For many years, it has been known that hypoxia affects the response to radiotherapy in human cancers. Hypoxic regions can develop as a tumor grows beyond the ability of its blood supply to deliver oxygen to the full extent of the tumor, exacerbated by vascular spasm or compression caused by increased interstitial fluid pressure. However, hypoxia is heterogeneous, and tumors that appear identical by clinical and radiographic criteria can vary greatly in their extent of hypoxia. Several invasive procedures to measure hypoxia in tumors have been developed and are predictive of response to therapy, but none of these is in routine clinical use because of technical complexity, inconvenience, and inability to obtain repeated measures. Noninvasive imaging with a hypoxia-directed radiopharmaceutical could be of great clinical utility. Most such radiopharmaceuticals under development use 2-nitroimidazole as the targeting moiety. 2-Nitroimidazole, which is selectively reduced and bound in hypoxic tissues, has been labeled with F-18, Cu-64/67, I-123, and Tc-99m. Of these, F-18-fluoromisonidazole and I-123-iodoazomycin arabinoside (IAZA) have been most widely studied clinically. Non-nitro-containing bioreductive complexes, such as the Cu-60/62/64 thiosemicarbazone ATSM and Tc-99m butylene amineoxime (BnAO or HL91), have also been evaluated. In particular, 1-123-IAZA and Cu-60-ATSM have shown correlation with response to radiotherapy in preliminary clinical studies. However, more preclinical studies comparing imaging with validated invasive methods and clinical studies with outcome measures are required. Nuclear medicine is poised to play an important role in optimizing the therapy of patients with hypoxic tumors.