Machine-Learning Prediction of Postoperative Pituitary Hormonal Outcomes in Nonfunctioning Pituitary Adenomas: A Multicenter Study

ObjectiveNo accurate predictive models were identified for hormonal prognosis in non-functioning pituitary adenoma (NFPA). This study aimed to develop machine learning (ML) models to facilitate the prognostic assessment of pituitary hormonal outcomes after surgery.MethodsA total of 215 male patients with NFPA, who underwent surgery in four medical centers from 2015 to 2021, were retrospectively reviewed. The data were pooled after heterogeneity assessment, and they were randomly divided into training and testing sets (172:43). Six ML models and logistic regression models were developed using six anterior pituitary hormones.ResultsOnly thyroid-stimulating hormone (p < 0.001), follicle-stimulating hormone (p < 0.001), and prolactin (PRL; p < 0.001) decreased significantly following surgery, whereas growth hormone (GH) (p < 0.001) increased significantly. The postoperative GH (p = 0.07) levels were slightly higher in patients with gross total resection, but the PRL (p = 0.03) level was significantly lower than that in patients with subtotal resection. The optimal model achieved area-under-the-receiver-operating-characteristic-curve values of 0.82, 0.74, and 0.85 in predicting hormonal hypofunction, new deficiency, and hormonal recovery following surgery, respectively. According to feature importance analyses, the preoperative levels of the same type and other hormones were all important in predicting postoperative individual hormonal hypofunction.ConclusionFluctuation in anterior pituitary hormones varies with increases and decreases because of transsphenoidal surgery. The ML models could accurately predict postoperative pituitary outcomes based on preoperative anterior pituitary hormones in NFPA.

Estrogen mediates sex differences in preoptic neuropeptide and pituitary hormone production in medaka

The preoptic area (POA) is one of the most evolutionarily conserved regions of the vertebrate brain and contains subsets of neuropeptide-expressing neurons. Here we found in the teleost medaka that two neuropeptides belonging to the secretin family, pituitary adenylate cyclase-activating polypeptide (Pacap) and vasoactive intestinal peptide (Vip), exhibit opposite patterns of sexually dimorphic expression in the same population of POA neurons that project to the anterior pituitary: Pacap is male-biased, whereas Vip is female-biased. Estrogen secreted by the ovary in adulthood was found to attenuate Pacap expression and, conversely, stimulate Vip expression in the female POA, thereby establishing and maintaining their opposite sexual dimorphism. Pituitary organ culture experiments demonstrated that both Pacap and Vip can markedly alter the expression of various anterior pituitary hormones. Collectively, these findings show that males and females use alternative preoptic neuropeptides to regulate anterior pituitary hormones as a result of their different estrogen milieu.

Effects of oral macimorelin on copeptin and anterior pituitary hormones in healthy volunteers

Purpose The test with the highest diagnostic accuracy for diabetes insipidus is copeptin measurement after hypertonic saline infusion. However, the procedure is cumbersome and unpleasant due to rapid sodium increase. An oral stimulation test would be highly desirable. Macimorelin, an oral ghrelin agonist, is a newly approved diagnostic test for growth hormone (GH) deficiency, but its effects on copeptin/vasopressin are unknown and the effects on other pituitary hormones only scarcely investigated. Methods In this prospective, interventional, proof-of-concept study Copeptin and anterior pituitary hormones were measured in 28 healthy volunteers on two test days at baseline, 30, 45, 60, 90 and 120 min after a single dose of macimorelin (first visit: 0.5 mg/kg, second visit: 0.75 mg/kg). Results Baseline copeptin levels were 5.26 pmol/L [1.57, 6.81] and did not change after macimorelin intake (0.5 mg/kg: maximal median change 0.40 [− 0.49, 0.65] pmol/L, p = 0.442; 0.75 mg/kg: − 0.13 [− 0.45, 0.17] pmol/L, p = 0.442. Median GH levels increased from 3.67 mU/L with a maximal median change of 94.66 [IQR 56.5; 110.96] mU/L, p < 0.001. No effect was seen on cortisol, ACTH, LH and FSH levels. Prolactin (max. median change 100 [2.5; 146.5] mU/L, p = 0.004) and free thyroxine (fT4) (0.5 [0.2; 0.8] pmol/L, p < 0.001) increased, whereas TSH decreased (− 0.18 [− 0.22, − 0.09] mU/L, p < 0.001). Conclusion We confirm an increase of GH upon macimorelin in healthy volunteers. However, macimorelin did not stimulate copeptin and therefore does not provide an oral test alternative for the diagnosis of diabetes insipidus. Additionally, a stimulatory effect was seen for prolactin and fT4, but not for ACTH and gonadotropic hormones. Registration The trial was registered on ClinicalTrials.gov (NCT03844217) on February 18, 2019.

