The HARP (Hemoglobin–Albumin–C-Reactive Protein) Index: A Biologically Designed Composite Biomarker for Continuous Prognostic Modeling in Locally Advanced Nasopharyngeal Carcinoma

Background/Objectives: To evaluate the prognostic significance of the novel hemoglobin–albumin–C-reactive protein (HARP) index in patients with locally advanced nasopharyngeal carcinoma (LANPC) treated with definitive concurrent chemoradiotherapy (CCRT). Methods: This retrospective study included 248 patients with LANPC treated with definitive CCRT between 2011 and 2020. The HARP index [hemoglobin × (albumin ÷ C-reactive protein)] was calculated using pretreatment laboratory values. Prognostic associations between HARP and survival outcomes were evaluated using continuous and categorical Cox regression analyses, restricted cubic spline modeling, receiver operating characteristic analyses, and bootstrap-based internal validation. Results: Lower pretreatment HARP values were independently associated with inferior progression-free survival (PFS) and overall survival (OS). In continuous Cox analyses, increasing HARP values were associated with reduced risks of mortality and disease progression (both p < 0.001). ROC analysis identified 3.2 as the exploratory cut-off value for clinical stratification. Patients with HARP ≥ 3.2 (n = 112) had significantly better outcomes than those with HARP < 3.2 (n = 136). Median PFS and OS were not reached in the high-HARP group, whereas they were 47.0 and 72.0 months, respectively, in the low-HARP group. Low HARP values were associated with inferior PFS (HR, 3.86; p < 0.001) and OS (HR, 3.06; p < 0.001). Bootstrap internal validation demonstrated stable model discrimination with minimal optimism. Conclusions: The novel HARP index is an independently associated and internally validated prognostic biomarker in patients with LANPC treated with definitive CCRT. HARP may provide a practical tool for improved risk stratification, pending prospective external validation.