Glucagon-Like Peptide-1 Receptor Agonists: Their Therapeutic Potential in Cystic Fibrosis

Abstract

Cystic fibrosis (CF) is a monogenic disorder leading to pulmonary disease, pancreatic insufficiency and cystic fibrosis-related diabetes (CFRD). Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are now being investigated in people with cystic fibrosis (pwCF) and CFRD. To date, their therapeutic potential has been almost exclusively studied in case reports or case series. These agents improved glycated haemoglobin (HbA_1c) and continuous glucose monitoring (CGM) parameters. Benefits were also observed in weight reduction, particularly for subjects on cystic fibrosis transmembrane conductance regulator (CFTR) modulator therapy elexacaftor/tezacaftor/ivacaftor (ETI). However, discordant results have also been reported. Moreover, GLP-1RAs have improved pulmonary function, even following lung transplantation. Importantly, the dual glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) receptor agonist tirzepatide has also yielded favourable outcomes. Finally, preliminary evidence suggests potential inhibition of bone resorption, pointing to a therapeutic perspective in cystic fibrosis-related bone disease (CFBD). However, potential adverse events should not be ignored. These include risk of acute pancreatitis, nausea/vomiting, nutritional depletion, bowel dysmotility and distal intestinal obstruction syndrome, as well as others. Adverse events should be addressed with caution, and dose adjustments may be useful. Large prospective multicentre studies are now required to validate these outcomes and to suggest implications for clinical practice.

Key Summary Points

Cystic fibrosis (CF) is a major monogenic disease and cystic fibrosis-related diabetes (CFRD) is a cardinal feature of the condition without any highly efficacious treatment addressing the rising prevalence of obesity in people with cystic fibrosis (pwCF) (increasingly on elexacaftor/tezacaftor/ivacaftor) and simultaneously optimising glycaemic control.

Accumulating evidence suggests that the incretin effect is dysregulated in pwCF and potential restoration is linked with beneficial effects in glucose control.

In experimental settings, limited data points to decreased expression of glucagon-like peptide-1 receptor (GLP-1R) in pancreatic islets from pwCF and upregulation of cystic fibrosis transmembrane conductance regulator (CFTR) upon treatment with liraglutide.

Clinical studies showed promising outcomes in glycaemic control, weight management and pulmonary function for pwCF (most on elexacaftor/tezacaftor/ivacaftor) treated with glucagon-like peptide 1 receptor agonists (GLP-1RAs) or the novel GLP-1R and glucose-dependent insulinotropic polypeptide (GIP) receptor agonist tirzepatide, as reduced glycated haemoglobin (HbA1c), body weight, body mass index (BMI) and improved continuous glucose monitoring (CGM) metrics, forced expiratory volume in the first second (FEV1) and forced vital capacity (FVC) were found.

Future trials should delineate the exact effects of GLP-1RAs in pwCF and identify which subgroups of pwCF will experience the greatest benefit.

Introduction

Cystic fibrosis (CF) is a multisystem genetic disorder attributed to genetic mutations of cystic fibrosis transmembrane conductance regulator (CFTR) [1–4]. It is estimated that at least 100,000 people worldwide are affected [1]. The condition is frequently accompanied by major comorbidities, including cystic fibrosis-related diabetes (CFRD) [1, 5]. The latter is attributed to insulin insufficiency due to destruction of pancreatic islets [6]. The cornerstone of medical management is individualised insulin therapy, with smaller requirements compared with type 1 diabetes mellitus (T1DM) [7]. Nutritional management is also an essential parameter in the comprehensive management of the condition [8]. Screening for CFRD is required for all people with cystic fibrosis (pwCF) [9].

In recent years, the use of dipeptidyl peptidase-4 inhibitors (DPP-4is) has been considered a reliable alternative to insulin for the management of CFRD or CF with abnormal glucose tolerance [10, 11]. Dipeptidyl peptidase-4 (DPP4), glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are strongly involved in the incretin effect [12]. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have reshaped pharmacotherapy in type 2 diabetes mellitus (T2DM) [13] and are increasingly being appreciated for obesity and other conditions [14]. Recent treatment guidelines for CFRD endorse the use of GLP-1RAs, although the latter have not been adequately studied in pwCF [15]. Non-insulin pharmacological options are gaining increasing attention, given the rising rate of pwCF living with concomitant obesity, particularly those on CFTR modulator therapy with elexacaftor/tezacaftor/ivacaftor (ETI) [15]. In this rationale, as the proportion of pwCF exceeding the predefined target weight is constantly increasing, the potential of weight-lowering agents, such as GLP-1RAs, merits further exploration [15].

The aim of this review article is to summarise experimental and clinical evidence on the potential of GLP-1RAs for pwCF and CFRD. This review is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Search Strategy

An electronic search was conducted using Scopus, PubMed and Google Scholar for articles published, using combinations of the following key words: “cystic fibrosis”, “cystic fibrosis-related diabetes” and “glucagon-like peptide-1 receptor agonists”. All types of articles (clinical trials, meta-analyses, case–control studies, observational studies, cross-sectional studies, prospective/retrospective studies, cohort studies, comparative studies, randomised studies, experimental studies) were included. Only articles written in English were considered.

