RESUMO
OBJECTIVES: SELECTTION was a multinational, prospective observational study to assess the choice of and the time from initiation of second-line treatment to treatment discontinuation for any reason in patients with non-small cell lung cancer. METHODS: Treatment cohorts were constructed based on prescribed second-line treatments that were at the discretion of the treating physicians, for 1013 patients enrolled in 11 countries. Propensity score analysis was conducted to assess whether the cohorts were comparable. Time from initiation of second-line treatment to treatment discontinuation was the primary endpoint. Reasons for treatment discontinuation, overall survival, progression-free survival and choice of second-line treatment were secondary endpoints. RESULTS: The treatment cohorts were pemetrexed (46.2%), docetaxel (22.9%), erlotinib (20.4%) and other treatments (10.5%). Analyses of baseline data and propensity scores showed that the erlotinib cohort comprised substantially different patients compared with the pemetrexed and docetaxel cohorts: patients in the erlotinib cohort were more likely to be women, never-smokers, have adenocarcinoma and worse performance status. Therefore, comparisons of outcomes between cohorts were not appropriate. Although disease progression was the most common reason for treatment discontinuation in all cohorts, erlotinib patients tended to continue treatment after disease progression, whereas in the docetaxel and pemetrexed cohorts, discontinuation occurred soon after disease progression. CONCLUSIONS: In real-world clinical practice the profiles of patients assigned to erlotinib were distinctly different from those assigned to pemetrexed or docetaxel. The most common reason for treatment discontinuation was progressive disease, reflecting adherence to clinical recommendations. It is difficult to extrapolate these findings to the present, as both clinical practice and the approved indications for NSCLC treatments have evolved substantially.