1. Background
Men and women with type 2 diabetes appear to respond differently to treatment with GLP‐1R agonists [1], with women showing more weight loss compared to men, whereas the glycaemic effects and cardiovascular benefits are similar [2]. The underlying mechanism explaining the sex differences in treatment response remains unclear.
Weight‐lowering effects of GLP‐1R agonists are assumed to be mediated via GLP‐1Rs in the brain, which are expressed in regions involved in the reward system, food intake and energy expenditure [3]. However, since it is unclear whether GLP‐1R agonists can cross the blood–brain barrier [4], stimulation of GLP‐1Rs in regions outside the blood–brain barrier, such as the pituitary, should be considered.
Previously, we showed [68Ga]Ga‐NODAGA‐exendin‐4 (radiolabelled exendin) uptake in the pituitary of individuals with type 2 diabetes using PET/CT, indicating the presence of GLP‐1Rs in the pituitary [5, 6]. One of these studies suggested higher uptake of radiolabelled exendin in men compared to women with type 2 diabetes [6]. We were interested in whether we could validate these sex differences in pituitary GLP‐1Rs in a larger cohort of individuals with type 2 diabetes using radiolabelled exendin PET/CT imaging. Therefore, we performed a pooled post hoc analysis to investigate sex differences in radiolabelled exendin uptake in the pituitary of individuals with type 2 diabetes.
2. Methods
2.1. Participants
Studies with individuals with type 2 diabetes who underwent [68Ga]Ga‐NODAGA‐exendin‐4 PET/CT were included in this post hoc analysis (NCT03923114, n = 19; NCT03182231, n = 10 and NCT05418907, n = 6). Inclusion and exclusion criteria, and ethical approval are described in the Supporting Information. The same methodology was applied to all data across the different studies.
2.2. PET/CT Acquisition
Participants were asked to fast for 4 h and those using short‐acting insulin were asked to stop its administration for at least 6 h prior to the scan to avoid hyperinsulinemia. Blood was collected to determine baseline fasting glucose, fasting C‐peptide and HbA1c levels. Subsequently, 100 ± 5 MBq of [68Ga]Ga‐NODAGA‐exendin‐4 was administered intravenously, followed by a PET/CT scan of the head. Details about the scanner, reconstruction and quantitative image analysis are described in the Supporting Information.
2.3. Statistical Analysis
Data were analysed using Prism (10.4.1; GraphPad Software, San Diego, CA). Independent t‐tests including Welch correction or Mann–Whitney U‐tests were performed to assess group differences. Further statistical analyses are described in the Supporting Information. p‐values < 0.05 were considered statistically significant.
3. Results
In total, 22 women and 13 men with type 2 diabetes were included for this analysis. Diabetes duration was longer in men compared to women (16(12–22) vs. 8.5(3.3–18) years, p = 0.027), but there were no differences with respect to age, BMI, HbA1c, fasting glucose, fasting C‐peptide, or the proportion treated with insulin (Table 1).
TABLE 1.
Participant characteristics.
| Variable | Women (n = 22) | Men (n = 13) | p |
|---|---|---|---|
| Age, y | 54.8 ± 9.8 | 60.1 ± 9.0 | 0.11 |
| BMI, kg/m2 | 35.9 ± 6.8 | 32.4 ± 6.0 | 0.12 |
| Diabetes duration, y | 8.5 (3.3–18.3) | 16.0 (12.0–21.5) | 0.027 |
| Fasting glucose, mmol/L | 8.8 ± 2.4 | 9.5 ± 3.3 | 0.49 |
| Fasting C‐peptide, nmol/L | 0.7 (0.4–2.1) | 1.2 (0.5–1.3) | 0.78 |
| HbA1c, mmol/mol (%) | 61.1 ± 13.5 (7.7 ± 1.0) | 61.1 ± 11.7 (7.7 ± 1.2) | 0.99 |
| Glucose lowering medication, n (%) | |||
| Insulin | 13 (54.5) | 10 (76.9) | 0.19 |
| GLP‐1RA | 5 (22.7) | 4 (30.8) | 0.11 |
| SGLT2i | 1 (4.5) | 0 (0) | 0.59 |
| SU derivatives | 7 (31.8) | 5 (38.5) | 0.078 |
| Metformin | 14 (63.6) | 12 (92.3) | 0.10 |
| Hormone replacement therapy, n (%) | 0 (0) | 0 (0) | — |
| Type 2 diabetes complications, n (%) | |||
| Retinopathy | 2 (9.1) | 5 (38.5) | 0.043 |
| Neuropathy | 0 (0) | 4 (30.8) | — |
| Nephropathy | 1 (4.5) | 2 (15.4) | 0.091 |
Note: Data presented as mean ± SD, median (IQR) or n (%). Significant associations in bold. Missing data in HbA1c (women n = 1).
