Site en français
SALMON Isabelle


Department of Pathology

Person in charge of the Unit : Oui

The main research themes of the laboratory focus on the identification and validation of new biomarkers in human cancers with diagnostic, prognostic and theragnostic purposes. The research activities combine fundamental and clinical aspects. For more than 15 years, our investigations have been focused on protein biomarkers in human tissue samples, animals and in vitro models. Immunohistochemistry (IHC) plays an essential role in the validation of these biomarkers because, as opposed to other biochemical approaches, this technology enables morphological control and thus protein localization at histological and cellular levels. A close collaboration with the Laboratory of Image Synthesis and Analysis (LISA, Ecole polytechnique, U.L.B., allows us to develop standardized tools for characterizing protein expression by using the multiple abilities provided by digital image analysis. From this collaboration was created the interfaculty unit, DIAPath (Digital Image Analysis in Pathology,, which is included in the Center for Microscopy and Molecular Imaging (CMMI, Biopark of Gosselies, know-how in biomarkers is often requested by other university and biotech research teams. These collaborations lead us to analyse human tumours from many origins as well as pathologic tissues from other diseases, such as inflammatory diseases, graft-versus-host or diabetes.

Digital Image Analysis in Pathology

Person in charge of the Unit : Oui

DIAPath is a transdisciplinary and interfaculty research unit (Faculties of Medicine and École polytechnique de Bruxelles) integrated into the "Center for Microscopy and Molecular Imaging" (CMMI, Biopark of Gosselies). This unit is the result of a long-standing collaboration between the Pathology Department of the Erasme Hospital and the Laboratory of Image Synthesis and Analysis (LISA, Ecole polytechnique, ULB). Thanks to this collaboration, DIAPath is developing an integrated computational pathology approach for the characterisation, validation and monitoring of histopathological biomarkers in animal and human tissues.

The approach developed by DIAPath uses histological, immunohistochemistry (IHC) and chromogenic in situ hybridisation (CISH) techniques. In addition, the unit has developed Whole Slide Imaging for the objective and quantitative characterisation of biomarkers using image analysis aided by artificial intelligence. These biomarkers can be morphological in nature or concern the expression, colocalisation or co-expression of antigens (or other labelled molecules), as well as their distribution in histological samples. Data analysis skills complete the set-up. The overall objective is to extract information useful for understanding disease processes and responses to treatment, as well as to identify and validate new biomarkers useful for diagnostic, prognostic and therapeutic purposes. 

DIAPath is continuing to develop its skills to extend its tissue labelling, imaging and analysis techniques to fluorescence.



A first objective of this project is to establish an accurate topography of the SARS-CoV-2 virus in the organs and various cells of patients who died in COVID-19 units, in order to contribute to a better understanding of the pathogenesis of COVID-19. A second objective is to provide a routine diagnostic test for the detection of SARS-CoV-2 in routine samples in pathological anatomy. DIAPath is particularly involved in the development of biomarkers (anti-SARS-Cov2 IHC, anti-ACE-2 IHC, anti-SARS-Cov-2 CISH).

Identification of biomarkers related to tumor invasion and angiogenesis in neuro-oncology

Gliobastoma (GBM) is the most frequent brain tumor in adults. Despite progress in surgery, radiotherapy and chemotherapy, the overall survival of patients with GBM remains extremely poor. The hallmark of GBM is local invasion of single tumor cells to adjacent and distant brain structures that renders complete tumor resection impossible and leads to tumor recurrence and death of the patient. In addition the particular GBM vasculatur has motivated us to study a number of angiogenesis actors. This is why we center this research subject around the two following aspects: (1) Study of actors involved in GBM cell migration. In the literature, many data support the implication of the IGF system in GBM pathogenesis. In addition, previous studies performed in our Department showed that IGFs promote proliferation and migration of GBM cells. To improve the treatment, it is imperative to better understand the molecular mechanisms involved in GBM pathogenesis, in order to develop new molecularly targeted therapies. The aim of this project is to evaluate if the migration of the GBM cells in response to IGF-I might be due to tenascin-C secretion. Tenascin-C is a component of the extracellular matrix that promotes the GBM cells migration. In a previous work, we showed that tenascin-C expression is associated with a worse prognosis in grade II astrocytoma. The molecular mechanisms of these processes are still in study.(2)  Study of actors involved in tumor angiogenesis. This project focuses on the study of isolated and coobined effects induced by galectins-1 and 3 on angiogenesis. Our results show a synergistic effect of these two galectins on endothelial cell growth and tube formation through activation of VEGFR1 and 2. We are now analyzing the signaling pathways that are potentially involved in these effects. From a clinical point of view, we also analyze VEGF and VEGFR expression in endothelial cells in a large series of normal, inflammatory and cancer tissue samples.


Development of a digital pathology platform within the CMMI (DIAPath department: Digital Image Analysis in Pathology)
Provide state-of-the-art expertise in the field of histopathology and related imaging technologies.
Meet the specific needs of scientific and industrial partners.

Development of a digital pathology platform within the CMMI

Molecular classification and identification of tumor stem cell biomarkers in neuro-oncology

GBM are considered by the WHO classification as a single histological entity. However, considerable variability in biologic behavior still exits within this entity, resulting in significant differences in terms of prognosis and response to treatment. In an attempt to better understand GBM, many groups have used high-scale molecular profiling studies. These studies have revealed different molecular GBM subtypes. Interestingly, these different subtypes show different prognosis and respond differently to intensive therapy. Furthermore, data in the literature suggest that the high resistance of GBM to adjuvant therapy is partly due to the presence of a tumorigenic subpopulation of cancer stem cells called glioma stem cells (GSC). This is why, in complement to the 1st research subject, we also study the two following aspects: (1) Molecular classification of glioblastomas. We are currently trying to approach the molecular classification of GBM by the use of immunohistochemistry (IHC), a method which is easily applicable on a routine basis. This method has allow us so far to distinguish two clinically relevant classes of GBM based on the analysis of only three biomarkers (EGFR, p53 and PDGFRA) using image analysis to quantify their expression. We are currently investigating the possibility to improve this classification with the addition of other biomarkers resulting from our different research subjects in neuro-oncology. (2) Characterization of cancer stem cells in GBM. The cancer stem cell hypothesis suggests that only a distinct population of tumor cells, so-called cancer stem cells (CSC), is able to give rise to malignancies and tumor growth. CSC are defined as tumor cells with stem cell properties (i.e. asymmetric cell division, infinite growth, multipotency and cancer-initiating ability upon orthotopic implantation). A therapy will ultimately fail if it does not eliminate CSC; thus efficacy of a therapeutic agent depends on its efficacy against CSC. No specific marker of glioma stem cells (GSC) is currently reported in literature. Several markers are used to isolate GSC from human GBM but their expression is heterogeneous. This point complicates the accurate identification of GSC within GBM, an essential step to develop targeted therapies against GSC. Our project proposes to better characterize human GSC by studying the expression of several markers.