Engagement of myelomonocytic Siglecs by tumor-associated ligands modulates the innate immune response to cancer.

Certain pathogenic bacteria are known to modulate the innate immune response by decorating themselves with sialic acids, which can engage the myelomonocytic lineage inhibitory receptor Siglec-9, thereby evading immunosurveillance. We hypothesized that the well-known up-regulation of sialoglycoconjugates by tumors might similarly modulate interactions with innate immune cells. Supporting this hypothesis, Siglec-9-expressing myelomonocytic cells found in human tumor samples were accompanied by a strong up-regulation of Siglec-9 ligands. Blockade of Siglec-9 enhanced neutrophil activity against tumor cells in vitro. To investigate the function of inhibitory myelomonocytic Siglecs in vivo we studied mouse Siglec-E, the murine functional equivalent of Siglec-9. Siglec-E-deficient mice showed increased in vivo killing of tumor cells, and this effect was reversed by transgenic Siglec-9 expression in myelomonocytic cells. Siglec-E-deficient mice also showed enhanced immunosurveillance of autologous tumors. However, once tumors were established, they grew faster in Siglec-E-deficient mice. In keeping with this, Siglec-E-deficient macrophages showed a propensity toward a tumor-promoting M2 polarization, indicating a secondary role of CD33-related Siglecs in limiting cancer-promoting inflammation and tumor growth. Thus, we define a previously unidentified impact of inhibitory myelomonocytic Siglecs in cancer biology, with distinct roles that reflect the dual function of myelomonocytic cells in cancer progression. In keeping with this, a human polymorphism that reduced Siglec-9 binding to carcinomas was associated with improved early survival in non-small-cell lung cancer patients, which suggests that Siglec-9 might be therapeutically targeted within the right time frame and stage of disease.

Machine learning approach for distinguishing malignant and benign lung nodules utilizing standardized perinodular parenchymal features from CT.

PURPOSE: Computed tomography (CT) is an effective method for detecting and characterizing lung nodules in vivo. With the growing use of chest CT, the detection frequency of lung nodules is increasing. Noninvasive methods to distinguish malignant from benign nodules have the potential to decrease the clinical burden, risk, and cost involved in follow-up procedures on the large number of false-positive lesions detected. This study examined the benefit of including perinodular parenchymal features in machine learning (ML) tools for pulmonary nodule assessment. METHODS: Lung nodule cases with pathology confirmed diagnosis (74 malignant, 289 benign) were used to extract quantitative imaging characteristics from computed tomography scans of the nodule and perinodular parenchyma tissue. A ML tool development pipeline was employed using k-medoids clustering and information theory to determine efficient predictor sets for different amounts of parenchyma inclusion and build an artificial neural network classifier. The resulting ML tool was validated using an independent cohort (50 malignant, 50 benign). RESULTS: The inclusion of parenchymal imaging features improved the performance of the ML tool over exclusively nodular features (P < 0.01). The best performing ML tool included features derived from nodule diameter-based surrounding parenchyma tissue quartile bands. We demonstrate similar high-performance values on the independent validation cohort (AUC-ROC = 0.965). A comparison using the independent validation cohort with the Fleischner pulmonary nodule follow-up guidelines demonstrated a theoretical reduction in recommended follow-up imaging and procedures. CONCLUSIONS: Radiomic features extracted from the parenchyma surrounding lung nodules contain valid signals with spatial relevance for the task of lung cancer risk classification. Through standardization of feature extraction regions from the parenchyma, ML tool validation performance of 100% sensitivity and 96% specificity was achieved.