Abbreviations: BMI, Body mass index; GLP‐1RA, glucagon‐like peptide‐1 receptor agonists; HbA1c, Haemoglobin A1c; SGLT2i, sodium‐glucose cotransporter 2 inhibitors; SU derivatives, sulfonylurea derivatives.
The uptake of radiolabelled exendin in the pituitary was significantly higher in men compared to women (SUVmax 5.3 (4.1–8.0) vs. 2.9 (2.1–4.7), p = 0.0054; Figure 1A,B), which did not change after correcting for diabetes duration (F(1,32) = 14.14, p = < 0.001; Table S1). The difference in medians of SUVmax between the two groups was 2.4 (95% CI [0.9–3.9]). When restricting the analysis to participants treated with insulin alone, the results were about similar with higher uptake of radiolabelled exendin in men compared to women (SUVmax 5.3 (4.3–7.5) vs. 2.8 (2.2–3.4), p = 0.030; Figure 1C). Finally, we compared pituitary tracer uptake in women below and above 55 years of age, as proxy for pre‐ and postmenopausal state, respectively, which revealed similar SUVmax in the pituitary in the two groups of women (SUVmax 2.9 (2.1–5.2) vs. 2.9 (2.1–4.1), p > 0.99; Figure 1D). If we excluded women aged between 50 and 60 to account for potential misclassification bias of the assumed menopause proxy, the results did not change (Figure S1).
FIGURE 1.
Radiolabeled exendin uptake in the pituitary of women and men with type 2 diabetes. (A) SUVmax in the pituitary in women and men with type 2 diabetes. (B) Sagittal slides of PET/CT showing uptake of radiolabeled exendin in the pituitary of a woman and man with type 2 diabetes, depicted by the white arrow. (C) SUVmax in the pituitary in women and men with type 2 diabetes on insulin treatment and (D) in women < 55 years versus ≥ 55 years as a surrogate marker for the menopausal state. SUVmax in the pituitary was determined based on a spheric volume of interest (VOI) with a radius of 5 mm on CT. Data presented as median (IQR).
Pituitary tracer uptake positively correlated with BMI in both sexes (women r = 0.60, 95% CI [0.23–0.82], p = 0.003; men r = 0.61, 95% CI [0.086–0.87], p = 0.027; Figure S2A), but this did not explain the different uptake in men compared to women. Diabetes duration showed an inverse trend with pituitary tracer uptake in both men and women, but neither correlation reached statistical significance (women r = −0.37, 95% CI [−0.69–0.079], p = 0.09; men r = −0.40, 95% CI [−0.78–0.19], p = 0.18; Figure S2B).
4. Conclusion
This post hoc analysis showed that the pituitary uptake of radiolabelled exendin was higher in men compared to women with type 2 diabetes. Although pituitary tracer uptake positively correlated with BMI in both sexes, this did not explain the difference between men and women, nor did the difference in diabetes duration play a role. In addition, pituitary tracer uptake was not influenced by insulin treatment or the assumed menopausal state in women with type 2 diabetes.
