PROJECT SUMMARY/ABSTRACT Head and neck squamous cell carcinoma (HNSCC) remains a leading cause of cancer deaths worldwide. Genotoxic agents, including radiation therapy (RT) and cisplatin (CDDP), are treatments that damage cellular DNA. RT and CDDP are the current standard of care in multiple solid tumors, including HNSCC. CDDP is the most commonly used chemotherapeutic agent in HNSCC proving superior to novel targeted agents in recent large randomized trials. Despite this, high rates of treatment failure persist in patients who develop resistance following this toxic chemotherapy. Treatment failure is uniformly fatal. However, no robust predictors of acquired cisplatin resistance or tumor response exist. Given this critical unmet need, we have focused our efforts on the assessment of tumor response using minimally invasive quantitative imaging (hyperpolarized magnetic resonance imaging; HP-MRI) while patients are undergoing therapy. We showed that CDDP and other genotoxic agents trigger measurable fluctuations in tumor cell metabolism detectable through HP-MRI with [1-13C]-pyruvate in real time (confirmed by conventional biochemical assays). Genotoxic stress suppresses the apparent rate of pyruvate conversion into lactate (kPL) via lactate dehydrogenase (LDH) in a manner that correlates with anti- tumor effectiveness. We therefore hypothesize that changes in kPL provide unique insight into metabolic changes induced by cisplatin that can be used to optimize response to therapy in HNSCC. In Aim 1, we will characterize baseline HP-MRI parameters such as kPL across the spectrum of HNSCC subtypes and validate the relationship between CDDP and associated shifts in carbon flux. We will also identify metabolomic differences in HNSCC models that affect baseline values of metabolic imaging biomarkers and modulate apparent changes induced by cisplatin. In Aim 2, we will integrate the dose-response data from Aim 1 to develop a predictive model of response to CDDP based on metabolic imaging parameters. We will use a simple algorithm to adjust therapeutic dose based on HP-MRI data in animal models of HNSCC to maximize tumor growth delay, and test whether thresholds suggestive of strong response can be used to select the more effective treatment regimen when multiple regimens are tested in parallel. In Aim 3, we will conduct a first-in- human evaluation of changes in HP-MRI to detect shifts in carbon flux following CDDP in HNSCC patients. We will correlate changes in metabolic imaging parameters with the baseline metabolic phenotype of tumors as determined from metabolomic analysis and direct measurements of tumor LDH. Successful completion of this study will establish HP-MRI as a non-invasive imaging modality able to predict response to treatment, which will be a noteworthy first step towards a precision oncology approach that we have been seeking for nearly half a century. Thus, the proposed research is relevant to the part of the NIH’s mission that pertain...