LAUSR

We examined the meaning of metabolically active lesions on 1 month restaging nuclear imaging of patients with relapsed/refractory (R/R) large B-cell lymphoma (LBCL) receiving axicabtagene ciloleucel (axi-cel) by assessing the relationship between total metabolic tumor volume (MTV) on positron emission tomography (PET) scans and circulating tumor DNA (ctDNA) in the plasma. In this prospective multicenter sample collection study, MTV was retrospectively calculated via commercial software at baseline, 1 and 3 months post chimeric antigen receptor (CAR) T-cell therapy; ctDNA was available pre and post axi-cel. Spearman correlation coefficient (rs) was used to study the relationship between the variables and a mathematical model was constructed to describe tumor dynamics 1 month post CAR T-cell therapy. The median time between baseline scan and axi-cel infusion was 33 (range, 1-137) days for all 57 patients. For 41 of the patients with imaging within 33 days of axi-cel or imaging before that time but no bridging therapy, the correlation at baseline became stronger (rs 0.61, P< 0.0001) compared to all patients (rs 0.38, P= 0.004). Excluding patients in complete remission with no measurable residual disease, ctDNA and MTV at 1 month did not correlate (rs 0.28, P= 0.11), but did correlate at 3 months (rs 0.79, P= 0.0007). Modeling of tumor dynamics, which incorporated ctDNA and inflammation as part of MTV, recapitulated outcomes of patients with positive radiologic 1-month scans. Our results suggested that non-progressing hypermetabolic lesions on 1 month PET represent ongoing treatment response and their composition may be elucidated by concurrent ctDNA.