Background:
Most survival data in colorectal cancer (CRC) is derived from clinical
trials or register-based studies. Hospital Biobanks, linked with
hospital electronic records, could serve as a data-gathering method
based on consecutively collected tumor samples. The aim of this Biobank
study was to analyze survival of colorectal patients diagnosed and
treated in a single-center university hospital over a period of 12
years, and to evaluate factors contributing to outcome.

Material and methods:
A total of 1777 patients with CRC treated during 2001–2012 were
identified from the Auria Biobank, Turku, Finland. Longitudinal clinical
information was collected from various hospital electronic records and
date and cause of death obtained from Statistics Finland.

Results:
Cancer-specific, overall and disease-free survival was higher in
patients diagnosed during 2004–2008 as compared with patients diagnosed
in 2001–2003. Further improvement was not seen during years 2009–2012.
Potential factors contributing to the improvement were introduction of
multidisciplinary meetings, centralization of rectal cancer surgery, use
of adjuvant chemotherapy and systematic preoperative radiotherapy of
rectal cancer. The proportion of patients with stage I–IV CRC remained
similar over the study period, but a marked decrease in non-metastatic
rectal cancer with biopsy only (locally advanced disease) was observed.
In stage I–III rectal cancer, Cox multivariate analysis suggested age,
comorbidity, R1 resection, T staging and tumor grade as prognostic
factors. In colon cancer, prognostic factors were age, comorbidity,
gender and presence of lymph node metastases.

Conclusions:
Organizational changes in the treatment of CRC patients made since 2004
coincide with improved survival in CRC and a marked reduction in
locally advanced rectal cancers. The clinical presentation of CRC has
remained similar between 2001 and 2012.