The First Cut: the Importance of the Initial Specimen Margin for Oral Cavity Cancer

Abstract

During the resection of an oral cavity squamous cell carcinoma (OCSCC) the pathologist may discover tumor approaching or at the surgical specimen edge(s) that are cleared by the surgeon by performing additional resections. It is not known how patients with margins that are negative after additional excision (NAAE) compare to those with initially negative or positive margins. We sought to determine the impact of NAAE and whether it is an indication for adjuvant radiation. Two pathologists re-reviewed all available pathologic specimens of patients with OCSCC treated with surgical resection at our institution. The distance of tumor to the edge of the main surgical specimen was categorized as positive (tumor < 0.1 cm) or negative (tumor ≥ 0.1 cm). For patients with positive margins of the primary specimen, if additional excisions were taken and found to clear the margins, they were categorized as NAAE margins. The following were considered adverse risk features: perineural invasion (PNI), lymphovascular invasion (LVI), stage T3-4, and stage N2-3. We used the 8th edition of the AJCC staging system, which incorporates tumor depth and extracapsular extension (ECE). Locoregional recurrence (LRR) differences were determined with Gray’s test and Fine-Gray models. Differences in Progression Free Survival (PFS) were determined with log-rank tests and Cox regression models. A conditional landmark analysis using 3 months after resection was used. We identified 416 adult patients with OCSCC treated with surgical resection from 1998-2014 with a median follow up of 39 months. 412 specimens were available for review. 198 patients had NAAE margins, 181 had negative margins, and 37 had positive margins. When used, the median dose for adjuvant radiation was 60 Gy (range 48-70.4 Gy). Compared to negative margins, NAAE margins were associated with having a T3-4 stage, an N2-3 stage, ECE, PNI, a larger depth of invasion, and adjuvant radiation/chemoradiation. For patients with negative, NAAE, and positive margins, the 2-year estimates of LRR were 27%, 36%, and 47% (p=0.06), and 2-year estimates of PFS were 65%, 50%, and 22% (p<0.001). On multivariate analysis, compared to NAAE margins, positive margins had a trend for worse LRR (HR 2.35, p=0.13) and inferior PFS (HR 2.10, p=0.0008); whereas outcomes were more favorable for negative margins but not statistically significant (LRR HR 0.83, p=0.34; PFS HR 0.85, p=0.32). Similar results were seen for patients with adverse risk features who received adjuvant radiation for negative compared to NAAE margins (LRR HR 0.70, p=0.22; PFS HR 0.74, p=0.20). Our institutional experience indicates that there may be a benefit to having negative margins of the initial specimen compared to margins that are NAAE. Positive margins for patients with OCSCC are associated with inferior outcomes.