Glasgow Prognostic Score and Gustave Roussy Immune Score in Hodgkin Lymphoma: Survival Associations and Limited Incremental Prognostic Value Beyond the International Prognostic Score

Background/Objectives: Although outcomes in Hodgkin lymphoma (HL) have improved substantially, patients with advanced-stage disease, comorbidities, or relapsed/refractory presentations can still fare poorly. Blood-based indices of systemic inflammation and nutrition are derived from routine tests, but their value beyond established prognostic models is uncertain. We examined the association of the baseline Gustave Roussy Immune Score (GRIm) and Glasgow Prognostic Score (GPS) with treatment response, progression-free survival (PFS), and overall survival (OS) in HL, focusing on their performance relative to the seven-factor International Prognostic Score (IPS-7). Methods: We retrospectively analysed 110 adults with histologically confirmed HL treated at a tertiary haematology centre between January 2015 and December 2025. GPS, GRIm, and IPS-7 were calculated from data recorded at diagnosis. Treatment response was classified as complete versus non-complete. Outcomes were assessed with Kaplan–Meier analysis, log-rank tests, Cox regression, Harrell’s C-index, and likelihood-ratio testing. Results: Most patients had advanced-stage disease (69.1%) and received ABVD-based treatment (94.5%); complete response was achieved in 90 (81.8%). GPS and GRIm were not significantly associated with non-complete response, whereas IPS-7 was. Over a median follow-up of 39.5 months, 28 patients (25.5%) progressed or died and 17 (15.5%) died. In univariable Cox analysis, high GRIm risk (HR = 2.68, 95% CI 1.17–6.14), higher GPS (HR = 2.18 per point, 95% CI 1.23–3.89), and higher IPS-7 (HR = 2.10 per point, 95% CI 1.59–2.77) predicted shorter PFS. For OS, GPS and IPS-7 were significant, whereas GRIm was not. After adjustment for IPS-7, neither GPS nor GRIm remained independently associated with PFS or OS, and adding either score to IPS-7 produced only small, non-significant gains in discrimination. Conclusions: Baseline GPS and GRIm were associated with survival on univariable analysis, particularly for PFS, but their incremental value beyond IPS-7 was limited. These scores may help describe baseline inflammatory and nutritional risk and should not be regarded as alternatives to established HL prognostic models. In particular, GPS and GRIm were not significantly associated with treatment response and should be viewed as supportive markers requiring external validation, rather than as tools that can independently guide treatment decisions.