In 2015, the Auvergne-Rhône-Alpes Regional Health Agency asked Santé publique France to investigate whether there were any excess cases of cancer among residents living near the Roussillon chemical plant (Isère department). Mortality and incidence data collected from Inserm’s CépiDC and the Isère Cancer Registry cover the period from 2003 to 2013. The SMR and SIR (standardized mortality and incidence ratios), obtained through indirect standardization, were analyzed for the six municipalities adjacent to the complex, using the Isère department as a reference. The study shows a slightly higher incidence of cancer among men (approximately 8% higher than that observed among men in the Isère department) and a cancer mortality rate that is not different from the departmental average. Among women, cancer incidence and cancer mortality across all six municipalities in the Roussillon area are comparable to those in the Isère department. Analyses by cancer site show that cancers affecting the respiratory tract—primarily pleural mesothelioma (cancer of the protective membrane surrounding the lung)—are overrepresented in the Roussillon area. For pleural mesothelioma, the risk—in terms of incidence and mortality—is more than five times higher among the male population in the study compared to the male population of Isère. To a lesser extent, a significant excess mortality from cancers of the larynx, trachea, bronchi, and lung is observed among both men and women in the area. No other cancer sites are found to be overrepresented among residents living near the Roussillon chemical plant. The overall excess of cancers observed among men is largely attributable to pleural mesothelioma, itself largely attributable to occupational exposure to asbestos used within the chemical complex. This exposure could also explain part of the excess deaths from cancers of the larynx, trachea, bronchi, and lung in the area. Thus, the source of the increased cancer risk in this study area appears to be occupational rather than environmental exposure, although the study does not allow for establishing causality in the observed relationship.