Pancreatic cancer is the most malignant gastrointestinal tumor worldwide, and its incidence is increasing every year. In the past two decades, there have been significant advances in diagnostic methods and surgical techniques for pancreatic cancer. However, frustratingly, little progress has been made in improving the prognosis of patients with pancreatic cancer, even after complete surgical resection. By analyzing 1312 radical resected pancreatic cancer patients with complete data for more than a decade, our team found that a decade of technological advances and the use of new drugs did not significantly improve the overall survival of patients. Relying on a single radical procedure simply does not seem to improve patient survival in pancreatic cancer. As Dr. Blake Cady, an American oncology authority, said, “Biology is King; Selection is Queen; Technical maneuvers are the Prince and Princess”. It is precisely because we know little about the biological characteristics of tumors, and the diagnosis and treatment are based on traditional anatomy, although there are great advances in diagnostic and treatment techniques, it is difficult to apply them “effectively” in the individualized treatment of pancreatic cancer. It is easy to see that new evaluation concepts and methods applied to the treatment of pancreatic cancer patients will undoubtedly greatly promote the development of precise individualized diagnosis and treatment. In the past 10 years, Prof. Liu Liang’s group has been trying to find ways to predict prognosis from tumor markers and traditional imaging, and has achieved remarkable results. For example, (1) combining microvascular density and microvascular intensity of tumor within postoperative resected pancreatic cancer samples to predict the efficacy of radical surgery for pancreatic cancer. It was found that in addition to the density of microvessels within the tumor providing a more abundant channel for cancer cells to metastasize hemorrhagically, the disruption of microvascular structure was also an important mechanism to increase the metastasis of pancreatic cancer. (2) In collaboration with the Department of Mathematics of Fudan University and the Department of Biostatistics of Shanghai Jiaotong University, we systematically analyzed the serological data and biological samples of 274 patients with radical pancreatic cancer in the past 5 years, and retrospectively analyzed the data of patients with locally progressive (257 cases) and metastatic (384 cases) pancreatic cancer in the past 2 years, and screened a set of meaningful “triple positive” indicators. We screened a set of meaningful “triple positive” indicators: “CEA+/CA125+/CA19-9 ≥1000U/ml” predicts that pancreatic cancer patients do not benefit from surgery (Liu et al., Int J Cancer, 2015), and for the first time objectively identified a subgroup of pancreatic cancer patients who “do not benefit from surgery” before surgery. The first objective identification of a subgroup of pancreatic cancer patients with “no benefit from surgery” in the preoperative period. (2) conducted studies on the pancreatic cancer microenvironment and found that the immune cell composition and distribution in the tumor microenvironment significantly predicted the outcome of pancreatic cancer surgery (Wang et al., Ann Surg, 2015). However, through continuous exploration and combining with the latest overseas research reports, our team has increasingly recognized the limitations of traditional evaluation methods, and as a highly heterogeneous tumor, the prognostic evaluation of pancreatic cancer needs to be expanded to new areas beyond genomics and proteomics. Based on this understanding, Prof. Liu Liang’s group also searched for a new method of preoperative prognosis prediction by combining tumor metabolic detection technology PET/CT and serum CA19-9, and this study has been published in full in international mainstream journals (Xu et al., Eur J Nucl Med Mol Imaging, 2014). The findings of this study confirm our previous speculations on the value of “tumor metabolic load” in the preoperative evaluation of pancreatic cancer. The concept of “tumor metabolism” was first introduced in the 1920s by Dr. Otto Warburg, the Nobel laureate who discovered the “Warburg effect”. In the following decades, the research has gradually expanded from tumor glucose metabolism to lipid metabolism, protein metabolism, amino acid metabolism, and other fields, and there are interconversions between different types of metabolism, which complete the process of replenishment of the tricarboxylic acid cycle (TCA) loop and provide material and energy support for tumor growth. Through a review of relevant studies in the past decade, Prof. Liu Liang’s team found that the indicators of “tumor metabolism” have predictive and even therapeutic value in many solid tumors. The “tumor metabolic load” is likely to become a new method to describe and evaluate pancreatic tumors in addition to TNM staging, and is also considered to be the fourth core mechanism to determine tumorigenesis and progression.