Personalized Simulations Can Predict Outcomes for Treatment in Patients With Diffuse Large B-Cell Lymphoma

Investigators at Brighton and Sussex Medical School (BSMS) have developed a new approach to predicting the effectiveness of treatments for patients with diffuse large B-cell lymphoma (DLBCL), according to research published in Blood Cancer Journal.¹

Image credit: David A Litman | stock.adobe.com

Norris et al utilized genomic sequencing data to manufacture personalized simulations of individual patients that can determine the impact a patient’s genetic mutations will have on the behavior of cancer cells. This development could revolutionize clinical decision-making and lead to further advancements in care for heterogeneous blood cancers.¹

“This study supports the integration of genetic sequencing at the diagnosis stage of DLBCL to better determine patient prognosis,” Simon Mitchell, reader in Cancer Systems Biology at BSMS, said in a news release. “As sequencing costs decrease, we hope that this approach will become a standard diagnostic practice, enabling precise identification of patients who might benefit from alternative treatments.”²

Typically, the most aggressive hematological malignancies contain genetic aberrations that affect multiple genes. The investigators noted the need for new approaches that can prospectively identify poor prognosis patients, with the goal to create more effective treatment strategies.¹

Computational models of molecular signaling in normal B cells have previously been used to predict the survival and proliferation of cells. Yet it is not known if this mutational data alone can enable in silico simulations to make predictions of prognosis in blood cancers.¹

In the study, the investigators used mechanistic computational models to simulate how mutations combine in B cell malignancies, developing a pipeline to create patient simulations and test if these models could generate clinically meaningful prognostic information.¹

The method was able to successfully identify patients with dismal, intermediate, and good prognoses across multiple datasets. For example, the investigators found that in multiple myeloma, when mutations impacted the cell cycle and apoptosis at the same time, they combined deleteriously in simulations and conferred poor prognosis.¹

This achievement was completed using data from whole-exome sequencing (WES) or targeted sequencing panels, which provided robust predictions despite the mutational heterogeneity.²

Unlike other statistical approaches, the predictive accuracy of the newly-created simulations were improved as larger validation datasets were applied, which emphasized the importance of integrating molecular network knowledge into the analysis.²

Interestingly, the personalized computational simulations were independent of the International Prognostic Index and stage and genetic clustering, and could also be combined with these factors to improve prognostic power.¹

The models were particularly able to identify patients with co-occurring mutations that were simultaneously pro-proliferative and anti-apoptotic. This observation could not be determined from mutational clustering or the appearance of specific mutations.¹

“This study marks a significant step forward in the quest for personalized cancer treatment,” Mitchell said. “By harnessing the power of computational modeling to place genomic data into context, we hope to pave the way for more accurate prognostic predictions and tailored therapeutic strategies.”²

Despite the success of their approach, the investigators cautioned that more work must be done if modeling is to become widely used for personalized medicine approaches. They further noted that performing large-sale computational simulations can be “computationally challenging,” and requires substantial resources.¹

The techniques demonstrated in the study could also apply to other types of cancer beyond DLBCL, specifically those that are characterized by genetic heterogeneity.²

“This study represents significant progress toward stratified medicine, enabling more targeted treatments that could lead to substantial improvements in the treatment of patients diagnosed with blood cancer,” said Simon Ridley, director of research and advocacy at Leukemia UK, in a press release.²

REFERENCES

1. Norris R, Jones J, Mancini E, et al. Patient-specific computational models predict prognosis in B cell lymphoma by quantifying pro-proliferative and anti-apoptotic signatures from genetic sequencing data. Blood Cancer J. 2024;14(105). doi:10.1038/s41408-024-01090-y

2. EurekAlert! Personalized simulations predict patient outcomes for blood cancer treatment in breakthrough study. News Release. Released July 15, 2024. Accessed July 24, 2024. https://www.eurekalert.org/news-releases/1051388