OBJECTIVES

: With the endorsement from the China government, the National Cancer Big Data Platform (NCBDP) Project was initiated in 2016 to establish a nation-wide platform to collect tumor patient data throughout the country. This real-world data has become part of the trend using medical big data to generate evidence and eventually benefit cancer patients in China. This abstract presents the current status of this national tumor databank.

METHODS

: The National Cancer Center of China has been designated as the repository to house all the collected data in this system. Digital China Health Technologies Corporation is the technology driver of the system. Raw data are collected from tumor specialty hospitals and tumor departments of general hospitals throughout China and then mapped a common data model with standard representations of clinical information. Data management also includes data security assurance and data quality control. The NCBDP contains diverse types of patient level information from both inpatient and outpatient settings, such as diagnosis codes, laboratory tests, medication information, and clinical documents.

RESULTS

: As of January 2019, raw clinical data were collected from 55 cancer hospitals in China, with over 10 million patients in total and an average time span of 5.4 years. Data from 42 hospitals went through initial cleaning processes (i.e., removing duplicated and sparse records), resulting in 3.6 million patients with 88 types of tumors. Eighty-two percent of the patients are from top 15 tumors. Lung cancer has most of the individual patients, which accounts 16.4% of the tumor data and breast cancer accounts the second largest number of patients (12.6%). Gastric malignant tumor, esophagus cancer, and cervical cancer are in the top third, fourth, and fifth places, respectively.

CONCLUSIONS

: NCBDP represents an evolving oncology real world database currently available in China for clinical research, drug development, and health policy generating.