To determine the impact of diagnostic TORS lingual tonsillectomy (DTLT) on objective swallowing measures for carcinoma of unknown primary (CUP).
Between 10/2016-1/2020, 27 patients with p16+ squamous cell carcinoma (SCC) level 2a nodal disease underwent DTLT and ipsilateral neck dissection for CUP. No patient had a history of cutaneous SCC. Patients participated in Modified Barium Swallow (MBS) three weeks post-TORS, which were then compared to those from a contemporaneous cohort of 40 patients with clinically-identified p16+ base of tongue (BOT) primary tumors. DIGEST scores were retrospectively calculated. Univariate and multivariate analysis performed, stratified by BOT glossectomy (n = 40) versus lingual tonsillectomy for CUP (n = 27). Radiation to the resected primary or potential primary sources was omitted if margins were ≥3 mm or if no primary identified.
Twenty-seven consecutive patients with clinical stage cT0N1 HPV-associated OPSCC had a BOT primary pathologically identified in 18/27 (67%). Univariate analysis of functional swallow assessment on MBSImP correlated with improved post-TORS DIGEST scores for CUP. On multivariate analysis (MVA) DIGEST safety scores were improved for CUP than cT1 BOT glossectomy [Odds Ratio (OR) 0.28, p = 0.038]. MVA on matched pT1 CUP (n = 27) vs. pT1 BOT (n = 19), OR of moderate/severe dysphagia for CUP was 0.54 [0.12-2.38, p = 0.417] for DIGEST safety scores and 0.27 [0.06-1.18, p = 0.082] for DIGEST efficiency scores. Moderate/severe dysphagia as determined by DIGEST overall scores for CUP compared to cT1 and pT1 yielded an OR of 0.39 (p = 0.081) and 0.42 (p = 0.195), respectively. Twenty-six total patients received adjuvant RT, and 18 (11 with ≥3 mm margins, 9 with negative specimens) were spared intentional RT to the oropharynx. Median follow-up was 22.6 months with 100% PFS.
Patients undergoing DTLT for CUP demonstrated acute swallow defecits in the post-operative setting. A comparison of long-term functional results between DTLT and elective irradiation of the primary site should be studied.
Level III.

Copyright © 2021 Elsevier Ltd. All rights reserved.