Introduction of a New Pathology Workup Protocol for Glottic Cancer Treated With Transoral Laser Microsurgery (TLM): Prospective Analysis of Oncological Outcomes and Matched Case-Control Study.

[BACKGROUND/PURPOSE] The value of margin status after TLM for glottic cancer is debatable, due to difficulties in specimen orientation and margin analysis. To reduce these difficulties, we recently introduced a standardized protocol of oriented fixation of TLM specimens. This proved feasible and resulted in high margin evaluability rates and a decreased rate of false positive deep margins, when compared to a historical TLM cohort. For the patients whose specimens were processed according to this protocol, we prospectively analyzed oncological outcomes, identified prognostic factors and assessed the influence of the protocol introduction on outcomes compared with a historical TLM cohort.

[METHODS] Ninety-six patients with glottic malignancies treated with TLM were included. Resection specimens were processed according to the new protocol. Descriptive statistics and survival analyses were used to determine oncological outcomes. To assess the effect of the protocol introduction on outcomes, a matched-case-control analysis was performed, using a historical TLM-cohort as controls. The Cox proportional hazards model was used to analyze prognostic effects of patient and treatment characteristics, including the pathology protocol introduction, on overall survival (OS), disease-specific survival (DSS), disease-free survival (DFS) and local recurrence-free survival (LRFS).

[RESULTS] Two-year outcomes were favorable: 88.5% OS, 97.0% DSS, and 87.6% LRFS. At multivariable analysis, the presence of multiple positive superficial margins was a negative prognosticator for OS (HR 4.102) and increasing cT classification proved a negative prognosticator for DFS (HR 2.828) and LRFS (HR 2.676). Matched case-control analysis did not reveal a significant difference in oncological outcomes between cohorts. Deep margin status had a strong differential effect for DFS (p-value for interaction = 0.0205) and for LRFS (p-value for interaction = 0.0176) between cohorts, indicating a prognostic effect of deep margin status on both outcomes in the current cohort, but not in the historical cohort.

[DISCUSSION/CONCLUSION] The introduction of a new standardized technique of oriented fixation of TLM specimens did not affect oncological outcomes when compared to a historical TLM cohort, but assigned a significant prognostic effect to deep margin status for DFS and LRFS, facilitating the decision making process with regards to planning of second-look procedures, administration of adjuvant radiotherapy or determination of follow-up intensity.