Diffuse large B-cell lymphoma (DLBCL) represents 30-40% of all non-Hodgkin lymphomas (NHL) and is a disease with an aggressive behavior. Because about one-third of DLBCL patients will be refractory or resistant to standard therapy, several studies focused on identification of new individual prognostic and risk stratification biomarkers and new potential therapeutic targets. In contrast to other types of cancers like carcinomas, where tumor microenvironment was widely investigated, its role in DLBCL pathogenesis and patient survival is still poorly understood, although few studies had promising results. The composition of TME and its interaction with neoplastic cells may explain the role of several genes (beta2-microglobulin gene, CD58 gene), receptor-like programmed cell death-1 (PD-1) and its ligand (PD-L1), or other cell components (Treg) in tumor evasion of immune surveillance, resulting in tumor progression. Also, it was found that “gene expression profile” of the microenvironmental cells, the phenotype of tumor-associated macrophages (TAM), the expression of matricellular proteins like SPARC and fibronectin, the overexpression of several types of matrix metalloproteinases (MMPs) like MMP-2 and MMP-9, or the tissue inhibitors of matrix metalloproteinases (TIMPs) may lead to a favorable or adverse outcome. With this review, we try to highlight the influence of microenvironment components over lymphoid clone progression and their prognostic impact in DLBCL patients.

Dendritic cells (DCs) are antigen-presenting cells with an important role in the innate and adaptive immune system. In skin lesions, cutaneous DCs (Langerhans cells, dermal DCs and plasmacytoid DCs) are involved in immune activation in inflammatory benign lesions, as well as in malignant lymphoid proliferations. Density and distribution of DCs in the dermal infiltrate can be helpful to differentiate benign, reactive infiltrate from malignant nature of the lymphoid population. We performed a retrospective study including 149 patients: 35 with mycosis fungoides, 35 with spongiotic dermatitis, 35 with psoriasis, 35 with lupus and 9 with cutaneous T-cell lymphomas (other than mycosis fungoides), diagnosed using histopathological and immunohistochemical stains. Density and distribution of DCs were evaluated using specific markers (CD1a, CD11c and langerin). In all cases, numerous DCs were identified in the dermal infiltrate. Their number was significantly increased in mycosis fungoides and T-cell lymphomas and moderately increased in inflammatory lesions. Variable patterns of distribution were identified such as clusters of DCs with arachnoid extension in mycosis fungoides, nodular pattern in inflammatory lesions and dispersed distribution with peripheric accumulation in T-skin lymphomas. Therefore, immunohistochemical characterization of DC distribution can be an adjuvant tool in differential diagnosis in inflammatory dermatosis and skin lymphomas.

Cutaneous melanoma is a significant immunogenic tumoral model, the most frequently described immune phenomenon being tumor regression, as a result of the interaction of tumoral antigens and stromal microenvironment. We present a retrospective cohort study including 52 cases of melanoma with regression. There were evaluated correlations of the most important prognostic factors (Breslow depth and mitotic index) with FOXP3 expression in tumor cells and with the presence of regulatory T cells and dendritic cells in the tumoral stroma. FOXP3 expression in tumor cells seems an independent factor of poor prognosis in melanoma, while regression areas are characterized by a high number of dendritic cells and a low number of regulatory T cells. FOXP3 is probably a useful therapeutical target in melanoma, since inhibition of FOXP3-positive tumor clones and of regulatory T cells could eliminate the ability of tumor cells to escape the immune defense of the host.

Mycobacteria identification is crucial to diagnose tuberculosis. Since the bacillus is very small, finding it in Ziehl–Neelsen (ZN)-stained slides is a long task requiring significant pathologist’s effort. We developed an automated (AI-based) method of identification of mycobacteria. We prepared a training dataset of over 260,000 positive and over 700,000,000 negative patches annotated on scans of 510 whole slide images (WSI) of ZN-stained slides (110 positive and 400 negative). Several image augmentation techniques coupled with different custom computer vision architectures were used. WSIs automatic analysis was followed by a report indicating areas more likely to present mycobacteria. Our model performs AI-based diagnosis (the final decision of the diagnosis of WSI belongs to the pathologist). The results were validated internally on a dataset of 286,000 patches and tested in pathology laboratory settings on 60 ZN slides (23 positive and 37 negative). We compared the pathologists’ results obtained by separately evaluating slides and WSIs with the results given by a pathologist aided by automatic analysis of WSIs. Our architecture showed 0.977 area under the receiver operating characteristic curve. The clinical test presented 98.33% accuracy, 95.65% sensitivity, and 100% specificity for the AI-assisted method, outperforming any other AI-based proposed methods for AFB detection.