Composite Nutritional–Inflammatory Indices and Overall Survival Across the Esophageal Cancer Treatment Continuum

Background: Malnutrition and systemic inflammation are near-universal in esophageal cancer and are captured by composite blood-based indices—the prognostic nutritional index (PNI), neutrophil-to-lymphocyte ratio (NLR), geriatric nutritional risk index (GNRI)/nutritional risk index (NRI), and the newer global immune-nutrition-inflammation index (GINI). Published syntheses have each addressed a single index, usually within a single treatment setting, leaving unclear whether prognostic performance is preserved as treatment has diversified from surgery alone to chemoradiotherapy- and immunotherapy-based regimens. Objectives: To compare, across the surgical, chemo(radio)therapy, and immunotherapy-based treatment settings, the association between an adverse composite nutritional-inflammatory index category and overall survival in esophageal cancer. Methods: We conducted a rapid evidence synthesis with exploratory random-effects meta-analysis. Because a fully systematic, protocol registered, dual-reviewer search was not undertaken, this work is deliberately reported as a rapid synthesis rather than as a PRISMA compliant systematic review. Eligible reports were cohort studies of patients with esophageal cancer providing a categorical (adverse vs favourable) hazard ratio (HR) with 95% confidence interval (CI) for overall survival according to pretreatment PNI, NLR, GNRI/NRI, or GINI. All estimates were direction-standardised to express the hazard associated with the adverse category. One estimate per independent cohort was pooled using a random-effects (DerSimonian–Laird) model, pre-specified by treatment setting; duplicate and institutionally overlapping cohorts were excluded from the primary analysis and examined in sensitivity analyses. Certainty was rated using GRADE. Results: Fifteen study-index observations derived from 14 unique cohorts (3,904 patients) were eligible for narrative synthesis; eight independent cohorts (2,450 patients) contributed to the primary meta analysis. An adverse index category was associated with worse overall survival (pooled HR 2.39, 95% CI 2.02–2.82; I²=0%), with consistent point estimates in the surgery-alone (HR 2.23, 95% CI 1.81–2.76), chemo(radio)therapy (HR 2.36, 95% CI 1.69–3.28), and immunotherapy based subgroups (HR 3.47, 95% CI 2.16–5.58). The estimate was stable on leave-one-out analysis (HR 2.27 2.45) and across three pre-specified sensitivity analyses (HR 2.39–2.45). However, the surgery-alone 95% prediction interval crossed unity (0.89–5.59), and the apparent absence of heterogeneity (I²=0%) is more plausibly an artifact of a small, restrictively selected estimate set than evidence of true homogeneity. Certainty of evidence was low. Conclusions: Adverse composite nutritional-inflammatory index status is associated with approximately a doubling of mortality hazard at every point on the esophageal cancer treatment continuum, including under immunotherapy. Given a non-exhaustive search, an exclusively retrospective and predominantly East Asian evidence base, study-specific cut-offs, and low GRADE certainty, these findings are hypothesis-generating and support prospective, protocol-driven validation rather than immediate clinical adoption.