Pancoast tumour is a rare neoplasia in which the optimal therapeutic management is still controversial. The traditional treatment of Pancoast tumour (surgery, radiotherapy or a combination of both) have led to an unsatisfactory outcome due to the high rate of incomplete resection and the lack of local and systemic control. The aim of the study was to determine the efficacy of the trimodality approach. Fifty-six patients (male/female ratio: 47/9, median age: 64 years) in stage IIB to IIIB were treated during a period between 1994 and 2013. Induction therapy consisted of 2-3 cycles of a platinum-based chemotherapy associated with radiotherapy (30-44 Gy). After restaging, eligible patients underwent surgery 2 to 4-week post-radiation. Thirty-two (57.1%) patients were cT3 and 24 (42.9%) cT4, 47 (83.9%) were N0 and 9 (16.1%) N+. Forty-eight (85.7%) patients underwent R0 resection and 10 (17.9%) had a complete pathological response (CPR). Thirty-day mortality rate was 5.4%, major surgical complications occurred in 6 (10.7%) patients. At the end of the follow-up, 17 (30.4%) patients were alive and 39 (69.6%) died (29 for cancer-related causes), with an overall 5-year survival of 38%. At statistical analysis, stage IIB (P = 0.003), R0 resection (P = 0.03), T3 tumour (P = 0.002) and CPR (P = 0.01) were significant independent predictors of better prognosis. This combined approach is feasible, and allows for a good rate of complete resection. Long-term survival rates are acceptable, especially for early stage tumours radically resected. Systemic control of disease still remains poor, with distant recurrence being the most common cause of death.