Effect of CF on Incretin Secretion

Pancreatic enzyme insufficiency is regarded a cardinal feature of CF. A study including five pwCF (mean age 25.8 years and mean body mass index [BMI] 20.0 kg/m²) with normal glucose tolerance (apart from one participant with impaired glucose tolerance) and six matched healthy controls examined the role of pancreatic enzyme supplementation in the incretin effect [16]. There was significant improvement in gastric emptying (p < 0.001) and glucose control (p < 0.001) [16]. Insulin levels were similar between the two groups [16]. Post-prandially, insulin-to-glucose ratio was significantly decreased in pwCF (p < 0.05), without improvement following enzyme supplementation [16]. The latter marginally improved post-prandial glucagon (p = 0.08) [16]. Baseline GLP-1 was insignificantly lower in pwCF compared with controls [16]. Nevertheless, post-prandial GLP-1 was diminished in pwCF, and this decrease was fully reversed through pancreatic enzyme supplementation [16]. GIP was similar at baseline, but post-prandially significantly decreased GIP (p < 0.001) and a lower peak value (p < 0.05) were found [16]. Enzyme supplementation was beneficial in increasing post-prandial GIP (p < 0.001), but the effect was more pronounced in the control arm (p < 0.01) [16].

Hillman et al. [17] examined the effect of CF on GLP-1: 29 pwCF, 19 subjects with CFRD and 18 individuals in the control arm received a breakfast of 23 g protein, 25 g fat and 76 g carbohydrates after an overnight fasting [17]. GLP-1 was measured at 15, 30, 45 and 60 min after meal in tubes prefilled with a DPP-4 inhibitor [17]. Significantly decreased GLP-1 was found in subjects with CFRD (p = 0.001) and in pwCF (p = 0.02) compared with the control arm [17]. GLP-1 did not differ between subjects with CFRD and pwCF (p = 0.06) [17]. C-peptide was significantly reduced in subjects with CFRD compared with both pwCF (p = 0.0001) and controls (p = 0.0001) [17]. There was no difference in C-peptide between pwCF without CFRD and controls [17].

Nyirjesy et al. [18] assessed four different treatment arms: 16 subjects with abnormal glucose tolerance (impaired glucose intolerance or CFRD) receiving GLP-1 (1.5 pmol/kg/min), 16 subjects with abnormal glucose tolerance or CFRD receiving GIP (4 pmol/kg/min), 4 pwCF, and 4 controls [18]. At baseline, subjects in all groups had comparable median age (18–43 years), BMI (18–34 kg/m²), glycated haemoglobin (HbA_1c) (4.8–6.2%), forced expiratory volume in the first second (FEV1) (32–44%) and forced vital capacity (FVC) (51–123%) [18]. Six subjects were on CFTR modulator treatment (ivacaftor or lumacaftor/ivacaftor) [18]. They used a hyperglycaemic clamp with a variable infusion rate of 20% glucose to achieve certain glucose levels (230 mg/dl and 340 mg/dl) [18]. Glucose was significantly lower following GLP-1 administration compared with GIP administration (p < 0.001) [18]. The same effect was observed in the hyperglycaemic clamp (p < 0.001) [18]. Insulin at 30 min following GLP-1 administration was insignificantly (p = 0.057) higher compared with 30 min following GIP administration [18]. During hyperglycaemic clamp, insulin and C-peptide were significantly increased (p < 0.001 and p < 0.001, respectively) in pwCF with concomitant pancreatic insufficiency and abnormal glucose tolerance receiving GLP-1 compared with controls [18]. These results were independent of CFTR modulator therapy, suggesting a potential neutral effect of the former on incretin secretion [18]. Certainly, the effect of ETI on incretin secretion merits further investigation and a potentiating effect would be hugely beneficial for pwCF. Glucagon was decreased at 30 min following GLP-1 administration and increased at 30 min following GIP administration (p < 0.001) [18].

Another double-blind randomised crossover trial focused on the effect of pancreatic enzyme replacement therapy on incretin hormones in adolescents with CF (mean age 13.1 years, mean FEV1 92.4%; two of them had CFRD) and seven adolescents in the control arm [19]. Pancreatic enzyme replacement therapy improved post-prandial glucose (p = 0.0005) [19]. Similar trends were noted at 240 min (p = 0.0002) [19]. Pancreatic enzyme replacement therapy significantly slowed gastric emptying (p = 0.003) without any correlation to GLP-1 or GIP levels [19]. Fasting GLP-1 (but not GIP) was significantly higher in the control arm (p = 0.002) [19]. Pancreatic enzyme replacement therapy increased the area under the curve [AUC] of GLP-1 (p = 0.009), so that it did not differ from controls (p = 0.18) [19]. This therapy increased post-prandial glucagon (p < 0.0001) but not insulin (p = 0.41) [19].

Lei et al. [20] presented a rationale for the therapeutic potential of incretin effect mediators in bone disease among pwCF based on a secondary analysis of a mechanistic double-blinded randomised placebo-controlled crossover trial with two groups. Group A included 25 pwCF aged 18–40 years with mean age 27.1 years, mean BMI 23.3 kg/m² and mean HbA_1c 5.6%; group B included three individuals without CF [20]. Following infusion, participants were randomly assigned to receive either a subsequent incretin infusion or placebo [20]. The primary endpoint was the effect of GIP infusion at 80 min and of GLP-1 infusion at 60 min on C-terminal telopeptide (CTX), a bone resorption marker [20]. In group A, significantly decreased CTX during GIP infusion was seen (p < 0.01) [20]. CTX at 30 min (p = 0.0002) and at 80 min (p = 0.012) was significantly lower compared with 5 min prior to infusion [20]. At the time span from 5 min prior to the infusion to 30 min after its initiation, significantly greater CTX increase was found with GIP infusion compared with placebo (p = 0.005) [20]. Significance was not confirmed for the time span to 80 min (p = 0.13) [20]. No such change could be observed, when comparing the group receiving GLP-1 with the control arm [20]. These preliminary outcomes support the emerging potential of GIP as a bone anti-resorptive agent in pwCF [20].

