Tirzepatide as an Insulin Sensitizer

Tirzepatide, a glucagon-like peptide 1/glucose-dependent insulinotropic peptide (GLP-1/GIP) receptor coagonist, offers a paradigm of multireceptor pharmacology for the treatment of metabolic diseases. A stronger agonist of the GIP than the GLP-1 receptor (1), tirzepatide potentiates GLP-1-induced weight loss without GIP itself reducing energy intake (2), and in preclinical models is associated with weight-independent insulin sensitization. This pattern of coagonism could not be easily predicated on the basis of the known human biology of GIP alone (3,4), but nonetheless carries major therapeutic potential in patients with type 2 diabetes. Excerpting from mechanistic studies (5), GLP-1/GIP synergism on weight loss is likely to operate at the level of the central nervous system, where both GLP-1 and GIP receptors are expressed on vascular, glial, and neuronal cells in the arcuate, dorsomedial, and paraventricular nuclei of the human hypothalamus (6). Weight-independent insulin sensitization, on the other hand, primarily occurs in adipose tissue, where GIP but not GLP-1 receptors are expressed, and, mostly indirectly, in skeletal muscle and liver (3-5). Making intelligent use of metabolomics, Pirro and colleagues (7) construe the metabolic model that mediated the physiologic actions of tirzepatide in a group of obese patients with type 2 diabetes (6). Thus, 3 main metabolite clusters were found to be altered by tirzepatide: (1) branched-chain amino acids and their ketoacids as well as products of their catabolism were dose-dependently lowered; (2) 2-hydroxybutyrate, a methionine/threonine end product strongly associated with glycemia, stood out as the metabolite with the largest percentage decrease from baseline; and (3) total triglycerides and diglycerides with fewer acyl carbons and fewer acyl double bonds were reduced. These changes were significantly different not only from baseline and placebo but also from the corresponding changes induced by dulaglutide, and correlated with the concomitant changes in disease phenotypes (glycated hemoglobin and Homeostatic Model Assessment 2 of Insulin Resistance). The interconnections among the circulating metabolite clusters can be best appreciated with reference to a metabolic model evolved from metabolomics studies of insulin-resistant subjects (8). Briefly, as a result of insulin resistance at the level of adipose tissue (unrestrained lipolysis), muscle tissue (reduced glucose uptake and amino acid catabolism), and liver (increased glucose, ketone, and triglyceride output), the tricarboxylic acid cycle is overloaded with primary substrates (glucose, free fatty acids, amino acids, and ketone bodies), exceeding its oxidative capacity. These substrates therefore build up in plasma along with their intermediates (branched-chain amino acids and ketoacids, lactate, alanine, glutamate, and other amino acids). Propionyl-coenzyme A, a direct product of the oxidation of odd-chain fatty acids and some amino acids (isoleucine, valine, methionine, and threonine) accumulates as its conversion to succinate to enter the tricarboxylic acid cycle is impeded. One result is that catabolism of 2-ketobutyrate through propionyl-coenzyme A is slowed down, and conversion to its reduced form, 2-hydroxybutyrate via lactate dehydrogenase is favored. Thus, the 2-hydroxybutyrate/2-ketobutyrate couple sits at the crossroads of amino acid catabolism and mitochondrial substrate spillover, thereby strongly signaling the overall metabolic disturbance in strong association with clinical phenotypes. Of note, the substrate overload enhances mitochondrial oxidative stress and consumes reducing equivalents, resulting in a deficit of glutathione and decreased availability of its component amino acid, glycine, for resynthesis. The metabolites changes induced by tirzepatide in the trial (7) are the exact reversal of the metabolic overload model delineated here: by inverting the pattern of its changes, one gets a faithful summary of the study results, which collectively amount to whole-body insulin sensitization. It is important to reflect that substantial weight loss would probably bring about a similar, if not identical, pattern of metabolic/metabolite effects; this possibility was not explored in the trial because the comparator, dulaglutide, did not induce the same weight loss as the 3 doses of tirzepatide tested. However, the authors calculate that the observed weight changes explained consistently low fractions of the variability of the concomitant metabolite changes. More cogently, in obese mice pair feeding, or preventing weight loss by GLP-1 receptor knockout, partly preserved the effect of tirzepatide on clamp-based insulin sensitivity (5). In this case, insulin sensitization is an emergent property, which neither component of the coagonist has. The root of tirzepatide sensitization appears to rest in the concurrence of GLP-1-driven weight loss, which relieves insulin inhibition of lipolysis, and GIP-driven increase in blood flow through fat depots, which delivers more and washes out more bloodborne substances. May the net balance of this traffic be the key to this pharmacological sensitization? A new, interesting model for further physiological investigation. All in all, pending more evidence in people, the therapeutic arsenal for type 2 diabetes may eventually feature a bona fide insulin sensitizer, competing in this capacity only with pioglitazone. The latter, however, causes dose-dependent weight gain by its very mechanism of action; regardless of its pathogenic significance, weight gain is unwelcome to most patients with diabetes. Also, insulin sensitization per se typically associates with improved blood pressure and lipid control and better ß-cell function as well as reduced proinflammatory and prothrombotic markers. The available clinical and metabolic evidence so far encourages to fully explore the therapeutic potential of tirzepatide—in the longer run, in nondiabetic obese individuals, in longstanding type 1 diabetes, and in patients with high cardiovascular risk and/or renal impairment. Funding: None. Disclosures: E.F. reports research support from Janssen, Boehringer Ingelheim, and Sanofi; serving on advisory panels: Boehringer Ingelheim, Lexicon, and Oramed; and receiving speaker’s honoraria from Boehringer Ingelheim, Lilly & Co., and MSD. Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.