Abstract: Background and purpose: Cancer is a global major public health concern, and survival rate is a key indicator in evaluating the treatment effect of cancer. This study described the survival of over 200 thousand cancer patients from a hospital-based cancer registry database, aiming to reflect current cancer treatment effect somehow with real world data. Methods: A total of 202 542 hospitalized patients in Fudan University Shanghai Cancer Center from Jan. 1 2008 to Dec. 31 2017 were included in the analysis. Medical records review, telephone visits and death registry data linkage were applied in collecting endpoint data. The last follow-up date was Nov. 30 2019. Life table was applied in evaluating the 1-year, 3-year and 5-year overall survival (OS) rate for all and stratified by gender, age group and periods. Kaplan-Meier method was applied in plotting the survival curves. Results: The 1-year, 3 year and 5-year OS of cancer patients were 89.8%, 77.4% and 71.0%. The 5-year OS of males and females were 58.8% and 78.7% respectively. Among common cancers, thyroid cancer held the highest 5-year survival rate 98.6% while pancreatic cancer the lowest, 11.4%. The 5-year survival rates of breast, lung and kidney cancer in 2013—2017 were 90.0%, 55.9% and 80.7% respectively, significantly higher than those in 2008—2012. No significant improvement was observed in other cancers. Conclusion: Most cancer patients can have a relatively better prognosis after standardized treatment. The survival of female is superior to male. The improvement of the survival of breast cancer and lung cancer could be contributed to novel treatment methods and early diagnosis.

Key words: Cancer, Survival rate, Hospital-based registry