Experimental Research

A histological study focused on pancreas-specific effects of GLP-1 in CF by assessing formalin-fixed paraffin-embedded pancreas specimens from children with CF, with CFRD and children without CF [21]. There were no differences in mean islet area between the study groups [21]. After stratifying tissue samples according to glucagon per islet area, the authors observed that the latter was elevated in CFRD compared with samples from controls (p = 0.01) [21]. Glucagon-like peptide-1 receptor (GLP-1R) expression per islet area (p = 0.004) and insulin per islet area (p = 0.03) were reduced in pwCFRD vs. controls [21] (Table 1).

Table 1.

Experimental studies on the effect of GLP-1RAs in CF

Study	Experimental model and study design	Major outcomes	Conclusions
Gharib et al. (2024) [21]	Formalin-fixed paraffin-embedded pancreas specimens 10 pwCF without CFRD 9 pwCF with CFRD 9 subjects without CF	Mean islet area was similar between groups, while glucagon per islet area was elevated in CFRD (Control 15.1 ± 2.3, CF 20.8 ± 3.1 and CFRD 29.5 ± 3.1%; p = 0.53 Control vs. CF, p = 0.01 Control vs. CFRD) Reduced GLP-1R by 77% area per islet area in pancreas from CFRD compared with control arm Significantly reduced insulin per islet area in CFRD compared with control arm (Control 52.5 ± 3.9 vs. CFRD 36.2 ± 4.9%, p = 0.03) Significantly reduced GLP-1R expression per insulin-positive islet area in CFRD (Control 36.2 ± 2.7 vs. CFRD 13.5 ± 3.3%, p = 0.004) GLP-1 from α-cells enhanced insulin secretion through binding to GLP-1R on β-cells 216 genes were significantly differentially expressed in α-cells (114 up, 102 down, p < 0.01) and 272 genes were significantly differentially expressed in β-cells (119 up, 153 down, p < 0.01) in CFRD In pwCFRD, upregulation of α-cell PCSK1N enzyme PC1/3 responsible for GLP-1 production and downregulation of PCSK1N	GLP-1R restoration may be beneficial for CFRD
Khan et al. (2019) [22]	C57BL/6 mice	CFTR inhibition was associated with non-significant weight gain for mice on CFTRi, but weight gain was significantly lower compared with control mice (60.3 ± 31.17% reduction, p < 0.01) Treatment with CFTRi and STZ significantly reduced fat mass, without significant effect on bone mineral density or lean mass No impact on glucose tolerance was found with CFTRi administration CFTRi administration reduced plasma insulin (36.40 ± 9.62% reduction, p < 0.01) and pancreatic insulin content (39.42 ± 9.74% reduction, p < 0.01), without any effect on glucagon and GLP-1 CFTRi administration was significantly correlated with islet area (69 ± 20% reduction, p < 0.001), β-cell area (72 ± 20% reduction, p < 0.001), and non-β-cell area (56 ± 23% reduction, p < 0.001) Endocrine/exocrine ratio was significantly lower compared with control arm (3.4 ± 0.6% vs. 7.7 ± 0.8%, p < 0.001) Histology: reduction (p < 0.001) in islet size and increase (p < 0.01) in ductal area	Islet cell number is maintained and treatment in CF should address β-cell mass preservation
Voetmann et al. (2025) [23]	C57BL/6 mice GLP-1R KO mice	682 genes and 1212 were affected by treatment with liraglutide at 2 h and 4 h, respectively A twofold upregulation of CFTR and fivefold upregulation of MUC5B was observed following liraglutide treatment Ren1, Vldlr and Il33 were upregulated Malat1 silencing was found in Brunner’s gland as a target of GLP-1 conjugated antisense oligonucleotide	Increased mucus response was found with GLP-1

CF cystic fibrosis, CFRD cystic fibrosis-related diabetes, GLP-1R glucagon-like peptide-1 receptor, GLP-1 glucagon-like peptide-1, PCSK1N proprotein convertase subtilisin/kexin type 1 inhibitor, PC1/3 proprotein convertase 1, CFTR cystic fibrosis transmembrane conductance regulator, CFTRi cystic fibrosis transmembrane conductance regulator inhibitor, STZ streptozocin, KO knockout

Khan et al. [22] studied pancreas-specific effects of CFTR inhibition in C57BL6 mice using CFTRinh172 [22]. CFTR inhibition resulted in increased weight gain, significantly reduced fat without any detrimental effects on bone mineral density or lean mass. It had no effect on glucose tolerance [22]. It reduced plasma insulin (p < 0.01) and overall pancreatic insulin content (p < 0.01) without changes in glucagon and GLP-1 [22]. Insulin release from islets was significantly increased (p < 0.001) [22]. However, there was no effect on messenger ribonucleic acid (mRNA) levels of key insulin-sensing genes (SCLC2A2, GCK, ABCC8, KCNJ11) from CFTRi-treated mice [22]. Decreased endocrine/exocrine ratio (p < 0.001) was seen [22]. Finally, histology showed significantly decreased islet size (p < 0.001) and increased ductal area (p < 0.01) in CFTRi-treated mice [22].

