Real-world CAR-T outcomes in third-line LBCL
| Real-world study design . | Constructs . | Patients . | Percent of patients ineligible for CAR-T clinical trials . | Response rates . | Survival outcomes . | Median duration of response . | Rates of CRS . | Rates of neurotoxicity . | Treatment-related mortality . | Conclusions . |
|---|---|---|---|---|---|---|---|---|---|---|
| Locke et al.28 Postapproval safety observational study | Axi-cel | 1343 | 38% aged ≥65, 4% with ECOG PS 2+, 13% cardiac comorbidities, 2% hepatic comorbidities, 2% renal comorbidities, 15% double/ triple-hit lymphoma, 66% refractory disease | ORR 74% (CR 56%) ORR 78% (CR 62%) for patients aged ≥65 ORR 57% (CR 29%) for hepatic comorbidities ORR 70% (CR 43%) for renal comorbidities ORR 47% (CR 20%) for ECOG 2-3 | 18 mo: PFS 42%, OS 52% | 18 mo: DOR 61% | 83% | 55% | Not assessed | Patients aged ≥65 with moderate to severe pulmonary disease and ECOG 2-3 had inferior ORR. Age ≥65 was not associated with inferior survival, but was associated with higher rates of CRS and ICANS. ECOG PS significantly affected all efficacy outcomes. |
| Jacobson et al.25 Postauthorization safety study through the CIBMTR registry | Axi-cel | 1297 | 57% were ineligible for ZUMA-1 | ORR 73% (CR 56%) | mPFS 8.6 mo mOS 21.8 mo | mDOR NR 24-mo DOR 57% | 83% (8% grade 3+) | 55% (24% grade 3+) | 3%, 2% died from CRS, 1% died from ICANS | Real-world response rates were similar to ZUMA-1, with similar DOR in patients who were and were not eligible for ZUMA-1. High-grade CRS and ICANS were lower in the real-world cohort. ECOG PS of 2 or greater was associated with inferior response rates, PFS, and OS. Patients age ≥65 had a higher ORR despite higher risk of CRS and ICANS as compared to younger patients. |
| Landsburg et al.29 Observational study through the CIBMTR registry | Tisa-cel | 1159 | 31% ineligible for JULIET | ORR 60% | 2 y: PFS 28%, OS 44% | 2 y: DOR 53% | 58% (6% grade 3+) | 23% (7% grade 3+) | Not assessed | Real-world outcomes were similar to JULIET. Patients with comorbidities (who were not eligible for JULIET) had similar efficacy outcomes. Patients with ECOG PS 2-4 had higher rates of high-grade CRS and ICANS but similar efficacy outcomes. |
| Bachy et al.30 Retrospective French DESCAR-T registry study with propensity score matching between axi-cel and tisa-cel | Axi-cel Tisa-cel | 452 axi-cel 277 tisa-cel | Not assessed | Axi-cel: ORR 80% (CR 60%) Tisa-cel: ORR 66% (CR 42%) | 1 y: PFS 47% axi-cel, 33% tisa-cel OS 64% axi-cel, 49% tisa-cel | 1 y: DOR 54% axi-cel, 42% tisa-cel | Axi-cel: 86% (5% grade 3+) Tisa-cel: 76% (9% grade 3+) | Axi-cel: 49% (14% grade 3+) Tisa-cel: 22% (3% grade 3+) | Axi-cel: no grade 5 CRS, 1 patient grade 5 ICANS Tisa-cel: 2 grade 5 CRS, no grade 5 ICANS No other treatment-related grade 5 AEs | Real-world response and survival rates were similar to clinical trials. Real-world ORR, CR, PFS, and OS better with axi-cel compared to tisa-cel, although axi-cel with more toxicity. |
| Chihara et al.31 Retrospective cohort from Medicare claims data | All CAR-T | 551 | 31% of patients in this cohort were age ≥75 | Not assessed | Age ≥75: mPFS 160 d mOS 403 d Age 70-74: mPFS 379 d mOS 603 d Age 65-69: mPFS 194 mOS 518 1 y PFS: Age ≥75 34% Age 70-74 52% Age 65-69 43% | Not assessed | Not assessed | Not assessed | Not assessed | Older patients, particularly those age ≥75, had significantly worse PFS and OS compared to younger patients as well as compared to clinical trials. |
