AKAurélie Kamoun

Aurélie Kamoun

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Aurélie Kamoun has a total of 198 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Deep learning assessment of metastatic relapse risk from digitized breast cancer histological slides

    by I. Garberis, V. Gaury, C. Saillard, D. Drubay, K. Elgui, B. Schmauch, A. Jaeger, L. Herpin, J. Linhart, M. Sapateiro, et al.

    Accurate risk stratification is critical for guiding treatment decisions in early breast cancer. We present an artificial intelligence (AI)-based tool that analyzes digitized tumor slides to predict 5-year metastasis-free survival (MFS) in patients with estrogen receptor-positive, HER2-negative (ER + /HER2 − ) early breast cancer (EBC). Our deep learning model, RlapsRisk BC, independently predicts MFS and provides significant prognostic value beyond traditional clinico-pathological variables (C-index 0.81 vs 0.76, p < 0.05). Applying a 5% MFS event probability threshold stratifies patients into low- and high-risk groups. After dichotomization, combining RlapsRisk BC with clinico-pathological factors increases cumulative sensitivity (0.69 vs 0.63) and dynamic specificity (0.80 vs 0.76) compared to clinical factors alone. Expert analysis of high-impact regions identified by the model highlights well-established morphological features, supporting its interpretability and biological relevance.


  • Top Cited Papers
    Open Access

    Pacpaint: a histology-based deep learning model uncovers the extensive intratumor molecular heterogeneity of pancreatic adenocarcinoma

    by Charlie Saillard, Flore Delecourt, Benoit Schmauch, Olivier Moindrot, Magali Svrcek, Armelle Bardier-Dupas, Jean Francois Emile, Mira Ayadi, Vinciane Rebours, Louis de Mestier, et al.

    Two tumor (Classical/Basal) and stroma (Inactive/active) subtypes of Pancreatic adenocarcinoma (PDAC) with prognostic and theragnostic implications have been described. These molecular subtypes were defined by RNAseq, a costly technique sensitive to sample quality and cellularity, not used in routine practice. To allow rapid PDAC molecular subtyping and study PDAC heterogeneity, we develop PACpAInt, a multi-step deep learning model. PACpAInt is trained on a multicentric cohort (n = 202) and validated on 4 independent cohorts including biopsies (surgical cohorts n = 148; 97; 126 / biopsy cohort n = 25), all with transcriptomic data (n = 598) to predict tumor tissue, tumor cells from stroma, and their transcriptomic molecular subtypes, either at the whole slide or tile level (112 µm squares). PACpAInt correctly predicts tumor subtypes at the whole slide level on surgical and biopsies specimens and independently predicts survival. PACpAInt highlights the presence of a minor aggressive Basal contingent that negatively impacts survival in 39% of RNA-defined classical cases. Tile-level analysis ( > 6 millions) redefines PDAC microheterogeneity showing codependencies in the distribution of tumor and stroma subtypes, and demonstrates that, in addition to the Classical and Basal tumors, there are Hybrid tumors that combine the latter subtypes, and Intermediate tumors that may represent a transition state during PDAC evolution.


  • Open Access

    Deep Learning Allows Assessment of Risk of Metastatic Relapse from Invasive Breast Cancer Histological Slides

    by I. Garberis, V. Gaury, C. Saillard, D. Drubay, K. Elgui, B. Schmauch, A. Jaeger, L. Herpin, J. Linhart, M. Sapateiro, et al.

    Background Correctly classifying early estrogen receptor-positive and HER2-negative (ER+/HER2) breast cancer (EBC) cases allows to propose an adapted adjuvant systemic treatment strategy. We developed a new AI-based tool to assess the risk of distant relapse at 5 years for ER+/HER2− EBC patients from pathological slides. Patients and Methods The discovery dataset (GrandTMA) included 1429 ER+/HER2− EBC patients, with long-term follow-up and an available hematoxylin-eosin and saffron (HES) whole slide image (WSI). A Deep Learning (DL) network was trained to predict metastasis free survival (MFS) at five years, based on the HES WSI only (termed RlapsRisk). A combined score was then built using RlapsRisk and well established prognostic factors. A threshold corresponding to a probability of MFS event of 5% at 5 years was applied to dichotomize patients into low or high-risk groups. The external validation, as well as assessment of the additional prognosis value of the DL model beyond standard clinico-pathologic factors were carried out on an independent, prospective cohort (CANTO, NCT01993498) including 889 HES WSI of ER+/HER2− EBC patients. Results RlapsRisk was an independent prognostic factor of MFS in multivariable analysis adjusted for established clinico-pathological factors (pConclusions Our deep learning model developed on digitized HES slides provided additional prognostic information as compared to current clinico-pathological factors and has the potential of valuably informing the decision making process in the adjuvant setting when combined with current clinico-pathological factors.