The effect of the glucagon-like peptide-1 receptor agonist (GLP-1RA) liraglutide on CFTR gene has also been studied [23]. A twofold upregulation of CFTR and fivefold upregulation of MUC5B, encoding mucin 5b, was observed following liraglutide treatment [23]. Ren1, Vldlr and Il33 were also upregulated [23]. Malat1 silencing was found in Brunner’s gland as a target of GLP-1 conjugated antisense oligonucleotide targeting Malat1 [23].

Clinical Studies

Clinical evidence on the use of GLP-1RAs in pwCF and/or pwCFRD is extremely limited and largely relies on case reports and case series (Table 2).

Table 2.

Clinical studies on the effect of GLP-1RAs in pwCF

Study	Study type	Study duration	Study population and design	Major outcomes	Conclusions
Geyer et al. (2019) [24]	Double-blind randomised crossover trial	2 days	6 pwCF with mean age of 17.1 years (aged between 11 and 24 years), BMI z-score of − 0.56 (− 1.15 to 0.13) and mean FEV1% of 82.25 (48–98.9)	Significantly lower AUC240 was found for blood glucose (1431 ± 54 vs. 1814 ± 109 mmol/L/min, p < 0.0001) and for peak glucose (7.65 ± 0.34 vs. 9.53 ± 0.63 mmol/L, p < 0.0001) for exenatide users compared with the control arm Significantly lower AUC240 was found for insulin (2516 ± 293 vs. 3581 ± 422 mU/L/min, p = 0.0016), C-peptide AUC240 (188,186 ± 24,103 vs. 264,763 ± 25,629 ng/dL/min, p < 0.0001), GLP-1 (4459 ± 561 vs. 5983 ± 382 pmol/L/min, p = 0.0083) and GIP AUC240 (8264 ± 1006 vs. 10,045 ± 846 pmol/L/min, p = 0.0045), without difference for glucagon Significantly delayed gastric emptying was found based on 10% gastric emptying (tlag 138.8 ± 33.0 vs. 54.7 ± 5.9 min, p = 0.0044), peak 13CO2 excretion (tmax 320.85 ± 59.6 vs. 143.8 ± 15.1 min, p = 0.0007) and gastric half-emptying (t1/2 357 ± 62 vs. 236 ± 48 min, p = 0.0757) on exenatide	Exenatide may be considered a therapeutic option for pwCF
Ahmed et al. (2024) [25]	Case series	24 months	5 pwCF aged between 22 and 63 years treated with semaglutide or exenatide	A 22-year-old female subject on ETI with baseline weight 127.6 kg, baseline BMI 45.1 kg/m2, baseline HbA1c 6.9% and baseline FEV1% 105 experienced reductions in weight (102.97 kg, − 19.3%), BMI (36.7 kg/m2) and baseline FEV1% (100) and an increase in HbA1c (13.9%) following semaglutide treatment for 24 months A 25-year-old male subject not on ETI with baseline weight 111.5 kg, baseline BMI 32.1 kg/m2, baseline HbA1c 8.9% and baseline FEV1% 82 experienced reductions in weight (103.8 kg, − 6.9%), HbA1c (8.5%), and increases in BMI (32.7 kg/m2) and FEV1% (94) following exenatide treatment for 24 months A 44-year-old male subject on ETI with baseline weight 109.4 kg, baseline BMI 33.7 kg/m2, baseline HbA1c 6.8% and baseline FEV1% 88 experienced reductions in HbA1c (6.6%) and baseline FEV1% (83) and an increase in weight (117.9 kg, + 7.8%) and BMI (36.3 kg/m2) following semaglutide treatment for 24 months A 63-year-old male subject on ETI with baseline weight 195.2 kg, baseline BMI 57 kg/m2, baseline HbA1c 8.6% and baseline FEV1% 54 experienced reductions in weight (178 kg, − 8.8%), BMI (50.4 kg/m2) and HbA1c (7.1%) and an increase in FEV1% (62) following semaglutide treatment for 24 months A 53-year-old female subject on ETI with baseline weight 86.6 kg, baseline BMI 31.78 kg/m2, baseline HbA1c 8.7% and baseline FEV1% 36 experienced reductions in weight (71.2 kg, − 17.7%), BMI (25.5 kg/m2) and HbA1c (6.0%) and an increase in FEV1% (52) following semaglutide treatment for 24 months	The therapeutic potential of GLP-1RAs in CF is notable
Park et al. (2025) [26]	Case series	6 months (median 1–50 months)	11 pwCF aged 22–63 years with median BMI 32.9 kg/m2 (24–47 kg/m2) treated with semaglutide or tirzepatide	Weight loss was found in all participants with median weight loss of 7.2 kg (− 2.1 kg to − 20.2 kg) and reduced BMI by 3 kg/m2 (− 0.9 to − 8.1 kg/m2) 8 out of 11 subjects experienced improvement in FEV1% (median + 3%, − 5 to + 18) and 9 out of 11 subjects improvement in FVC% (median + 6%, + 1 to + 26) Total insulin dose was reduced by 31.5% (mean) in 7 subjects with CFRD TIR was improved by 18.5% (mean) (range + 13% to + 22%) in 5 subjects with CFRD 2 subjects discontinued treatment due to severe nausea/vomiting, 1 subject due to self-perceived lack of benefit and 1 subject due to insurance transition	Weight loss and insulin dose reduction were achieved by all subjects treated with GLP-1RAs
Horvit et al. (2025) [27]	Cohort study	16 months (3–50 months)	13 pwCF aged between 23 and 46 years with median BMI of 26.2 kg/m2 treated with semaglutide or tirzepatide	Participants experienced a median weight loss of 11.1 kg (IQR 8.5–18.8 kg) corresponding to 15.5% (IQR − 21.6 to − 13.6%) At baseline, 2 subjects had normal weight, 8 were overweight and 3 were obese Median BMI was reduced from baseline to 21.7 kg/m2 FEV1 increased by 10.8% (IQR 7.3–19.4) in 11 participants FVC increased by 10.6% (IQR 6.0–13.4%) Decreases by 16.7% in HbA1c were noted; HbA1c was reduced from 7.0% (IQR 6.1–7.7%) to 6% (IQR 5.8–6.5%)	Promising outcomes were reported with GLP-1RAs in pwCF
Sydney et al. (2025) [28]	Case reports	N/A	(1) A 62-year-old white female subject with CF (heterozygous carrier of the delta F508 deletion of the CFTR gene) treated with liraglutide/dulaglutide, metformin, repaglinide, canagliflozin, basal and prandial insulin (2) A 65-year-old white subject with CF (9 poly-T allele pathogenic variant (5 T/7 T/8 T) and concomitant haemochromatosis treated with liraglutide, dulaglutide, oral semaglutide, metformin, repaglinide, and empagliflozin	Acute pancreatitis may be attributed to treatment with GLP-1RAs	pwCF treated with GLP-1RAs may have increased risk of acute pancreatitis
Gnanapragasam et al. (2020) [29]	Case report	6 months	21-year-old male subject with CFRD treated with semaglutide 0.13 to 0.16 mg, insulin glargine and lispro	HbA1c decreased from 9.1% to 6.8% following treatment initiation with semaglutide Improvement was noted in CGM metrics (TIR 68%, TAR 32%), especially after uptitration of semaglutide (TIR 77%, TAR 23%) Mean glucose was 142 mg/dl (SD 51 mg/dL) C-peptide was mildly increased following treatment (0.77 ng/mL vs. 0.64 ng/mL) There was no nausea, upper abdominal pain/discomfort, or diarrhoea Weight was mildly reduced from 77 to 75 kg	Semaglutide treatment may fully replace prandial insulin
Auth et al. (2025) [30]	Case report	800 days	32-year-old male subject on ETI with poorly controlled CFRD with BMI of 38.5 kg/m2 and baseline HbA1c between 9.4% and 11.4%	HbA1c decreased from 10.5% to 5.7% and BMI to 33.4 kg/m2 CGM-based mean glucose decreased from 278 to 129 mg/dl Insulin total daily dose decreased from 230 to 90 units	Semaglutide was associated with positive effects without any detrimental changes in pulmonary function