| Nastoupil et al.32 Retrospective cohort from the US Lymphoma CAR-T Consortium | Axi-cel | 298 | 43% were ineligible for ZUMA-1 due to comorbidities | ORR 82% (CR 64%) | 1 y: PFS 47% OS 68% PFS 34% for ZUMA-1 ineligible patients | mDOR NR (median follow-up of 12.9 mo) | 91% (7% grade 3+) | 69% (31% grade 3+ | 4.4%, 1 death due to HLH, 1 death due to ICANS | Real-world outcomes and safety were comparable to ZUMA-1, although ECOG PS 2-4 had worse PFS, OS, and toxicities. Patients ineligible for ZUMA-1 had inferior PFS and OS. |
| Sano et al.33 Retrospective cohort from the US Lymphoma CAR-T Consortium | Axi-cel | 272 30% were aged ≥65 | Not assessed | Age ≥65 ORR 84% (CR 71%) Age ≤65 ORR 82% (CR 51%) | Age ≥65 mPFS 9.2 mo, mOS not assessable Age ≤65 mPFS 7.4 mo, mOS 18.7 mo | Not assessed | Age ≥65 92% (7% grade 3+) Age ≤65 91% (7% grade 3+) | Age ≥65 78% (35% grade 3+) Age ≤65 65% (31% grade 3+) | 2 deaths (1 in age ≥65 and 1 age <65) | Rate of complete response was higher in patients aged ≥65 compared to age <65. There was no difference in PFS, OS, and toxicities between patients aged <65 and ≥65. |
| Bethge et al.34 Retrospective cohort from the German Registry for Stem Cell Transplantation | Axi-cel Tisa-cel | 173 axi-cel 183 tisa-cel | 13% eligible for ZUMA-1 89% eligible for JULIET | Axi-cel: ORR 74% (CR 42%) Tisa-cel: ORR 53% (CR 32%) | Axi-cel: 1 y PFS 35%, OS 55% Tisa-cel: 1 y PFS 24%, OS 53% | Not assessed | Axi-cel: 81% (10% grade 3+) Tisa-cel: 65% (13% grade 3+) | Axi-cel: 44% (16% grade 3+) Tisa-cel: 22% (7% grade 3+) | Axi-cel: 2 y 10.4% Tisa-cel: 2 y 3.5% | Real-world ORR, CR, and OS were comparable to clinical trials, but real-world PFS was worse. Higher rates of delayed infection-related NRM in real-world cohort. |
| Riedell et al.35 Retrospective cohort of 8 US centers | Axi-cel Tisa-cel | 168 axi-cel 92 tisa-cel | 61% axi-cel ineligible for ZUMA-1 43% tisa-cel ineligible for JULIET | Axi-cel: ORR 52% (CR 44%) Tisa-cel: ORR 41% (CR 35%) | Axi-cel: 1 y PFS 42%, OS 62% Tisa-cel: 1 y PFS 32%, OS 59% | Axi-cel: 1 y DOR 70% Tisa-cel: 1 y DOR 75% | Axi-cel: 85% (9% grade 3+) Tisa-cel: 39% (1% grade 3+) mDOR NR for either cohort | Axi-cel: 56% (38% grade 3+) Tisa-cel: 11% (1% grade 3+) | Axi-cel: 9%, 4 deaths from ICANS Tisa-cel: 7% | Real-world outcomes were slightly lower compared to ZUMA-1 and JULIET, although safety outcomes were comparable. Axi-cel and tisa-cel had comparable efficacy, although less CRS and ICANS with tisa-cel. |
| Pasquini et al.36 Postauthorization safety study through the CIBMTR registry | Tisa-cel | 155 | Not assessed | ORR 62% (CR 40%) | 1 y: EFS 52% OS 77% | 1 y DOR 61% | 45% (4.5% grade 3+) | 18% (5.1% grade 3+) | 1.2% | Similar efficacy and improved safety compared to JULIET. |
| Kittai et al.37 Retrospective cohort from 4 centers | Axi-cel Tisa-cel | 94 axi-cel 36 tisa-cel | 57% of patients had high comorbidity based on CIRS score | ORR 68% (CR 42%) | mPFS 6.7 mo mOS NR 1 y OS 60% | Not assessed | 79% | 58% | 12.3%, 2 deaths due to CRS, 1 death due to ICANS | Worse ECOG PS and comorbidities by CIRS score were associated with worse survival. Presence of more comorbidities was not associated with worse CRS or ICANS. |