Evaluation of anterior pituitary hormonal profile in patients with first seizure of tonic-clonic and temporal lobe epilepsies in Al-Zahra hospital of Isfahan, Iran, during 2014-2016: comparison of hormonal profiles in 2 types of epilepsy

Background and aims: Temporal lobe and tonic-clonic seizures are the most common types of epilepsy. It is a disease that disrupts the regulation of the hypothalamic pituitary adrenal (HPA) axis. The disturbance in the regulation of this axis affects the secretion of hormones from the anterior pituitary. Due to the different origins of these 2 types of epilepsy listed above, this study aimed to examine their differences in the level of anterior pituitary hormones. Methods: This cross-sectional study was conducted in Al-Zahra hospital of Isfahan, Iran, from 2014 to 2016. A total of 36 patients with the idiopathic first seizure were selected for the examination including 20 patients with tonic-clonic seizure and 16 patients with temporal lobe seizure. The diagnosis of epilepsy was confirmed by a neurologist along with electroencephalogram (EEG) findings. The levels of anterior pituitary hormones including thyroid-stimulating hormone (TSH), adrenocorticotrophic hormone (ACTH), GH (growth hormone), follicle-stimulating hormone (FSH), luteinizing hormone (LH) and PRL (prolactin) were measured and compared using ELISA method in both groups. Results: The results indicated that all the anterior pituitary hormones except TSH and ACTH had higher means in the temporal lobe epilepsy than in the tonic-clonic epilepsy, but a significant difference (mean ± SD) was reported only in GH (4.59±2.95) (2.55±1.87) (P<0.005) and FSH (19.4±14.06) (4.34±2.37) (P<0.033) respectively in the temporal lobe and tonic-clonic seizures. Meanwhile, there was a significant relationship between seizure, regardless of its type, and the level of growth hormone and FSH. Conclusion: In the future, hormonal changes may be used as an adjunct to diagnose the type of epilepsy. More studies are required to confirm the findings.

The Epidemiology and Hormonal Assessment of Patients with Empty Sella on Pituitary Magnetic Resonance: A Retrospective Single Centre Study in Saudi Community Based Hospital

Background: No large enough series have been reported to allow a detailed definition of the epidemiology of primary empty sella (PES) and diagnostic anterior pituitary hormones abnormalities in Saudi Arabia. Objective: To evaluate the epidemiological and hormonal profile of patients with PES. Design: Retrospective analysis of pituitary Magnetic Resonance Image (MRI) and hormonal data of patients with PES between January 2008 and December 2017. Settings: Departments of Endocrinology and Radiology at King Fahad Armed Forces Hospital, Jeddah, Saudi Arabia. Patients: 537 patients with clinical, hormonal and radiological data. Main outcome measures: The frequency of PES in patients and associated abnormal pituitary hormones. Results: Over the 9-year period, Out of 537 patients; 161 (30 %) were male and 376 (70 %) were female with mean age 35.4 ± 13.7. Among them we found 50 (9.3%) cases with PES on MRI. 11 (22 %) cases were male and 39 (78 %) were female with female to male ratio of 3.5 to 1. Cases with PES were significantly older than cases with pituitary adenoma and normal MRI, 45.3 ± 19.6 vs. 35.3 ± 13.9, p value <0.0001. PES occurred with the greatest frequency at the fourth decade. During the hormonal assessment, 34 (68%) cases of PES were found to have some degree of anterior pituitary hormones abnormailities. No anterior pituitary hormones abnormalities were found in 16 (32%) of cases. Hyperprolactinemia was found in 26 (52%) of cases. Cases of hypogonadotropic hypogonadism and central hypothyroidism were equally distributed, 4 cases ( 8% ). Female predominance was found. Conclusion: In our study, PES was more common in middle aged female. The high incidence of endocrine abnormalities in patients with PES mandates that these patients should routinely be subjected to endocrine evaluation to detect these deficiencies early. Limitations: Question of clustering of cases within the study region and limited study sample size.

