Background: Estrogen may play an important role in the pathogenesis of non-small cell lung cancer (NSCLC), especially in women. Preclinical studies have shown that estrogen can modulate gene expression important in cell proliferation in NSCLC cells. The estrogen receptor has been identified on NSCLC cells and a functional aromatase enzyme is present in 86% of NSCLC. In women over 65, lower levels of aromatase expression in NSCLC tumors was associated with improved survival. Treatment with estrogen plus progestin in postmenopausal women increased the number of deaths from NSCLC. A phase I study combining fulvestrant and gefitinib in women with NSCLC showed the combination was well-tolerated and demonstrated antitumor activity. There is a need for further studies of consolidation/maintenance therapy in patients with advanced NSCLC who have received induction therapy. Methods: We designed a phase II randomized trial of post-induction consolidation treatment in postmenopausal women with advanced NSCLC. Eligible patients have an objective response or stable disease after 6 cycles of first-line platinum-based chemotherapy (carboplatin + gemcitabine, paclitaxel or pemetrexed +/- bevacizumab). 100 patients will be randomized (3 to 1 ratio) to receive either AF and best supportive care (BSC) or BSC alone. AF treatment consists of anastrozole 1 mg by mouth once daily and fulvestrant 250 mg intramuscular monthly after a loading dose of 500 mg on day 1 and 250 mg on day 14 of cycle 1, repeated every 4 weeks until disease progression. Concurrent use of bevacizumab is allowed in both arms. This provides 80% power for the primary comparison of progression-free survival (PFS) at the 0.05 significance level, projecting an improvement in median PFS from 3 months in the BSC arm to 5.7 months in the AF arm. We plan to measure the plasma levels of 17b-estradiol, vascular endothelial growth factor (VEGF), E-selectin, thrombospondin-1 and insulin growth factor (IGF)-1 and baseline tumor tissue expression of ERa, ERb, PR, VEGF, and aromatase to correlate with efficacy parameters.