Surgical margins in a single‐modality transoral robotic surgery: A conundrum—Reply

Abstract

We appreciate the authors thoughtful commentary on our work and the opportunity to reply to the concerns raised about both our manuscript and this topic in general. We agree that surgical margins in human papillomavirus-related oropharynx squamous cell carcinoma (HPV+OPSCC) is an area of contention and one deserving of further investigation. The authors are correct in noting that slightly over half of the patients in our study had margins less than 3 mm. However, they fail to note that although only 13 patients had margins ranging from 3 to 4.9 mm, an additional 30 patients (30.3%) had wider margins and were included in comparative analyses. While individual subgroups are small, no statistical comparisons are made in the manuscript between the mentioned subgroups; thus, the focus on these individual subgroups is somewhat misleading. Nonetheless, our results demonstrate that local control did not differ between clear and close margins. We concede that the overall sample is small as is the event rate of local recurrence and that presenting negative results in small studies is subject to type 2 error. The median duration of follow-up of 21 months is also highlighted in this letter as a limitation of our study. Although this period is relatively limited in duration, multiple studies have demonstrated short median times to local recurrence in surgically treated HPV+OPSCC ranging from 0.7 to 1.6 years.1–3 In our study, no patient experienced local recurrence beyond 18 months of follow-up. While we noted in our study discussion that these patients remain at risk for recurrence, this study duration appears adequate to draw conclusions regarding local control in this patient population. We will continue to follow this cohort and plan to present long-term data when mature. The authors comment on specimen versus tumorbased margins, noting superiority of specimen-based margin assessment. The authors reference an article on oral cavity cancer, for which multiple recent studies support specimen-based margins. While this may hold true in oral cavity cancer, we are unable to identify any studies that examined differences between these approaches in HPV+OPSCC in a dedicated fashion. In a meta-analysis of 24 studies on surgically managed HPV+OPSCC by Gorphe et al., 10 studies used specimen-based margins, four employed defect-based margins, and 10 did not describe their methodology. They were unable to identify differences in outcomes based on this approach and no conclusive evidence supports either approach in OPSCC. In our study, there were significant differences by institution in use of specimen verses defect-based frozen section analysis, and this heterogeneity is further highlighted in the study by Gorphe et al. The reviewers comment on secondary re-resection for margin clearance again references a study on oral cavity cancer, and while it is logical that tissue distortion and potential for sampling error may limit the effectiveness of re-resection in HPV+OPSCC as well, it is notable that no patients undergoing delayed re-resection in our study (n = 13, median follow-up 21 months) experienced local recurrence. Moore et al. did identify lower regional control, disease-specific survival, and overall survival among patients requiring margin re-resection in TORS in univariate analysis. However, 3-year local recurrencefree survival exceeded 96% in the group requiring reresection (n = 79). It is unclear why re-resection was associated with regional recurrence rather than local recurrence in their study, although they postulate that the need for re-resection may be reflective of a more aggressive tumor biology. We agree with the authors of this letter that caution should be employed when considering the validity of re-resection margins. The authors also comment on the relatively low rate of contralateral neck dissection performed in base of tongue (BOT) tumors. This intervention is unrelated to the primary study outcome of local control, but it is important to note that controversy remains in the literature regarding the necessity of contralateral neck dissection in lateralized BOT HPV+OPSCC. In the cited study, only 5.6% of patients undergoing contralateral neck dissection for BOT HPV+OPSCC were clinically N0, and 37% were clinically N2-N3. In their study, only 15 patients had lateralized tumors with clinically N0 contralateral necks, of which four (26.7%) ultimately had contralateral pathologic N+ status. It is unclear what the nodal status of the ipsilateral neck was in these 15 patients, which is of great importance in making the Received: 30 June 2021 Accepted: 26 July 2021