Traditional radiation treatment planning relies on density imaging such as Computed Tomography for anatomic information of various structures of interest including target and normal tissues. However, the difficulties to distinguish malignant from normal tissue on CT slides often leads to inaccurate outlining of the GTV and/or to geographic misses. 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) has shown an increase in both sensitivity and specificity over CT in locoregional staging of patients with non-small cell lung cancer (NSCLC). The co registration of FDG-PET images to the data of the CT planning offers the radiation oncologist the possibility to include functional information into the target outlining. For the treatment of patients with NSCLC, it has been shown that the use of FDG-PET images: 1) modified the shape and volume of radiation fields in 22-62% of cases, mainly due to a better nodal staging and distinction of atelectasis from tumor and; 2) significatively reduced the interobserver and intraobserver variability. This paper reviews the results reported in the literature. Challenges and proposed solutions are discussed.