Machine learning-based identification of colorectal advanced adenoma using clinical and laboratory data: a phase I exploratory study in accordance with updated World Endoscopy Organization guidelines for noninvasive colorectal cancer screening tests.

Objective: The recent World Endoscopy Organization (WEO) guidelines now recognize precursor lesions of colorectal cancer (CRC) as legitimate screening targets. However, an optimal screening method for detecting advanced adenoma (AA), a significant precursor lesion, remains elusive. Methods: We employed five machine learning methods, using clinical and laboratory data, to develop and validate a diagnostic model for identifying patients with AA (569 AAs vs. 3228 controls with normal colonoscopy). The best-performing model was selected based on sensitivity and specificity assessments. Its performance in recognizing adenoma-carcinoma sequence was evaluated in line with guidelines, and adjustable thresholds were established. For comparison, the Fecal Occult Blood Test (FOBT) was also selected. Results: The XGBoost model demonstrated superior performance in identifying AA, with a sensitivity of 70.8% and a specificity of 83.4%. It successfully detected 42.7% of non-advanced adenoma (NAA) and 80.1% of CRC. The model-transformed risk assessment scale provided diagnostic performance at different positivity thresholds. Compared to FOBT, the XGBoost model better identified AA and NAA, however, was less effective in CRC. Conclusion: The XGBoost model, compared to FOBT, offers improved accuracy in identifying AA patients. While it may not meet the recommendations of some organizations, it provides value for individuals who are unable to use FOBT for various reasons.

Adenoma location, size, and morphology are risk factors for FOBT false-negative results in inpatients with advanced colorectal adenoma.

Recently, advanced adenoma (AA) has been recognized as a target for colorectal cancer (CRC) screening. However, the fecal occult blood test (FOBT), the primary non-invasive screening method, shows limited sensitivity in detecting AA. This study investigates the relationship between adenoma characteristics and FOBT false-negative results. In a retrospective cohort study conducted from 2015 to 2022, we examined 342 inpatients with AA who underwent colonoscopy and received qualitative FOBT. FOBT sensitivity was analyzed about various adenoma characteristics, and logistic regression models were employed to investigate the relationship between adenoma features and FOBT false-negative outcomes. FOBT sensitivity in AA inpatients was 52.63%. Significant differences in sensitivity were observed based on adenoma location (left vs. right), morphology (with or without pedunculation), and size (≤ 10 mm vs. > 10 mm). After adjusting for several potential confounders, FOBT showed a reduced false-negative rate in AA with large-sized (OR, 0.49; 95% CI 0.31-0.77), left-sided location (OR, 0.53; 95% CI 0.31-0.89), and pedunculated morphology (OR, 0.73; 95% CI 0.43-1.24). AA with large size, left-sided location, and pedunculated morphology independently contribute to a decreased rate of FOBT false-negative results. However, these adenoma characteristics are not actively modifiable. Therefore, novel non-invasive methods are needed to improve AA detection accuracy.

Monitoring the Stimulated Uncapping Process of Gold-Capped Mesoporous Silica Nanoparticles.

To establish a new method for tracking the interaction of nanoparticles with chemical cleaving agents, we exploited the optical effects caused by attaching 5-10 nm gold nanoparticles with molecular linkers to large mesoporous silica nanoparticles (MSN). At low levels of gold loading onto MSN, the optical spectra resemble colloidal suspensions of gold. As the gold is removed, by cleaving agents, the MSN revert to the optical spectra typical of bare silica. Time-lapse images of gold-capped MSN stationed in microchannels reveal that the rate of gold release is dependent on the concentration of the cleaving agent. The uncapping process was also monitored successfully for MSN endocytosed by A549 cancer cells, which produce the cleaving agent glutathione. These experiments demonstrate that the optical properties of MSN can be used to directly monitor cleaving kinetics, even in complex cellular settings.

Mesenchymal Stem Cells Immunosuppressed IL-22 in Patients with Immune Thrombocytopenia via Soluble Cellular Factors.

Mesenchymal stem cells are immunoregulation cells. IL-22 plays an important role in the pathogenesis of immune thrombocytopenia. However, the effects of mesenchymal stem cells on IL-22 production in patients with immune thrombocytopenia remain unclear. Flow cytometry analyzed immunophenotypes of mesenchymal stem cells; differentiation of mesenchymal stem cells was observed by oil red O and Alizarin red S staining; cell proliferation suppression was measured with MTS; IL-22 levels of cell-free supernatants were determined by ELISA. Mesenchymal stem cells inhibited the proliferation of activated CD4(+)T cells; moreover, mesenchymal stem cells immunosuppressed IL-22 by soluble cellular factors but not PGE2. These results suggest that mesenchymal stem cells may be a therapeutic strategy for patients with immune thrombocytopenia.

Luciferase and luciferin co-immobilized mesoporous silica nanoparticle materials for intracellular biocatalysis.

We report a gold nanoparticle (AuNP)-capped mesoporous silica nanoparticle (Au-MSN) platform for intracellular codelivery of an enzyme and a substrate with retention of bioactivity. As a proof-of-concept demonstration, Au-MSNs are shown to release luciferin from the interior pores of MSN upon AuNP uncapping in response to disulfide-reducing antioxidants and codeliver bioactive luciferase from the PEGylated exterior surface of Au-MSN to Hela cells. The effectiveness of luciferase-catalyzed luciferin oxidation and luminescence emission in the presence of intracellular ATP was measured by a luminometer. Overall, the chemical tailorability of the Au-MSN platform to retain enzyme bioactivity, the ability to codeliver enzyme and substrate, and the potential for imaging tumor growth and metastasis afforded by intracellular ATP- and glutathione-dependent bioluminescence make this platform appealing for intracellular controlled catalysis and tumor imaging.

Interaction of mesoporous silica nanoparticles with human red blood cell membranes: size and surface effects.

The interactions of mesoporous silica nanoparticles (MSNs) of different particle sizes and surface properties with human red blood cell (RBC) membranes were investigated by membrane filtration, flow cytometry, and various microscopic techniques. Small MCM-41-type MSNs (∼100 nm) were found to adsorb to the surface of RBCs without disturbing the membrane or morphology. In contrast, adsorption of large SBA-15-type MSNs (∼600 nm) to RBCs induced a strong local membrane deformation leading to spiculation of RBCs, internalization of the particles, and eventual hemolysis. In addition, the relationship between the degree of MSN surface functionalization and the degree of its interaction with RBC, as well as the effect of RBC-MSN interaction on cellular deformability, were investigated. The results presented here provide a better understanding of the mechanisms of RBC-MSN interaction and the hemolytic activity of MSNs and will assist in the rational design of hemocompatible MSNs for intravenous drug delivery and in vivo imaging.