COPEPTIN AS A SURROGATE MARKER OF ARGININE VASOPRESSIN:CLINICAL DIAGNOSTIC AND PROGNOSTIC APPLICATIONS IN ENDOCRINE AND NON-ENDOCRINE DISEASES

BACKGROUND AND AIMS. Copeptin is the C-terminal peptide of pre-pro-arginine vasopressin and it is synthetized and released in equimolar amounts as arginine vasopressin (AVP) from the posterior pituitary gland. Due to greater feasibility of laboratory assessment, it is used as a surrogate marker of AVP in several endocrine and non-endocrine disorders. AVP deficiency (previously known as central diabetes insipidus) is a disorder characterized by hypotonic polyuria with polydipsia. It may ensue after hypothalamus-pituitary surgery and its timely recognition and management is paramount to limit complications like dehydration and acute hypernatremia. In the outpatient setting, AVP deficiency has to be differentiated from primary polydipsia to avoid erroneous treatments, in both adult and pediatric patients. To date, hypertonic saline-stimulated copeptin is considered the gold standard for the diagnosis of AVP deficiency, but it is a cumbersome test requiring close monitoring. Alternative tests like arginine-stimulated and glucagon-stimulated copeptin could allow a simpler diagnostic workup in this setting but have not been validated yet. AVP and copeptin are involved in acute and chronic non-endocrine diseases as well, and a diagnostic and prognostic role of copeptin has been reported in conditions like acute myocardial infarction, ischemic stroke, congestive heart failure, sepsis. The aim of this dissertation was to investigate usefulness of copeptin measurement, under baseline and stimulated conditions, in different clinical settings. In study A, we aimed to investigate the accuracy of unstimulated copeptin assessed 2 days post-operatively in the diagnosis of AVP deficiency occurring after pituitary surgery. Study B was a multi-center study comparing arginine-stimulated copeptin with the gold standard hypertonic saline test in the differential diagnosis of polyuria-polydipsia syndrome in adults. Study C was a proof-of-concept study to investigate the response of copeptin to glucagon administration in a paediatric cohort. Finally, study D investigated copeptin secretion in Coronavirus Disease 2019 (COVID-19) hospitalized patients and correlation of copeptin with disease severity and in-hospital outcomes. METHODS AND RESULTS. Study A was a single-center, retrospective case-control study on 47 patients (F/M 27/20, mean age 55 ± 15 years) undergoing neurosurgical procedure for hypothalamus—pituitary disease at Ospedale Maggiore Policlinico in Milan, between March 2016 and October 2018. Copeptin was assessed on second post-operative day and compared between patients developing (n=16) and not developing (n=31) AVP deficiency. Median copeptin was significantly lower in patients who developed AVP deficiency than in patients who did not (3.2 [interquartile range 2.4-3.7] vs. 4.5 [3.6-7.4] pmol/L, p=0.005). Receiver Operator Curve (ROC) analysis showed that copeptin had modest accuracy in diagnosing AVP deficiency (area under the curve 0.75, p=0.005), and the best copeptin cut-off of 3.75 pmol/L, as derived according to Youden’s J statistics, had a negative predictive value of 87%. Study B was an international, non-inferiority trial. Adult patients with polydipsia and hypotonic polyuria or a known diagnosis of AVP deficiency underwent hypertonic saline stimulation and arginine stimulation. Two endocrinologists independently made the final diagnosis of AVP deficiency or primary polydipsia with use of clinical information, treatment response, and the hypertonic saline test results. The primary outcome was the overall diagnostic accuracy according to prespecified copeptin cutoff values of 3.8 pmol/L for arginine and 4.9 pmol/L for hypertonic saline. Arginine-stimulated copeptin failed to demonstrate non-inferiority compared to hypertonic saline stimulation (accuracy 74.4%, 95% confidence interval [CI], 67.0 to 80.6 for arginine-stimulated copeptin; 95.6%, 95% CI, 91.1 to 97.8 for hypertonic