P2.02-054 Impact of Prognostic Nutrition Index for Induction Chemoradiotherapy Followed by Surgery in Locally Advanced Non-Small Lung Cancers

Abstract

fractions (range 50-75Gy, 3-35 fractions). 106 were stage I, 56 were stage II and 232 were stage III. Average age was 68 years. 26 patients received stereotactic body radiotherapy. ECOG, stage and weight loss were all associated with survival (p¼0.007, p¼0.022, p¼0.032 respectively). Modified comorbidity score and interval between diagnosis and start of radiotherapy were not (p¼0.101 and p¼0.353 respectively). A retrospectively assigned Charlston comorbidity score was available for 271 of the 394 patients who were analyzed using the same variables as above but substituting the Charlston score for the modified comorbidity score. ECOG PS and Charlston score were significantly associated with survival (p¼0.012 and p¼0.046 respectively, but stage, weight loss and interval to radiotherapy were not (p¼0.129, p¼0.150 and p¼0.09 respectively). When Colinet score was introduced to the analysis it did not reach statistical significance (p¼0.729) but Charlston score retained it (p¼0.042). Conclusion:Delay from date of diagnosis to initiation of radiotherapy could not be demonstrated to correlate negatively with survival, at least in the time range encountered in our patients. Although counterintuitive, this is in agreement with the limited published data which are retrospective to our knowledge. The modified comorbidity score was not specifically developed by Colinet et al to prognosticate for lung cancer and indeed appears inadequate for this purpose. The Charlston score however, is confirmed as prognostic in our study and although more complex and time consuming to determine, should be the preferred comorbidity index for prognostication of radically treated non-small cell lung cancer patients.