Cushing's disease: predicting long-term remission after surgical treatment

Abstract

<jats:p>Cushing's disease (CD) is a state of excess glucocorticoid production resulting from an adrenocorticotropic hormone (ACTH)–secreting pituitary adenoma. The gold-standard treatment for CD is transsphenoidal adenomectomy. In the hands of an experienced neurosurgeon, gross-total resection is possible in the majority of ACTH-secreting pituitary adenomas, with early postoperative remission rates ranging from 67% to 95%. In contrast to the strong data in support of resection, the clinical course of postsurgical persistent or recurrent disease remains unclear. There is significant variability in recurrence rates, with reports as high as 36% with a mean time to recurrence of 15–50 months. It is therefore important to develop biochemical criteria that define postsurgical remission and that may provide prognosis for long-term recurrence. Despite the use of a number of biochemical assessments, there is debate regarding the accuracy of these tests in predicting recurrence. Here, the authors review the various biochemical criteria and assess their utility in predicting CD recurrence after resection.</jats:p>

Prediction of Recurrence after Transsphenoidal Surgery for Cushing’s Disease: The Use of Machine Learning Algorithms

“…Incorporating feature selection and univariate analysis, some results corroborated the previous studies, suggesting that younger age [43, 44], higher postoperative morning serum cortisol [6, 9-13], and higher postoperative morning ACTH [11, 14, 45, 46] had a detrimental effect on recurrence.…”

“…Previously, the postoperative serum cortisol nadir was the common risk factor to predict recurrence after TSS for CD; however, it had poor accuracy according to many studies [8, 11, 15, 41, 42]. We showed that the performance of postoperative morning serum cortisol nadir in predicting CD recurrence was poor (AUC = 0.635) and significantly worse than that of most ML-based predictive models.…”

“…Some studies suggested that patients with subnormal postoperative cortisol levels [6, 9-13], low postoperative UFC levels [11], or low postoperative ACTH levels [14] tended to have a lower risk of recurrence after initial TSS for CD, while others indicated that there were no reliable risk factors [41, 42]. A meta-analysis [15] also concluded that no reliable predictors of recurrence after TSS were found.…”

“…However, other studies demonstrated that there was no relationship between age and recurrence [4, 6, 11, 15, 47]. In most of these studies, the Student t test was used to compare age between the recurrence group and non-recurrence group, and no significant difference was found.…”

“…Some studies have shown that patients with subnormal postoperative cortisol levels [6, 9-13], low postoperative urinary free cortisol (UFC) levels [11], or low postoperative ACTH levels [14] tended to face a lower risk of recurrence after initial TSS for CD. However, no clear predictive models for recurrence were developed.…”