Effect of erlotinib and celecoxib as short-course neoadjuvant therapy in operable oral cavity squamous cell carcinoma: a randomized controlled window-of-opportunity trial

Background Delays in definitive surgery for operable oral cavity squamous cell carcinoma (OCSCC) may allow tumor progression. This randomized phase II trial evaluated a short-course neoadjuvant combination of erlotinib and celecoxib during the preoperative waiting period. Methods Sixty-three patients with resectable OCSCC were randomized to celecoxib, erlotinib, combination therapy, or observation for 21 days before surgery. The primary endpoint was biomarker modulation; secondary endpoints were tumor response, safety, and feasibility. Biomarker modulation was analyzed using two-way ANCOVA of post-treatment levels adjusted for baseline expression. Clinical and radiological changes in the longest tumor dimension (LTD) were assessed. Results Erlotinib exposure was independently associated with significant reductions in CD31 (p < 0.001), CD34 (p = 0.006), and cytoplasmic AKT (p = 0.006). Celecoxib exposure was also associated with reduced CD31 expression (p = 0.003). No significant effects were observed on EGFR, COX-2, CD44, HIF1α, or PI3K. Erlotinib-containing regimens reduced or stabilized tumor size, whereas the observation arm showed tumor progression during the wait time for surgery. Partial clinical responses occurred in 43.7% with erlotinib and 60% with combination therapy. Toxicities were predominantly grade 1–2, mainly acneiform rash and transient hepatic dysfunction, and did not delay surgery. Combination therapy showed a trend toward reduced angiogenesis. Conclusions Short-course neoadjuvant erlotinib, particularly with celecoxib, is feasible, safe, and biologically active in operable OCSCC, supporting further evaluation to prevent progression during unavoidable surgical delays.