Random selection of a chemotherapy regimen improves responses and survival only modestly in patients with advanced non-small cell lung cancer. Chemotherapy that is selected based on the molecular determinants of the tumor may further augment response rates and survival. This requires an in-depth understanding of the prognostic and predictive significance of the molecular determinants. The ultimate clinical utility of these molecular determinants will depend on the feasibility and ease of estimating these parameters using, in primary tumors, techniques that are widely applicable and relatively inexpensive. Using our work with excision repair cross-complementation group 1 (EERC1) and ribonucleotide reductase M1 (RRM1) as a paradigm, we demonstrate the step-wise development of molecular determinants as tools to aid in the selection of chemotherapy. We show that molecular determinant-based selection of chemotherapy is both feasible in the clinical setting and suggests clinical benefit. These findings are currently being confirmed in a Phase III trial. This paradigm could be used for developing customized treatment strategies for other cancers with other chemotherapeutic and targeted agents. In the future, identifying a molecular determinant could be an integral part of drug development. Developing functional imaging techniques for widely used molecular determinants would also mitigate the need for repeated invasive biopsies, allowing us to evaluate the change in the status of molecular determinants in response to treatment.