CF cystic fibrosis, pwCF people with cystic fibrosis, BMI body mass index, IQR interquartile range, FEV1 forced expiratory volume in the first second, FVC forced vital capacity, HbA_1c glycated haemoglobin, GLP-1RAs glucagon-like peptide 1 receptor agonists, CFTR cystic fibrosis transmembrane conductance regulator, ETI elexacaftor/tezacaftor/ivacaftor, AUC₂₄₀ area under the curve over 240 min, SE standard error, SD standard deviation, CFRD cystic fibrosis-related diabetes, TAR time above range, TIR time in range, CGM continuous glucose monitoring, N/A not applicable

In a double-blind randomised crossover trial, Geyer et al. [24] evaluated the therapeutic potential of exenatide in six pwCF with mean age of 17.1 years (from 11 to 24 years) [24]. Over 2 days, subjects received either exenatide 2.5 μg or placebo (0.9% saline) subcutaneously 15 min before the ingestion of a pancake meal labelled with¹³C-octanoate in addition to pancreatic enzyme replacement following overnight fasting [24]. Exenatide treatment resulted in significantly lower area under the curve over 240 min (AUC₂₄₀) for blood glucose (p < 0.0001) and for peak glucose (p < 0.0001) compared with the control arm [24]. Significantly lower AUC₂₄₀was seen for insulin (p = 0.0016), C-peptide (p < 0.0001), GLP-1 (p = 0.0083) and GIP (p = 0.0045) [24]. Gastric emptying was significantly delayed vs. controls, as based on three indices: 10% gastric emptying (t_lag 138.8 ± 33.0 vs. 54.7 ± 5.9 min, p = 0.0044), peak ¹³CO₂ excretion (t_max 320.85 ± 59.6 vs. 143.8 ± 15.1 min, p = 0.0007) and gastric half-emptying (t_1/2 357 ± 62 vs. 236 ± 48 min, p = 0.0757) [24].

In a case series, two male and two female pwCF treated with semaglutide and one male subject treated with exenatide were monitored for 24 months [25]. All had CFRD with pancreatic insufficiency [25]. Four of them were on ETI for a period ranging from 6 to 15 months [25]. One participant had achieved a significant 10% increase in body weight with ETI before initiating GLP-1RAs [25]. The other participants had experienced a markedly lower weight gain, ranging from 0.3% to 3% [25]. Of note, this case series study included individuals with very high weight reaching even 195.2 kg (corresponding to a BMI of 57 kg/m²) [25]. The greatest absolute weight loss reported was 24.63 kg corresponding to 19.3% weight loss and a reduction in BMI by − 8.4 kg/m² [25]. One subject on semaglutide gained 8.5 kg (+ 7.8%) [25]. Increases in BMI were reported not only for the latter (+ 2.6 kg/m²) but also for the participant on exenatide (+ 0.6 kg/m²) who experienced a more moderate weight loss of 7.7 kg (6.9%) [25]. Reduced HbA_1c from 0.2% to 1.7% was found in all participants except for a 7% increase in HbA_1c in the participant experiencing the greatest weight loss [25]. Changes in lung function were highly variable, ranging from a decrease by 5% to an increase by 16% [25]. Improvement was found for subjects with relatively low FEV1% at baseline [25]. Treatment discontinuation was reported for one participant due to diminished appetite and nausea [25]. Insulin requirements either decreased or remained stable [25].

