The Brain-Body Axis at SfN 2026: GLP-1, Addiction, and Alzheimer's Research


At the Society for Neuroscience (SfN) annual meeting in Washington, D.C., two events explored how signals from the body may shape brain function. A roundtable on obesity and addiction focused on GLP-1, whose role in both appetite regulation and drug reward has helped spur interest in incretin therapies for opioid use disorder [1].
A separate SfN briefing approached the brain-body axis through Alzheimer’s disease [2]. Earlier this year, two trials reported that semaglutide did not slow clinical decline in early Alzheimer’s disease [3]. The work presented at SfN explored a different route through the gut: probiotics combined with a prebiotic improved spatial learning in a rat model of early tau pathology [2,4]. For drug developers, the brain-body axis broadens the search for targets beyond the brain itself. Preclinical studies are essential to determine how these pathways influence brain function and whether the evidence supports clinical testing.
Although these research areas involve different disease mechanisms, they highlight a shared translational challenge: understanding how peripheral metabolic, immune, and gut-derived signals influence central nervous system function. For drug developers, this requires preclinical models and study designs that can distinguish target-specific neural effects from broader systemic changes.
![Figure 1. The research presented at Neuroscience 2026 showcases cutting-edge insights about contributions from the gut, vasculature and the brain's plumbing system [2].](https://web.cdn.cyagen.com/cyagen-en-v2/pictures/5076dc1adbb8cfb7603081267f78de89.webp)
GLP-1 and Incretin Signaling in Preclinical Addiction Research
One roundtable focused on incretin pharmacology in substance use disorders. “Overlap of GLP-1 Circuitry in Addiction and Obesity: Treatment Implications,” chaired by the director of NIDA, examined the overlap in reward circuitry [1]. In a rat model of cocaine seeking, GLP-1 signaling engaged GABA neurons in the ventral tegmental area and modulated mesolimbic dopamine activity [5]. Obesity and addiction converge on shared circuitry at the junction of homeostatic and hedonic control. Human evidence in opioid use disorder (OUD) remains limited.
In earlier work from Penn State College of Medicine, male Sprague-Dawley rats trained to self-administer fentanyl were tested for cue-, drug-, and stress-induced seeking. Acute subcutaneous pretreatment with tirzepatide (0.05 mg/kg), the dual GLP-1/GIP receptor agonist, significantly reduced infusion attempts during cue-induced seeking (p = 0.0065), after a noncontingent fentanyl infusion (p = 0.0035), and in a separate yohimbine challenge, with pretreatment repeated after seven days of abstinence (p = 0.0075). These comparisons pooled rats with different histories of fentanyl intake (tirzepatide, n = 16; saline, n = 8) [6].
![Figure 2. Study design for the experiment testing the effects of tirzepatide on cue-, drug-, and stress-induced fentanyl seeking in male rats following fentanyl self-administration [6].](https://web.cdn.cyagen.com/cyagen-en-v2/pictures/463ba8af32b5870afdc50b00d77bf7a5.webp)
A 2026 Biological Psychiatry review from the same group adds three points for discovery teams [7]. The 0.05 mg/kg dose was the most effective of the three tirzepatide doses tested (0.05, 0.1, and 0.15 mg/kg) in this rat paradigm. In a separate rat study, titration to 0.3 mg/kg liraglutide reduced heroin seeking, whereas titration to 0.6 mg/kg did not. The authors proposed that reduced responsiveness following higher-dose exposure could reflect receptor desensitization or internalization, although this mechanism was not directly established in the study. Preclinical studies also suggest that GIP receptor agonism can attenuate nausea-related responses to GLP-1 receptor agonists. Some observational studies have associated incretin-based treatment with lower rates of opioid overdose or OUD-related hospitalization. Estimates vary by outcome and comparator. The review identified one completed pilot trial reporting lower opioid craving and three other trials ongoing at the time of writing [7].
The Gut-Brain Axis and Systemic Factors in Alzheimer's disease
Another focus was the systemic contribution to neurodegeneration. SfN's press conference on brain-body connections in Alzheimer's disease, moderated by a co-director of the University of Rochester's Center for Translational Neuromedicine, brought together four distinct lines of research linking systemic physiology, vascular health, and brain function. These findings included associations between intestinal barrier-related symptoms and circulating amyloid, links between brain waste-clearance efficiency and cognitive performance in older adults, the detection of plastic-like particles around cerebral blood vessels in postmortem AD tissue, and preclinical evidence from a probiotic-plus-prebiotic intervention in a rat model of early tau pathology[2].
