Neoadjuvant chemotherapy and radiotherapy followed by surgery in stage IIIA non-small-cell carcinoma of the lung: report of a Cancer and Leukemia Group B phase II study.

PURPOSE: This phase II trial was designed to evaluate the feasibility, toxicity, response rates, and survival for neoadjuvant chemotherapy and radiotherapy (RT) followed by surgical resection in newly diagnosed patients with surgically staged IIIA non-small-cell lung carcinoma (NSCLC). PATIENTS AND METHODS: Previously untreated patients with NSCLC underwent bronchoscopy, chest and abdominal computed tomography (CT), bone scan, and surgical staging of the mediastinum. Neoadjuvant treatment consisted of concurrent chemotherapy and RT. Patients then underwent surgical resection, which was followed in turn by additional chemotherapy and RT. Chemotherapy included cisplatin 100 mg/m2 on days 1 and 29, vinblastine 3 mg/m2 on days 1 and 3 and 29 and 31, and fluorouracil (5-FU) 30 mg/kg/d by infusion on days 1 to 3 and 29 to 31 (FVP). RT began on day 1 and included 3,000 cGy in 15 fractions. Surgery took place on day 55, and one more cycle of chemotherapy and an additional 3,000 cGy of RT began on day 85. RESULTS: Forty-one eligible patients (median follow-up, 53 months) were studied. N2 disease was present in 80%, whereas 20% had T3N0 or T3N1 lesions. Response to neoadjuvant chemotherapy and RT included no complete responses (CR), 21 (51%) partial responses (PR) or regressions, 19 (46%) stable disease (SD), and one (2%) progressive disease (PD). Thirty-one patients underwent surgery, and 25 were resected. In four of the 25 resection specimens, no viable tumor was present, whereas in three of the six unresectable patients, extensive biopsy results demonstrated only necrotic tumor. The maximum response achieved using all protocol treatment was 27 (66%) CRs, seven (17%) PRs or regression, six (15%) SDs, and one (2%) PD. Toxicity was substantial and primarily hematologic. There were six (15%) treatment-related deaths, which included three perioperative deaths and three chemotherapy-related toxicity deaths. The Kaplan-Meier curve indicated a 1-year survival of 58% and a median survival of 15.5 months. Nine patients (22%) remain disease-free. CONCLUSIONS: There was a reasonably high rate of PR associated with concurrent neoadjuvant chemotherapy and RT, and a high percentage of patients who ultimately were rendered completely disease-free. However, treatment-related morbidity and mortality was common. Median survival seemed to be only modestly improved beyond that achieved with less intensive means of treatment. However, a group has emerged of patients who enjoy prolonged disease-free survival and possible cure.