Scientific and social questions are currently being raised regarding the possible environmental origins of cancers whose causes are unknown or poorly explained by conventional factors. To improve our understanding, it is necessary to establish health surveillance linked to the environment and to conduct specific studies. The objective was to identify which cancer sites should be prioritized for this surveillance and these studies. A prioritization method was developed step-by-step through scientific consensus. A composite scale comprising 16 criteria, with weights ranging from 1 to 3, was constructed, and an indicator was identified for each criterion to establish a score for each cancer site. The criteria were chosen to address the following three concepts: suspected or proven link to the environment, public health significance, and social perception. Data were extracted from 28 semi-structured interviews with clinicians and scientific experts, as well as from the literature. Application of the method demonstrates that it is feasible and discriminatory. A clear ranking of the 24 selected sites was obtained, with scores ranging from 6.3 to 30. The “central nervous system” site ranked first, followed by “lung,” “non-Hodgkin’s malignant lymphoma,” “pleural mesothelioma,” “leukemias,” and “skin.” These six sites consistently rank within the top seven regardless of the set of criteria considered. Given the inherent limitations of constructing a composite scale, the method successfully identified a group of six priority sites with a certain degree of robustness.