This is the first study investigating sex differences in pituitary GLP‐1Rs of people with type 2 diabetes. Given that treatment with GLP‐1R agonists has been shown to cause more weight loss in women compared to men, we expected more pituitary GLP‐1Rs in women than in men, but our results show the opposite. Therefore, these findings do not support a primary role for the pituitary to explain sex differences in weight loss upon treatment with GLP‐1R agonists, and potentially for pituitary GLP‐1Rs to be clinically relevant in mediating weight loss altogether. However, whether our findings suggest that men are less sensitive to GLP‐1 and therefore need to express more pituitary GLP‐1Rs as a compensatory mechanism requires further study.
We showed a positive correlation between BMI and pituitary tracer uptake regardless of sex. This may suggest that people with obesity express more pituitary GLP‐1Rs potentially as a compensatory response to decreased sensitivity to GLP‐1 and/or GLP‐1 expression. In individuals with type 2 diabetes and obesity, understimulation of the reward system in the central nervous system compared to healthy lean individuals has been shown, but GLP‐1R agonist treatment improved this stimulation, which was associated with weight loss [7]. In addition, individuals with obesity who were brain insulin‐resistant in the orbitofrontal cortex, a major reward area, still could respond to GLP‐1 to exert a food response [8]. These studies support that obesity may lead to upregulation of GLP‐1Rs in the brain as a compensatory mechanism. Remarkably, a previous immunohistochemical analysis on postmortem tissue showed an inverse correlation between GLP‐1R immunoreactivity in the lateral hypothalamus and BMI [9]. However, in this previous study, GLP‐1R expression was assessed through immunohistochemistry on postmortem tissue, which clearly differs from our in vivo imaging technique. The complexity of these findings indicates that further research is required to understand the association between BMI and pituitary GLP‐1Rs.
At this time, it remains challenging to find possible explanations for sex differences in pituitary GLP‐1Rs, since data about the function of pituitary GLP‐1Rs are still very limited. The pituitary is involved in several hormonal pathways. We previously excluded the role of the hypothalamus‐pituitary–adrenal axis in differential GLP‐1R expression in the pituitary [6]. Only ACTH was suggested to have a link with pituitary GLP‐1Rs. A possible explanation may be the interplay with sex hormones through the hypothalamus‐pituitary‐gonadal axis. A synergistic effect has been described between GLP‐1 and oestrogen in energy expenditure and glucose metabolism [10]. Although oestrogen receptor (ER) signalling is required for proper GLP‐1R signalling on food‐reward behaviour in both sexes [11], menopause may impact ER expression in the brain [12]. However, we showed that pituitary tracer uptake was not influenced by the assumed menopausal state in women with type 2 diabetes. Besides oestrogen, testosterone may also play a role as GLP‐1R agonist treatment can increase testosterone in men [13]. These hormonal pathways need to be further elucidated.
Our analysis has strengths and limitations. This is the first in‐human study showing sex differences in GLP‐1Rs in the pituitary with a large sample size that is relatively large for this type of PET/CT imaging studies. In addition, the groups were similar with respect to age, BMI and glycaemic parameters. Although diabetes duration differed between sexes, its correlation with pituitary tracer uptake was not statistically significant and showed an inverse trend. Given the inverse relation, any adjustment might have contributed to a stronger effect in sex differences in tracer uptake rather than diminishing this effect. The major limitation is that we do not have data regarding weight loss upon GLP‐1R agonist treatment to understand the association with pituitary GLP‐1Rs. Another limitation includes the limited hormonal data available, making it unclear whether different stages of the menstrual cycle, menopause or testosterone levels influence pituitary GLP‐1R expression. The last limitation is that we cannot differentiate between the two lobes of the pituitary on CT to assess sex‐based differences in PET signal localisation; co‐registration with MR imaging could be useful in future studies. Besides lobular localisation, cellular localisation also remains unresolved; therefore, GLP‐1R expression in different pituitary cell populations should be assessed in post‐mortem tissue.
In conclusion, our data on the uptake of radiolabelled exendin in the pituitary do not support a major role for pituitary GLP‐1Rs in explaining sex differences following GLP‐1R agonist treatment. Further research is required to better understand the underlying mechanisms that may contribute to these sex differences.
Funding
This study received institutional funding from the Radboud University Medical Center, Nijmegen, the Netherlands.
