Background: Constitutive activation of the NRF2 stress response pathway is a common occurrence in head and neck squamous cell carcinomas (HNSCC), promoting metabolic reprogramming, immune evasion, and resistance to radiation therapy. Through integrative omics, we identified pyrimethamine (PYR), an FDA-approved antiprotozoal agent, as a potent degrader of NRF2 protein via inhibition of dihydrofolate reductase (DHFR). Preclinical studies in NRF2-active mouse models demonstrated that PYR has biologic efficacy and a safe toxicity profile. Methods: This single-arm phase 2 window-of-opportunity trial administered PYR (50mg PO daily x 14 days) before surgery in patients with locoregionally advanced HNSCC. The primary efficacy hypothesis was that PYR would result in a 50% increase in tumor DHFR protein, a surrogate of DHFR inhibition and on-target biologic efficacy. The primary safety endpoint was proportion of patients who experienced ≥ grade 3 adverse events (AE) due to PYR or treatment-related delay in surgery by ≥ 10 days. Secondary endpoints included safety, tolerability, and NRF2 pathway suppression among tumors classified as NRF2-active in pre-treatment samples. In parallel, preclinical studies with mouse syngeneic oral cancer (MOC1) cell lines and tumor grafts were performed to assess DHFR inhibition as a therapeutic strategy to suppress one carbon metabolism and sensitize tumors to radiation therapy. Results: 22 patients were enrolled, 20 of whom completed the trial with clinical stage III (n=6, 30%) and stage IV (n=14, 70%) disease. Two patients withdrew after enrollment, one prior to starting treatment and one after the first dose. Tumor DHFR increased by 50% or more in all 20 post-treatment samples compared to baseline (mean increase: 1,282%, range: 335-5263%, adjusted p-value: 0.00032), demonstrating potent on-target activity. Treatment-related ≥ grade 3 AEs and surgery delays ≥ 10 days did not occur. Two patients demonstrated high baseline NRF2 pathway activity but expression changes among NRF2 target genes were heterogeneous. Preclinical studies demonstrated DHFR inhibition was associated with NRF2 suppression across multiple HNSCC cell lines. Treatment of MOC1 tumor-bearing C57Bl/6 mice demonstrated that DHFR inhibition enhanced sensitivity to radiation therapy (n=10 per cohort, p=0.01). Conclusions: Neoadjuvant PYR achieved robust DHFR inhibition in human HNSCC tumors with an acceptable safety profile. Parallel preclinical data showed DHFR inhibition-driven NRF2 suppression and radiosensitization. This first-in-class trial and preclinical studies are strong rationale for a phase 2 trial integrating PYR with definitive radiation-based therapy for NRF2-active HNSCCs.