597	70.22 ± 30.28	2.54 ± 20.38
Lipase (U/l)	714	37.62 ± 26.15	5.36 ± 28.98	640	38.85 ± 22.44	1.41 ± 48.91

Data are mean ± SD for participants with a baseline value and ≥1 on-treatment value.

ALT: alanine aminotransferase; AST: aspartate aminotransferase; eGFR: estimated glomerular filtration rate; GGT: gamma-glutamyl transferase; GOT: glutamic oxaloacetic transaminase; GPT: glutamate pyruvate transaminase.

Hepatic injury was less common with linagliptin (1.99% of patients) than with placebo (2.92%) (Table 4). Mean changes from baseline in liver enzymes and bilirubin were small and generally similar in both groups (Table 5).

One patient experienced interstitial lung disease (an individual receiving placebo) and one experienced pancreatitis (an individual in the linagliptin group with acute pancreatitis) (Table 4). Mean changes in amylase and lipase levels were higher in the linagliptin group than the placebo group (Table 5) but the percentage of patients with amylase or lipase levels greater than the upper limit of normal at end of treatment was low and similar in each group (Figure 1). No patients experienced pancreatic cancer. Similarly, no patients in either treatment group experienced bullous pemphigoid or other skin lesions.

Figure 1. Proportion of Asian patients with last value on treatment for amylase and lipase relative to the upper limit of normal (ULN).

Overall, mean values at baseline and end of treatment were within the normal range for all laboratory parameters.

The general trends for adverse events of special interest in the overall Asian cohort were very similar in the East Asian subgroup (Supplemental Table S4).

4. Discussion

Our comprehensive analysis demonstrates the safety and tolerability profile of the DPP-4 inhibitor linagliptin in Asian patients. This analysis pooled data for 4457 patients from 21 randomized, double-blind, placebo-controlled clinical trials of linagliptin published up to 2021, which includes all eligible trials from the now-completed, company-sponsored clinical development program for this medication. Overall, the percentage of Asian patients experiencing an adverse event with linagliptin was no greater than with placebo, as was also the case for serious adverse events, drug-related events, and events causing treatment discontinuation. Furthermore, there were no substantial differences in the incidence of most specific types of adverse event with linagliptin compared with placebo.

DPP-4 inhibitors are widely prescribed in Asian countries, in general, and are the most prescribed class of oral glucose-lowering drug in Japan in particular [43–45]. Consequently, it is important to characterize their safety profiles in Asian patients, who have historically been a minority in multinational trials of glucose-lowering drugs. Unlike older classes of glucose-lowering drugs such as sulfonylureas and insulin, DPP-4 inhibitors have an inherently low risk for causing hypoglycemia, due to their glucose-dependent insulinotropic mechanism of action mediated via the incretin hormone glucagon-like peptide-1 (GLP-1) [3–6]. Previous pooled analyses of clinical trials demonstrated the safety and tolerability profile of linagliptin in multinational patient cohorts [7,46–48] whereas a pooled analysis of Asian patients focused primarily on efficacy [49] – our pooled analysis exclusively focuses on safety and tolerability findings in a larger cohort of Asian patients that includes the most recent trials. The safety and tolerability profile observed in our analysis is consistent with the adverse events listed in the current prescribing information for linagliptin [17–19].

In our pooled analysis, the incidence of hypoglycemia was slightly higher with linagliptin (7.3%) than placebo (6.0%), which may reflect the greater use of sulfonylurea-containing regimens in the linagliptin group. This minor imbalance between treatment arms in the use of sulfonylureas was mainly due to the inclusion of Asian patients from one study: a multinational phase III trial in which patients with insufficient glycemic control despite using the maximum tolerated doses of metformin and a sulfonylurea were randomized in a 3:1 ratio to linagliptin or placebo [25]. Increased incidence of hypoglycemia with concomitant use of sulfonylureas occurs with all DPP-4 inhibitors, as well as GLP-1 receptor agonists, and is thought to result from a pharmacodynamic interaction in which sulfonylureas uncouple the insulinotropic action of GLP-1 from its dependence on ambient glucose levels [50]. For this reason, the prescribing information for linagliptin and other incretin therapies in most territories suggests reducing the dose of sulfonylurea when co-administered.