Another case series evaluated semaglutide and tirzepatide in 11 pwCF over a median period of 6 months (from 1 to 50 months): 7 participants had concomitant CFRD [26]. All participants had a BMI at baseline corresponding to overweight or obesity. Median BMI was 32.9 kg/m² [26]. Two pwCF had homozygous F508del mutations, seven had F508del heterozygous mutations, and two had other mutations [26]. With the exception of two pwCF with lung transplantation, all pwCF with F508del mutations received ETI treatment [26]. One subject discontinued treatment with ETI because of weight gain of 16 kg [26]. Weight gain was attributed mainly to ETI, while three subjects received prednisone following lung transplantation [26]. Regarding semaglutide and tirzepatide dosing, lower doses than those usually prescribed were given as a result of adequate weight loss and in order to minimise the risk for gastrointestinal adverse events [26]. Weight loss was achieved by all participants. Median weight loss was 7.2 kg (from 2.1 to 20.2 kg) [26]. Maximum weight loss was achieved by a participant on GLP-1RA for 50 months [26]. No other weight-lowering agents were used [26]. Reduced BMI by 3 kg/m² (from − 0.9 to − 8.1 kg/m²) was also observed [26]. Moreover, improvement in FEV1% was found for 8 out of 11 pwCF (median + 3%, − 5 to + 18) and in FVC for 9 out of 11 pwCF (median + 6%, + 1 to + 26) [26]. For one subject with a high FEV1% value at baseline (137%), a negligible decline by 2% was reported following GLP-1RAs treatment [26]. Decreases by 1% in FVC were reported for two participants as well [26]. No participant on GLP-1RAs experienced more than one pulmonary exacerbation requiring outpatient management or hospitalisation for a period over 6 months [26]. Decreased total insulin dose by 31.5% (mean) was observed in seven subjects with CFRD. Continuous glucose monitoring (CGM) showed increased time-in-range (TIR) by 18.5% (mean) (range 13–22%) in five subjects with CFRD [26]. Time with high glucose was reduced (median − 15.5%, range − 12% to − 17%) and the same was seen for time with very high glucose (median − 5.5%, range − 1% to − 16%) [26]. The effects on HbA_1c were not so pronounced: four subjects experienced a decrease ranging from 0.3% to 2.6%, and three subjects experienced slight increases between 0.1% and 0.4% [26]. No hypoglycaemic events were reported [26]. Two subjects discontinued treatment as a result of severe nausea and vomiting, one subject because of self-perceived lack of benefit and one subject because of insurance issues [26]. One participant on 0.5 mg semaglutide weekly was admitted to hospital for nausea and vomiting: after exclusion of acute pancreatitis, a dose reduction to 0.25 mg semaglutide weekly was decided [26]. Another participant could not sustain treatment with either semaglutide (for 23 months) or dulaglutide (for 19 months) [26]. No incident of acute pancreatitis was reported [26].

A recent cohort study monitored the therapeutic effect of semaglutide and tirzepatide for a median period of 16 months [27]. It included 13 pwCF (12 female subjects, all non-Hispanic), 11 of whom had concomitant CFRD [27]. Overall, 10 participants were on ETI therapy over a median period of 43.4 months [27]. Seven subjects received tirzepatide (2.5 mg, 5 mg and 10 mg) and 6 subjects semaglutide (0.25 mg, 1 mg, 1.5 mg and 2 mg) [27]. Median weight loss was 11.1 kg (interquartile range [IQR] 8.5–18.8 kg) corresponding to a percentage weight loss of 15.5% (IQR − 21.6% to − 13.6%) [27]. With GLP-1RAs, all subjects had normal weight by the end of the study period [27]. At baseline, subjects with normal BMI received GLP-1RAs for optimising glycaemic control [27]. HbA_1c decreased by 1.0% corresponding to a decrease by 16.7% [27]. Median BMI was reduced from 26.2 to 21.7 kg/m² [27]. Importantly, most subjects on GLP-1RAs experienced improvement in lung function: FEV1 increased by 10.8% (IQR 7.3–19.4%) in 11 out of 13 participants. FVC was increased by 10.6% (IQR 6.0–13.4%) [27]. However, one subject receiving the lowest semaglutide dose (0.25 mg) experienced a slight decrease in FEV1 [27]. Both subjects with a history of lung transplantation experienced benefit in lung function [27].

Two further case reports focused on adverse outcomes with GLP-1RAs in CFRD [28]. The first subject received liraglutide/dulaglutide, metformin, repaglinide, canagliflozin, basal and prandial insulin [28]. The second subject received liraglutide, dulaglutide, oral semaglutide, metformin, repaglinide, and empagliflozin [28]. Both experienced acute pancreatitis [28].

The beneficial effects of semaglutide in carefully selected individuals with CFRD were shown in two additional case reports [29, 30] (Table 2).

Discussion

This review summarises evidence on the therapeutic potential of GLP-1RAs and of the dual agent tirzepatide in pwCF and CFRD and provides insights into practical recommendations and arising considerations for clinical practice.

