The main finding of our study is that in contrast to a significant increase of copeptin levels upon glucagon, there is no effect of glucagon on plasma OT levels.

Recent data indicated that the social and emotional deficits experienced by patients with AVP deficiency might be explained, at least partly, by an additional OT deficiency [6,8,17,18,19,20,21,22,23]. The fact that the observed changes in psychopathology are present in patients with or without additional anterior pituitary dysfunction strengthens this hypothesis [6, 8]. However, measuring and replacing OT is not currently done in clinical routine [19]. Various diagnostic challenges make it difficult to measure basal OT levels accurately and detect a potential deficiency, such as its short half-life in plasma, which lasts only a few minutes [24] and conflicting data regarding whether a single peripheral OT measurement reflects central OT activity [2, 14, 25, 26]. Therefore, an alternative approach, such as provocative testing, might be more suitable to reveal an insufficient OT release [19, 27]. We recently investigated other established pituitary provocation tests, such as the hypertonic saline, the macimorelin or the arginine infusion test, hypothesizing an OT-stimulating effect. Our results, however, indicated that neither of these tests does sufficiently stimulate OT levels in healthy volunteers [28]. MDMA has recently been shown to be a strong stimulus for OT; however, its implication in clinical practice might be challenging. Therefore, an alternative stimulation test is desirable.

Hypoglycaemia strongly stimulates the anterior pituitary, leading to increased levels of growth hormone, cortisol, prolactin and also the posterior pituitary with AVP release [16, 29]. Some studies have indicated that hypoglycaemia induced by the insulin hypoglycaemia test (IHT) also leads to OT release [10, 11]. Chiodera et al. have shown a doubling in plasma OT levels in healthy volunteers following 30 min of profound hypoglycaemia. Mechanistically, it is postulated that OT is involved in the central regulation of pancreatic secretion and its release might be stimulated by brain-glucose-sensing neurons located in the hypothalamus and can be activated by the sympathetic autonomic branch [15]. We recently provided evidence for a strong effect of glucagon on plasma copeptin levels as a surrogate marker of AVP [12]. One possible explanation for this stimulating effect might be the dynamics of glucose levels after glucagon injection. Initially, plasma glucose levels increase from baseline and reach a peak, followed by a rapid decrease, bringing glucose levels back to low or normal levels [9]. We demonstrated a clear correlation between the drop in glucose levels and the increase in copeptin levels upon glucagon stimulation supporting this hypothesis [9]. Therefore, it is possible that the rapid decrease in elevated glucose levels might mimic – in analogy to IHT – an acute relative hypoglycaemic state without inducing absolute hypoglycaemia [9]. However, although it seems that this postulated effect of relative hypoglycaemia is potent enough to stimulate copeptin, it appears that OT is only stimulated by absolute hypoglycaemia, for which glucagon is not a strong enough stimulus. Alternatively, the hypothesised effect of glucagon might have been attenuated by concurrent administration of ondansetron, a serotonin (5-HT) receptor antagonist. In support of this, animal studies indicate that 5-HT and receptor agonists stimulate AVP and OT release [30, 31]. Conversely, in 1998, Volpi et al. demonstrated in a study of 12 participants that ondansetron partially reduced AVP response to hypoglycaemia but interestingly not OT response [32]. However, due to the limited sample size the possibility of ondansetron exerting a similar suppressive release on OT cannot be completely excluded. Moreover, as we only measured OT up to 120 min after glucagon injection, contrary to 180 min in the study of Volpi et al., we could have missed the stimulatory effect. Furthermore, despite the hypothesis of a glucagon-induced relative hypoglycaemia-mediated OT effect, the exact mechanisms remain unclear. Some previous research also discussed stimulating active peptides fragmented directly from glucagon [33, 34]. In support of this, various application routs of glucagon have been shown to stimulate the pituitary gland differently [35]. Currently, further agents are under investigation for OT stimulation, i.e., GLP-1 analogues, melatonin, and corticotropin-releasing-hormone (clinicaltrials.gov NCT04897802, NCT04902235).

Considering the so far negative results on OT levels for most stimulation tests, and taking into account the strong stimulus, MDMA remains currently the most promising test to diagnose possible OT deficiency. As MDMA is currently in phase 3 trials for the development as a therapeutic agent for post-traumatic stress disorders, the use of MDMA as a diagnostic agent will be directly accessible to clinicians in the near future. Therefore, further developing the MDMA test seems to be the most promising method for clinical routine. Moreover, for clinical practice, addressing preanalytical and analytical challenges associated with OT measurements are crucial before implementing diagnostic tests. The reliability of currently available assays, including EIA and RIA, remains a subject of ongoing debate. Consensus among experts emphasizes standardized analytical procedures to preserve accuracy and reproducibility, entailing cooled plasma samples, batch measurement, and sample extraction—methods followed in our study [36]. Nevertheless, as promising methods (liquid chromatography/mass spectrometry, LC/MS) are in development, available assays should be used with their respective limitations. Consequently, as suggested, OT results should be interpreted not in absolute terms but rather in relation to relative changes [37].

Some limitations should be considered for our study. First, this is a secondary analysis of a prospective diagnostic study. Despite careful blood sampling, immediate centrifugation, and quick storage at –80 °C, a certain decay of OT cannot be excluded, presenting a certain bias as the OT assay requires standardized procedures in blood sampling and storage. Considering the poor response of OT to glucagon stimulation, the limited sensitivity of the test and its challenging handling characteristics might have influenced the findings of the study. Second, the sample size was limited and did not allow for sub-group analysis. Third, although all samples were measured in one batch, there is ongoing discussion on the currently available OT enzyme immunoassay (EIA) and radioimmunoassay (RIA). In conclusion, our study demonstrates that glucagon does not stimulate OT levels.