Classification and regression tree analysis of patients with non-small-cell lung cancer treated with gefitinib after chemotherapy.

Abstract

Many randomized studies have shown that epidermal growth factor receptor (EGFR-) tyrosine kinase inhibitors (TKIs) are apparently advantageous over standard chemotherapy in non-small-cell lung cancer (NSCLC) patients with EGFR active mutation in front-line treatment. But which subgroup of advanced NSCLC could benefit from EGFR-TKIs in the second-or third-line setting remains elusive. To explore predictive factors of advanced NSCLC patients with the unknown status of EGFR mutation treated by gefitinib in the second-or third-line setting, we used classification and regression tree (CART) analysis to screen which patients would benefit more. One hundred and fifty-five patients with advanced NSCLC previously unsuccessfully treated with platinum-based chemotherapy were included in this study. Patients received gefitinib as part of the Expanded Access Program of the China Charity Federation between 2 March 2005 and 11 May 2011. Multivariate analysis of progression-free survival (PFS) was performed using CART analysis. This method uses recursive partitioning to assess the effect of specific variables on PFS, thereby ultimately generating groups of patients with similar clinical features on PFS. The median PFS in patients with NSCLC who were treated with gefitinib after prior chemotherapy was 16 months (95% confidence interval [CI] 13.44-18.56). CART was performed with an initial split on adenocarcinoma differentiation, and four terminal subgroups were formed. The median PFS of the four subsets ranged from 12 to 42 months. Adenocarcinoma differentiation, brain metastasis and prior thoracic radiotherapy are predictors of PFS in previously treated NSCLC patients. CART can be used to identify homogeneous patient populations in clinical practice and future clinical trials.