| Cook et al.38 Meta-analysis of prospective and retrospective studies of CAR-T in primary and secondary CNS lymphoma | Multiple CAR-T products | 30 patients with primary CNS lymphoma (PCNSL) 98 patients with secondary CNS lymphoma (SCNSL) | Not assessed | PCNSL: ORR 64% (CR 56%) SCNSL: ORR 57% (CR 47%) | PCNSL: 6-mo PFS 37% SCNSL: 6-mo PFS 37% | PCNSL: mDOR 9 mo SCNSL: mDOR 4.6 mo | PCNSL: 70% (13% grade 3+) SCNSL: 72% (11% grade 3+) | PCNSL: 53% (18% grade 3+) SCNSL: 48% (26% grade 3+) | Not assessed | Incidence of CRS and ICANS in real-world cohort of PCNSL and SCNSL is comparable to clinical trials. Approximately one-third of patients with PCNSL and SCNSL had 6-mo complete responses, which could be durable with longer follow-up. |
| Lin et al.50 Single-center retrospective cohort of older adults who did and did not receive CAR-T | Axi-cel and tisa-cel | 24 older adults received CAR-T 18 older adults did not receive CAR-T 25 younger adults (age <65) received CAR-T | Not assessed | 100-d CR 51% (for 49 patients who received CAR-T) | 6-mo PFS 48%, OS 71% (for 49 patients who received CAR-T) Older adults who received CAR-T had a significantly lower risk of death (HR 0.31) compared to older adults who did not receive CAR-T No difference in PFS or OS between older or younger adults who received CAR-T by chronological age, functional limitations, cognitive impairment, or comorbidity burden | Not assessed | 83% (8% grade 2±) | 54% (25% grade 2±) | 2 deaths in older adults after CAR-T | Older adults who receive CAR-T have better postrelapse OS compared to patients who instead receive chemotherapy and/or supportive care. PFS, OS, and rates of CRS and ICANS are not different between older and younger adults who receive CAR-T. |
| Real-world study design . | Constructs . | Patients . | Percent of patients ineligible for CAR-T clinical trials . | Response rates . | Survival outcomes . | Median duration of response . | Rates of CRS . | Rates of neurotoxicity . | Treatment-related mortality . | Conclusions . |
|---|---|---|---|---|---|---|---|---|---|---|
| Locke et al.28 Postapproval safety observational study | Axi-cel | 1343 | 38% aged ≥65, 4% with ECOG PS 2+, 13% cardiac comorbidities, 2% hepatic comorbidities, 2% renal comorbidities, 15% double/ triple-hit lymphoma, 66% refractory disease | ORR 74% (CR 56%) ORR 78% (CR 62%) for patients aged ≥65 ORR 57% (CR 29%) for hepatic comorbidities ORR 70% (CR 43%) for renal comorbidities ORR 47% (CR 20%) for ECOG 2-3 | 18 mo: PFS 42%, OS 52% | 18 mo: DOR 61% | 83% | 55% | Not assessed | Patients aged ≥65 with moderate to severe pulmonary disease and ECOG 2-3 had inferior ORR. Age ≥65 was not associated with inferior survival, but was associated with higher rates of CRS and ICANS. ECOG PS significantly affected all efficacy outcomes. |
| Jacobson et al.25 Postauthorization safety study through the CIBMTR registry | Axi-cel | 1297 | 57% were ineligible for ZUMA-1 | ORR 73% (CR 56%) | mPFS 8.6 mo mOS 21.8 mo | mDOR NR 24-mo DOR 57% | 83% (8% grade 3+) | 55% (24% grade 3+) | 3%, 2% died from CRS, 1% died from ICANS | Real-world response rates were similar to ZUMA-1, with similar DOR in patients who were and were not eligible for ZUMA-1. High-grade CRS and ICANS were lower in the real-world cohort. ECOG PS of 2 or greater was associated with inferior response rates, PFS, and OS. Patients age ≥65 had a higher ORR despite higher risk of CRS and ICANS as compared to younger patients. |
