To evaluate whether age modifies the association between the geriatric nutritional risk index (GNRI) and overall survival (OS) in patients aged ≥ 18 years with newly diagnosed diffuse large B-cell lymphoma (DLBCL), we conducted a multi-centre retrospective study of 552 patients. Multivariable Cox regression with restricted cubic spline (RCS) modelling showed that GNRI was significantly associated with OS, but the relationship was non-linear (P for non-linearity = 0.0158). A three-dimensional Cox-RCS model revealed that older age was associated with a higher mortality risk. The association between lower GNRI and worse OS was consistent across all age groups, including younger patients. Among patients with low GNRI (< 92), the most common pattern of malnutrition involved normal albumin (Alb) and body mass index (BMI). This pattern accounted for approximately 40% of low-GNRI patients, regardless of age group (based on thresholds of 65, 70, or 75 years). Survival curves demonstrated that a low GNRI (< 92) predicted poor prognosis, regardless of whether Alb or BMI was the main contributing factor. There was no evidence of interaction between age and GNRI on prognosis. Malnutrition, as measured by GNRI, adversely affects survival in both younger and older DLBCL patients.
Keywords: Diffuse large B-cell lymphoma; Effect modification; Geriatric nutritional risk index; Older adult; Younger adult.
© 2025. The Author(s).