This study seeks to examine the extent to which cancer services are geographically located according to cancer incidence, and assess the association with cancer survival. We identified hospital sites serving English PCTs (Primary Care Trusts) with the management and treatment of breast, lung and colorectal cancer. Geographical access was estimated as travel time in minutes from LSOAs (Lower Super Output Areas) to the nearest hospital site and aggregated to PCT level. Correlations between PCT level mean travel times and cancer cases were estimated using Spearman's rank correlation. Associations between PCT level mean travel times and cancer relative survival rates were estimated using linear regression with adjustment for area deprivation and for a PCT level measure of the reported ease of obtaining a doctor's appointment. We found that cancer services tended to be located farther from areas with more cancer cases, and longer average travel times are associated with worse survival after adjustment for age, sex, year and area deprivation. This suggests that geographical access to cancer services remains a concern in England.