What is the place of clinical variables in advanced non-small cell lung cancer (NSCLC) patients treated with chemotherapy ?

International audienceBackground: Chemotherapy (CT) has shown its effectiveness in symptom control and quality of life improvement in advanced NSCLC patients. The therapeutic strategy and some clinical variables could have a major impact on outcome. Methods: Our retrospective analysis evaluated the impact on overall survival (OS) of the clinical benefit (CB), ECOG performance status (PS) and toxicity, function of treatment. CB was defined as disease-related symptoms improvement according to hospitalization report. Only grade III-IV CTC-NCI version 2 toxicities have been considered. OS was calculated between start of CT and death or last follow-up. Multivariate Cox regression analysis including CB, PS, toxicity and age, stratified by AJCC initial stage was used. Results: Data of 68 consecutive stage IIIB-IV patients treated in a single French centre were analyzed. Chemotherapy was platinum-salt based in 88, 45 and 25% of pts for the first, second and third-line, respectively. Median age was 61 years, 37% were women. More than half (66%) were metastatic and 14% were previously irradiated. Median survival was 14 months (95% CI, 6.1-21.8), 53 % of patients are dead. The risk of death PS-related was multiplied by 2.3, 2.4 and 5.3 for the first, second and third-line of CT, respectively. PS and CB were initially associated with OS (first and second-line CT), but after the third-line of CT only PS was significantly related with OS. The risk of death reduction induced by a CB was 59, 82 and 29%, respectively. Less toxicities during CT were associated with a better OS (an unsignificant 20- 30% risk of death reduction), independently of the chronology of CT. Older pts >70 years have a higher risk of death (HR=1.87), independently of the CB and treatment-related toxicities in the multivariate analysis (P=0.18), sex-adjusted. Conclusions: No matter how many lines of CT are used for a specified patient, the ECOG PS was a patient-related variable with a dominant impact on the outcome. CT must be less toxic in order to achieve a CB and ameliorate the PS