LAUSR

Tumor resistance to radiation therapy remains a significant obstacle to long-term cancer patient survival, especially for head and neck squamous cell carcinoma (HNSCC), a cancer type with poor long-term outcomes. NADPH metabolism is a viable target for personalized radiation-sensitizing chemotherapies due to the reliance of tumors on NADPH for DNA repair, antioxidation, and cell survival following ionizing radiation. We have previously shown that genome-scale metabolic models that incorporate transcriptomic, kinetic, thermodynamic, and metabolite concentration data can make accurate predictions of NADPH production and sensitivity to redox-cycling chemotherapeutics such as β-lapachone in radiation-sensitive and -resistant HNSCC cell lines. Through the incorporation of transcriptomic data from the Cancer Genome Atlas, we have extended our modeling approach to develop personalized flux balance analysis-based metabolic models of HNSCC patient tumors. This collection of 132 models predicts that radiation-resistant tumors display greater conversion of NADP+ to NADPH than radiation-sensitive tumors, consistent with our prior work in matched HNSCC cell lines. Differences in NADPH production are manifested by the re-routing of metabolic flux through alternate NADPH-producing pathways in the human metabolic network. In addition, simulated gene knockdown of isocitrate dehydrogenase (IDH) isoforms has a more pronounced effect on NADPH production in radiation-resistant tumors, supporting prior evidence of IDH as a promising chemotherapeutic target for multiple cancer types. Finally, we discovered a subset of radiation-sensitive patients that displayed significantly lower NADPH-producing flux throughout the metabolic network as well as greater sensitivity to knockdown of NADPH production genes. Almost all of these TCGA patients with lower fluxes had tumors of the larynx as opposed to other head and neck cancer regions, indicating that NADPH metabolism in HNSCC tumors may depend greatly on the tumor’s anatomical location. This modeling approach can be extended to investigate the synergistic action of redox-based chemotherapeutics and radiation therapy to treat radiation-resistant HNSCC patients.