Abstract Purpose: The study aims to discuss various types of cancer clinical trial data with potential statistical solutions. Background: Cancer clinical trial allows researchers discovering new strategies to prevent, detect, diagnose, or treat cancer. Each trial collects many types of patient data to evaluate various treatment characteristics, such as tumor response data, adverse event (AE) data, demographic data, and follow-up data. Each data type is often stored in a multi-dimensional and irregularly longitudinal layout with potential hierarchical structure, therefore generating ‘big data' phenomena. Such complexities often scare away researchers and lead to use naïve descriptive statistics for analysis. As a result, it loses valuable information and becomes a major hurdle to advance cancer precision medicine. Here we share our thoughts for two primary data types: tumor response data and safety data. Tumor Response Data: Tumor response data is a key component to evaluate treatment efficacy in clinical trial for solid tumor. Response Evaluation Criteria for Solid Tumors (RECIST) has been a standard tool to assess treatment effect. RECIST measures three types of lesion to determine treatment efficacy: target lesion, non-target lesion, and new lesion. Target lesion is the primary lesion to determine response (tumor shrinkage), progression (tumor outgrowth), or stable disease. However, studies have raised issues regarding RECIST being applied for incorrect determination of response, which resulted in premature termination of therapy and imprecise efficacy. One key limitation of RECIST is the use of sum of all target lesion sizes for clinical decision-making. While this approach is straightforward for calculation, it fails to address lesion variability of tumor growth in an individual patient. Statistical mixed effect model could be a powerful tool to fully utilize all lesion data to provide overall assessment of treatment effect and individual lesion variability. Safety Data: Safety data is a critical component of clinical trial to protect patients from unnecessary risk and to develop safety profile of the drug for benefit-risk assessment. The safety data contains many types of AE to monitoring drug toxicity. Evaluation metrics include grade, duration, and likelihood of the treatment attribution; therefore it forms a giant complicated structure. However, safety report in clinical trials is often presented in summary descriptive statistics, such as yes/no of occurrence or frequency of occurrence for each AE. This type of report has been challenged due to inadequate information or lack of details. Statistical opportunities include integration of various safety metrics to generate more useful toxicity variables, such as AE duration adjusted with different grade. Conclusion: Clinical trial collects valuable big data and statistical methods have potential to transform the rich but complicated data into informative evidence. Citation Format: Dung-Tsa Chen, Ben Creelan, Trevor Rose, Wenyaw Chan. Understanding of cancer clinical trial data and statistical opportunities [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 763.