RESUMO
Artificial intelligence (AI)-powered approaches are becoming increasingly used as histopathologic tools to extract subvisual features and improve diagnostic workflows. On the other hand, hi-plex approaches are widely adopted to analyze the immune ecosystem in tumor specimens. Here, we aimed at combining AI-aided histopathology and imaging mass cytometry (IMC) to analyze the ecosystem of non-small cell lung cancer (NSCLC). An AI-based approach was used on hematoxylin and eosin (H&E) sections from 158 NSCLC specimens to accurately identify tumor cells, both adenocarcinoma and squamous carcinoma cells, and to generate a classifier of tumor cell spatial clustering. Consecutive tissue sections were stained with metal-labeled antibodies and processed through the IMC workflow, allowing quantitative detection of 24 markers related to tumor cells, tissue architecture, CD45+ myeloid and lymphoid cells, and immune activation. IMC identified 11 macrophage clusters that mainly localized in the stroma, except for S100A8+ cells, which infiltrated tumor nests. T cells were preferentially localized in peritumor areas or in tumor nests, the latter being associated with better prognosis, and they were more abundant in highly clustered tumors. Integrated tumor and immune classifiers were validated as prognostic on whole slides. In conclusion, integration of AI-powered H&E and multiparametric IMC allows investigation of spatial patterns and reveals tissue relevant features with clinical relevance. SIGNIFICANCE: Leveraging artificial intelligence-powered H&E analysis integrated with hi-plex imaging mass cytometry provides insights into the tumor ecosystem and can translate tumor features into classifiers to predict prognosis, genotype, and therapy response.
Assuntos
Carcinoma Pulmonar de Células não Pequenas , Neoplasias Pulmonares , Humanos , Carcinoma Pulmonar de Células não Pequenas/patologia , Neoplasias Pulmonares/patologia , Inteligência Artificial , Ecossistema , Citometria por ImagemRESUMO
Introduction: Metabolic reprogramming is a hallmark feature of pancreatic ductal adenocarcinoma (PDAC). A pancreatic juice (PJ) metabolic signature has been reported to be prognostic of oncological outcome for PDAC. Integration of PJ profiling with transcriptomic and spatial characterization of the tumor microenvironment would help in identifying PDACs with peculiar vulnerabilities. Methods: We performed a transcriptomic analysis of 26 PDAC samples grouped into 3 metabolic clusters (M_CL) according to their PJ metabolic profile. We analyzed molecular subtypes and transcriptional differences. Validation was performed by multidimensional imaging on tumor slides. Results: Pancreatic juice metabolic profiling was associated with PDAC transcriptomic molecular subtypes (p=0.004). Tumors identified as M_CL1 exhibited a non-squamous molecular phenotype and demonstrated longer survival. Enrichment analysis revealed the upregulation of immune genes and pathways in M_CL1 samples compared to M_CL2, the group with worse prognosis, a difference confirmed by immunofluorescence on tissue slides. Enrichment analysis of 39 immune signatures by xCell confirmed decreased immune signatures in M_CL2 compared to M_CL1 and allowed a stratification of patients associated with longer survival. Discussion: PJ metabolic fingerprints reflect PDAC molecular subtypes and the immune microenvironment, confirming PJ as a promising source of biomarkers for personalized therapy.
RESUMO
Transcatheter aortic valve implantation (TAVI) is a minimally invasive intervention for the treatment of severe aortic valve stenosis. The main cause of failure is the structural deterioration of the implanted prosthetic leaflets, possibly inducing a valvular re-stenosis 5-10 years after the implantation. Based solely on pre-implantation data, the aim of this work is to identify fluid-dynamics and structural indices that may predict the possible valvular deterioration, in order to assist the clinicians in the decision-making phase and in the intervention design. Patient-specific, pre-implantation geometries of the aortic root, the ascending aorta, and the native valvular calcifications were reconstructed from computed tomography images. The stent of the prosthesis was modeled as a hollow cylinder and virtually implanted in the reconstructed domain. The fluid-structure interaction between the blood flow, the stent, and the residual native tissue surrounding the prosthesis was simulated by a computational solver with suitable boundary conditions. Hemodynamical and structural indicators were analyzed for five different patients that underwent TAVI - three with prosthetic valve degeneration and two without degeneration - and the comparison of the results showed a correlation between the leaflets' structural degeneration and the wall shear stress distribution on the proximal aortic wall. This investigation represents a first step towards computational predictive analysis of TAVI degeneration, based on pre-implantation data and without requiring additional peri-operative or follow-up information. Indeed, being able to identify patients more likely to experience degeneration after TAVI may help to schedule a patient-specific timing of follow-up.