Nodal yield and log odds ratio are associated with survival outcomes in early stage clinically node-negative oral cavity squamous cell carcinoma

Presentation: C007
Topic: Oral Cancer
Type: Poster
Date: Thursday, April 19, 2018
Session: 9:00 AM - 7:00 PM
Authors: Anuraag S Parikh, MD, Ayaz Khawaja, MD, Besher Assi, Joseph Zenga, MD, Mark A Varvares, MD
Institution(s): Massachusetts Eye and Ear Infirmary

Background: The number of lymph nodes (LN) harvested during lymphadenectomy, as well as the log of the ratio of positive to negative LN (known as the log odds ratio), has previously been associated with improved overall survival and locoregional control. We sought to evaluate whether a nodal yield (NY) and the log odds ratio (LODDS) may be associated with survival outcomes in early stage, clinically N0 oral cavity squamous cell carcinoma (OCSCC).

Methods: Treatment-naïve patients treated surgically for clinically node-negative T1 or T2 OCSCC between 2005-2014 were identified using the institutional cancer registry. Clinical and pathologic data were collected retrospectively from patient charts. All patients underwent surgical resection of the primary tumor, as well as upfront neck dissection. LODDS was calculated as log [(No. positive nodes + 0.5) / (No. total nodes – No. positive nodes + 0.5)] and stratified as LODDS < -1.5 (category A), -1.5 < LODDS < -1.0 (category B), and -1.0 < LODDS < 0 (category C). Primary outcomes were overall (OS) and disease-free survival (DFS).

Results:A total of 164 patients were identified. Mean age of the patients was 63.8 years (SD 13.8). Forty patients had had positive nodes. Average NY was 21 (range 1-76; median 19). Cutoffs of 15, 18, and 21 were used to assess the association of high NY with survival outcomes. DFS at 2 and 5 years was was 76% and 58% in patients with NY ≥ 18 and 63% and 37% in patients with NY < 18 (p=0.01). Similar differences were observed with cutoffs of 15 (p=0.03) and 21 (p=0.04) for NY. For all three cutoffs, there was no statistically significant difference in OS.

The LODDS ratio was significantly associated with both OS and DFS. For patients in categories A, B, and C, 2-year OS was 89%, 77% and 64%, while 5-year OS was 77%, 62%, and 47% respectively (p=0.02). Similarly, the 2-year DFS was 78%, 70% and 47%, while 5-year DFS was was 61%, 44%, and 25%, respectively (p=0.001).

Conclusions: A higher NY was associated with an improvement in locoregional control, but not OS. The LODDS ratio was similarly associated with better locoregional control, but was also associated with improved OS. Thus, a more complete neck dissection may improve prognosis in early stage, clinically N0 OCSCC.