Validation of the prognostic value of lymph node ratio in patients with cutaneous melanoma: A population-based study of 8,177 cases

Background. The proportion of positive among examined lymph nodes (lymph node ratio [LNR]) has been recently proposed as an useful and easy-to-calculate prognostic factor for patients with cutaneous melanoma. However; its independence from the standard prognostic system TAW has not been formally proven, in a large series of patients. Methods. Patients with histologically proven cutaneous melanoma were identified from the Surveillance Epidemiology End Results database. Disease-specific survival was the clinical outcome of interest. The prognostic ability of conventional factors and LAIR was assessed by multivariable survival analysis using the Cox regression model. Results. Eligible patients = 8,177) were diagnosed with melanoma between 1998 and 2006. Among lymph node-positive cases (n = 3,872), most LAIR values ranged from 1% to 10% (n = 2,187). In the whole series (>= 5 lymph nodes examined) LAIR significantly contributed to the Cox model independently of the TNM effect on survival (hazard ratio, 1.28; 95% confidence interval, 1.23-1.32; P < .0001). On subgroup analysis, the significant and independent prognostic value of LAIR was confirmed both in patients with a >= 10 lymph nodes examined (11 =,4,381) and in those with TNM stage III disease (n = 3,658). In all cases, LAIR increased the prognostic accuracy of the survival model. Conclusion. In this large series of patients, time LAIR independently predicted disease-specific survival, improving the prognostic accuracy of the TNM system. Accordingly, the LAIR should be taken into account for the stratification of patients' risk, both in clinical and research settings. (Surgery 2011;150:83-90.)