Our results with concurrent biochemotherapy in patients with stage IV melanoma have been encouraging. Based on these data, we conducted a phase II study to determine the clinical and histological response rate to neoadjuvant concurrent biochemotherapy in patients with local-regional metastases of cutaneous melanoma (stage III). A total of 65 patients with biopsy-proven, measurable and potentially resectable local-regional disease (nodal, satellite/in-transit metastases and/or local recurrence) were treated with cisplatin 20 mg/m2 intravenously (i.v.) on days 1 to 4, vinblastine 1.5 mg/m2 i.v. on days 1 to 4, dacarbazine 800 mg/m2 i.v. on day 1 only, interleukin-2 9 MIU/m2 per day i.v. by 96 h continuous infusion on days 1 to 4, and interferon-alpha 2a 5 MU/m2 subcutaneously on days 1 to 5, repeated every 3 weeks. Patients underwent surgery after two to four courses of biochemotherapy. Those with tumour regression after two preoperative courses received two additional postoperative courses. Of the 64 patients assessable for clinical response, 28 (44%) had a partial response. Of the 62 patients whose response was assessed histologically, four (6.5%) had no evidence of viable tumour in the surgical specimen (pathological complete remission, pCR) and 27 (43.5%) had a partial response, giving an overall response rate of 50%. Tumour burden did not correlate with response, although patients who achieved a pCR had a significantly lower tumour burden (P = 0.02). Our phase II study indicates that neoadjuvant biochemotherapy is an active treatment for melanoma patients with local-regional metastases. However, it is unclear if biochemotherapy is more active than chemotherapy alone; phase III randomized trials are ongoing to answer this question in patients with stage IV disease.