We have been investigating intrapleural malignancy as a “proof of concept” clinical model for the gene therapy of non-small cell lung cancer (NSCLC). Our focus has been to use malignant pleural effusions (MPE) to model an in vivo milieu, and to assess the transduction efficiency of tumor cells in this milieu by viral vectors. Consequently, our preclinical studies have used effusions derived from patients in transduction bioassays, and by targeting cell lines derived from MPE for gene-transfer studies. These studies have produced observations that have broad relevance with respect to in vivo gene therapy. Indeed, these studies assert that viral vectors confront a number of hurdles within the extracellular matrix (ECM) and the soluble component of MPE in situ , long before they get the opportunity to interact with the cells targeted for transduction. In essence, the extracellular milieu serves as a sink for gene-transfer vectors, thus contributing to the global inefficiency of gene transfer observed by vectors in vivo.