The existing in vitro models for antitumor drug screening have great limitations. Many compounds that inhibit 2D cultured cells do not exhibit the same pharmacological effects in vivo, thereby wasting human and material resources as well as time during drug development. Therefore, developing new models is critical. The 3D bioprinting technology has greater advantages in constructing human tissue compared with sandwich culture and organoid construction. Here, we used 3D bioprinting technology to construct a 3D model with HepG2 cells (3DP-HepG2). The biological activities of the model were evaluated by immunofluorescence, real-time quantitative PCR, and transcriptome sequencing. Compared with the traditional 2D cultured tumor cells (2D-HepG2), 3DP-HepG2 showed significantly improved expression of tumorrelated genes, including ALB, AFP, CD133, IL-8, EpCAM, CD24, and β-TGF genes. Transcriptome sequencing analysis revealed large differences in gene expression between 3DP-HepG2 and 2D-HepG2, especially genes related to hepatocyte function and tumor. We also compared the effects of antitumor drugs in 3DP-HepG2 and 2D-HepG2, and found that the large differences in drug resistance genes between the models may cause differences in the drugs' pharmacodynamics.

Background: Systemic immune-inflammation index (SII) is considered to be a prognostic marker in several cancers. However, the prognostic value of baseline pre-operative SII in gallbladder carcinoma (GBC) has not been evaluated. This study aimed to determine the prognostic significance of SII and generate a predictive nomogram. Methods: We retrospectively studied 142 GBC patients who underwent surgical resection at the Peking Union Medical College Hospital between 2003 and 2017. SII, neutrophil-to-lymphocyte ratio (NLR), and lymphocyte-to-monocyte ratio (LMR) were evaluated for their prognostic values. A multivariate Cox proportional hazards model was used for the recognition of significant factors. Then, the cohort was randomly divided into the training and the validation set. A nomogram was constructed using SII and other selected indicators in the training set. C-index, calibration plots, and decision curve analysis were performed to assess the nomogram's clinical utility in both the training and the validation set. Results: The predictive accuracy of SII (Harrell's concordance index [C-index]: 0.624), NLR (C-index: 0.626), and LMR (C-index: 0.622) was evaluated. The multivariate Cox model showed that SII was a superior independent predictor than NLR and LMR. SII level (≥600) (hazard ratio [HR]: 1.694, 95% confidence interval [CI]: 1.069-2.684, p = 0.024), carbohydrate antigen (CA) 19-9 level (≥37 U/ml) (HR: 2.407, 95% CI: 1.472-3.933, p < 0.001), and TNM stage (p = 0.026) were selected to construct a nomogram for predicting overall survival (OS). The predictive ability of this model was assessed by C-index (0.755 in the training set, 0.754 in the validation set). Good performance was demonstrated by the calibration plot. A high net benefit was proven by decision curve analysis (DCA). Conclusion: SII is an independent prognostic indicator in GBC patients after surgical resection, and the nomogram based on it is a useful tool for predicting OS.

No abstract