Lymphoma tumor burden before chimeric antigen receptor T-Cell treatment: RECIL vs. Lugano vs. metabolic tumor...

High tumor burden has emerged as a negative predictor of efficacy in chimeric antigen receptor T-cell therapy (CART) in patients with refractory or relapsed large B-cell lymphoma. This study analyzed the deviation among imaging-based tumor burden (TB) metrics and their association with progression-free (PFS) and overall survival (OS).

In this single-center observational study, we included all consecutively treated patients receiving CD19 CART with available baseline PET-CT imaging. Imaging-based TB was determined based on response evaluation criteria in lymphoma (RECIL), the Lugano criteria, and metabolic tumor volume. Total, nodal and extranodal TB were represented, according to the respective criteria, by sum of longest diameters (TB_RECIL), sum of product of perpendicular diameters (TB_Lugano), and metabolic tumor volume (TB_MTV). Correlation statistics were used for comparison. Proportional Cox regression analysis studied the association of TB metrics with PFS and OS.

34 consecutive patients were included (median age: 67 years, 41% female) with total median baseline TB_RECIL of 12.5 cm, TB_Lugano of 4,030 mm² and TB_MTV of 330 mL. The correlation of TB_RECIL and TB_Lugano with TB_MTV was strong (ρ=0.744, p<0.001 and ρ=0.741, p<0.001), with lowest correlation for extranodal TB_RECIL with TB_MTV (ρ=0.660, p<0.001). Stratification of PFS was strongest by total TB_MTV>50% (HR=2.915, p=0.042), whereas total TB_RECIL>50% and total TB_Lugano>50% were not significant (both p>0.05). None of the total TB metrics were associated with OS (all p>0.05).

Pre-CART TB metrics vary significantly based on the assessment method, impacting their association with survival outcomes. The correlation between TB_RECIL, TB_Lugano and TB_MTV was influenced by disease phenotype and prior bridging therapy. TB method of assessment must be considered when interpreting the impact of TB on outcomes in clinical trials. Considering the heterogeneity, our results argue for standardization and harmonization across centers.