PURPOSE: This study was tried to evaluate the potential benefits of concurrent chemoradiation therapy (low dose daily cisplatin combined with split course radiation therapy) compared with conventional radiation therapy alone in stage III non-small cell lung cancer. The end points of analyses were response rate, overall survival, survival without locoregional failure, survival without distant metastasis, prognostic factors affecting survival and treatment related toxicities. MATERIAL AND METHODS: Between April 1992 and March 1994, 32 patients who had stage III non-small cell lung cancer were treated with concurrent chemoradiation therapy. Radiation therapy for 2 weeks (300cGy given 10 times up to 3000cGy) followed by a 3 weeks rest period and then radiation therapy for 2 more weeks (250cGy given 10 times up to 2500cGy) was combined with 6mg/M2 of cisplatin. Follow-up period ranged from 13 months to 48 months with median of 24 months. Historical control group consisted of 32 patients who had stage III non-small cell lung cancer were received conventionally fractionated (daily 170-200cGy) radiation therapy alone. Total radiation dose ranged from 5580cGy to 7000cGy with median of 5940 cGy. Follow-up period ranged from 36 months to 105 months with median of 62 months. RESULTS: Complete reponse rate was higher in chemoradiation therapy (CRT) group than radiation therapy (RT) group (18.8% vs. 6.3%). CRT group showed lower in-field failure rate compared with RT group (25% vs. 47%). The overall survival rate had no significant differences in between CRT group and RT group (17.5% vs. 9.4% at 2 years). The survival without locoregional failure (16.5% vs. 5.3% at 2 years) and survival without distant metastasis (17% vs. 4.6% at 2 years) also had no significant differences. In subgroup analyses for patients with good performance status (Karnofsky performance scale 80), CRT group showed significantly higher overall survival rate compared with RT group (62.5% vs. 15.6% at 2 years). The prognostic factors affecting survival rate were performance status and pathologic subtype (squamous cell cancer vs. nonsquamous cell cancer) in CRT group. In RT alone group, performance status and stage (IIIa vs IIIb) were identified as a prognostic factors. RTOG/EORTC grade 2-3 nausea and vomiting (22% vs. 6%) and bone marrow toxicities (25% vs. 15.6%) were significantly higher in CRT group compared with RT alone group. The incidence of RTOG/EORTC grade 3-4 pulmonary toxicity had no significant differences in between CRT group and RT group (16% vs. 6%). The incidence of WHO grade 3-4 pulmonary fibrosis also had no significant differences in both group (38% vs. 25%). In analyses for relationship of field size and pulmonary toxicity, the patients who treated with field size beyond 200cm2 had significantly higher rates of pulmonary toxicities. CONCLUSION: The CRT group showed significantly higher local control rate than RT group. There were no significant differences of survival rate in between two groups. The subgroup of patients who had good performance status showed higher overall survival rate in CRT group than RT group. In spite of higher incidence of acute toxicities with concurrent chemoradiation therapy, the survival gain in subgroup of patients with good performance status were encouraging. CRT group showed higher rate of early death within 1 year, higher 2 year survival rate compared with RT group. Therefore, to evaluate the accurate effect on survival of concurrent chemoradiation therapy, systematic follow-up for long term survivors are needed.