Objective: Regarding vulvar cancer, a nomogram has been suggested for the prediction of relapse-free survival (RFS). While the nomogram has been developed and validated in a Western study, there was no validation in Korean population. Thus, we have undertaken the study to assess the applicability of nomogram for predicting RFS in Korean patients with vulvar cancer.
Methods: A total of 204 cases newly diagnosed as vulvar cancer between 1982 and 2006 were identified. Among them 70 cases were not eligible due to inappropriate cell type (40 cases) and radiation as primary therapy (30 cases). Forty-four cases were not evaluable due to inadequate data and persistent disease. Finally a total of 90 patients primarily treated by surgery were included for analysis. Variables including age and the characteristics of primary tumor, nodal status, and surgical margin were collected for predicting RFS based on nomogram, which was compared with actual RFS. A calibration plot was drawn showing the actual versus predicted probability for 6 groups of patients segregated according to their predicted probabilities. In addition, discrimination of the nomogram was quantified with the concordance index.
Results: Patients' mean age was 58 years and mean follow-up period was 47.9 months. Observed 2y- and 5y-RFS rates were 81% and 68%, respectively, corresponding to 79% and 72% in the original cohort. The trend line in calibration plot showed comparable concordance with an ideal line, having a slope of 1.04 for 2y-RFS (R(2)=.35) and 0.98 for 5y-RFS (R(2)=.80), respectively. The concordance index was 0.79 in the KGOG data set, which was improved to 0.82 with the data set limited to squamous cell carcinoma.
Conclusion: The nomogram provides the predictive capacity for relapse-free survival in Korean patients with vulvar cancer.
Keywords: External validation; Nomogram; Prediction; Vulvar cancer.