Studies have examined the incretin effect and its major mediators in pwCF [16–20]. It is well established that pancreatic insufficiency is encountered in about 85% of people with CF [1]. Accordingly, pancreatic enzyme supplementation is often required [1]. Accumulating evidence points to significantly increased GLP-1 and GIP secretion following pancreatic enzyme supplementation [16, 19]. Supplementation improved gastric emptying and reduced post-prandial hyperglycaemia [16, 19]. The latter is associated with restored glucagon and GLP-1 secretion and with delayed gastric emptying [19]. The presence of active GLP-1 may be considered an indicative marker for progression to CFRD [17]. Importantly, GLP-1 may improve glucose-dependent insulin secretion, thus strengthening the clinical perspective of GLP-1RAs in pwCF and CFRD [18].

Several GLP-1RAs have been investigated: exenatide [24, 25], semaglutide [24, 26, 27], dulaglutide [26–28], liraglutide [28] and tirzepatide [26, 27]. In general, treatment discontinuation was not very common. Reasons for discontinuation included diminished appetite and nausea, insurance issues, accelerated weight loss, planned pregnancy and acute pancreatitis [25–28]. Slowed gastric emptying was confirmed in a clinical trial of three subjects on exenatide [24]. These adverse events are consistent with the safety profile of GLP-1RAs [31–33].The gastrointestinal adverse effects of GLP-1RAs pose an additional burden. They range from mild conditions, such as gastrointestinal reflux disease, to severe conditions, such as pancreatitis [34–36]. Common disorders include oesophageal dysmotility and subsequent gastroparesis, small intestine dysmotility, distal idiopathic obstruction syndrome, constipation small intestinal bacterial overgrowth syndrome, flatulence, abdominal distension, fatty stool, and elevated risk for gastrointestinal malignancies [37–40]. The simultaneous presence of CFRD might further exacerbate the burden of gastrointestinal manifestations [41].

Nutritional status is another crucial parameter for pwCF, and so caution is required, because GLP-1RAs may indirectly provoke nutritional deficiencies [42, 43]. There has also been a recent report of a 45-year-old man with CFRD carrying a homozygous ΔF508 mutation, who received ETI and pancreatic enzyme supplementation and began treatment with 0.25 mg semaglutide [44]. He suffered distal intestinal obstruction syndrome twice within weeks [44]. While pwCF are predisposed to such gastrointestinal adverse events, drugs such as semaglutide may increase this risk and thus particular caution is required [45, 46].

The relationship between GLP-1RAs use and acute pancreatitis prevalence is ambiguous, and the latest evidence even suggests a protective effect [47–49]. Granted that recent reports have documented acute pancreatitis cases for pwCF on CFTR modulator therapy, GLP-1RA treatment should be utilised even more cautiously [50, 51]. Physicians are encouraged to gradually adjust and titrate doses of GLP-1RAs in order to minimise adverse effects [52]. GLP-1RAs may be safely titrated among pwCF, as shown in case reports and case series [52].

Furthermore, pulmonary exacerbations pose a major clinically diverse challenge to pwCF and may require hospitalisation [53–55]. Hyperglycaemia commonly accompanies such exacerbations [56]. On the basis of currently available knowledge on the use of GLP-1RAs in acute conditions, physicians are encouraged to discontinue treatment with GLP-1RAs [57].

For pwCF and/or CFRD, GLP1-RAs are used at smaller doses compared with T2DM, in order to minimise adverse events and to maximise compliance [26, 29, 30]. This strategy has proved successful in most cases reported, and so a conservative and cautious stepwise uptitration may be optimal in pwCF. The study on exenatide was conducted in young pwCF, the youngest being 11 years of age, and yielded promising outcomes [24]. Accumulating evidence suggests that certain GLP-1RAs may be safely used for children older than 6 years of age, and this consideration should be addressed for children with CF as well [58–60].

Use of GLP-1RAs was associated with beneficial effects on weight management and glycaemic control [25–27]. Weight loss varied considerably across studies. Maximum loss was 24.6 kg in absolute terms (21.6% of baseline weight) and the greatest decrease in BMI was 8.4 kg/m² [25–27]. Changes in HbA_1c varied considerably, with most individuals experiencing slight decreases [25, 27]. Decreased insulin dose or even discontinuation of prandial insulin as well as reductions of time with high glucose and time with very high glucose have also been documented [26, 29, 30]. The use of CGM has not been widely studied among pwCF, and therefore clinical decisions regarding CFRD management should not be exclusively based on CGM metrics [61]. An effect of ETI on CGM metrics was considered unlikely in a 3-month and 6-month period after treatment initiation [62].

Limited evidence from two case series is available on the potential of GLP-1RAs for pwCF without concomitant CFRD [26, 27]. In the first series, 4 out of 11 without CFRD experienced weight reduction by 2.1–11.7 kg (corresponding to decreases by 2.0–14.3% from baseline weight and reductions in BMI by 0.9–4.1 kg/m²) [26]. The most pronounced weight decrease among subjects without CFRD was seen in the single participant on ETI [26]. The same held true for the improvement in pulmonary function (increase in FEV1% by 7%) [26]. In the second cases series, two participants on 10 mg tirzepatide without CFRD, BMI was reduced by 2.5–4.7 kg/m² and FEV1% was increased by 4–9%, pointing to promising outcomes [27].

Beyond glycaemic control, use of GLP-1RAs among pwCF correlated with improvements in lung function, as evidenced by increased FEV1% and FVC [25–27]. Nevertheless, inconsistent findings were found for subjects with very high baseline FEV1% [27]. Furthermore, the incidence of pulmonary exacerbations was decreased, particularly for subjects on long-term use of GLP-1RAs [26, 30]. Improved lung function was also seen in subjects with prior lung transplantation [27].

