Advancing Education, Research, and Quality of Care for the Head and Neck oncology patient.
Background: Given the co-morbidity associated with cachexia, previous literature has accentuated the significance of greater body mass index (BMI) in improving patient outcomes despite co-morbidities associated with obesity. However, there is little data on the impact of glucagon-like peptide-1 receptor agonists (GLP1-RA) on outcomes for obese patients diagnosed with head and neck cancer. GLP1-RAs improve diabetes control and chronic weight loss management, but we seek to evaluate the association of GLP1-RA use and overall survival, functional outcomes, and post-operative complications in head and neck cancer (HNC) patients.
Design Type: National multicenter cohort study
Methods: Population-level data was queried utilizing the TriNetX Research Network with Natural Language Processing. TriNetX collects real-time multicenter data from 95 healthcare organizations in the U.S. with deidentified information for over 130 million patients representing diverse socioeconomic and demographic backgrounds. Patients with ICD-10 HNC diagnoses and BMI ≥ 30 (before and after HNC diagnosis and treatment) were divided into the following groups: never prescribed GLP1, prescribed a GLP1 at any time, prescribed GLP1 before HNC diagnosis, and prescribed GLP1 after HNC diagnosis. The following GLP1-RA were queried: semaglutide, liraglutide, exenatide, and tirzepatide. Demographic and co-morbidity data was collected and utilized to propensity score match cohorts. Overall survival and risk for functional deficit outcomes (dysphagia, gastrostomy tube placement/dependence, tracheostomy placement/dependence) was compared between GLP1 users and never users. Post-operative (within 1 month of surgery) complications were compared between patients with recent GLP1 use (1 year before and after surgery for HNC) and without. The following complications were evaluated: surgical site/post-operative infection, postoperative hemorrhage/hematoma, emergency room visits, and thromboembolic events. Univariate analysis was performed for descriptive statistics, odds ratios (ORs), and Cox proportional hazards ratio (HR) calculations.
Results: 10,495 patients were diagnosed with HNC, were obese, and used GLP1-RA sometime in their lives. 6,182 were prescribed before HNC diagnosis; 4,569 were prescribed GLP1 after; and 1,341 were prescribed GLP1 within 1 year of HNC surgical treatment. Before matching, GLP1 patients tended to be younger (56.2 vs 60.4), female, and white. They were also significantly more like to have co-morbidities, including hypertension, diabetes, liver disease, kidney disease, asthma, and more. After matching, recent GLP1 use was associated with significantly fewer peri-operative emergency visits and lower risk (not significant) of postoperative complications: emergency visits [OR = 1.80, 95% CI (1.24,2.61)], thromboembolic event [1.20,(0.517,2.80)], postoperative hemorrhage/hematoma [2.02,(0.941,4.34)], postoperative/surgical site infection [1.92,(0.887,4.14)]. GLP1 use was also associated with improved functional outcomes and 5-year survival than no GLP1 use, regardless of timing of GLP1 prescription (before or after HNC diagnosis/treatment): HR for 5-year mortality between no GLP1 use and recent GLP1 use [1.86 (1.61,2.14)], dysphagia [1.72,(1.56,1.90)], gastrostomy tube placement/dependence [2.80,(2.04,3.84)], tracheostomy tube placement/dependence [2.59,(2.07,3.24)].
Conclusion: GLP1-RA use is associated with fewer peri-operative emergency visits, better 5-year survival, and better functional outcomes than non-users within the obese HNC patient population. While comparisons of obese vs non-obese patient outcomes have been studied, this is the first study to our knowledge that compares outcomes within the obese population when it comes to GLP1-RA use.