At SfN 2026, researchers from Memorial University of Newfoundland reported that combining probiotics with a prebiotic improved spatial learning in a rat model of pretangle tau pathology [2]. In a January 2025 study, the same group found that supplementation significantly improved performance on a spontaneous location recognition task. In that study, the improvement in Y-maze performance did not reach statistical significance, and supplementation did not significantly improve odor discrimination [4]. Other studies highlighted in the same SfN briefing linked intestinal barrier-related symptoms to circulating amyloid in patients with long COVID, found an association between brain waste clearance efficiency and cognition in older adults, and identified plastic-like particles around blood vessels in brain tissue from people who had died with Alzheimer’s disease [2]. These observations raise questions about how gut barrier function, brain waste clearance, and vascular exposures relate to cognitive aging and Alzheimer’s disease.
![Figure 3. Schematic of experimental design and the effects of probiotic supplementation on behavioral tasks [4].](https://web.cdn.cyagen.com/cyagen-en-v2/pictures/193b0f4f096939b3afc0f9b0e1ca86e5.webp)
What the EVOKE and EVOKE+ Trials Reveal About Semaglutide in AD
Enthusiasm needs calibrating against two large randomized trials. The evoke and evoke+ Phase 3 trials of oral semaglutide in early Alzheimer's disease, enrolling 3,808 amyloid-confirmed participants across 566 sites and published in The Lancet in March 2026, did not slow progression. Mean CDR-SB change to week 104 was 2.3 and 2.2 on semaglutide, versus 2.3 and 2.1 on placebo, respectively (p = 0.57 in evoke and p = 0.46 in evoke+). Neither trial met its primary endpoint, so formal confirmatory testing of secondary endpoints did not proceed [3].
The biomarker data make the failure instructive. At week 78, several CSF tau measures were up to 10% lower with semaglutide than with placebo in an exploratory substudy, without adjustment for multiple comparisons. Circulating hsCRP also declined. At week 104, plasma neurofilament light was about 5% higher with semaglutide in evoke+, and GFAP was about 4% higher in both trials [3]. The clinical significance of these differences remains uncertain. Treatment may have begun too late in the disease course, earlier observational associations may partly reflect differences between patients who received GLP-1 therapies and those who did not, and peripheral effects alone may be insufficient to alter established brain pathology.

Earlier intervention and combination therapy remain open questions. For brain–body research, these results underscore the importance of testing how changes in a proposed pathway relate to outcomes that matter to patients. In OUD, studies can test whether the reductions in drug seeking observed in rats translate into less illicit opioid use in patients.
Preclinical Models and Behavioral Endpoints for Brain-Body Research
The studies above pose different questions for model selection. The tirzepatide study used an opioid self-administration paradigm to measure seeking under three conditions and tracked changes in body weight [7]. The probiotic work used a pretangle tau model and behavioral tasks sensitive to early deficits in spatial learning [4].
Three model categories can support these studies. Off-the-shelf genetically engineered mouse models, including established amyloid and tau lines, obesity models, and incretin receptor knockouts, provide characterized baselines for cross-study comparison, and sourcing validated strains rather than building de novo can shorten study preparation. Intercrossed strains can address specific brain-body questions. Examples include an amyloid or tau line on a metabolic background, or a circuit-restricted GLP-1R or GIPR deletion layered onto a neurodegeneration model. Humanized target alleles can help when a compound shows species differences at the receptor of interest. Cyagen's hutau/hGLP1R mouse (I001221) combines a humanized MAPT locus with a humanized GLP1R allele through breeding of the hutau(MAPT) and hGLP1R lines. This model provides a humanized target background for studies involving GLP-1R pharmacology and human MAPT expression. However, human MAPT replacement alone does not establish tau aggregation or an AD-like disease phenotype. Studies evaluating therapeutic effects in tauopathy would require an appropriately validated disease model or additional pathological induction, together with relevant molecular and functional endpoints.

Model availability is necessary but not sufficient. Tirzepatide-treated rats lost significant weight relative to saline controls [7], so any reward-specific claim must survive a metabolic control analysis. Measures of food intake, activity, and responding for a non-drug reward would help assess selectivity. Depending on the study question, relevant capabilities include behavioral batteries spanning self-administration, reinstatement, cognition, and motivation; metabolic profiling of body composition, glucose-insulin dynamics, and energy expenditure; PK/PD with CNS exposure; and in vivo pharmacology able to run chronic arms long enough to detect tolerance. Acute-only dosing cannot establish whether the behavioral effect persists with repeated treatment. Sourcing strains, intercrosses, and phenotyping from one partner, as Cyagen's combined model and metabolic and neurology CRO platform allows, can help coordinate measurements. Collecting metabolic and behavioral data in the same animals allows investigators to test whether the two outcomes covary within individuals.