Conflicts of Interest
Rick I. Meijer is supported by a Dutch Diabetes Foundation Senior Fellowship. The other authors declare no conflicts of interest.
Supporting information
Data S1: dom70600‐sup‐0001‐Supinfo.docx.
Data Availability Statement
The datasets analysed during the current study are available from the corresponding author on reasonable request.
References
- 1. Börchers S. and Skibicka K. P., “GLP‐1 and Its Analogs: Does Sex Matter?,” Endocrinology 166, no. 2 (2025): bqae165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Rentzeperi E., Pegiou S., Koufakis T., Grammatiki M., and Kotsa K., “Sex Differences in Response to Treatment With Glucagon‐Like Peptide 1 Receptor Agonists: Opportunities for a Tailored Approach to Diabetes and Obesity Care,” Journal of Personalized Medicine 12, no. 3 (2022): 454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kabahizi A., Wallace B., Lieu L., et al., “Glucagon‐Like Peptide‐1 (GLP‐1) Signalling in the Brain: From Neural Circuits and Metabolism to Therapeutics,” British Journal of Pharmacology 179, no. 4 (2022): 600–624. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Buller S. and Blouet C., “Brain Access of Incretins and Incretin Receptor Agonists to Their Central Targets Relevant for Appetite Suppression and Weight Loss,” American Journal of Physiology. Endocrinology and Metabolism 326, no. 4 (2024): E472–E480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Deden L. N., Booij J., Grandjean J., et al., “Brain Imaging of the GLP‐1 Receptor in Obesity Using 68Ga‐NODAGA‐Exendin‐4 PET,” Brain Sciences 11, no. 12 (2021): 1647. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Tokgöz S., Boss M., Jansen T. J. P., et al., “Activation of the HPA Axis Does Not Explain Nonresponsiveness to GLP‐1R Agonist Treatment in Individuals With Type 2 Diabetes,” Diabetes 74, no. 2 (2024): 212–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. ten Kulve J. S., Veltman D. J., van Bloemendaal L., et al., “Endogenous GLP1 and GLP1 Analogue Alter CNS Responses to Palatable Food Consumption,” Journal of Endocrinology 229, no. 1 (2016): 1–12. [DOI] [PubMed] [Google Scholar]
- 8. Heni M., Kullmann S., Gallwitz B., Häring H. U., Preissl H., and Fritsche A., “Dissociation of GLP‐1 and Insulin Association With Food Processing in the Brain: GLP‐1 Sensitivity Despite Insulin Resistance in Obese Humans,” Molecular Metabolism 4, no. 12 (2015): 971–976. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Psilopanagioti A., Nikou S., Logotheti S., Arbi M., Chartoumpekis D. V., and Papadaki H., “Glucagon‐Like Peptide‐1 Receptor in the Human Hypothalamus Is Associated With Body Mass Index and Colocalizes With the Anorexigenic Neuropeptide Nucleobindin‐2/Nesfatin‐1,” International Journal of Molecular Sciences 23, no. 23 (2022): 14899. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Finan B., Yang B., Ottaway N., et al., “Targeted Estrogen Delivery Reverses the Metabolic Syndrome,” Nature Medicine 18, no. 12 (2012): 1847–1856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Richard J. E., Anderberg R. H., López‐Ferreras L., Olandersson K., and Skibicka K. P., “Sex and Estrogens Alter the Action of Glucagon‐Like Peptide‐1 on Reward,” Biology of Sex Differences 7, no. 1 (2016): 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ishunina T. A. and Swaab D. F., “Alterations in the Human Brain in Menopause,” Maturitas 57, no. 1 (2007): 20–22. [DOI] [PubMed] [Google Scholar]
- 13. Salvio G., Ciarloni A., Ambo N., et al., “Effects of Glucagon‐Like Peptide 1 Receptor Agonists on Testicular Dysfunction: A Systematic Review and Meta‐Analysis,” Andrology 13, no. 8 (2025): 2022–2034. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: dom70600‐sup‐0001‐Supinfo.docx.
Data Availability Statement
The datasets analysed during the current study are available from the corresponding author on reasonable request.