It is important to establish the cardiorenal safety of glucose-lowering drugs in Asian patients, given the high rates of cardiovascular and kidney complications in people with T2D, the latter of which are more common in Asian patients than white patients [12,13,51–54]. In our pooled analysis, the incidence of cardiovascular and kidney-related adverse events was comparable between the linagliptin and placebo groups, as was the change in eGFR; however, the number of events was very low. Unlike the trials in our pooled analysis, the placebo-controlled CARMELINA trial was designed to robustly evaluate cardiovascular and kidney outcomes with linagliptin, and found no increased risk of cardiorenal events in either the overall trial cohort or the Asian subgroup. Notably, there was a modestly reduced risk for albuminuria progression and a numerically lower incidence of hospitalization for heart failure with linagliptin compared to placebo [8,10]. Furthermore, a three-year post-marketing surveillance study of linagliptin in routine clinical practice in Japan, which included over 2200 T2D patients, found that kidney function remained stable during the study period with no new safety concerns regardless of eGFR level [55].

Pancreatitis is a known but rare adverse event associated with DPP-4 inhibitors [56]. In our pooled analysis, there was one case of pancreatitis (an acute episode) with linagliptin among the 2712 treated patients and none with placebo. Observational studies in Japan, Taiwan and South Korea have not found an increased risk for pancreatitis with DPP-4 inhibitors in clinical practice [57–59]. Descriptive mean changes in lipase and amylase levels were higher in the linagliptin group compared with the placebo group. In addition, the percentage of patients in the linagliptin group who had levels more than three times greater than the upper limit of normal at last observation was similar or slightly higher numerically than in the placebo group (amylase: linagliptin 0.12% [n/N = 3/2549], placebo 0.13% [n/N = 2/1597]; lipase: linagliptin 0.56% [n/N = 4/714], placebo 0.31% [n/N = 2/640]).

DPP-4 is a ubiquitous multifunctional protein whose diverse biological activities include enzymatic cleavage of various growth factors and chemokines in addition to GLP-1, as well as non-enzymatic roles in immunomodulation, inflammation, and antioxidative response – together, these suggest a potential role in cancer biology [60]. Some preclinical studies have found an association between DPP-4 inhibitors and cancer, but clinical evidence to date has not found a clear association between cancer incidence or prognosis in T2D patients [60]. Potential effects on pancreatic cancer are of particular interest, given the activity of GLP-1 in the pancreas. In our pooled analysis, there were no cases of pancreatic cancer; however, the observation period was too short to evaluate cancer risk. The CARMELINA and CAROLINA trials were much longer (median durations of 2.2. and 6.3 years, respectively). In CARMELINA, there were no cases of pancreatic cancer with linagliptin and one case with placebo (0.4%) among the Asian participants [10]; in CAROLINA, there were three cases with linagliptin (0.6%) and two with glimepiride (0.4%) in Asians [11]. Overall, the risk of pancreatic cancer with DPP-4 inhibitors remains uncertain [61], especially given the lengthy latency period of this tumor type and the complexity of the potential association, with some risk factors for pancreatic cancer being common in T2D patients.

In 2015, a safety signal for bullous pemphigoid with DPP-4 inhibitors was identified from a disproportionality analysis of the US Food and Drug Administration Adverse Event Reporting System. Consequently, US prescribing information for DPP-4 inhibitors was updated to include bullous pemphigoid, an autoimmune skin disorder characterized by blisters. Japanese prescribing information was updated similarly in 2016 following a request from the Pharmaceuticals and Medical Devices Agency. In our analysis, we did not detect any cases of bullous pemphigoid or other types of skin lesions. However, bullous pemphigoid occurs mainly in patients over the age of 70, whereas patients in our analysis were younger on average (mean of 55 years). Notably, no cases of bullous pemphigoid were detected in Asian participants in the CARMELINA trial where the average age was higher (65 years), despite a numerical imbalance in the overall trial cohort with seven cases in the linagliptin group (0.2%) and none in the placebo group [10]. There were no cases of pemphigoid with linagliptin monotherapy in 2235 patients in a three-year post-marketing surveillance study in Japan [62], while a similar three-year post-marketing surveillance study in Japan observed two cases in 3372 patients (0.06%) receiving linagliptin added to other glucose-lowering drugs [63].

