Gastric cancer is a heterogeneous cancer, making treatment responses difficult to predict. Here we show that we identify two distinct molecular subtypes, mesenchymal phenotype (MP) and epithelial phenotype (EP), by analyzing genomic and proteomic data. Molecularly, MP subtype tumors show high genomic integrity characterized by low mutation rates and microsatellite stability, whereas EP subtype tumors show low genomic integrity. Clinically, the MP subtype is associated with markedly poor survival and resistance to standard chemotherapy, whereas the EP subtype is associated with better survival rates and sensitivity to chemotherapy. Integrative analysis shows that signaling pathways driving epithelial-to-mesenchymal transition and insulin-like growth factor 1 (IGF1)/IGF1 receptor (IGF1R) pathway are highly activated in MP subtype tumors. Importantly, MP subtype cancer cells are more sensitive to inhibition of IGF1/IGF1R pathway than EP subtype. Detailed characterization of these two subtypes could identify novel therapeutic targets and useful biomarkers for prognosis and therapy response.

Fully Homomorphic encryption (FHE) has been gaining in popularity as an emerging means of enabling an unlimited number of operations in an encrypted message without decryption. A major drawback of FHE is its high computational cost. Specifically, a bootstrapping step that refreshes the noise accumulated through consequent FHE operations on the ciphertext can even take minutes of time. This significantly limits the practical use of FHE in numerous real applications.By exploiting the massive parallelism available in FHE, we demonstrate the first instance of the implementation of a GPU for bootstrapping CKKS, one of the most promising FHE schemes supporting the arithmetic of approximate numbers. Through analyzing CKKS operations, we discover that the major performance bottleneck is their high main-memory bandwidth requirement, which is exacerbated by leveraging existing optimizations targeted to reduce the required computation. These observations motivate us to utilize memory-centric optimizations such as kernel fusion and reordering primary functions extensively.Our GPU implementation shows a 7.02× speedup for a single CKKS multiplication compared to the state-of-the-art GPU implementation and an amortized bootstrapping time of 0.423us per bit, which corresponds to a speedup of 257× over a single-threaded CPU implementation. By applying this to logistic regression model training, we achieved a 40.0× speedup compared to the previous 8-thread CPU implementation with the same data.

Hepatocellular carcinoma (HCC) is one of the most common malignancies and causes of death worldwide. In this study, we assessed the correlation between clinicopathologic factors with programmed cell death protein 1 (PD-1) and programmed cell death ligand-1 (PD-L1), and cytotoxic T lymphocyte-associated molecule-4 (CTLA-4) expressions. Furthermore, we analyzed the prognostic significance of these proteins in a subgroup of patients. We retrospectively evaluated the PD-1, PD-L1, and CTLA-4 expressions in 294 HCC tissue microarray samples using immunohistochemistry. PD-1 and PD-L1 expressions were significant related to high CD8+ tumor-infiltrating lymphocytes (TILs) (r = 0.664, p < 0.001 and r = 0.149, p = 0.012). Only high Edmondson-Steiner grade was statistically related to high PD-1 expression. High PD-L1 expression was demonstrated as an independent poor prognostic factor for disease-free survival in addition to previous known factors, size >5 cm and serum albumin ≤3.5 g/dL in high CD8+ TILs group. We have demonstrated that the combined high expression of PD-L1 and CD8+ TIL is an important prognostic factor related to the immune checkpoint pathway in HCC and furthermore, there is a possibility that it could be used as a predictor of therapeutic response. Also, this result would be helpful in evaluating the applicable group of PD-1/PD-L1 blocking agent for HCC patients.

Objective Pressure injuries are common and serious complications for hospitalized patients. The pressure injury rate is an important patient safety metric and an indicator of the quality of nursing care. Timely and accurate prediction of pressure injury risk can significantly facilitate early prevention and treatment and avoid adverse outcomes. While many pressure injury risk assessment tools exist, most were developed before there was access to large clinical datasets and advanced statistical methods, limiting their accuracy. In this paper, we describe the development of machine learning-based predictive models, using phenotypes derived from nurse-entered direct patient assessment data. Methods We utilized rich electronic health record data, including full assessment records entered by nurses, from 5 different hospitals affiliated with a large integrated healthcare organization to develop machine learning-based prediction models for pressure injury. Five-fold cross-validation was conducted to evaluate model performance. Results Two pressure injury phenotypes were defined for model development: nonhospital acquired pressure injury (N = 4398) and hospital acquired pressure injury (N = 1767), representing 2 distinct clinical scenarios. A total of 28 clinical features were extracted and multiple machine learning predictive models were developed for both pressure injury phenotypes. The random forest model performed best and achieved an AUC of 0.92 and 0.94 in 2 test sets, respectively. The Glasgow coma scale, a nurse-entered level of consciousness measurement, was the most important feature for both groups. Conclusions This model accurately predicts pressure injury development and, if validated externally, may be helpful in widespread pressure injury prevention.

