DoctorateOpen Access

Towards Leveraging Inhibition State of the Kinome for Precision Oncology

Is this your thesis?

This record came from a bulk archive import. If it’s yours, link it to your profile.

2023
0 views
0 downloads

Abstract (EN)

Protein phosphorylation forms the most common method of regulation in eukaryotes, and kinases are enzymes that chiefly enable its application. Due to their central role in physiology, dysregulation of the kinome is implicated in a myriad of diseases, particularly cancer. This dissertation demonstrates that the measured inhibition of the kinome (the kinome inhibition state) by cancer targeted therapies can be predictive of cell line and patient-derived xenograft (PDX) tumor responses to treatment by that therapy using interpretable machine learning models. The predictive capability of kinome inhibition states with currently used baseline genomics for monotherapy cancer cell line responses across diverse cancer types is demonstrated first using multi-dose kinome inhibition states, and second using multi-assay single-dose data. Then, the predictive value of kinome inhibition states is extended to kinase inhibitor combination therapies, demonstrating that combined kinome inhibition states can accurately predict cancer cell line sensitivity and synergy to combination treatments, providing the basis for rational kinome-informed drug combination selection. Finally, the predictive capacity of kinome inhibition states is demonstrated for PDX tumor responses in five common solid tumor types, confirming the generalizability of kinome inhibition-based prediction models in a preclinical setting, and emphasizing their potential for clinical translation and application in precision oncology. Overall, this dissertation provides compelling evidence that integrating kinome inhibition states in machine learning models can enhance the prediction of cancer cell line and PDX tumor responses. This work shows that kinome inhibition data has potential to be included in precision oncology platforms alongside baseline genomic profiling, aiding in the identification of effective therapeutic strategies and ultimately improving patient outcomes.

Author

Chinmaya U. Joisa

How to Cite

Chinmaya U. Joisa (Doctorate thesis). Towards Leveraging Inhibition State of the Kinome for Precision Oncology, 2023, The University of North Carolina at Chapel Hill, Biomedical Engineering.

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from The University of North Carolina at Chapel Hill