Background: Transoral robotic surgery (TORS) with or without postoperative adjuvant radiation (RT) or chemoradiation (CRT) has become an acceptable alternative to definitive RT or CRT for select oropharyngeal cancers (OPSCC). The historically reported rates of subclinical cervical nodal metastases in OPSCC predate the emergence of Human Papilloma Virus (HPV) as the predominant causative agent. The rate of occult nodal disease in the current era of OPSCC dominated by HPV+ is not known, and consequently it is challenging to predict which patients will be upstaged postoperatively and require adjuvant therapy. Knowledge of the rate post-operative upstaging due to occult cervical disease or pathologic extranodal extension (ENE) would have a significant impact on treatment regimens.
Objective: The goal of this study was to determine the rate of nodal upstaging and occult ENE in a multi-institutional population of OPSCC patients treated with TORS and neck dissection (ND).
Methods: This retrospective, multicenter study examined the rate of post-operative pathologic upstaging for patients undergoing TORS with ND for OPSCC performed at the participating institutions: Princess Margaret Cancer Center, Toronto; Icahn School of Medicine at Mount Sinai, New York City; and Montefiore Medical Center, New York City. A neuroradiologist at each site blinded to final pathological diagnosis reviewed pre-operative imaging. These findings were compared to operative pathology. The AJCC 7th Edition was applied for staging.
Results: Ninety-five patients were identified that met inclusion, 85% of which were p16+. The mean age at surgery was 60 (SD = 10.5). There were 78 (82%) male and 17 (18%) female patients. The preoperative radiographic nodal stages were N2b n = 45 (47%), N0 n=19 (20%), N1 n=13 (14%), N2a n=9 (9%), N2c n= 8 (8%) and N3 n= 1 (1%). Thirty-three (35%) patients underwent bilateral neck dissections, and 62 (65%) underwent ipsilateral neck dissection only. Five of 19 (26%) patients with no evidence of nodal disease on imaging had occult nodal disease. Eight of 32 (25%) patients presenting with radiographic evidence of N0 or N1 disease on imaging were upstaged to more advanced disease, indicating implications for additional adjuvant treatment not predicted on a priori imaging. Nineteen patients (20%) were nodally upstaged postoperatively, and 27 (28%) were nodally downstaged postoperatively. Fifty-three patients (56%) had radiographic ENE or were suspicious for ENE preoperatively. Twenty-four patients (26%) had pathologic ENE in the cohort. Five of 42 patients (12%) had occult ENE in the absence of radiographic evidence. No patients with radiographic N0 disease had occult ENE. The sensitivity and specificity for pathologic ENE based on preoperative imaging were 79% (95% CI = 58%-93%) and 54% (95% CI = 42%-67%), respectively. The presence of radiographic nodal positivity (p=0.0053) and >1 suspicious node (p=0.0071) were associated with occult pathologic ENE on univariate analysis.
Conclusion: Predicting pathologic staging preoperatively for OPSCC patients undergoing TORS and neck dissection remains a challenge. Our findings suggest a small proportion of patients would require further adjuvant therapies not predicted on preoperative imaging based on occult nodal disease and ENE, and nearly one-quarter of patients will be down-staged postoperatively.