Abstract

It is unclear whether the log odds of positive lymph nodes (LODDS) outperforms the number of positive lymph nodes (LN+) in predicting the overall survival (OS) of oral squamous cell carcinoma (OSCC) patients. The specific aim of this study was to compare the prognostic predictive performance of LN+ with LODDS in OSCC patients. This prospective cohort study was conducted in Fujian, China, from December 2005 to January 2017. Patients' characteristics and clinicopathologic data were obtained through medical records, and follow-up data were obtained by telephone interviews. Cox proportional hazards models were used to evaluate the association between LN+or LODDS and OS in OSCC. Finally, the Harrell concordance index, Akaike information criterion, and area under the receiver operating characteristic curve were adopted as criteria for assessing the predictive performance of lymph node models. For all 706 patients, the 5-year survival rate was 65.69% (95% confidence interval, 0.61 to 0.70) and the mean age at diagnosis was 57.32±11.80years. Of the patients, 456 were men and 250 were women (ratio of 1.82:1). LN+ and LODDS were significantly associated with a poor prognosis of OSCC patients (all P values for trend<.001). Furthermore, the prognostic value of LODDS was not better than that of LN+. An interesting finding was that there was a J-shaped relationship between the number of negative lymph nodes and OS. The hazard ratio was reduced with each additional negative lymph node dissected up to 24 negative lymph nodes, with no improvement in prognosis beyond this number. Moreover, when the number of negative lymph nodes was greater than 40, the negative lymph nodes suggested a worse prognosis for OSCC patients. Our study suggests that the discriminatory capability of LODDS was not superior to that of LN+. An interesting finding was that, when the number of negative lymph nodes was greater than 40, the predictive power of LODDS was reduced tremendously.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call