Predicting Response to Neuromodulators or Prokinetics in Patients with Suspected Gastroparesis Using Machine Learning: the "BIDnD" Model.

INTRODUCTION: Pharmacologic therapies for symptoms of gastroparesis have limited efficacy and it is difficult to predict which patients will respond. In this study, we implemented a machine-learning model to predict the response to prokinetics and/or neuromodulators in patients with gastroparesis-like symptoms. METHODS: Subjects with suspected gastroparesis underwent simultaneous gastric emptying scintigraphy (GES) and wireless motility capsule (WMC) and were followed for 6 months. Subjects were included if they were started on neuromodulators and/or prokinetics. Subjects were considered responders if their Gastroparesis Cardinal Symptom Index (GCSI) at 6 months decreased by ≥1 from baseline. A machine-learning model was trained using lasso regression, ridge regression or random forest. Five-fold cross-validation was used to train the models and the area under the receiver operator characteristic curve (AUC-ROC) was calculated using the test set. RESULTS: Of the 150 patients enrolled, 123 patients received either a prokinetic and/or a neuromodulator. Of the 123, 45 were considered responders and 78 were non-responders. A ridge regression model with the variables: BMI, Infectious prodrome, delayed GES, no diabetes (BIDnD), had the highest AUC-ROC of 0.72. The model performed well for subjects on prokinetics without neuromodulators (AUC-ROC of 0.83) but poorly for those on neuromodulators without prokinetics. A separate model with GET, duodenal MI, no diabetes, and functional dyspepsia performed better (AUC-ROC of 0.75). DISCUSSION: This machine learning model has an acceptable accuracy in predicting those who will respond to neuromodulators and/or prokinetics. If validated, our model provides valuable data in predicting treatment outcomes in patients with gastroparesis-like symptoms.

Aprepitant Has Mixed Effects on Nausea and Reduces Other Symptoms in Patients With Gastroparesis and Related Disorders.

BACKGROUND & AIMS: There are few effective treatments for nausea and other symptoms in patients with gastroparesis and related syndromes. We performed a randomized trial of the ability of the neurokinin-1 receptor antagonist aprepitant to reduce symptoms in patients with chronic nausea and vomiting caused by gastroparesis or gastroparesis-like syndrome. METHODS: We conducted a 4-week multicenter, double-masked trial of 126 patients with at least moderate symptoms of chronic nausea and vomiting of presumed gastric origin for a minimum of 6 months. Patients were randomly assigned to groups given oral aprepitant (125 mg/day, n = 63) or placebo (n = 63). The primary outcome from the intention-to-treat analysis was reduction in nausea, defined as a decrease of 25 mm or more, or absolute level below 25 mm, on a daily patient-reported 0-to-100 visual analog scale (VAS) of nausea severity. We calculated relative risks of nausea improvement using stratified Cochran-Mental-Haenszel analysis. RESULTS: Aprepitant did not reduce symptoms of nausea, based on the primary outcome measure (46% reduction in the VAS score in the aprepitant group vs 40% reduction in the placebo group; relative risk, 1.2; 95% CI, 0.8-1.7) (P = .43). However, patients in the aprepitant group had significant changes in secondary outcomes such as reduction in symptom severity (measured by the 0-5 Gastroparesis Clinical Symptom Index) for nausea (1.8 vs 1.0; P = .005), vomiting (1.6 vs 0.5; P = .001), and overall symptoms (1.3 vs 0.7; P = .001). Adverse events, predominantly mild or moderate in severity grade, were more common in aprepitant (22 of 63 patients, 35% vs 11 of 63, 17% in the placebo group) (P = .04). CONCLUSIONS: In a randomized trial of patients with chronic nausea and vomiting caused by gastroparesis or gastroparesis-like syndrome, aprepitant did not reduce the severity of nausea when reduction in VAS score was used as the primary outcome. However, aprepitant had varying effects on secondary outcomes of symptom improvement. These findings support the need to identify appropriate patient outcomes for trials of therapies for gastroparesis, including potential additional trials for aprepitant. ClinicalTrials.gov no: NCT01149369.

Clinical presentation and pathophysiology of gastroparesis.

Gastroparesis is a heterogeneous disorder defined by delay in gastric emptying. Symptoms of gastroparesis are nonspecific, including nausea, vomiting, early satiety, bloating, and/or abdominal pain. Normal gastric motor function and sensory function depend on a complex coordination between the enteric and central nervous system. This article discusses the pathophysiology of delayed gastric emptying and the symptoms of gastroparesis, including antropyloroduodenal dysmotility, impaired gastric accommodation, visceral hypersensitivity, and autonomic dysfunction. The underlying pathophysiology of gastroparesis is complex and multifactorial. The article discusses how a combination of these factors leads to symptoms of gastroparesis.