What Brain-Body Research Means for Drug Discovery
The axis presents a clear asymmetry. In neurodegeneration the hurdles are real, both semaglutide trials missed their primary endpoint despite changes in exploratory biomarkers [3], and mechanisms such as gut barrier integrity and glymphatic clearance remain hypothesis-generating. In OUD, rodent studies and an early human craving signal support clinical testing of effects on opioid use [6].
Together, these findings support an indication-specific approach to preclinical research rather than a one-size-fits-all model strategy. For systemic interventions, researchers need to determine whether observed behavioral changes reflect target-specific effects on neural circuits or broader changes in appetite, body weight, metabolism, and general motivation. Strong preclinical programs will be those that connect systemic pharmacological effects with measurable CNS outcomes and use appropriate controls to establish the mechanisms underlying observed behavioral changes.
👉 [Explore Cyagen's One-stop Neuroscience CRO Solutions]
👉 [Explore Cyagen's Preclinical CRO Solutions for Metabolic Disease Research]
👉 [Discuss Your Brain-Body Axis Related Project with Our Scientists]
References
[1] Volkow ND, chair. Overlap of GLP-1 circuitry in addiction and obesity: treatment implications. Basic-Translational-Clinical Roundtable presented at: Neuroscience 2026, the annual meeting of the Society for Neuroscience; 2026 Nov 16; Washington, DC. Session BTCR01.
[2] Society for Neuroscience. New brain-body connections expand understanding of Alzheimer's disease. Medical Xpress [Internet]. 2026 Nov 14 [cited 2026 Nov 19].
[3] Cummings JL, Atri A, Sano M, Zetterberg H, Scheltens P, Knop FK, Johannsen P, Wichmann CA, Abschneider RM, Leon T, Feldman HH. Efficacy and safety of oral semaglutide 14 mg (flexible dose) in early-stage symptomatic Alzheimer's disease (evoke and evoke+): two phase 3, randomised, placebo-controlled trials. Lancet. 2026 May 30;407(10544):2167-2179. doi: 10.1016/S0140-6736(26)00459-9. Epub 2026 Mar 19. PMID: 41865758.
[4] Flynn CM, Omoluabi T, Janes AM, Rodgers EJ, Torraville SE, Negandhi BL, Nobel TE, Mayengbam S, Yuan Q. Targeting early tau pathology: probiotic diet enhances cognitive function and reduces inflammation in a preclinical Alzheimer's model. Alzheimers Res Ther. 2025 Jan 18;17(1):24. doi: 10.1186/s13195-025-01674-1. PMID: 39827356; PMCID: PMC11742226.
[5] Volkow ND, Wargo EM. GLP-1 receptor agonists in addiction therapy. Science. 2026 Nov 5;394(6824):610-611. doi: 10.1126/science.aej9836. PMID: 42986956.
[6] Moran-Bariso CFP, Whitfield C, Brandl C, Acharya N, Grigson PS, Nyland JE. Acute treatment with tirzepatide reduces cue-, drug-, and stress-induced fentanyl seeking in male rats. Poster presented at: Consortium on Substance Use and Addiction 7th Annual Conference; 2026 May 4; State College, PA.
[7] Grigson PS, Bunce SC, Brick TR, Freet CS, Nyland JE, Kawasaki SS, Brandl CG, Shealy E, Huddleston K. Glucagon-Like Peptide-1 Receptor Agonists as a Candidate Treatment for Opioid Use Disorder: From Rats to Humans. Biol Psychiatry. 2026 Oct 15;100(8):823-830. doi: 10.1016/j.biopsych.2026.06.033. Epub 2026 Jul 21. PMID: 42480803.
[8] Stapleton HM, Borges DS, Trindade EBSM, Yuan Q. Sex-dependent locus coeruleus vulnerability in Alzheimer's disease: gut dysbiosis as a driver and probiotic intervention as rescue. Biol Sex Differ. 2026 Jan 27;17(1):34. doi: 10.1186/s13293-026-00834-8. PMID: 41593789; PMCID: PMC12918643.




