Abstract 623MO
Background
Diffuse Large B-Cells Lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma in adults (30-40%). In the 2016 World Health Organization (WHO), DLBCL are classified into 3 molecular subtypes according to cell of origin (COO), germinal-center, activated B-cell like and unclassified, based on gene-expression profiling (GEP). In daily routine, COO classification is replaced by immunochemistry (IHC) stains using the Hans algorithm based on the expression of CD10, BCL6 and MUM1 proteins. In addition, co-expression of BCL2 and MYC proteins is of prognostic value and defines the double-protein expressors (DPE) subtypes, associated with worse prognosis. Fluorescent In Situ Hybridization (FISH) is mandatory in the workup of DLBCL to detect MYC and BCL2 and/or BCL6 rearrangements.
Methods
565 whole-slide images (WSI) stained with hematoxylin/eosin from the LYSA trial “GHEDI” (Deciphering the Genetic Heterogeneity of Diffuse large B-cell lymphoma in the rituximab era) dataset were analyzed. A Deep Learning (DL) model was trained to predict COO and DPE status, the presence of MYC rearrangements (no MYC rearrangement/MYC-Single Hit or HGBL-Double Hit/Triple Hit) and expression of BCL6, CD10 and MUM1 proteins from WSI. Performance was evaluated using several repetitions of stratified five-fold cross-validation.
Results
The DL model achieved a ROC AUC of 0.624 for GC, 0.687 for DPE, and 0.675 for MYC rearrangement. Using Cox proportional hazard model, predictions of DPE status (HR=0.38, P=.016), and MYC rearrangements (HR=5.23, P<.001) and MUM1 expression (HR=2.80, P=.027) were associated with worse overall survival.
Conclusions
Our study demonstrates the predictive power of DL applied to WSI to predict DLBCL subtypes. Such predictive models could be used to augment pathologists analysis capacities, especially when IHC staining or FISH are not available.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Institut Carnot CALYM.
Funding
Owkin.
Disclosure
J. Schiratti: Financial Interests, Institutional, Full or part-time Employment: Owkin. E. Brion: Financial Interests, Institutional, Full or part-time Employment: Owkin. C. Maussion: Financial Interests, Institutional, Full or part-time Employment: Owkin. L. Danneaux: Financial Interests, Institutional, Full or part-time Employment: Owkin. All other authors have declared no conflicts of interest.