BackgroundSilent mating type information regulation 2 homolog 1 (SIRT1), an NAD+-dependent deacetylase, might act as a tumor promoter by inhibiting p53, but may also as a tumor suppressor by inhibiting several oncogenes such as β-catenin and survivin. Deleted in breast cancer 1 (DBC1) is known as a negative regulator of SIRT1.MethodsImmunohistochemical expressions of SIRT1, DBC1, β-catenin, surviving, and p53 were evaluated using 2 mm tumor cores from 349 colorectal cancer patients for tissue microarray.ResultsOverexpression of SIRT1, DBC1, survivin, and p53 was seen in 235 (67%), 183 (52%), 193 (55%), and 190 (54%) patients, respectively. Altered expression of β-catenin was identified in 246 (70%) patients. On univariate analysis, overexpression of SIRT1 (p=0.029) and altered expression of β-catenin (p=0.008) were significantly associated with longer overall survival. Expression of SIRT1 was significantly related to DBC1 (p=0.001), β-catenin (p=0.001), and survivin (p=0.002), but not with p53. On multivariate analysis, age, tumor stage, differentiation, and expression of SIRT1 were independent prognostic factors significantly associated with overall survival.ConclusionsSIRT1 overexpression is a good prognostic factor for colorectal cancer, and SIRT1 may interact with β-catenin and survivin rather than p53.

Plexiform angiomyxoid myofibroblastic tumor (PAMT) of the stomach is a recently recognized entity. Because of its rarity, only 22 cases have been reported in the English-language literature and most of these are single case reports. We report two cases of gastric PAMT. The tumor cells were bland and plexiform arranged in a myxoid stroma, which was positive for alcian blue. Immunohistochemically, the tumor cells were positive for smooth muscle actin, but negative for c-kit, CD34, desmin, S-100 protein, epithelial membrane antigen, neurofilament, and protein kinase C-theta. Mutation analyses for exon 9, 11, 13, and 17 of KIT genes and 12, 14, and 18 of the platelet-derived growth factor receptor alpha (PDGFRA) genes were performed and the tumors were wild-type for mutation.

Homomorphic encryption (HE) draws huge attention as it provides a way of privacy-preserving computations on encrypted messages. Number Theoretic Transform (NTT), a specialized form of Discrete Fourier Transform (DFT) in the finite field of integers, is the key algorithm that enables fast computation on encrypted ciphertexts in HE. Prior works have accelerated NTT and its inverse transformation on a popular parallel processing platform, GPU, by leveraging DFT optimization techniques. However, these GPU-based studies lack a comprehensive analysis of the primary differences between NTT and DFT or only consider small HE parameters that have tight constraints in the number of arithmetic operations that can be performed without decryption. In this paper, we analyze the algorithmic characteristics of NTT and DFT and assess the performance of NTT when we apply the optimizations that are commonly applicable to both DFT and NTT on modern GPUs. From the analysis, we identify that NTT suffers from severe main-memory bandwidth bottleneck on large HE parameter sets. To tackle the main-memory bandwidth issue, we propose a novel NTT-specific on-the-fly root generation scheme dubbed on-the-fly twiddling (OT). Compared to the baseline radix-2 NTT implementation, after applying all the optimizations, including OT, we achieve 4.2× speedup on a modern GPU.

IL-6 and TNFα were significantly increased in the bone marrow aspirate samples of patients with active multiple myeloma (MM) compared to those of normal controls. Furthermore, MM patients with advanced aggressive disease had significantly higher levels of IL-6 and TNFα than those with MM in plateau phase. TNFα increased interleukin-6 (IL-6) production from MM cells. However, the detailed mechanisms involved in signaling pathways by which TNFα promotes IL-6 secretion from MM cells are largely unknown. In our study, we found that TNFα treatments induce MEK and AKT phosphorylation. TNFα-stimulated IL-6 production was abolished by inhibition of JAK2 and IKKβ or by small interfering RNA (siRNA) targeting TNF receptors (TNFR) but not by MEK, p38, and PI3K inhibitors. Also, TNFα increased phosphorylation of STAT3 (ser727) including c-Myc and cyclin D1. Three different types of JAK inhibitors decreased the activation of the previously mentioned pathways. In conclusion, blockage of JAK/STAT-mediated NF-κB activation was highly effective in controlling the growth of MM cells and, consequently, an inhibitor of TNFα-mediated IL-6 secretion would be a potential new therapeutic agent for patients with multiple myeloma.