Repeat biopsy of patients with acquired resistance to EGFR TKIs: implications of biopsy-related factors on T790M mutation detection

Abstract

To find predictors of non-diagnostic repeat biopsy specimen acquisition for mutational analysis and detection of epidermal growth factor receptor (EGFR) T790M mutation. We retrospectively reviewed 90 non-small cell lung cancer patients harbouring EGFR mutations who underwent repeat cone-beam CT-guided transthoracic needle biopsy. Clinical characteristics as well as biopsy-related factors were compared between patients with and without diagnostic specimen acquisition and between patients with and without T790M mutation. After univariate analysis, multivariate logistic regression analysis was performed to reveal independent predictors. Diagnostic biopsy specimens for mutational test were obtained in 90% (81/90) of patients, of which 62% (50/81) possessed T790M mutation. None of the analysed variables were significantly associated with non-diagnostic specimen acquisition. For T790M detection, duration of EGFR tyrosine kinase inhibitor treatment (p = 0.066), duration of total chemotherapy (p = 0.026), tumour size (p = 0.066), and metastatic lung lesion as a biopsy target (p = 0.029) showed p values less than 0.10. Multivariate analysis revealed that target tumour size (odds ratio, 0.765; p = 0.031) was an independent predictor of T790M mutation. Metastatic lesions as biopsy targets (odds ratio, 4.194; p = 0.050) showed marginal statistical significance. Non-diagnostic repeat biopsy specimen acquisition was not related to the clinical or technical factors. However, detection of T790M at repeat biopsy might be associated with smaller target tumour size and selection of metastatic lesions as biopsy targets.