塚本 徹哉

BACKGROUND: The expression patterns and functions of Sushi Domain Containing 2 (SUSD2) differ among various malignancies. This research aims to investigate the expression of SUSD2 and the role of the SUSD2+ cancer-associated fibroblasts (CAFs) for immunotherapy in gastric cancer. METHODS: The expression of SUSD2 and specific markers (CD4, CD8, PD-1, TIGIT, TIM-3 and CD163) was determined using immunohistochemistry and multiplex immunofluorescence (mIHC) on paraffin sections. Flow cytometry and western blot were used to assess the expression of SUSD2 in fibroblasts from fresh samples. Also, analysis of single-cell and bulk RNA sequencing was employed to confirm the presence and characterize the function of SUSD2+ CAFs. The predictive power of indicators for neoadjuvant immunotherapy was evaluated via ROC curve analysis. Animal experiment was employed to validate the immunosuppressive effect of SUSD2+ CAFs. RESULTS: SUSD2 is mainly expressed on fibroblasts within the tumors and the high infiltration of SUSD2+ CAFs went together with a poor survival and a more advanced tumor stage. Significantly, the joint use of SUSD2+ CAFs and CD8+ T cells demonstrated a remarkable ability to predict the efficacy of neoadjuvant immunotherapy superior to PD-L1 combined positive score. High SUSD2+ CAFs was correlated with resistance to immunotherapy as well as low CD8+ T infiltration and high exhausted T cell infiltration. CONCLUSIONS: We have identified a novel subset of CAFs that could predict the survival and response to neoadjuvant immunotherapy of patients. The SUSD2+ CAFs have the potential to serve as a predictive biomarker and a promising target for immunotherapy.

Cytology, a type of pathological examination, involves sampling cells from the human body and observing the morphology of the nucleus, cytoplasm, and cell arrangement. In developing classification AI technologies to support cytology, it is essential to collect and utilize a diverse range of images without bias. However, this is often challenging in practice because of the epidemiologic bias of cancer types and cellular characteristics. The main aim of this study was to develop a method to generate cytological diagnostic images from image findings using text-to-image technology in order to generate diverse images. In the proposed method, we collected Papanicolaou-stained specimens derived from the lung cells of 135 lung cancer patients, from which we extracted 472 patch images. Descriptions of the corresponding findings for these patch images were compiled to create a data set. This dataset was then utilized to finetune the Stable Diffusion (SD) v1 and v2 models. The cell images generated by this method closely resemble real images, and both cytotechnologists and cytopathologists provided positive subjective evaluations. Furthermore, SDv2 demonstrated shapes and contours of nuclei and cytoplasm that were more similar to real images compared to SDv1, showing superior performance in quantitative evaluation metrics. When the generated images were utilized in the classification tasks for cytological images, there was an improvement in classification performance. These results indicate that the proposed method may be effective for generating high-quality cytological images, which enables the image classification model to learn diverse features, thereby improving classification performance.

BACKGROUND: The contribution of the tumor microenvironment and extracellular matrix to the aggressive biology of Gastric Cancer (GC) has been recently characterized; however, the role of EMILIN-1 in this context is unknown. EMILIN-1 is an essential structural element for the maintenance of lymphatic vessel (LV) integrity and displays anti-proliferative properties as demonstrated in skin and colon cancer. Given the key role of LVs in GC progression, the aim of this study was to investigate the role of EMILIN-1 in GC mouse models. METHODS: We used the syngeneic YTN16 cells which were injected subcutaneously and intraperitoneally in genetically modified EMILIN-1 mice. In alternative, carcinogenesis was induced using N-Methyl-N-nitrosourea (MNU). Mouse-derived samples and human biopsies were analyzed by IHC and IF to the possible correlation between EMILIN-1 expression and LV pattern. RESULTS: Transgenic mice developed tumors earlier compared to WT animals. 20 days post-injection tumors developed in EMILIN-1 mutant mice were larger and displayed a significant increase of lymphangiogenesis. Treatment of transgenic mice with MNU associated with an increased number of tumors, exacerbated aggressive lesions and higher levels of LV abnormalities. A significant correlation between the levels of EMILIN-1 and podoplanin was detected also in human samples, confirming the results obtained with the pre-clinical models. CONCLUSIONS: This study demonstrates for the first time that loss of EMILIN-1 in GC leads to lymphatic dysfunction and proliferative advantages that sustain tumorigenesis, and assess the use of our animal model as a valuable tool to verify the fate of GC upon loss of EMILIN-1.

Abstract Approximately 660,000 women are diagnosed with cervical cancer annually. Current screening options such as cytology or human papillomavirus testing have limitations, creating a need to identify more effective ancillary biomarkers for triage. Here, we evaluated whether metabolomic analysis of cervical mucus metabolism could be used to identify biomarkers of cervical intraepithelial neoplasia (CIN) and cervical cancer. The case–control group consisted of 181 CIN, 69 squamous cell carcinoma (SCC) patients, and 48 healthy controls in the primary cohort. We undertook metabolomic analyses using ultra‐HPLC–tandem mass spectrometry. Univariate and multivariate analyses were carried out to profile metabolite characteristics, and receiver operating characteristic (ROC) analysis identified biomarker candidates. Five metabolites conferred the highest discriminatory power for SCC: oxidized glutathione (GSSG) (area under the ROC curve, 0.924; 95% confidence interval, 0.877–0.971), malic acid (0.914, 0.859–0.968), kynurenine (0.884, 0.823–0.945), GSSG/glutathione (GSH) (0.936, 0.892–0.979), and kynurenine/tryptophan (0.909, 0.856–0.961). Malic acid was the best marker for detection of CIN2 or worse (0.858, 0.793–0.922) and was a clinically useful metabolite. We confirmed the reproducibility of the results by validation cohort. Additionally, metabolomic analyses revealed eight pathways strongly associated with cervical neoplasia. Of these, only the tricarboxylic acid cycle was strongly associated with all CINs and cancer, indicating active energy production. Aberrant arginine metabolism by decreasing arginine and increasing citrulline might reduce tumor immunity. Changes in cysteine‐methionine and GSH pathways might drive the initiation and progression of cervical cancer. These results suggest that metabolic analysis can identify ancillary biomarkers and could improve our understanding of the pathophysiological mechanisms underlying cervical neoplasia.