Postoperative radiotherapy in radically resected non-small cell lung cancer

Purpose: To evaluate the value of adjuvant postoperative external-beam radiation (EBR) in patients with radically resected non-small cell lung cancer (NSCLC) pT1-3 pN0-2 compared to patients with resected NSCLC without adjuvant EBR. Materials and methods: In 155 patients (121 male, 34 female; mean age, 59 years) 105 lobectomies, 12 bilobectomies, and 38 pneumonectomies with radical lymph node dissection of the ipsilateral side were performed. Postoperative staging was done according to the TNM system and was as follows: pT1 (n=38), pT2 (n=89), pT3 (n=28), pN0 (n=39), pN1 (n=67) and pN2 (n=49). Histopathologic study revealed 68 squamous cell carcinomas, 53 adenocarcinomas, 21 large cell carcinomas, 6 adenosquamous cell carcinomas, and 7 bronchioloalveolar cell carcinomas. All patients were randomly assigned into two treatment groups: 72 patients with no further treatment (control group), and 83 patients (EBR group) with adjuvant postoperative EBR of the bronchial stump and mediastinum (50 to 56 Gy, 8 or 23 MV photons, 2 Gy/d, 5 d/wk) beginning 4 to 6 weeks after surgery. Results: The overall 5-year survival rate (median observation time, 43 months) of all patients was 24.1% (EBR group, 29.7%; control group, 20.4%; p>0.05, not significant). The relative risk of the EBR group was 0.85 with a two-sided confidence interval of 0.66 to 1.09. The overall 5-year recurrence-free survival was 20.6% (EBR, 27.1%; control group, 15.6%; p=0.07). The relative risk of the EBR group was 0.80 with a confidence interval of 0.63 to 1.01. The rate of local recurrences at the bronchial stump and/or mediastinum was significantly smaller in the EBR group (n=5) than in the control group (n=17) (p<0.01). Multivariate analysis (chi(2) test) demonstrated an independent influence of postoperative EBR on the incidence of local recurrences. The incidence of distant metastases was slightly but not significantly higher in patients without EBR (38 patients) compared to those who had EBR (32 patients). Conclusion: High-dose postoperative EBR to the mediastinum significantly reduces the risk of local recurrence at the bronchial stump and/or mediastinum. Age, sex, histological subtype, tumor size, surgical approach, or extent of lymph node involvement had no prognostic value-only postoperative EBR had an independent influence on the risk of local recurrence. The effect of postoperative EBR was on the verge of significance with respect to recurrence-free survival and showed the same tendency in overall survival, however with an attenuated relative risk.