To compare patterns of recurrence and disease-free survival (DFS) of node-positive and node-negative patients with advanced vulval squamous cell carcinoma (SCC). Fifty-five patients with FIGO stage III/IVA vulval SCC who had surgery at the Queensland Centre for Gynaecological Cancer from 1989 to 1999 were included. Patients were grouped as follows: Group A, pT3 N0; Group B, pT3 N1; Group C, pT4 N2. Treatment included surgery +/- postoperative radiotherapy. Multivariate Cox models were calculated to identify independent prognostic factors. After a median follow-up of 96 months, 25 patients (45.5%) experienced recurrence at the vulva (n = 2), pelvis (n = 8), or distant sites (n = 15). Recurrence in the pelvis and at distant sites was more likely for patients in groups B and C (P 0.003). At 5 years the probability of DFS was 66.6%, 35.3%, and 39.8% for patients in groups A, B, and C, respectively (P 0.085). Patients with negative nodes (n = 15), one microscopic positive node (n = 11), and two or more positive nodes (n = 29) had a probability of DFS of 66.6%, 67.3%, and 26.1% at 5 years, respectively (P 0.005). Patients with > or =2 positive groin nodes are at risk for distant failure. The DFS of patients with negative groin nodes and those with only one microscopic positive node is very similar. The prognosis of patients with > or =2 positive unilateral or bilateral groin nodes is similar. The current FIGO staging system inaccurately reflects prognosis for patients with advanced vulval cancer. Clinical trials are warranted to investigate the benefit of systemic treatment.