AUTHORS
Max Farber Kelsten, MD, Nicole Altomare, MD2, Omar Elghawy, MD3, Colin J. Thomas, MD4, Ari R. Pelcovits, MD5, Joanna M Rhodes, MD, MSCE6, Adam Kidwell, MD7, Matthew J. Cortese, MD, MPH8, James A Davis, PharmD, BCOP9, Deborah Stephens, DO10, Tamara K Moyo, MD,PhD11, Laura Quillen, MD12, Lindsey Fitzgerald, MD13, Jonathon B. Cohen, MD, MS14, Megan Melody, MD, MS15, Brian Hess, MD16, Rahul Kumar Thakur, MD17, Megan M Herr, PhD18, Thomas Ollila, MD19, Daniel Antonio, MPH20, Adit Dharia, MD21, Matthew Matasar, MD22, Hiffsa Taj, MD, MBBS23, Natalie Jo Hawes, BSN24, Yasmin H Karimi, MD25, Jonathan Moreira, MD26, Shuo Ma, MD27, Jane N Winter, MD28, Nirav N. Shah, MD, MS29, Leo Gordon, MD30, Narendranath Epperla, MD, MS, FACP31, Natalie S Grover, MD32, Reem Karmali, MD33
AFFILIATIONS
1 Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
2 Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
3 Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, United States
4 Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, United States
5 Brown University Health Cancer Institute, Brown University, Providence, Rhode Island, United States
6 Rutgers Cancer Institute, New Brunswick, New Jersey, United States
7 MCW Cancer Center, Medical College of Wisconsin, Milwaukee, Wisconsin, United States
8 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
9 Hollings Cancer Center, Medical University of South Carolina, Charleston, South Carolina, United States
10 UNC Chapel Hill, Chapel Hill, North Carolina, United States
11 Levine Cancer Institute, Atrium Health, Charlotte, North Carolina, United States
12 Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, United States
13 Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, United States
14 Winship Cancer Institute, Emory University, Atlanta, Georgia, United States
15 Tampa General Hospital Cancer Institute, Tampa, Florida, United States
16 Hollings Cancer Center, Medical University of South Carolina, Charleston, South Carolina, United States
17 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
18 Roswell Park Comprehensive Cancer Center, Buffalo, New York, United States
19 Brown University Health Cancer Institute, Brown University, Providence, Rhode Island, United States
20 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois, United States
21 HCA Florida Oak Hill Hospital/ USF Morsani College of Medicine, Tampa, Florida, United States
22 Rutgers Cancer Institute, New Brunswick, New Jersey, United States
23 University of Kentucky Markey Cancer Center, Lexington, Kentucky, United States
24 University of Kentucky Markey Cancer Center, Lexington, Kentucky, United States
25 University of Michigan Rogel Cancer Center, Ann Arbor, Michigan, United States
26 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois, United States
27 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois, United States
28 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois, United States
29 MCW Cancer Center, Medical College of Wisconsin, Milwaukee, Wisconsin, United States
30 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois, United States
31 Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, United States
32 UNC Chapel Hill, Chapel Hill, North Carolina, United States
33 Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois, United States
ABSTRACT
Background:
CART and BsAb are effective treatments for patients (pts) with R/R LBCL. CART access and manufacturing latency may encourage earlier use of BsAb, with unknown impact on subsequent CART efficacy. In this analysis, we compared outcomes in R/R LBCL pts who received CART with or without prior BsAb.
Methods:
We identified pts with R/R LBCL treated with a CD19-directed CART at 14 US centers from 2015-2026. Pts were classified as having received anti-CD20 BsAb before CART (BBC) or CART alone (CART Only). BsAb bridging to CART was excluded. Baseline characteristics were compared with the Pearson Chi-squared test and included demographics, pt characteristics, disease features, and treatment characteristics.
