Purpose : Since February 1991 a Prospective study for non-small cell lung cancer patients who underwent radical resection and had a risk factor of positive resection margin or regional lymph node metastasis has been conducted to evaluate the effect of MVP chemotherapy and radiotherapy on the pattern of failure, disease free and overall survival. and tolerance of combined treatment. Materials and Methods: Twenty nine patients were registered to this study until Sep. 1993; of these 26 received planned therapy Within 3 weeks after radical resection, two cycles of MVP(Mitomycin C $6mg/m^2,$ Vinblastin $6mg/m^2,$ Cisplatin $60mg/m^2$) chemotherapy was given with 4 weeks intervals. Radiotherapy (5040cGy tumor bed dose and 900cGy boost to high risk area) was started 3 to 4 weeks after chemotherapy. Results: One and two year overall survival rates were $76.5\%;and;58.6\%$ respectively. Locoregional failure developed in 6 patients$(23.1\%)$ and distant failure in 9 patients$(34.6\%)$ Number of involved lymph nodes, resection margin positivity showed some correlation with failure pattern but T-stage and N-stage showed no statistical significance. The group of patients who received chemotherapy within 2 weeks postoperatively and radiotherapy within 70 days showed lower incidence of distant metastasis. Postoperative combined therapy were well tolerated without definite increase of complication rate, and compliance rate in this study was $90\%$. Conclusion: 1) MVP chemotherapy showed no effect on locoregional recurrence, but appeared to decrease the distant metastasis rate and 2) combined treatments were well tolerated in all patients. 3) The group of patients who received chemotherapy within 2 weeks postoperatively and radiotherapy within 70days showed lower incidence of distant metastasis. 4) Addition of chemotherapy to radiotherapy failed to increase the overall or disease free survival.