Non-small cell lung cancer (NSCLC) is the major determinant of overall cancer mortality in the developed world. Although the brain metastasis (BM), one of the major relapse in NSCLC, contributes significantly to the morbidity and mortality of lung cancer, relevant driver mechanisms associated are largely unknown. The elucidation of pre-determining signature which is highly prognostic for the development of brain or general metastasis and overall survival may enable better stratification of NSCLC patients at high risk of BM and identify novel therapeutic candidates targeting BM. In this study, first of all, we evaluated copy number alterations (CNA) in 12 ADCs adenocarcinomas (ADCs) and 6 squamous cell carcinomas (SQCCs) formaldehyde fixed and paraffin embedded (FFPE) samples to identify distinct pattern of chromosomal aberrations depending on histologic-subtype. The most frequent CNAs detected in primary ADCs were gains of 3q, 5p, 5q, 6p, 8q, 9p, 11p, 15q, 17q, 17q, and losses of 10q and 22q whereas MIP analysis on primary SQCCs revealed gains in 4q and 12q and loss in 9q. In addition, the patterns of CNAs in 11 matched NSCLC cases composed of primary tumor and BM were analyzed with MIP array. The BM carried the majority of genetic alterations present in the corresponding primary tumor but with a significantly higher frequency including gain of 9p11.2 and loss of 2p11.2-11.1 in ADC cases. Finally, to uncover predetermining genetic signatures that can predict the risk of BM in especially ADC which is characterized by the early development of BM, comparative MIP was performed in primary ADC of 4 cases with synchronous and 8 cases with metachronous BM. Selectively amplified regions of 5q35, 10q23 and 17q23-24 were found to contain the putative candidate genes including NeurL1B, ACTA, FAS and ICAM2. These genomic signatures elucidated in our study may help to generate useful prognostic biomarkers associated NSCLC metastasis to brain.