Search

Additional file 1: Figure S1. Waterfall plot of tumor size change. Figure S2. Survival in different clinical/pathological responses. Figure S3. Survival in different doses of apatinib. Figure S4. Survival analysis for adverse events. Figure S5. Survival analysis for apatinib application.

Additional file 2: Table S1. Details about RECIST evaluation.

Wei Li,1 Huiqin Jiang,1 Yiyi Yu,1 Yan Wang,1 Zhiming Wang,1 Yuehong Cui,1 Kuntang Shen,2 Zhenbin Shen,2 Yong Fang,2 Tianshu Liu1,3 1Department of Oncology, Zhongshan Hospital, Fudan University, Shanghai, China; 2Department of General Surgery, Zhongshan Hospital, Fudan University, Shanghai, China; 3Center of Evidence-Based Medicine, Fudan University, Shanghai, China Purpose: Chemotherapy is the standard care for patients with incurable advanced gastric cancer. Whether or when the addition of gastrectomy to chemotherapy improves survival of advanced gastric cancer patients with a single noncurable factor remains controversial. We aimed to evaluate the superiority of gastrectomy following chemotherapy vs chemotherapy alone regarding overall survival (OS) in these patients.Patients and methods: Patients with advanced gastric cancer from January 2008 to December 2014 were retrieved from our prospectively acquired database and retrospectively analyzed. The patients with a single noncurable factor were grouped in terms of cancer treatment: chemotherapy alone or gastrectomy following chemotherapy.Results: Four hundred and fourteen patients (333 chemotherapy alone and 81 gastrectomy following chemotherapy) were included in this study. Kaplan–Meier survival curve showed a significant difference on median OS between chemotherapy-alone group and the gastrectomy plus chemotherapy group (10.9 vs 15.9 months, P

Wei Li,1 Qian Li,1 Yiyi Yu,1 Yan Wang,1 Erbao Chen,1 Lingli Chen,2 Zhiming Wang,1 Yuehong Cui,1 Tianshu Liu1,3 1Department of Medical Oncology, Zhongshan Hospital, Fudan University, Shanghai, People’s Republic of China; 2Department of Pathology, Zhongshan Hospital, Fudan University, Shanghai, People’s Republic of China; 3Center of Evidence-Based Medicine, Fudan University, Shanghai, People’s Republic of ChinaCorrespondence: Tianshu Liu Department of OncologyZhongshan Hospital, Center of Evidence-Based Medicine, Fudan University, Shanghai, People’s Republic of ChinaEmail liu.tianshu@zs-hospital.sh.cnPurpose: Alpha-fetoprotein-producing gastric cancer (AFPGC) and hepatoid adenocarcinoma of stomach (HAS) are rare types of gastric cancer, with specific clinical manifestations and poor prognosis. The standardized treatment process of such cancers remains elusive. We aim to investigate the efficacy of immunotherapy combined with chemotherapy on patients with AFPGC or HAS.Patients and Methods: AFPGC and HAS patients who underwent immunotherapy and/or chemotherapy as the first-line treatment at our institute from June 2016 to December 2018 were enrolled in this observational study. Their clinicopathological characteristics, serum AFP level and treatment methods were collected. The progression-free survival (PFS) and overall survival (OS) were analyzed and compared between patients who received immunotherapy plus chemotherapy and those received chemotherapy.Results: A total of 21 patients with advanced AFPGC or HAS were included in the study and the median follow-up time was 28.0 months. Of the 21 patients, 7 patients received immunotherapy of PD-1 antibody (nivolumab) plus chemotherapy and 14 patients as control received chemotherapy with or without Herceptin/Apatinib. The median progression-free survival (mPFS) time was 5.0 months (4.3 months in the control group and 22.0 months in the immunotherapy group). The median overall survival (mOS) time of the control group was 16.0 months (14.0 months in chemotherapy alone subgroup, 20.0 months in chemotherapy plus Apatinib or Herceptin subgroup), while the mOS of patients receiving immunotherapy was not reached.Conclusion: This study suggested PD-1 checkpoint inhibitor plus chemotherapy could benefit AFPGC and HAS patients. Its mechanism of action warrants further investigation.Keywords: gastric cancer, immunotherapy, alpha-fetoprotein-producing gastric cancer, hepatoid adenocarcinoma of stomach