  • Top Cited Papers
    Open Access

    A deep learning model to predict RNA-Seq expression of tumours from whole slide images

    by Benoît Schmauch, Alberto Romagnoni, Elodie Pronier, Charlie Saillard, Pascale Maillé, Julien Calderaro, Aurélie Kamoun, Meriem Sefta, Sylvain Toldo, Mikhail Zaslavskiy, et al.

    Deep learning methods for digital pathology analysis are an effective way to address multiple clinical questions, from diagnosis to prediction of treatment outcomes. These methods have also been used to predict gene mutations from pathology images, but no comprehensive evaluation of their potential for extracting molecular features from histology slides has yet been performed. We show that HE2RNA, a model based on the integration of multiple data modes, can be trained to systematically predict RNA-Seq profiles from whole-slide images alone, without expert annotation. Through its interpretable design, HE2RNA provides virtual spatialization of gene expression, as validated by CD3- and CD20-staining on an independent dataset. The transcriptomic representation learned by HE2RNA can also be transferred on other datasets, even of small size, to increase prediction performance for specific molecular phenotypes. We illustrate the use of this approach in clinical diagnosis purposes such as the identification of tumors with microsatellite instability.


  • Reply To Kenneth B. Yatai, Mark J. Dunning, Dennis Wang. Consensus Genomic Subtypes of Muscle-invasive Bladder Cancer: A Step in the Right Direction but Still a Long Way To Go. Eur Urol 2020;77:434–5

    by Aurélie Kamoun, Aurélien de Reyniès, Yves Allory, Gottfrid Sjödahl, A. Gordon Robertson, Roland Seiler, Katherine A. Hoadley, Clarice S. Groeneveld, Hikmat Al-Ahmadie, Woonyoung Choi, et al.


  • Top Cited Papers
    Open Access

    A Consensus Molecular Classification of Muscle-invasive Bladder Cancer

    by Aurélie Kamoun, Aurélien de Reyniès, Yves Allory, Gottfrid Sjödahl, A. Gordon Robertson, Roland Seiler, Katherine A. Hoadley, Clarice S. Groeneveld, Hikmat Al-Ahmadie, Woonyoung Choi, et al.


  • CDKN2A homozygous deletion is a strong adverse prognosis factor in diffuse malignant IDH-mutant gliomas

    by Romain Appay, Caroline Dehais, Claude-Alain Maurage, Agusti Alentorn, Catherine Carpentier, Carole Colin, François Ducray, Fabienne Escande, Ahmed Idbaih, Aurélie Kamoun, et al.

    The 2016 World Health Organization (WHO) classification of central nervous system tumors stratifies isocitrate dehydrogenase (IDH)–mutant gliomas into 2 major groups depending on the presence or absence of 1p/19q codeletion. However, the grading system remains unchanged and it is now controversial whether it can be still applied to this updated molecular classification. In a large cohort of 911 high-grade IDH-mutant gliomas from the French national POLA network (including 428 IDH-mutant gliomas without 1p/19q codeletion and 483 anaplastic oligodendrogliomas, IDH-mutant and 1p/19q codeleted), we investigated the prognostic value of the cyclin-dependent kinase inhibitor 2A (CDKN2A) gene homozygous deletion as well as WHO grading criteria (mitoses, microvascular proliferation, and necrosis). In addition, we searched for other retinoblastoma pathway gene alterations (CDK4 amplification and RB1 homozygous deletion) in a subset of patients. CDKN2A homozygous deletion was also searched in an independent series of 40 grade II IDH-mutant gliomas. CDKN2A homozygous deletion was associated with dismal outcome among IDH-mutant gliomas lacking 1p/19q codeletion (P < 0.0001 for progression-free survival and P = 0.004 for overall survival) as well as among anaplastic oligodendrogliomas, IDH-mutant + 1p/19q codeleted (P = 0.002 for progression-free survival and P < 0.0001 for overall survival) in univariate and multivariate analysis including age, extent of surgery, adjuvant treatment, microvascular proliferation, and necrosis. In both groups, the presence of microvascular proliferation and/or necrosis remained of prognostic value only in cases lacking CDKN2A homozygous deletion. CDKN2A homozygous deletion was not recorded in grade II gliomas. Our study pointed out the utmost relevance of CDKN2A homozygous deletion as an adverse prognostic factor in the 2 broad categories of IDH-mutant gliomas stratified on 1p/19q codeletion and suggests that the grading of these tumors should be refined.