| Landsburg et al.29 Observational study through the CIBMTR registry | Tisa-cel | 1159 | 31% ineligible for JULIET | ORR 60% | 2 y: PFS 28%, OS 44% | 2 y: DOR 53% | 58% (6% grade 3+) | 23% (7% grade 3+) | Not assessed | Real-world outcomes were similar to JULIET. Patients with comorbidities (who were not eligible for JULIET) had similar efficacy outcomes. Patients with ECOG PS 2-4 had higher rates of high-grade CRS and ICANS but similar efficacy outcomes. |
| Bachy et al.30 Retrospective French DESCAR-T registry study with propensity score matching between axi-cel and tisa-cel | Axi-cel Tisa-cel | 452 axi-cel 277 tisa-cel | Not assessed | Axi-cel: ORR 80% (CR 60%) Tisa-cel: ORR 66% (CR 42%) | 1 y: PFS 47% axi-cel, 33% tisa-cel OS 64% axi-cel, 49% tisa-cel | 1 y: DOR 54% axi-cel, 42% tisa-cel | Axi-cel: 86% (5% grade 3+) Tisa-cel: 76% (9% grade 3+) | Axi-cel: 49% (14% grade 3+) Tisa-cel: 22% (3% grade 3+) | Axi-cel: no grade 5 CRS, 1 patient grade 5 ICANS Tisa-cel: 2 grade 5 CRS, no grade 5 ICANS No other treatment-related grade 5 AEs | Real-world response and survival rates were similar to clinical trials. Real-world ORR, CR, PFS, and OS better with axi-cel compared to tisa-cel, although axi-cel with more toxicity. |
| Chihara et al.31 Retrospective cohort from Medicare claims data | All CAR-T | 551 | 31% of patients in this cohort were age ≥75 | Not assessed | Age ≥75: mPFS 160 d mOS 403 d Age 70-74: mPFS 379 d mOS 603 d Age 65-69: mPFS 194 mOS 518 1 y PFS: Age ≥75 34% Age 70-74 52% Age 65-69 43% | Not assessed | Not assessed | Not assessed | Not assessed | Older patients, particularly those age ≥75, had significantly worse PFS and OS compared to younger patients as well as compared to clinical trials. |
| Nastoupil et al.32 Retrospective cohort from the US Lymphoma CAR-T Consortium | Axi-cel | 298 | 43% were ineligible for ZUMA-1 due to comorbidities | ORR 82% (CR 64%) | 1 y: PFS 47% OS 68% PFS 34% for ZUMA-1 ineligible patients | mDOR NR (median follow-up of 12.9 mo) | 91% (7% grade 3+) | 69% (31% grade 3+ | 4.4%, 1 death due to HLH, 1 death due to ICANS | Real-world outcomes and safety were comparable to ZUMA-1, although ECOG PS 2-4 had worse PFS, OS, and toxicities. Patients ineligible for ZUMA-1 had inferior PFS and OS. |
| Sano et al.33 Retrospective cohort from the US Lymphoma CAR-T Consortium | Axi-cel | 272 30% were aged ≥65 | Not assessed | Age ≥65 ORR 84% (CR 71%) Age ≤65 ORR 82% (CR 51%) | Age ≥65 mPFS 9.2 mo, mOS not assessable Age ≤65 mPFS 7.4 mo, mOS 18.7 mo | Not assessed | Age ≥65 92% (7% grade 3+) Age ≤65 91% (7% grade 3+) | Age ≥65 78% (35% grade 3+) Age ≤65 65% (31% grade 3+) | 2 deaths (1 in age ≥65 and 1 age <65) | Rate of complete response was higher in patients aged ≥65 compared to age <65. There was no difference in PFS, OS, and toxicities between patients aged <65 and ≥65. |
| Bethge et al.34 Retrospective cohort from the German Registry for Stem Cell Transplantation | Axi-cel Tisa-cel | 173 axi-cel 183 tisa-cel | 13% eligible for ZUMA-1 89% eligible for JULIET | Axi-cel: ORR 74% (CR 42%) Tisa-cel: ORR 53% (CR 32%) | Axi-cel: 1 y PFS 35%, OS 55% Tisa-cel: 1 y PFS 24%, OS 53% | Not assessed | Axi-cel: 81% (10% grade 3+) Tisa-cel: 65% (13% grade 3+) | Axi-cel: 44% (16% grade 3+) Tisa-cel: 22% (7% grade 3+) | Axi-cel: 2 y 10.4% Tisa-cel: 2 y 3.5% | Real-world ORR, CR, and OS were comparable to clinical trials, but real-world PFS was worse. Higher rates of delayed infection-related NRM in real-world cohort. |
