Author: ISAAC Amoah, Jackie Austin, Charlotte Block, Kaylee Brilz, Dylan Bui, Andrew E. Ekpenyong, Jayce Hughes, Pralhad Itani, Natasha Ratnapradipa, Sara Strom, Jacob Woolf 👨🔬
Affiliation: Creighton University 🌍
Purpose:
Glioblastoma (GBM) is the most common and aggressive primary brain tumor in adults, with a median survival of approximately 15 months despite the current standard of care, which includes surgery, radiotherapy, and chemotherapy with temozolomide (TMZ). The inherent resistance of GBM to conventional treatments calls for innovative therapeutic approaches. This study investigates the use of the immunotherapeutic drug, Lenalidomide, in combination with radiotherapy to improve therapeutic outcomes. Additionally, MATLAB-based unsupervised machine learning is employed to cluster and analyze cellular and organoid imaging data, providing a deeper understanding of treatment effects at the cellular level.
Methods:
Glioblastoma cell lines (T98G and U87-MG) are treated with Lenalidomide and subjected to clinically relevant radiation doses using a cell irradiator (Faxitron CellRad). Cell migration is monitored in real time using an Electric Cell Impedance Sensor (ECIS), and survival analysis is conducted using cloud-based clonogenic assays (Axion Omni). Images of treated cells and organoids, obtained from Axion Omni assays, are analyzed using unsupervised machine learning algorithms implemented in MATLAB to cluster morphological changes. This approach facilitates the identification of patterns and correlations between imaging data and treatment efficacy.
Results:
Our results demonstrate that Lenalidomide, in combination with radiotherapy, significantly enhances the antitumor effects on GBM cells by reducing their migratory capacity and improving radiosensitivity. Machine learning analysis of imaging data reveals distinct clustering patterns corresponding to treatment-induced morphological changes.
Conclusion:
This study highlights the potential of Lenalidomide-based radioimmunotherapy as a promising strategy to overcome GBM’s resistance to conventional treatments. By integrating advanced imaging analysis and unsupervised machine learning, we offer a robust method for evaluating therapeutic responses and optimizing treatment protocols. Ongoing efforts aim to analyze more data and expand this approach to other cancer types in preparation for personalized treatment strategies.