Association of Core Biopsy With Extranodal Extension in Surgically Treated Human Papillomavirus–Associated Oropharyngeal Squamous Cell Carcinoma

Abstract

Extranodal extension (ENE) is an adverse feature in human papillomavirus (HPV)-associated oropharyngeal squamous cell carcinoma (OPSCC) and is an indication for adjuvant treatment escalation. Preoperative core needle biopsy (CNB) may cause lymph node capsule disruption related to ENE development; however, evidence regarding this association in OPSCC is lacking. To assess whether preoperative nodal CNB is associated with presence of ENE in final pathology findings among patients with HPV-associated OPSCC targeted for primary surgical resection. This retrospective cohort study was conducted at a single academic tertiary care center from 2012 to 2022. All patients with OPSCC treated with transoral robotic surgery were assessed for eligibility, and primary surgical patients with HPV-associated OPSCC and node-positive disease confirmed on neck dissection were included in analyses. Data were analyzed from November 28, 2022, to May 21, 2023. Preoperative nodal CNB. The primary outcome was presence of ENE in final pathology findings. Secondary outcomes included adjuvant chemotherapy and recurrence rates. Outcomes of interest were investigated against patient demographic, clinical, and pathologic features. Of 106 patients (mean [SD] age, 60.2 [10.9] years; 99 [93.4%] men) included in analyses, 23 patients (21.7%) underwent CNB. Mean (range) preoperative node size was 3.0 (0.9-6.0) cm. Pathologic node class was pN1 in 97 patients (91.5%) and pN2 in 9 patients (8.5%). A total of 49 patients (46.2%) had ENE identified in final pathology analysis. Of 94 patients who received adjuvant therapy, 58 (61.7%) underwent radiation therapy and 36 (38.3%) underwent chemoradiation therapy. There were 9 recurrences (8.5%). In univariate analysis, CNB was associated with ENE (odds ratio [OR], 2.70; 95% CI, 1.03-7.08), but there was no association in a multivariable model including pN class and preoperative node size (OR, 2.56; 95% CI, 0.97-7.27). Compared with pN1 class, pN2 class was associated with ENE (OR, 10.93; 95% CI, 1.32-90.80). There were no associations of ENE with preoperative node size, presence of cystic or necrotic nodes, fine needle aspiration, tobacco or alcohol exposure, pathologic T class, prior radiation, or age. Furthermore, use of CNB was not associated with macroscopic ENE, adjuvant chemotherapy, or recurrence. This cohort study of patients with HPV-associated OPSCC found that preoperative nodal CNB was strongly associated with ENE in final pathology, supporting the possibility of an artifactual ENE component in this population.