Back

Early Measurable Residual Disease Detection after CAR-T is Associated with Poor Outcomes in Patients with Large B-cell Lymphoma

Krasnow, N.; Maurer, K.; Song, C.; Rhoades, J.; Xiong, K.; Crnjac, A.; Blewett, T.; Gao, L.; Jacene, H.; Merryman, R. W.; Gohil, S. H.; Duffy, C.; Guerrero, L. I.; Dela Cruz, J.; McDonough, M.; Wolff, J. O.; Redd, R.; Mattie, M.; Miles, B.; Makrigiorgos, G. M.; Neuberg, D. S.; Rodig, S. J.; Armand, P.; Jacobson, C. A.; Adalsteinsson, V. A.; Wu, C. J.

2024-11-28 oncology
10.1101/2024.11.27.24318095 medRxiv
Show abstract

Despite responses of chimeric antigen receptor (CAR)-T cells in relapsed/refractory (R/R) large B cell lymphoma (LBCL) patients, over half of patients eventually relapse. Methods to detect early disease persistence are needed to identify patients at high-risk of treatment failure. We recently developed MAESTRO, an ultrasensitive, tumor-informed measurable residual disease (MRD) assay, which can detect parts-per-million (ppm) levels of circulating tumor DNA (ctDNA) using minimal sequencing. We applied MAESTRO to 140 samples from 28 patients (15 durable responders at 12 months, 13 nonresponders) to identify treatment failure following axicabtagene ciloleucel (axi-cel) administered at our institution between 2018 and 2022. Responder and nonresponder patients had similar baseline tumor burden. By 1 week after infusion, responders had marked ctDNA reduction compared to nonresponders, p<0.001. At weeks 2 and 4, responders had ctDNA levels approaching 0 ppm, while nonresponders had persistence of ctDNA, each p<0.001. At day 0, 21% of patients had ctDNA fractions below 0.01%, hence these individuals would not have qualified for ctDNA monitoring with a less sensitive test. Our results confirm feasibility of highly sensitive MRD detection by ctDNA for early identification of patients at high risk of disease progression from axi-cel.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.