Development and validation of a clinical prediction-score model for distant metastases in major salivary gland carcinoma.

ABSTRACT

BACKGROUND: The most common pattern of failure in major salivary gland carcinoma (SGC) is development of distant metastases (DMs). The objective of this study was to develop and validate a prediction score for DM in SGC. PATIENTS AND METHODS: Patients with SGC treated curatively at four tertiary cancer centers were divided into discovery (n = 619) and validation cohorts (n = 416). Multivariable analysis using competing risk regression was used to identify predictors of DM in the discovery cohort and create a prediction score of DM; the optimal score cut-off was determined using a minimal P value approach. The prediction score was subsequently evaluated in the validation cohort. The cumulative incidence and Kaplan-Meier methods were used to analyze DM and overall survival (OS), respectively. RESULTS: In the discovery cohort, DM predictors (risk coefficient) were positive margin (0.6), pT3-4 (0.7), pN+ (0.7), lymphovascular invasion (0.8), and high-risk histology (1.2). High DM-risk SGC was defined by sum of coefficients greater than two. In the discovery cohort, the 5-year incidence of DM for high- versus low-risk SGC was 50% versus 8% (P < 0.01); this was similar in the validation cohort (44% versus 4%; P < 0.01). In the pooled cohorts, this model performed similarly in predicting distant-only failure (40% versus 6%, P < 0.01) and late (>2 years post surgery) DM (22% versus 4%; P < 0.01). Patients with high-risk SGC had an increased incidence of DM in the subgroup receiving postoperative radiation therapy (46% versus 8%; P < 0.01). The 5-year OS for high- versus low-risk SGC was 48% versus 92% (P < 0.01). CONCLUSION: This validated prediction-score model may be used to identify SGC patients at increased risk for DM and select those who may benefit from prospective evaluation of treatment intensification and/or surveillance strategies.