Predicting Everolimus Response in Rare Metastatic Thymic Epithelial Tumors
Thymic tumors, while rare, are classified as orphan diseases due to their infrequency. Despite their rarity, they are the predominant tumors found in the anterior mediastinal compartment among adults.
Thymic tumors represent a wide variety of tumors and the most recent histologic classification classifies them as thymomas, thymic carcinomas (TC), and neuroendocrine thymic tumors (NETT). Although thymic tumors have traditionally been considered orphan diseases, in the last few decades there has been an increased interest in thymic malignancies in the scientific community.
Optimal treatment in the relapsed-refractory setting for patients with advanced thymic tumors needs to be better understood. Cyclophosphamide/doxorubicin/cisplatin (CAP) and carboplatin/paclitaxel are options for thymic tumors. In the relapsed-refractory thymic tumor, several options for treatment exist, but the optimal sequence of therapy or choice of agent is not clear. The rarity of the disease has resulted in limiting the research to a few small single-arm Phase II studies, as larger randomized trials are more challenging to conduct.
The Cellworks Platform and the Computational Biology Model (CBM) can identify patterns and key pathways in an individual tumor and across tumors. By integrating a large volume of signaling, metabolic, and transcriptional networks from available published data in collaboration with tumor-specific genomic information, the Cellworks Platform can create an individual patient’s personalized tumor-specific network model in silico. The Cellworks Platform then biosimulates the impact of a single drug or a combination of drugs on the patient’s personalized disease model, based on the known mechanism of action of the drug(s). Here, we tested the effect of everolimus on thymic carcinoma patients and used our technology to predict response to therapy.
The Cellworks Platform identified key responsible pathways for the response and non-response of everolimus, as explained by network analysis.
Key Identified Pathway in Responder Patient
Loss of NF1 leads to HRAS-mediated activation of the AKT-MTOR pathway. R140Q mutation of IDH2 identified in this patient leads to 2 Hydroxy-ketoglutarate mediated activation of HIF1A. MTOR inhibitors nullify the activation of HIF1A via the MTOR pathway and are the key reason for response. T670I missense mutation of KIT, R348fs* mutation of PIK3R1, and CNV deletion of TET2 leads to activation of the AKT1-MTOR pathway are additional reasons for the response. The Cellworks Platform predicted this patient as a responder for Everolimus. Identified PFS post everolimus treatment of this patient was 22.96 Months
Key Identified Pathway in Non-Responder Patient
The primary cause of non-responsiveness in this patient, as identified by CBM, was the CNV deletion of MTOR. However, additional factors contributing to the downregulation of the AKT-MTOR pathway were also found, including CNV deletions of MET, IGF1R, MAGI1, and MAGI2. A high copy number of the STK11 gene was also identified which leads to an increase in PRKAA1. Increased activation of PRKAA1 leads to inactivation of the mTOR pathway. The Cellworks Platform predicted this patient as a non-responder for Everolimus. Identified PFS post everolimus treatment of this patient was 2.56 Months.
In Conclusion
Personalized Therapy Biosimulation, made possible using the Cellworks Platform and CBM integrates multiple genomic abnormalities in patients to identify disease-specific biomarkers and network characteristics and identify novel drug combinations with therapeutic potential. This approach also identified personalized drug regimens in all profiles, including those without actionable genetic mutations present. This predictive technology can improve the utility of tumor profiling and offer novel clinical treatments for patients who would otherwise have no options. This study is an example of how Cellworks personalized therapy biosimulation provides insights for rare diseases like thymoma, where the small number of patients makes it difficult to draw definitive conclusions using conventional methods.