Transoral robotic surgery in HPV+ oropharyngeal cancer of unknown primary.

ABSTRACT

PURPOSE: This study was designed to assess trends in and outcomes associated with TORS-treated HNCUP using a large national database. MATERIALS AND METHODS: HPV+ oropharyngeal HNCUPs were isolated from the 2004-2017 National Cancer Database. Overall survival (OS) was assessed, with patients stratified by 1) use of TORS and 2) whether the occult tumor was ultimately located. Demographic and oncologic predictors of survival were evaluated on regression. RESULTS: The cohort contained 284,734 cases, of which 8336 were HNCUPs. HNCUPs represented 2.49 % of all HNSCC in 2010 versus 3.13 % in 2017. 3897 (46.7 %) of these unknown primaries were ultimately identified. The proportion of cases treated with TORS increased from 6.9 % in 2010 to 18.1 % in 2017 (p < 0.001). Kaplan-Meier analysis of 2991 HPV+ oropharyngeal HNCUPs demonstrated higher 5-year overall survival (OS) for patients treated with robotic surgery versus no robotic surgery (95.4 % ± 1.7 % standard error [SE] versus 84.0 % ± 0.9 % SE; p < 0.001). Patients with primary tumors identified during treatment had improved OS compared to those whose tumors were not located (5-year OS was 90.5 % ± 0.9 % SE and 77.3 % ± 1.5 % SE, respectively; p < 0.001). For patients in which the primary tumor was found, those who received robotic surgery survived longer than those who did not (96.5 % ± 1.4 % SE versus 89.1 % ± 1.0 % SE 5-year OS; p < 0.001). The relationship between TORS and OS remained significant on Cox regression controlling for confounders. CONCLUSIONS: Use of TORS in the workup for HPV+ HNCUP is associated with higher rates of tumor identification and improved OS.