DVH Analysis for Predicting Radiation Pneumonitis in Cases of Advanced Lung Cancer Treated with Concurrent Chemo-Radiotherapy

Abstract

Purpose: Recently several DVH parameters have been reported as useful indicators for the occurrence of radiation pneumonitis (RP) after thoracic radiotherapy. On the other hand, in the treatment of unresectable non-small cell lung cancer, recent RTOG report showed concurrent chemo-radiotherapy yielded better outcome with higher non-hematologic toxicity than sequential therapy. The purpose of this study is to clarify if percent of pulmonary volume irradiated exceeding 20Gy (V20) is associated with the incidence and grade of radiation pneumonitis (RP) in cases of advanced lung cancer treated with concurrent chemo-radiotherapy. Materials and Methods: Seventy-one patients with advanced lung cancer who were treated with conventionally fractionated radiotherapy concurrently with systemic chemotherapy from Jan 1999 to Sep 2000, and were followed more than three months were analyzed in this study. There were 58 males and 13 females and the median age was 67 (range 42-79). Sixty-five were NSCLC and other 6 were SCLC. Clinical stages were IIIA in 21, IIIB in 47, and IV in 3patients. Chemotherapy was given with radiotherapy concurrently in initial treatment in 65, and other 6 received induction chemotherapy followed by concurrent chemo-radiotherapy. Chemotherapy were consisted of cisplatin or carboplatin based regimens combined with taxane in 36, and other agents such as vinolerbine / CPT11 in 35patients. Radiotherapy was delivered in 1.8-2.0 Gy/fr qd to total doses of 48-66Gy (median 60Gy). RP were graded according to NCI-Common Toxicity Criteria Ver.2.0. The relationship between RP grade and V20 were analyzed. Univariate and multivariate analyses were performed to assess other patient and treatment related factors as well as V20. Results: The median follow-up period was 7.5 months (range 3-19 months). Grade 0, 1, 2, 3, 5 RP were observed in 16, 35,17, 1, 2 patients respectively. The mean V20 was 23.0% in all patients, while the mean V20 were 20.1, 22.0, 26.3, 27.0, and 34.5% in grade 0, 1, 2, 3, and 5 patients respectively. The six month cumulative incidences of RP greater than grade 2 were 8.7%, 18.3%, 51%, and 85% in patients whose V20 was <20%, 21-25%, 26-30% and >31% respectively (p<0.0001). In the two patients died from RP, V20 were 32 and 37%. In both univariate and multivariate analysis, V20 was the only significant factor (p<0.0001, p=0.0022 respectively) associated with the incidence of grade2 or higher RP while other factors including age, gender, smoking history, pulmonary function (FEV 1.0L), location of the tumor (upper lobe vs. middle/lower lobe), chemotherapy regimen (platinum+taxane vs. others), total dose, overall radiation periods were not significant. Conclusion: The incidence and grade of RP following concurrent chemo-radiation of lung cancer were significantly related to V20. Therefore V20 is considered to be useful predicting factor of RP. The V20 less than 25% showed relatively low incidence of grade2 or higher RP while V20 greater than 30% showed very high incidence. We recommend keeping V20 less than 25% in concurrent chemo-radiation of lung cancer.