| Riedell et al.35 Retrospective cohort of 8 US centers | Axi-cel Tisa-cel | 168 axi-cel 92 tisa-cel | 61% axi-cel ineligible for ZUMA-1 43% tisa-cel ineligible for JULIET | Axi-cel: ORR 52% (CR 44%) Tisa-cel: ORR 41% (CR 35%) | Axi-cel: 1 y PFS 42%, OS 62% Tisa-cel: 1 y PFS 32%, OS 59% | Axi-cel: 1 y DOR 70% Tisa-cel: 1 y DOR 75% | Axi-cel: 85% (9% grade 3+) Tisa-cel: 39% (1% grade 3+) mDOR NR for either cohort | Axi-cel: 56% (38% grade 3+) Tisa-cel: 11% (1% grade 3+) | Axi-cel: 9%, 4 deaths from ICANS Tisa-cel: 7% | Real-world outcomes were slightly lower compared to ZUMA-1 and JULIET, although safety outcomes were comparable. Axi-cel and tisa-cel had comparable efficacy, although less CRS and ICANS with tisa-cel. |
| Pasquini et al.36 Postauthorization safety study through the CIBMTR registry | Tisa-cel | 155 | Not assessed | ORR 62% (CR 40%) | 1 y: EFS 52% OS 77% | 1 y DOR 61% | 45% (4.5% grade 3+) | 18% (5.1% grade 3+) | 1.2% | Similar efficacy and improved safety compared to JULIET. |
| Kittai et al.37 Retrospective cohort from 4 centers | Axi-cel Tisa-cel | 94 axi-cel 36 tisa-cel | 57% of patients had high comorbidity based on CIRS score | ORR 68% (CR 42%) | mPFS 6.7 mo mOS NR 1 y OS 60% | Not assessed | 79% | 58% | 12.3%, 2 deaths due to CRS, 1 death due to ICANS | Worse ECOG PS and comorbidities by CIRS score were associated with worse survival. Presence of more comorbidities was not associated with worse CRS or ICANS. |
| Cook et al.38 Meta-analysis of prospective and retrospective studies of CAR-T in primary and secondary CNS lymphoma | Multiple CAR-T products | 30 patients with primary CNS lymphoma (PCNSL) 98 patients with secondary CNS lymphoma (SCNSL) | Not assessed | PCNSL: ORR 64% (CR 56%) SCNSL: ORR 57% (CR 47%) | PCNSL: 6-mo PFS 37% SCNSL: 6-mo PFS 37% | PCNSL: mDOR 9 mo SCNSL: mDOR 4.6 mo | PCNSL: 70% (13% grade 3+) SCNSL: 72% (11% grade 3+) | PCNSL: 53% (18% grade 3+) SCNSL: 48% (26% grade 3+) | Not assessed | Incidence of CRS and ICANS in real-world cohort of PCNSL and SCNSL is comparable to clinical trials. Approximately one-third of patients with PCNSL and SCNSL had 6-mo complete responses, which could be durable with longer follow-up. |
| Lin et al.50 Single-center retrospective cohort of older adults who did and did not receive CAR-T | Axi-cel and tisa-cel | 24 older adults received CAR-T 18 older adults did not receive CAR-T 25 younger adults (age <65) received CAR-T | Not assessed | 100-d CR 51% (for 49 patients who received CAR-T) | 6-mo PFS 48%, OS 71% (for 49 patients who received CAR-T) Older adults who received CAR-T had a significantly lower risk of death (HR 0.31) compared to older adults who did not receive CAR-T No difference in PFS or OS between older or younger adults who received CAR-T by chronological age, functional limitations, cognitive impairment, or comorbidity burden | Not assessed | 83% (8% grade 2±) | 54% (25% grade 2±) | 2 deaths in older adults after CAR-T | Older adults who receive CAR-T have better postrelapse OS compared to patients who instead receive chemotherapy and/or supportive care. PFS, OS, and rates of CRS and ICANS are not different between older and younger adults who receive CAR-T. |
AE, adverse event; CIBMTR, Center for International Blood and Marrow Transplant Research; ECOG PS, Eastern Cooperative Oncology Group Performance Status; NRM, nonrelapse mortality.