Modification of Lugano criteria by pre-infusion tumor kinetics improves early survival prediction for patients with lymphoma under chimeric antigen receptor T-cell therapy.

ABSTRACT

BACKGROUND: Chimeric antigen receptor T-cell therapy (CART) is effective for patients with refractory or relapsed lymphoma with prolongation of survival. We aimed to improve the prediction of Lugano criteria for overall survival (OS) at 30-day follow-up (FU1) by including the pre-infusion tumor growth rate (TGRpre-BL) and its early change to 30-day FU1 imaging (TGRpost-BL). METHODS: Consecutive patients with pre-baseline (pre-BL), baseline (BL) and FU1 imaging with CT or positron emission tomography/CT before CART were included. TGR was defined as change of Lugano criteria-based tumor burden between pre-BL, BL and FU1 examinations in relation to days between imaging examinations. Overall response and progression-free survival were determined based on Lugano criteria. Proportional Cox regression analysis studied association of TGR with OS. For survival analysis, OS was analyzed using Kaplan-Meier survival curves. RESULTS: Fifty-nine out of 81 patients met the inclusion criteria. At 30-day FU1 8 patients (13.6%) had a complete response (CR), 25 patients (42.4%) a partial response (PR), 15 patients (25.4%) a stable disease (SD), and 11 patients (18.6%) a progressive disease (PD) according to CT-based Lugano criteria. The median TGRpre-BL was -0.6 mm2/day, 24.4 mm2/day, -5.1 mm2/day, and 18.6 mm2/day and the median TGRpost-BL was -16.7 mm2/day, -102.0 mm2/day, -19.8 mm2/day and 8.5 mm2/day in CR, PR, SD, and PD patients, respectively. PD patients could be subclassified into a cohort with an increase in TGR (7 of 11 patients (64%), PD TGRpre-to-post-BL INCR) and a cohort with a decrease in TGR (4 of 11 patients (36%), PD TGRpre-to-post-BL DECR) from pre-BL to post-BL. PD TGRpre-to-post-BL DECR patients exhibited similar OS to patients classified as SD, while PD TGRpre-to-post-BL INCR patients had significantly shorter OS (65 days vs 471 days, p<0.001). CONCLUSION: In the context of CART, the additional use of TGRpre-BL and its change to TGRpost-BL determined at 30-day FU1 showed better OS prognostication for patients with overall PD according to Lugano criteria. Therefore, this modification of the Lugano classification should be explored as a potential novel imaging biomarker of early response and should be validated prospectively in future studies.