Gender differences in hormonal statuses of patients with esophageal squamous cell carcinoma.

85 Background: Esophageal squamous cell carcinoma (ESCC) is 4-8 times more frequent in men than in women, with poorer prognosis. The purpose of the study was to analyze systemic and local hormonal statuses in patients with ESCC in dependence on their gender. Methods: The study included 34 ESCC (G2, рT1-2N0-1M0) patients: 21 men and 13 women. 11 men (52.4%) had lymph node metastases. Levels of sex steroids – total (T) and free (fT) testosterone, progesterone (P4), estradiol (E2), estrone (E1), prolactin (PRL), LH, FSH and sex steroid-binding globulin (SSBG) – were studied in the peripheral blood, tumor and resection line by the standard RIA and ELISA methods. Results: Men showed a decrease in the blood T concentration by 1.4 times, P4 by 2 times, and E2 increase by 2.5 times, while women showed 1.3 times increase in T and 2 times decrease in P4. Metastases in men were accompanied by an increase in blood levels of anterior pituitary hormones: PRL by 2.5 times, LH by 2.3 times and FSH by 2.2 times compared to patients without metastases. Analysis of the local saturation with sex hormones in visually unchanged esophageal tissues of ESCC patients did not reveal any significant gender difference. However, women were characterized by a higher content of anterior pituitary hormones, PRL, LH and FSH, in esophageal tissues compared to men. Men with metastases showed local hyper-androgenization of tumor and surrounding tissues, as the fT to E1 ratio was increased. Conclusions: The results confirmed the data on T-stimulation of malignant growth in the esophagus and the possible protective effect of estrogens. The free testosterone to estrogens ratio in esophageal tissues can probably serve as one of prognostic factors for the disease course.

Steroidogenic Factor 1, Pit-1, and Adrenocorticotropic Hormone: A Rational Starting Place for the Immunohistochemical Characterization of Pituitary Adenoma

Context.—Pituitary adenoma classification is complex, and diagnostic strategies vary greatly from laboratory to laboratory. No optimal diagnostic algorithm has been defined. Objective.—To develop a panel of immunohistochemical (IHC) stains that provides the optimal combination of cost, accuracy, and ease of use. Design.—We examined 136 pituitary adenomas with stains of steroidogenic factor 1 (SF-1), Pit-1, anterior pituitary hormones, cytokeratin CAM5.2, and α subunit of human chorionic gonadotropin. Immunohistochemical staining was scored using the Allred system. Adenomas were assigned to a gold standard class based on IHC results and available clinical and serologic information. Correlation and cluster analyses were used to develop an algorithm for parsimoniously classifying adenomas. Results.—The algorithm entailed a 1- or 2-step process: (1) a screening step consisting of IHC stains for SF-1, Pit-1, and adrenocorticotropic hormone; and (2) when screening IHC pattern and clinical history were not clearly gonadotrophic (SF-1 positive only), corticotrophic (adrenocorticotropic hormone positive only), or IHC null cell (negative-screening IHC), we subsequently used IHC for prolactin, growth hormone, thyroid-stimulating hormone, and cytokeratin CAM5.2. Conclusions.—Comparison between diagnoses generated by our algorithm and the gold standard diagnoses showed excellent agreement. When compared with a commonly used panel using 6 IHC for anterior pituitary hormones plus IHC for a low-molecular-weight cytokeratin in certain tumors, our algorithm uses approximately one-third fewer IHC stains and detects gonadotroph adenomas with greater sensitivity.