Splenic FDG PET uptake and CT volume as prognostic biomarkers in diffuse large B cell lymphoma

To evaluate whether baseline splenic ^18F-FDG PET uptake and CT-derived spleen volume predict early progression (≤ 36 months) and progression-free survival (PFS) in diffuse large B cell lymphoma (DLBCL), and whether they interact.

Retrospective cohort of adults with newly diagnosed DLBCL undergoing baseline ^18F-FDG PET/CT and CT. Splenic PET positivity was defined visually (uptake exceeding hepatic background); spleen volume was measured semi-automatically. Early progression/relapse was analysed as a 36-month milestone endpoint and PFS as time-to-event. Multivariable logistic and Cox models included IPI, sex, treatment, splenic PET, spleen volume, and a PET × volume interaction. A prespecified PFS sensitivity analysis re-included event-free patients censored before 36 months.

124 patients were included (68 men; median age 68 years [IQR 57-77]); 45/124 (36.3%) were PET-positive; median spleen volume was 3.13 dL (IQR 1.92-5.79). IPI predicted ≤ 36-month progression (OR 1.52; p = 0.021) and PFS (HR 1.44; p = 0.004). Splenic PET positivity predicted ≤ 36-month progression (OR 2.83; p = 0.048) but not PFS (p = 0.103). The PET × volume interaction was significant (OR 0.70; p = 0.021; HR 0.81; p = 0.033). In PET-negative patients, larger spleen volume predicted higher risk (OR 1.42; p = 0.027; HR 1.22; p = 0.042), whereas volume was neutral in PET-positive cases. In the PFS sensitivity analysis (N = 145), the interaction attenuated (HR 0.83; p = 0.057) and the PET-negative volume effect was borderline (HR 1.21; p = 0.053).

Splenic PET positivity is a dominant risk marker in DLBCL in this cohort. CT-derived spleen volume may add conditional prognostic information in PET-negative patients, but the incremental value of the PET × volume interaction is modest and requires external validation.