Combination chemotherapy versus temozolomide for patients with methylated MGMT (m-MGMT) glioblastoma: results of computational biological modeling to predict the magnitude of treatment benefit.

BACKGROUND: A randomized trial in glioblastoma patients with methylated-MGMT (m-MGMT) found an improvement in median survival of 16.7 months for combination therapy with temozolomide (TMZ) and lomustine, however the approach remains controversial and relatively under-utilized. Therefore, we sought to determine whether comprehensive genomic analysis can predict which patients would derive large, intermediate, or negligible benefits from the combination compared to single agent chemotherapy. METHODS: Comprehensive genomic information from 274 newly diagnosed patients with methylated-MGMT glioblastoma (GBM) was downloaded from TCGA. Mutation and copy number changes were input into a computational biologic model to create an avatar of disease behavior and the malignant phenotypes representing hallmark behavior of cancers. In silico responses to TMZ, lomustine, and combination treatment were biosimulated. Efficacy scores representing the effect of treatment for each treatment strategy were generated and compared to each other to ascertain the differential benefit in drug response. RESULTS: Differential benefits for each drug were identified, including strong, modest-intermediate, negligible, and deleterious (harmful) effects for subgroups of patients. Similarly, the benefits of combination therapy ranged from synergy, little or negligible benefit, and deleterious effects compared to single agent approaches. CONCLUSIONS: The benefit of combination chemotherapy is predicted to vary widely in the population. Biosimulation appears to be a useful tool to address the disease heterogeneity, drug response, and the relevance of particular clinical trials observations to individual patients. Biosimulation has potential to spare some patients the experience of over-treatment while identifying patients uniquely situated to benefit from combination treatment. Validation of this new artificial intelligence tool is needed.

Computational modeling of early T-cell precursor acute lymphoblastic leukemia (ETP-ALL) to identify personalized therapy using genomics.

Early T-cell precursor acute lymphoblastic leukemia (ETP-ALL) is an aggressive hematological malignancy for which optimal therapeutic approaches are poorly characterized. Using computational biology modeling (CBM) in conjunction with genomic data from cell lines and individual patients, we generated disease-specific protein network maps that were used to identify unique characteristics associated with the mutational profiles of ETP-ALL compared to non-ETP-ALL (T-ALL) cases and simulated cellular responses to a digital library of FDA-approved and investigational agents. Genomics-based classification of ETP-ALL patients using CBM had a prediction sensitivity and specificity of 93% and 87%, respectively. This analysis identified key genomic and pathway characteristics that are distinct in ETP-ALL including deletion of nucleophosmin-1 (NPM1), mutations of which are used to direct therapeutic decisions in acute myeloid leukemia. Computational simulations based on mutational profiles of 62 ETP-ALL patient models identified 87 unique targeted combination therapies in 56 of the 62 patients despite actionable mutations being present in only 37% of ETP-ALL patients. Shortlisted two-drug combinations were predicted to be synergistic in 11 profiles and were validated by in vitro chemosensitivity assays. In conclusion, computational modeling was able to identify unique biomarkers and pathways for ETP-ALL, and identify new drug combinations for potential clinical testing.

In silico-based high-throughput screen for discovery of novel combinations for tuberculosis treatment.

There are currently 18 drug classes for the treatment of tuberculosis, including those in the development pipeline. An in silico simulation enabled combing the innumerably large search space to derive multidrug combinations. Through the use of ordinary differential equations (ODE), we constructed an in silico kinetic platform in which the major metabolic pathways in Mycobacterium tuberculosis and the mechanisms of the antituberculosis drugs were integrated into a virtual proteome. The optimized model was used to evaluate 816 triplets from the set of 18 drugs. The experimentally derived cumulative fractional inhibitory concentration (∑FIC) value was within twofold of the model prediction. Bacterial enumeration revealed that a significant number of combinations that were synergistic for growth inhibition were also synergistic for bactericidal effect. The in silico-based screen provided new starting points for testing in a mouse model of tuberculosis, in which two novel triplets and five novel quartets were significantly superior to the reference drug triplet of isoniazid, rifampin, and ethambutol (HRE) or the quartet of HRE plus pyrazinamide (HREZ).