Studies in model systems have demonstrated that tumour pH can be a determinant of treatment response. The potential that tumour pH differs from that of normal tissues may provide a basis for selective killing of tumour cells. Although the data are limited, pH measurements in humans indicate a difference between tumour and normal tissues. In general, electrode pH (generally considered to reflect primarily extracellular pH, pHe) is lower in tumour than normal tissue. However, pH measured by magnetic resonance spectroscopy (MRS) or positron emission tomography (PET; both are generally considered to reflect primarily intracellular pH, pHi) is equal to or slightly higher in tumours than normal tissues. Hence, not only may pHe and pHi differ between normal and malignant tissues, but the pH gradient (which determines the distribution of chemotherapeutic agents that are weak acids or bases) is also reduced or reversed in tumours. To date, the majority of treatment‐related studies conducted have focused on hyperthermia (combined with radiotherapy) due to the recognized importance of acidic pH as a thermal sensitizer. However, the results have been somewhat surprising: patients with a better response to hyperthermia–radiotherapy have higher pH (as measured by electrode or 31P MRS) prior to treatment.