Characterizing Voice Changes and Primary Aberrant Features in Survivors of Locally Advanced Head and Neck Cancer Treated with Radiation Therapy
A. Hubler,C. Cooper,K. Heinzman,E. Hapner,W. Boswell,A.M. McDonald
DOI: https://doi.org/10.1016/j.ijrobp.2024.01.108
IF: 8.013
2024-04-01
International Journal of Radiation Oncology*Biology*Physics
Abstract:Purpose/Objective(s) Voice changes are common in head and neck cancer (HNC) survivors who undergo radiation therapy (RT). Most studies have assessed changes with validated patient reported outcomes measures (PROMS). There is a need to combine PROMS with patient voice recordings. The purpose of this study was to characterize voice changes using a standardized protocol and describe primary aberrant features using expert listeners. Materials/Methods This study enrolled long-term survivors (>2 years) of locally advanced (T3+ or N+) HNC who received ≥60 Gy at a single academic center. Patients completed a standardized protocol of several vocal tasks. Recordings included a single reading of the first 4 sentences of the Rainbow Passage and sustained /a/ and /i/ edited to the first 3 seconds of each phoneme. Nine blinded speech language pathologists with voice specialization rated dysphonia using the CAPE-V visual analog scale. Scores ranged 1-100 with higher scores indicating greater dysphonia. Mean scores across the 9 listeners were calculated and severity of dysphonia was classified for this study as normal [0-12), mild [12-20), mild-moderate [20-30), moderate [30-45), moderate-severe [45-60), severe [60-85), and severe-profound [85-100]. A 10% repeat assessment was used for intra-rater reliability. Cronbach's alpha was calculated to assess inter-rater consistency. Primary aberrant features were roughness, breathiness, tremor, spasm, strain, and asthenia. Changes in voice resonance were indicated with a binary yes/no. Results Two hundred one patients were enrolled with 198 recordings available for analysis. Median time from completion of RT was 5.6 years (range 1.7-28.9 years). Most common primary disease sites were oropharynx (51%), larynx (12%), and oral cavity (11%). Seventy-two percent of patients were male. Surgery and systemic therapy were included in 58% and 63% of treatment courses, respectively. Mean CAPE-V score for all patients was 31.9 (IQR 19.9 – 38.1). Dysphonia was classified as normal in 6.6%, mild in 18.7%, mild-moderate in 28.8%, moderate in 27.2%, moderate-severe in 11.1%, severe in 7.1%, and severe-profound in 0.5% of patients. In the validation cohort, the median percent difference in CAPE-V scores between initial and repeat assessments was 5.5% (IQR 2.0-12.5%). Cronbach's alpha was 0.95. In 64.6% of patients, at least 50% of listeners reported voice roughness as the primary aberrant feature, while breathiness, tremor, spasm, strain, and asthenia were reported in 2%, 0.5%, 0%, 7.6%, and 1.5%, respectively. Resonance changes were noted in 20.7% of patients by over half of listeners. Conclusion Dysphonia was present in many long term HNC survivors following RT with 45% having at least moderate dysphonia. Voice roughness was the most common primary aberrant feature with many bearing changes in resonance. Further understanding of clinical factors predictive of vocal changes and treatment factors mediating changes is needed to develop strategies to mitigate this late toxicity.
oncology,radiology, nuclear medicine & medical imaging