Photo Credit: iStock.com/libre de droit

The following is a summary of “Accuracy and Prognosis of Extranodal Extension on Radiologic Imaging in Human Papillomavirus-Mediated Oropharyngeal Cancer: a Head and Neck Cancer International Group (HNCIG) real-world study,” published in the April 2025 issue of the International Journal of Radiation Oncology, Biology, Physics by Mehanna et al.

Extranodal extension on radiology (iENE) has been suggested as a negative prognostic factor in human papillomavirus-mediated oropharyngeal cancer (HPV+ OPC) and is often a key criterion in surgical decision-making. However, existing single-center studies report wide variability in the sensitivity, specificity, and interobserver reliability of iENE assessments. This study aimed to evaluate the prognostic significance, diagnostic accuracy, and clinical implications of iENE in real-world practice.

A retrospective cohort analysis was conducted, including 821 consecutive patients with p16-positive OPC treated with either surgery or chemoradiotherapy (CRT) across 13 secondary care centers in nine countries between January 1, 1999, and December 31, 2020. After exclusions, 638 patients were included in the final analysis. Outcomes of interest were sensitivity, specificity, and OS, with imaging assessments performed by radiologists blinded to patient outcomes.

Among the cohort, 394 patients were classified as iENE-negative; however, 27.7% (109/394) of these were found to have pENE upon histologic examination. Conversely, 56.8% (109/192) of patients with histologic pENE were misclassified as iENE-negative on imaging. The overall sensitivity and specificity of iENE detection were 44.5% (95% CI, 37.8–51.4%) and 87.6% (95% CI, 84.1–90.6%), respectively, though these measures varied considerably between centers. The negative predictive value was 75.3% (95% CI, 72.3–77.5%).

Subgroup analysis revealed that the combination of both CT and MRI imaging significantly improved diagnostic performance. Sensitivity and specificity increased to 84.6% (95% CI, 65.1–95.6%; p < 0.001) and 94.5% (95% CI, 82.3–99.4%; p = 0.022), respectively, compared to the use of CT or MRI alone. Assessments by specialist head and neck radiologists demonstrated higher specificity (89.1% vs. 46.7%, p < 0.001) while maintaining similar sensitivity relative to non-specialists.

Multivariable analysis indicated that iENE positivity was not an independent predictor of overall survival (adjusted hazard ratio [aHR], 1.50; 95% CI, 0.97–2.32; p = 0.071) or disease-free survival (aHR, 1.41; 95% CI, 0.95–2.09; p = 0.089). Proposed modifications to the TNM staging system incorporating iENE did not substantially enhance prognostic performance.

In conclusion, iENE demonstrated modest and variable accuracy across institutions and did not independently predict outcomes in HPV+ OPC. Diagnostic performance improved with combined imaging modalities and specialized radiologic expertise. The findings underscore the urgent need for validated, standardized diagnostic criteria to optimize the clinical utility of iENE. Until such protocols are established, clinicians should exercise caution when basing treatment decisions on radiologic iENE findings.

Source: redjournal.org/article/S0360-3016(25)00370-0/abstract