Median duration of response (mDOR), duration of complete response (mDOCR), progression free survival (mPFS), and overall survival (mOS) were estimated using Kaplan-Meier method and assessed from CART infusion in months (mo) to event of interest or date of last follow-up. A sensitivity analysis was performed excluding pts that did not reach an event of interest and had <12 mo of follow-up.
Univariable (UVA) and multivariable (MVA) analyses were performed for CART PFS and OS. Variables included in UVA were prior BsAb, age >60, ECOG performance status (PS), double-hit lymphoma (DHL), primary refractory disease within 6 (PRD6) or 12 (PRD12) months of frontline treatment initiation, CART line of therapy (LOT), elevated LDH, extranodal disease, stage at CART, and CART product. Variables with p<0.10 in UVA were included in MVA, with prior BsAb included a priori. All but age >60, PRD12, and elevated LDH were included in MVA for PFS and OS. Age >60 was included in MVA for OS only. As BsAb use before CART was strongly associated with later CART LOT, mediation analysis via a Cox outcome and linear mediator model was performed to quantify how CART LOT mediated the impact of treatment sequence on OS. A p<0.05 was significant in all analyses.
Results:
We identified 947 CART pts: 904 CART Only and 43 BBC. De novo DLBCL was the most common histology in both groups (65%). BBC had a greater proportion of pts age >60 (78 vs 60%, p=0.04). As BsAb therapy, BBC pts received glofitamab (47%), epcoritamab (31%), mosunetuzumab (19%), and other (3%), and 36 pts received BsAb combination therapy. There was more use of axi-cel (52 vs 30%) in CART Only and liso-cel (57 vs 35%) in BBC (p=0.01). BBC pts received CART later (median LOT 4th vs 3rd, p<0.001). There were no other between-group differences in baseline characteristics, including PS, cell of origin (COO), LDH or bulky disease at CART, DHL, PRD6, or PRD12.
There were no differences in response rates, mDOR, mDOCR, or mPFS between groups. CART Only had a superior mOS (36.9 vs 9.2 mo, p=0.004). In regression analysis, prior BsAb was significant in UVA for OS (HR 1.99, 95% CI 1.24-3.21). CART LOT was significant in MVA for PFS (HR 1.14, 95% CI 1.03-1.26) and OS (HR 1.28, 95% CI 1.14-1.44). Mediation analysis showed a significant total effect of prior BsAb on OS (HR 2.18, 95% CI 1.11-4.29, p=0.02) mediated by CART LOT (indirect HR 1.35, 95% CI 1.14-1.60, p<0.001; proportion mediated 71%, 95% CI 3%-139%, p=0.04). The direct effect was not significant.
In the sensitivity analysis, CART Only included 769 pts and BBC included 23 pts. BBC had significantly more non-germinal center COO (80 vs 53%, p=0.02) and DHL (30 vs 14%, p=0.03). The groups were otherwise similar to the complete cohort. There was no difference in response rates, mDOR, or mDOCR between groups. CART Only had a superior mPFS (8.3 vs 3.1 mo, p=0.02) and mOS (32.0 vs 5.9 mo, p<0.001). In regression analysis, prior BsAb was significant in UVA for both PFS (HR 1.75, 95% CI 1.08-2.84) and OS (HR 2.68, 95% CI 1.67-4.31). CART LOT was significant in MVA for both PFS (HR 1.11, 95% CI 1.01-1.24) and OS (HR 1.23, 95% CI 1.10-1.38). Mediation analysis revealed the same pattern as the complete cohort, though with a non-significant proportion mediated.
Conclusions:
Pts receiving BBC are more likely to be age >60 and receive liso-cel than CART Only pts. Receiving CART after BsAb was associated with inferior mOS. This effect was mediated by LOT, revealing reduced efficacy of CART in later lines. Prospective trials that examine sequencing and are accompanied by correlative studies of T-cell subsets and functional profiles could both clarify this relationship and offer mechanistic understanding to inform therapeutic decision making.
DISCLOSURES
Max Farber Kelsten, MD: No relevant conflicts of interest to declare.