In our pooled analysis, we also did not observe any liver safety signal, with hepatic injury reported for ~2% of the linagliptin group and ~3% of the placebo group, and no clinically meaningful changes in liver enzymes or bilirubin. This is an important observation, given the high prevalence of liver disease in both T2D patients generally and the Asian region in particular, and it is consistent with a pooled analysis of patients with hepatic disorders receiving linagliptin [64].

The strengths and limitations of our analysis should be noted. The major limitation is that the safety and tolerability data are derived from clinical trials lasting less than 52 weeks that were primarily designed to evaluate the glycemic efficacy of linagliptin. Hence, the data are derived from descriptive, exploratory analyses, and do not provide insights into long-term safety. Conversely, the main strength is that the underlying trials were robustly controlled with randomization, placebo comparators and double-blinding, which enables confidence that the findings reflect the inherent safety and tolerability profile of linagliptin, albeit there were minor imbalances in some patient characteristics between treatment groups.

5. Conclusions

This comprehensive, up-to-date pooled analysis of clinical trial data demonstrates the safety and tolerability profile of linagliptin in Asian patients, including a large subgroup of East Asians. The results are consistent with older pooled analyses of overall trial populations including several races/ethnicities [7,46] and the current prescribing information for linagliptin [17–19], with no new safety signals identified. These findings complement recent subgroup analyses of the CARMELINA and CAROLINA outcomes trials demonstrating the cardiorenal safety profile of linagliptin in Asian patients [10,11], as well as post-marketing surveillance studies in Japan [55,62,63]. Taken together, linagliptin appears to be well tolerated in Asian patients with T2D, including East Asians.

Declaration of interests

K. Kanasaki has received lecture fees from Boehringer Ingelheim, Eli Lilly and Sanofi, and is under contract for consultancy with Boehringer Ingelheim; Mitsubishi Tanabe Pharma, Boehringer Ingelheim, and Ono Pharmaceutical contributed to establishing the Division of Anticipatory Molecular Food Science and Technology, Medical Research Institute, Kanazawa Medical University. F. Yamamoto is an employee of Nippon Boehringer Ingelheim Co. Ltd. C. Schepers and R. Sani Simões are employees of Boehringer Ingelheim. D. Yabe has received consulting/lecture fees from Eli Lilly Japan K.K., Novo Nordisk Pharma Ltd., Ono Pharmaceutical Co. Ltd., and Takeda Pharmaceutical Company Limited, and grants from Arkray Inc., Novo Nordisk Pharma Ltd., Nippon Boehringer Ingelheim, Ono Pharmaceutical Co. Ltd., Taisho Pharmaceutical Co. Ltd., Takeda Pharmaceutical Company Limited, and Terumo Corporation during the conduct of the study. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Author contributions

All authors participated in the interpretation of study results, and in the drafting, critical revision, and approval of the final version of the manuscript. All authors agree to be accountable for all aspects of this work.

Role of the sponsor

The Boehringer Ingelheim and Eli Lilly and Company Diabetes Alliance was involved in the study design, data collection, data analysis, and preparation of the manuscript.

Data availability

The sponsor of the clinical trials (Boehringer Ingelheim) is committed to responsible sharing of clinical study reports, related clinical documents, and patient-level clinical study data. Researchers are invited to submit inquiries via the following website (https://trials.boehringer-ingelheim.com/).

Acknowledgments

The authors thank the physicians and patients who participated in the clinical trials analyzed in this manuscript. The authors also thank the following individuals for their assistance and advice: Tomoo Okamura (Nippon Boehringer Ingelheim Co, Ltd, Tokyo, Japan), Atsushi Taniguchi, MSc (Nippon Boehringer Ingelheim Co, Ltd, Tokyo, Japan), Søren S. Lund (Boehringer Ingelheim International GmbH, Ingelheim, Germany), and Thomas Meinicke (Boehringer Ingelheim International GmbH, Ingelheim, Germany).

Supplementary materials

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14740338.2022.1999409.

Additional information

Funding

The studies that provided data for this analysis were sponsored by the Boehringer Ingelheim and Eli Lilly and Company Diabetes Alliance. Medical writing assistance was provided by Giles Brooke of Elevate Scientific Solutions during the preparation of this manuscript, and was funded by Nippon Boehringer Ingelheim Co. Ltd.