Objective measures of trismus and salivation poorly predict patient-reported outcomes following radiation for head and neck cancer

Presentation: AHNS17
Topic: Functional Outcomes / Quality
Type: Oral
Date: Wednesday, April 27, 2022
Session: 2:00 PM – 2:45 PM Quality of Life and Functional Outcomes
Authors: Sallie M Long, MD1; Annu Singh, BDS2; Amy L Tin, MA3; Bridget O'Hara, BSN, RN2; Marc A Cohen, MD, MPH4; Nancy Lee, MD5; David G Pfister, MD6; Tony Hung, MD, MBA, MSCR6; Richard J Wong, MD4; Andrew J Vickers, PhD3; Cherry L Estilo, DMD2; Jennifer R Cracchiolo, MD4
Institution(s): 1NewYork-Presbyterian Hospital/Weill Cornell; 2Dental Service, Memorial Sloan Kettering Cancer Center; 3Health Outcomes Research Group, Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center; 4Head and Neck Service, Department of Surgery, Memorial Sloan Kettering Cancer Center; 5Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center; 6Department of Medicine, Medical Oncology, Memorial Sloan Kettering Cancer Center


Xerostomia and reduced mouth opening negatively impact quality-of-life (QoL) following radiation treatment (RT) for head and neck cancer, but studies directly correlating quantitative measures of function with patient-reported outcomes (PROs) are lacking. The primary aim of this study is to correlate objective measures of salivary gland and oral cavity functions (stimulated and unstimulated salivary flow, maximal interincisal opening [MIO]) with subjective PRO scales on salivation and eating using a validated PROs instrument. A secondary aim is to describe trends in objective and PRO measures over time following RT.

Methods: 116 patients who underwent RT for head and neck cancer between January 2016 and March 2021 were identified. Patients had pre-treatment MIO and saliva measurements, at least one post-RT measurement, and a completed PROs questionnaire within 6 months of the post-RT measurement. Three independently scored PROs scales from the FACE-Q Head and Neck Cancer measure were analyzed (with higher scores reflecting better outcome): 1) Eating and Drinking, 2) Eating Distress, and 3) Salivation. To determine how much of the variation in PROs scores could be explained by objective measures, univariable linear regression models were performed for each PRO against each objective measure, and coefficients of determination (R2) were reported. Trends in objective and subjective measures over time were plotted using local weighted smoothing regression analysis.

Results: 113 patients were analyzed with a male predominance (n=87, 77%) and a median age of 61 years (IQR 53, 68). The majority had oropharynx tumors (n=64, 57%). Approximately one-third of patients underwent neck dissection (n=33, 29%), of whom 21 patients had dissection of level 1B. The R2 for Eating Distress and Eating and Drinking paired with stimulated saliva was 5.0 and 9.6%, respectively. R2 for unstimulated saliva was 5.3 and 6.8%, respectively. For the Salivation scale, stimulated and unstimulated saliva correlations were 8.5 and 13.8%, respectively. R2 for MIO against the PRO scores was 8.3% for Eating Distress and 10.4% for Eating and Drinking. MIO and saliva production (stimulated and unstimulated) remained relatively stable over time following RT. Eating and Drinking scores increased from 62 (95% CI 50, 74) immediately after RT to 88 (95% CI 80, 96) at year 1 prior to stabilizing. Eating Distress scores similarly increased from 55 (95% CI 42, 69) immediately after RT to 90 (95% CI 81, 100) at year 1 prior to stabilizing. Salivation scores remained constant over time (64 [95% CI 49, 79] immediately after RT and 64 [95% CI 54, 75] at year 1 after RT).

Conclusion: Objective measures of oral cavity and salivary function only explain a small fraction of changes in PRO scales built to measure these endpoints. Objective measures are relatively stable over time following RT, while PROs on Eating and Drinking and Eating Distress scales improve for the first year before stabilizing. This study highlights the importance of integrating PRO measures in head and neck cancer care. Future directives include creation of normative data trends and targeted interventions.