Shan Yu,1 Yan Wang,1 Xi Cheng,2 Minzhi Lv,3 Yuehong Cui,1 Wei Li,1 Yiyi Yu,1 Qian Li,1 Tianshu Liu1,4 1Department of Medical Oncology, Zhongshan Hospital, Fudan University, Shanghai, People’s Republic of China; 2Department of Medical Oncology, Sir Run Run Shaw Hospital, Zhejiang University, Hangzhou, People’s Republic of China; 3Department of Biostatistics, Zhongshan Hospital, Fudan University, Shanghai, People’s Republic of China; 4Center of Evidence-Based Medicine, Fudan University, Shanghai, People’s Republic of ChinaCorrespondence: Tianshu Liu Department of Medical OncologyCenter of Evidence-Based Medicine, Zhongshan Hospital, Fudan University, 180 Fenglin Road, Shanghai 200032, People’s Republic of ChinaEmail liu.tianshu@zs-hospital.sh.cnIntroduction: To compare the prognosis of adjuvant SOX (S-1 and oxaliplatin) vs XELOX (capecitabine and oxaliplatin) chemotherapy in Chinese patients with gastric cancer (GC) after D2 gastrectomy.Methods: This was a real-world study of patients with GC (stages II–III) who underwent D2 gastrectomy and received adjuvant SOX or XELOX between 01/2010 and 06/2017 in Zhongshan Hospital affiliated to Fudan University. The patients were matched by propensity score matching. The primary and secondary endpoints were disease-free survival (DFS) and overall survival (OS), respectively. Adverse events (AEs) were compared.Results: A total of 552 patients were included. The median follow-up time was 24.9 months. There were no differences in DFS (median, 44.4 vs 41.2 months; HR=1.17, 95% CI: 0.92– 1.48) and OS (median, 61.5 vs 65.3 months; HR=1.01, 95% CI: 0.73– 1.39) between the XELOX and SOX groups. Both DFS and OS had no significant differences between SOX and XELOX for all subgroups based on sex (P=0.949, P=0.990), age (P=0.303, P=0.392), Lauren type (P=0.362, P=0.573), type of gastrectomy (P=0.607 P=0.989), and pathological TNM stage (P=0.899, P=0.888). A total of 86 patients in the SOX subgroup (34.2%) experienced AEs, similar to the rate found in the XELOX subgroup (104 patients or 41.4%; P=0.098).Discussion: The results suggested that adjuvant SOX chemotherapy has similar survival benefits compared to XELOX chemotherapy in Chinese patients with pathological stage II or III GC after D2 gastrectomy.Keywords: gastric cancer, gastrectomy, adjuvant chemotherapy, capecitabine, oxaliplatin, S-1, XELOX

Colorectal cancer (CRC) is a common malignant tumor worldwide. The solute carrier family 25 member 18 (SLC25A18) transports glutamate across the inner mitochondrial membrane and involves some non-tumor diseases, yet little is known about its role in malignancy. Here, we studied the function and mechanism of SLC25A18 in CRC. We conducted a bioinformatic analysis of the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases to identify the correlation of SLC25A18 expression with clinic-pathological characteristics. Function experiments were implemented to estimate the variation of aerobic glycolysis and cell proliferation due to in vitro and in vivo up- or down-regulation of SLC25A18. Immunohistochemical staining of SLC25A18 was performed on a tissue microarray of 106 patients with primary or metastatic CRC to evaluate its predictive and prognostic value. SLC25A18 expression was low in the CRC samples and was negatively correlated with stage, age and serum carcinoembryonic antigen levels. High expression of SLC25A18 indicated longer disease-free survival time after surgery. Exogenous overexpression of SLC25A18 decreased glucose consumption, lactate production, intracellular ATP concentration and cell proliferation and abrogated expression of CTNNB1, PKM2, LDHA and MYC. Inhibition of Wnt/β-catenin restored SLC25A18-repressed cellular activities. SLC25A18 clinically predicted a longer survival time after surgery or medicine treatment. These results showed that increased SLC25A18 expression inhibits Warburg effect and cell proliferation via Wnt/β-catenin cascade, and suggest a better prognosis after treatment.

The purpose of this paper is to infer a global (collective) model of time-varying responses of a set of nodes as a dynamic graph, where the individual time series are respectively observed at each of the nodes. The motivation of this work lies in the search for a connectome model which properly captures brain functionality upon observing activities in different regions of the brain and possibly of individual neurons. We formulate the problem as a quadratic objective functional of observed node signals over short time intervals, subjected to the proper regularization reflecting the graph smoothness and other dynamics involving the underlying graph's Laplacian, as well as the time evolution smoothness of the underlying graph. The resulting joint optimization is solved by a continuous relaxation and an introduced novel gradient-projection scheme. We apply our algorithm to a real-world dataset comprising recorded activities of individual brain cells. The resulting model is shown to not only be viable but also efficiently computable.

We present a novel adversarial framework for training deep belief networks (DBNs), which includes replacing the generator network in the methodology of generative adversarial networks (GANs) with a DBN and developing a highly parallelizable numerical algorithm for training the resulting architecture in a stochastic manner. Unlike the existing techniques, this framework can be applied to the most general form of DBNs with no requirement for back propagation. As such, it lays a new foundation for developing DBNs on a par with GANs with various regularization units, such as pooling and normalization. Foregoing back-propagation, our framework also exhibits superior scalability as compared to other DBN and GAN learning techniques. We present a number of numerical experiments in computer vision as well as neurosciences to illustrate the main advantages of our approach.

Gaussian Process Factor Analysis (GPFA) has been broadly applied to the problem of identifying smooth, low-dimensional temporal structure underlying large-scale neural recordings. However, spike trains are non-Gaussian, which motivates combining GPFA with discrete observation models for binned spike count data. The drawback to this approach is that GPFA priors are not conjugate to count model likelihoods, which makes inference challenging. Here we address this obstacle by introducing a fast, approximate inference method for non-conjugate GPFA models. Our approach uses orthogonal second-order polynomials to approximate the nonlinear terms in the non-conjugate log-likelihood, resulting in a method we refer to as \textit{polynomial approximate log-likelihood} (PAL) estimators. This approximation allows for accurate closed-form evaluation of marginal likelihoods and fast numerical optimization for parameters and hyperparameters. We derive PAL estimators for GPFA models with binomial, Poisson, and negative binomial observations and find the PAL estimation is highly accurate, and achieves faster convergence times compared to existing state-of-the-art inference methods. We also find that PAL hyperparameters can provide sensible initialization for black box variational inference (BBVI), which improves BBVI accuracy. We demonstrate that PAL estimators achieve fast and accurate extraction of latent structure from multi-neuron spike train data.