P2.05-050 Impact of Inflammation and Sarcopenia on Outcomes after Stereotactic Body Radiotherapy for T1N0M0 Non-Small Cell Lung Cancer: Topic: Toxicities

Abstract

The purpose was to evaluate impact of systemic inflammation and sarcopenia on outcomes after stereotactic body radiotherapy (SBRT) for T1N0M0 non-small cell lung cancer (NSCLC) as a supplementary analysis of Japan Clinical Oncology Group (JCOG) study JCOG0403. Pretreatment serum C-reactive protein (CRP) was used as a marker for systemic inflammation. Patients were divided into high and low CRP groups with a threshold value of 0.3 mg/dL. Paraspinous musculature area (PMA) at a level of the 12th thoracic vertebra was measured on simulation CT with thresholding Hounsfield Units between -29 and 150. When PMA was lower than the gender-specific median, the patient was classified as sarcopenia. Toxicities, overall survival (OS) and cumulative incidence of cause-specific death were compared between groups. Kaplan-Meier method and cumulative incidence function were applied to estimate proportion of OS and cumulative incidence of cause-specific death, respectively. Of 169 patients enrolled into JCOG0403, 60 operable and 92 inoperable patients were included into this study after excluding 5 patients ineligible for JCOG0403 and 12 patients whose simulation CT images were unavailable or unsuitable for the PMA measurement. Forty-two patients were classified as high CRP. Medians of PMA were 31.6 cm2 (range, 12.6-52.9) and 25.1 cm2 (range, 3.4-38.5) in male and female, respectively. Proportions of toxicities Grade 3-4 were 19.1% and 10.9% in the high and low CRP groups; and 17.1% and 9.2% in the sarcopenia and non-sarcopenia groups, respectively. In the operable patient cohort, OS significantly differed between the CRP groups (log-rank test P=0.009; hazard ratio of high CRP 2.43, 95% confidence interval 1.23-4.80; 3-year OS of 58.8% and 83.6% for high and low CRP, respectively). This difference in OS was mainly contributed by difference in lung cancer death (Gray’s test P=0.070; 3-year cumulative incidence of 29.4% and 7.1%, respectively). No impact of sarcopenia on OS was observed in operable patients. In the inoperable patient cohort, OS did not differ between the CRP groups (log-rank test P = 0.925). No significant difference was observed in OS between the sarcopenia groups, either. The present study suggests that systemic inflammation may provide prognostic information for operable patients receiving SBRT for early-stage NSCLC. Further studies are warranted to confirm these findings.