gms | German Medical Science

Background: The purpose was to determine whether survival has improved in line with treatment advances in a population-based cohort of patients with pancreatic, gastric, oesophageal and colorectal cancer.

Patients and Methods: Patients with an invasive tumour diagnosed between 1978 and 2004 - recorded by the Munich Cancer Registry - were included in this analysis (pancreatic cancer n = 2707, gastric cancer n = 6160, oesophageal cancer n = 2334, rectal cancer n = 8537, colon cancer n = 12741). The 5- and 10-year relative survival (RS) have been analysed for each cohort in total as well as stage-specific.

Results: Pancreatic cancer: No better RS in the whole sample as well as stage-specific can be seen during different time periods. 5- and 10-year RS before and after 1988 were under 10%. Gastric cancer: 5- and 10-year RS for gastric cancer patients improved: it was 28.4% and 24.4% before versus 35.2% and 30.0% after 1988 for the whole sample. No stage-specific survival improvement, however, could be seen. Therefore, the better survival in all is a result of stage shift to more favourable stages due to early detection. Oesophageal cancer: The oesophageal cancer cohort demonstrated a similar pattern: 5- / 10-year RS before and after 1988 were 12.4% / 10.3% and 22.0% / 15.7%. In terms of stage patients showed no improvement. As opposed to gastric cancer no better stage distribution occurred. The survival improvement, therefore, is more likely a consequence of a change to more distal adenocarcinoma. Rectal Cancer: 5- / 10-year RS of rectal cancer patients diagnosed before 1988 was noticeably lower than of those diagnosed after 1988 (51.4% / 40.0% versus 61.6% / 51.7%). The stage-specific survival also improved in UICC I-III. Both, improvement in survival of the whole sample and a stage-specific improvement suggest an effect of better treatment in the last years. Colon cancer: 5- / 10-year RS before and after 1988 were 53.5% / 45.4% and 61.2% / 53.8%. An additional stage-specific better survival (UICC I-III) can also be seen as a therapy-effect.

Conclusions: Improvements in survival after diagnosis of cancer have different reasons: early detection (gastric cancer), changes of site and histology (oesophageal cancer) and better therapy (colorectal cancer) can be mentioned. Survival for pancreatic cancer patients, however, could not be influenced. There is also no improvement of life expectancy after metastases in all four gastrointestinal tumours in the past two decades.