Patient-reported symptom scores may help identify oropharyngeal cancer survivors at higher risk for delayed hypoglossal nerve neuropathy, according to findings published in JAMA Otolaryngology–Head & Neck Surgery.
In a retrospective analysis of prospectively collected patient-reported outcomes data, researchers evaluated 1,297 patients with oropharyngeal cancer treated with curative intent at The University of Texas MD Anderson Cancer Center from 2015 to 2023. Neuropathy status was retrospectively classified through structured clinical record review based on documented examination or imaging findings, introducing potential ascertainment bias. Eligible patients completed symptom assessments at baseline and during follow-up, including at least one assessment at 18 to 24 months or 60 months.
The researchers analyzed six items from the MD Anderson Symptom Inventory–Head and Neck module: chewing or swallowing difficulty, choking, speech or voice changes, mucus, fatigue, and dry mouth. These items were averaged to create the MDASI-HN-NERVE score, a composite intended to reflect bulbar-related symptom burden rather than diagnose cranial nerve XII neuropathy.
Clinically detected hypoglossal nerve neuropathy was identified through structured medical record review. Patients were classified as positive if they had at least two documented abnormalities, such as lingual atrophy with deviation. Patients classified as having “possible” neuropathy based on a single abnormal finding were grouped with negative cases in time-to-event analyses, a conservative approach that may have attenuated observed associations if some represented evolving neuropathy.
At 5 years, the cumulative incidence of clinically detected hypoglossal nerve neuropathy was approximately 4%. Higher MDASI-HN-NERVE scores were associated with greater neuropathy risk; each 1-point increase in the composite score was associated with a 1.35-fold higher hazard of neuropathy. The time-varying Cox model demonstrated moderate discrimination, with a C-index of 0.71.
Thresholds identified using maximally selected rank statistics within the same cohort also stratified risk. Patients with MDASI-HN-NERVE scores of at least 3.4 at baseline or at least 3.5 at 3 to 6 months following treatment had lower neuropathy-free survival through 60 months than patients with lower scores. The researchers cautioned that internally optimized thresholds may overestimate discriminatory performance and require external validation.
A single symptom item also showed prognostic value. Patients reporting speech or voice scores of at least 2 at 3 to 6 months had a 13.18-fold higher hazard of hypoglossal nerve neuropathy than patients with lower scores. However, the researchers noted that the threshold was optimized within the same cohort and that positive predictive values were modest because hypoglossal nerve neuropathy remained uncommon.
The study also included a sensitivity analysis excluding fatigue and dry mouth from the composite score. Similar association patterns were observed using the more bulbar-focused symptom construct, supporting the robustness of the findings while limiting mechanistic overinterpretation of the full six-item composite.
Symptom trajectories differed by neuropathy status. Patients with confirmed neuropathy had higher baseline symptom burden and more persistent or worsening symptoms over time, particularly in speech or voice, chewing or swallowing, mucus, and choking domains. Patients without neuropathy generally maintained low, stable scores.
The findings support symptom-based surveillance as a potential risk-stratification tool, not as diagnostic confirmation. The researchers noted that the small number of confirmed neuropathy cases limited precision and that symptom-informed diagnostic bias was possible because patients with speech or swallowing complaints may have undergone more detailed neurologic evaluation.
“MDASI-HN speech/voice symptoms and the MDASI-HN-NERVE score may provide a practical, patient-centered approach to aid early detection and monitoring of CN XII neuropathy risk,” wrote Wenye Song, MD, of The University of Texas MD Anderson Cancer Center, and colleagues.
The researchers suggested that symptom-based surveillance may ultimately prove most useful for identifying patients at low risk who may not require intensive neurologic follow-up, rather than confirming neuropathy in patients with persistent symptoms.
Disclosures: The study was supported by the NIH, National Cancer Institute, National Institute of Dental and Craniofacial Research, MD Anderson institutional resources, and the Charles and Daneen Stiefel MD Anderson Oropharyngeal Fund. Several researchers reported grants, personal fees, royalties, honoraria, or other relationships outside the submitted work.
