Patient-specific computational models predict prognosis in B cell lymphoma by quantifying pro-proliferative and anti-apoptotic signatures from genetic sequencing data.

Genetic heterogeneity and co-occurring driver mutations impact clinical outcomes in blood cancers. Grouping tumours into clusters based on genetic alterations is prognostically informative. However, predicting the emergent effect of co-occurring mutations that impact multiple complex and interacting signalling networks remains challenging. Here, we used mathematical models to predict the impact of co-occurring mutations on cellular signalling and cell fates in diffuse large B cell lymphoma (DLBCL) and multiple myeloma (MM). Simulations predicted adverse impact on clinical prognosis when combinations of mutations induced both pro-proliferative and anti-apoptotic signalling. So, we established a pipeline to integrate patient-specific mutational profiles into personalised lymphoma models. Using this approach, we identified a subgroup (19%) of patients characterised by simultaneous upregulation of anti-apoptotic and pro-proliferative (AAPP) signalling. AAPP patients have dismal prognosis and can be identified within all current genomic and cell-of-origin classifications. Combining personalised molecular simulations with mutational clustering enabled stratification of patients into clinically informative prognostic categories: good (80% progression-free survival at 120 months), intermediate (median progression-free survival of 93 months), and poor (AAPP, median progression-free survival of 26 months). This study shows that personalised computational models enable identification of novel high-risk patient subgroups, providing a valuable tool for future risk-stratified clinical trials.