[Measurement of tumour response to cancer treatment: morphologic imaging role]

New forms of cancer chemotherapy are tested in therapeutic trials (phase I, phase II or phase III) while chemotherapeutic agents whose efficacy has already been demonstrated are used, in routine clinical practice, in the context of protocols. The overall survival rate is the best objective parameter of efficacy of the treatments tested, but this parameter is obtained too late, as the effect on the tumour must be determined as soon as possible in order to institute another treatment if necessary. Tumour response, or objective response, is based on changes in the number and size of measurable primary or secondary tumour "targets". These parameters are obtained more rapidly than survival data, but their reliability is highly dependent on the quality of comparative clinical and especially radiological measurements of tumour targets. Medical imaging plays an essential role in these assessments. The absence of standardized techniques, poor selection of targets and inaccurate measurements can bias the results, particularly those of therapeutic trials. In view of the economic, scientific and patient-related stakes involved, a very rigorous approach is essential, directly implicating the responsibility of radiologists performing assessment examinations. The World Health Organization (WHO) guidelines defining the method of measurement of solid tumours and response criteria are no longer adapted to technical progress in imaging. Recently these guidelines have been updated and a new set of criteria has been proposed by the Response Evaluation Criteria in Solid Tumors (RECIST) Group, taking into account progress in imaging. They remain based on measurement of the size of the target lesion. The use of this single criterion of size to evaluate response to treatment needs to be discussed in the light of new technologies able to provide information on tumour composition, metabolism or neovascularization, modifications of which reflect response to treatment before a reduction in tumour volume can be detected.