Population-based period analysis of survival among colon cancer patients by the data from Arkhangelsk regional cancer registry

Objective. To estimate survival among patients with colon cancer (CC) in 2000-2013 in the Arkhangelsk region (AR) by time periods using the data of Arkhangelsk regional cancer registry (ARCR). Materials and methods. Data on all cases of CC (C18.0-C18.9) in the AR in 2000-2013 were extracted from the database of the ARCR. Disease-specific survival (DSS) in the periods 2000-2006 and 20072013 was calculated using Kaplan-Meier method with log-rank test. Adjustment for initial factors was performed by Cox regression with step-wise imputation. Results. Altogether, 4134 patients were selected, 2569 (62.1%) of them died during the period of analysis. The proportion of elderly patients, urban habitants, those with right-sided, Stage I, Stage IV tumors and being radically treated with surgical or combined approach was significantly greater in the later period. Median DSS was 16.5 (95% confidence interval (CI) 14,1-19,0) months and 22.2 (95% CI 20,3-25,7) months, respectively, χ2 = 8,01, hazard ratio (HR) in univariate model 0,89 (95% CI 0.83-0.97), p = 0,005. According to multivariate analysis, inclusion of age, stage, and inhabitance variables didn't change regression coefficients, while “type of treatment” variable made the difference insignificant, HR 0.95 (95% CI 0.87-1.03), р=0.18. Conclusion. DSS of the CC patients in AR is lower comparing to most developed countries. Small, but statistically significant increase of DSS in 2007-2013 compared to 2000-2006 may be explained rather by more extensive use of surgical treatment at the later time period then change in stage distribution.