saline– stimulated copeptin; estimated difference, −21.2%; 95% CI, −28.7 to −14.3). Adverse events were generally mild with the two tests. Arginine-stimulated copeptin ≤ 3.0 pmol/L led to a diagnosis of AVP deficiency with a specificity of 90.9%, whereas levels ≥ 5.2 pmol/L led to a diagnosis of primary polydipsia with a specificity of 91.4%. Study C was an observational, cross-sectional study. Twenty children, recruited among those undergoing glucagon stimulation test for suspected growth hormone deficiency and with no polyuria-polydipsia syndrome, were included (9 females, mean age 10.3±2.9 years). Blood samples for copeptin measurement were drawn before and 60, 90, 120, 150 and 180 minutes after glucagon administration. Variations of copeptin concentrations over time were evaluated. Median basal copeptin levels were 4.1 pmol/L [3.3-6.7]. During glucagon test, a significant increase of copeptin was recorded 120, 150 and 180 minutes after stimulation (p<0.01 vs. baseline for all), with peak concentrations recorded at 150 minutes (median 10.6, interquartile range 5.4-17.9 pmol/L). Copeptin response did not differ between patients with and without growth hormone deficiency. Study D included 116 patients admitted to medium intensity-of-care COVID-19 departments of Ospedale Maggiore Policlinico in Milan, between March 2020 and June 2020 and enrolled within the COVID-19 Network registry. Copeptin was assayed in blood samples drawn at hospital admission and stored in the institutional biobank, and clinical data were retrieved retrospectively from an online database. AVP – copeptin system activation was arbitrarily defined as copeptin values above the laboratory upper limit of normal (12.0 pmol/L). Association was tested between copeptin and (i) clinical and biochemical markers of disease severity at admission; (ii) in-hospital outcomes, i.e. in-hospital mortality, length of hospital stay, in-hospital complications. Median copeptin concentration at admission was 13.2 [6.3-30.8] pmol/L. Copeptin was above the laboratory upper limit of normal in 60 patients (52%) of the unsorted cohort, in 16/21 (76%) non-survivors (76%) and 44/95 (46%) survivors (p=0.02). Copeptin lacked association with need for supplemental oxygen or ventilation, arterial oxygen saturation, or radiological findings on chest imaging studies. Conversely, it was significantly associated with length of hospital stay (r=0.42, p<0.001), sepsis (odds ratio, OR 1.019, 95% CI 1.004 – 1.041, p=0.03) and acute kidney injury (OR 1.012, 95% CI 1.001 – 1.026, p=0.04). The ROC AUC was 0.87 for development of acute kidney injury and 0.95 for sepsis. Best cut-offs of 45.8 and 21.3 pmol/L respectively showed high negative predictive value (100% for both). CONCLUSION. While basal copeptin demonstrated modest diagnostic performance, stimulated copeptin assessment, in relation to osmotic triggers or stress, could be successfully applied in endocrine and non-endocrine conditions. Consistently, unstimulated copeptin assessed on second post-operative day had modest accuracy in diagnosing AVP deficiency after pituitary surgery, although accuracy could be improved through different timing of sampling. Arginine and glucagon were tested as non-osmotic stimuli to AVP-copeptin release. In the differential diagnosis of polyuria-polydipsia syndrome, hypertonic saline stimulation was confirmed as the gold standard test, although arginine stimulation could be regarded as a well-tolerated, feasible alternative in adults. Glucagon was able to stimulate copeptin release in children with intact posterior pituitary and should be further evaluated for the diagnosis of AVP deficiency. Finally, as a model of stress-related AVP-copeptin activation, promising results in terms of prognostic stratification were obtained when admission copeptin was assessed in COVID-19 hospitalized patients during the first Italian outbreak. These results need to be re-evaluated in the current epidemiological and clinical setting. Overall, implementation of copeptin measurement is warranted in clinical practice.