Noninvasive determination of tissue oxygen concentration by overhauser enhanced magnetic resonance imaging

Thomlinson and Gray originally proposed the concept of tumor hypoxia in the 1950s (1). They based this notion on examinations of histological sections of human lung tumors, which revealed a constant distance in the range of 100 to 150 mm between blood vessels and the edge of necrotic zones (1). This hypothesis was veri?ed in other human tumors by measuring pO2 values by inserting polarographic electrodes in accessible sites such as head and neck tumors (2). These measurements revealed signi?cant levels of hypoxia (10mmHg), which was correlated with treatment failure in fractionated radiotherapy. It is estimated that approximately one half of solid human tumors have median oxygen levels <10mmHg, prior to therapy (3,4). Subsequent studies, using oxygen electrodes inserted into various of human tumors, have demonstrated that signi?cant levels of hypoxia are associated with compromised response to radiation treatment for breast cancer, head and neck cancer, sarcomas, cervix cancer, and for prostate cancer (5-9). The radiobiological basis for the failure of fractionated radiation treatment of hypoxic tumors is well documented and is attributed to its ef?cient chemical repair of radiation-induced lesions on DNA (10). More recent clinical studies have noted that patients with tumors with low pO2 levels exhibited higher incidence of loco-regional failures (with or without distant metastases), irrespective of whether surgery or radiotherapy was performed (8). Not only did the presence of hypoxic cells in the tumor compromise therapy, but also their presence denoted a more aggressive disease.