  • Open Access

    Cystathionine as a marker for 1p/19q codeleted gliomas by in vivo magnetic resonance spectroscopy

    by Francesca Branzoli, Clément Pontoizeau, Lucien Tchara, Anna Luisa Di Stefano, Aurélie Kamoun, Dinesh K Deelchand, Romain Valabrègue, Stéphane Lehéricy, Marc Sanson, Chris Ottolenghi, et al.

    Codeletion of chromosome arms 1p and 19q (1p/19q codeletion) highly benefits diagnosis and prognosis in gliomas. In this study, we investigated the effect of 1p/19q codeletion on cancer cell metabolism and evaluated possible metabolic targets for tailored therapies. We combined in vivo 1H (proton) magnetic resonance spectroscopy (MRS) measurements in human gliomas with the analysis of a series of standard amino acids by liquid chromatography–mass spectroscopy (LC-MS) in human glioma biopsies. Sixty-five subjects with low-grade glioma were included in the study: 31 underwent the MRI/MRS examination, 47 brain tumor tissue samples were analyzed with LC-MS, and 33 samples were analyzed for gene expression with quantitative PCR. Additionally, we performed metabolic tracer experiments in cell models with 1p deletion. We report the first in vivo detection of cystathionine by MRS in 1p/19q codeleted gliomas. Selective accumulation of cystathionine was observed in codeleted gliomas in vivo, in brain tissue samples, as well as in cells harboring heterozygous deletions for serine- and cystathionine-pathway genes located on 1p: phosphoglycerate dehydrogenase (PHGDH) and cystathionine gamma-lyase (CTH). Quantitative PCR analyses showed 40–50% lower expression of both PHGDH and CTH in 1p/19q codeleted gliomas compared with their non-codeleted counterparts. Our results provide strong evidence of a selective vulnerability of codeleted gliomas to serine and glutathione depletion and point to cystathionine as a possible noninvasive marker of treatment response.


  • Open Access

    Recurrent activating mutations of PPARγ associated with luminal bladder tumors

    by Natacha Rochel, Clémentine Krucker, Laure Coutos-Thévenot, Judit Osz, Ruiyun Zhang, Elodie Guyon, Wayne Zita, Séverin Vanthong, Oscar Alba Hernandez, Maxime Bourguet, et al.

    The upregulation of PPARγ/RXRα transcriptional activity has emerged as a key event in luminal bladder tumors. It renders tumor cell growth PPARγ-dependent and modulates the tumor microenvironment to favor escape from immuno-surveillance. The activation of the pathway has been linked to PPARG gains/amplifications resulting in PPARγ overexpression and to recurrent activating point mutations of RXRα. Here, we report recurrent mutations of PPARγ that also activate the PPARγ/RXRα pathway, conferring PPARγ-dependency and supporting a crucial role of PPARγ in luminal bladder cancer. These mutations are found throughout the protein—including N-terminal, DNA-binding and ligand-binding domains—and most of them enhance protein activity. Structure-function studies of PPARγ variants with mutations in the ligand-binding domain allow identifying structural elements that underpin their gain-of-function. Our study reveals genomic alterations of PPARG that lead to pro-tumorigenic PPARγ/RXRα pathway activation in luminal bladder tumors and may open the way towards alternative options for treatment.


  • Open Access

    The consensus molecular classification of muscle-invasive bladder cancer

    by Aurélie Kamoun, Aurélien de Reyniès, Yves Allory, Gottfrid Sjödahl, A. Gordon Robertson, Roland Seiler, Katherine A. Hoadley, Hikmat Al-Ahmadie, Woonyoung Choi, Clarice S. Groeneveld, et al.

    Muscle-Invasive Bladder Cancer (MIBC) is a molecularly diverse disease with heterogeneous clinical outcomes. Several molecular classifications have been proposed, yielding diverse sets of subtypes, which hampers the clinical implications of such knowledge. Here, we report the results of a large international effort to reach a consensus on MIBC molecular subtypes. Using 1750 MIBC transcriptomes and a network-based analysis of six independent MIBC classification systems, we identified a consensus set of six molecular classes: Luminal Papillary (24%), Luminal Non-Specified (8%), Luminal Unstable (15%), Stroma-rich (15%), Basal/Squamous (35%), and Neuroendocrine-like (3%). These consensus classes differ regarding underlying oncogenic mechanisms, infiltration by immune and stromal cells, and histological and clinical characteristics. This consensus system offers a robust framework that will enable testing and validating predictive biomarkers in future clinical trials.