Treatment with ETI has decisively reshaped the therapeutic landscape for pwCF, offering a unique treatment choice for subjects with a single Phe508del CFTR mutation and a minimal function CFTR mutation [63]. Despite the enormous therapeutic benefits, the use of ETI has been associated with increased body weight in multiple studies [64–67]. An annual increase in BMI by 1.47 kg/m²with ETI has been suggested [64]. Increased frequency of metabolic syndrome for subjects on ETI has been reported as well, suggesting that an increase in cardiovascular morbidity cannot be excluded [65]. Some studies demonstrated that ETI therapy increases fat mass [66, 67].

Beyond CFRD, cystic fibrosis-related bone disease (CFBD) is also commonly encountered [68]. A study has reported that GIP infusion had a beneficial effect on CTX, thus showing the inhibitory effect of GIP in bone resorption in pwCF [20]. Such potential therapeutic benefits of GLP-1RAs on CFBD merit further investigation.

The strength of this review is discussion of available clinical and basic research in pwCF. Nevertheless, there are certain limitations. First, evidence largely relies on case reports and very small case series. No randomisation has been applied. Secondly, study duration was relatively short and highly variable. Finally, ethnic diversity was not considered. Consequently, large multicentre randomised prospective clinical trials with long-term follow-up are needed to confirm encouraging results and to clarify which subgroups of pwCF will get the greatest benefit from GLP-1RAs.

Currently, a few clinical trials are ongoing to elucidate the exact utility of GLP-1RAs in pwCF and pwCFRD. A phase 2, randomised, open-label, cross-over trial has enrolled to date about 30 participants treated with 0.75 mg dulaglutide to assess insulin secretion in adult individuals with pre-existing pancreatic insufficiency (NCT04731272). Another open-label, single-arm pilot study will assess the effect of semaglutide as an add-on treatment to insulin in individuals with CFRD and concomitant overweight/obesity (NCT05788965). An additional trial will examine the response of beta cells to GLP-1 and GIP (NCT01851694). An interesting new vista would be assessment of other dual agonists, such as glucagon receptor and GLP-1RAs [69, 70], as well as of triple agonists [71] in pwCF or CFRD.

Conclusions

Earliest clinical evidence supports the use of GLP-1RAs in pwCF and/or CFRD. The benefits are not restricted to glycaemic control and weight management but may be extended to pleiotropic effects, regarding lung function and CFBD. However, potential adverse events should not be ignored. These include risk of acute pancreatitis, nausea/vomiting, nutritional depletion, bowel dysmotility and distal intestinal obstruction syndrome, as well as others. Adverse events should be addressed with caution, and dose adjustments may be useful to minimise their frequency and to maximise benefits. Large randomised trials are required to incorporate these promising outcomes into clinical practice.

Author Contributions

Theodoros Panou, Evanthia Gouveri, Djordje S. Popovic and Nikolaos Papanas contributed to the article in a substantive and meaningful manner. Theodoros Panou searched and interpreted literature and wrote the first draft. Evanthia Gouveri, Djordje S. Popovic, and Nikolaos Papanas were involved in article conception and design. Nikolaos Papanas finalized the draft. Theodoros Panou, Evanthia Gouveri, Djordje S. Popovic and Nikolaos Papanas have approved the final form of this manuscript and agree to its submission.

Funding

No funding or sponsorship was received for this study or publication of this article.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Declarations

Conflict of Interest

Theodoros Panou has nothing to disclose. Evanthia Gouverihas attended conferences sponsored by Berlin-Chemie, Sanofi, AstraZeneca, Novo Nordisk, Lilly, Boehringer Ingelheim, Menarini, Petsiavas and KRKA Hellas; received speaker honoraria by AstraZeneca, Boehringer Ingelheim, Sanofi, Menarini. Djordje S. Popovic declares associations with: Abbott (speaker honoraria), ADOC Pharma (sponsored conferences) Alkaloid (speaker honoraria), Amicus (speaker honoraria, sponsored conferences), AstraZeneca (speaker honoraria), Boehringer Ingelheim (speaker honoraria), Berlin-Chemie (speaker honoraria), Eli Lilly (advisory board member, speaker honoraria, sponsored conferences), Galenika (speaker honoraria), Krka (speaker honoraria), Merck (speaker honoraria, sponsored conferences), Novo Nordisk (advisory board member, speaker honoraria, sponsored conferences, participation in sponsored studies), PharmaSwiss (speaker honoraria), Sanofi-Aventis (speaker honoraria, sponsored conferences, participation in sponsored studies), Servier (speaker honoraria), Viatris (speaker honoraria), and Wörwag Pharma (speaker honoraria, sponsored conferences). Nikolaos Papanas has been an advisory board member of AstraZeneca, Bayer, Boehringer Ingelheim, Menarini, MSD, Novo Nordisk, Pfizer, Roche, Takeda and TrigoCare International; has participated in sponsored studies by AstraZeneca, Eli Lilly, GSK, MSD, Novo Nordisk, Novartis and Sanofi-Aventis; has received honoraria as a speaker for AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, ELPEN, Galenica, KRKA, Menarini, MSD, Mylan, Novo Nordisk, Pfizer, Sanofi-Aventis, Takeda, Viatris and Vianex; and attended conferences sponsored by TrigoCare International, Bayer, Eli Lilly, Galenica, Novo Nordisk, Pfizer, Viatris and Sanofi-Aventis.

Ethical Approval

This review is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.