A Real-World Comparison of CAR T-Cell Therapy Versus a Historical Standard-of-Care Approach for Relapsed-Refractory Large B-Cell Lymphoma in Ontario, Canada
Bibliographic record
Abstract
Background Anti-CD19 Chimeric Antigen Receptor T-cell (CAR-T) immunotherapies have been funded in Canada for patients with relapsed-refractory large B-cell lymphoma (RR LBCL) after two lines of systemic therapy since December 2019. However, real-world evidence of CAR-T outcomes has been limited to small series with little comparison to prior standard-of-care management. We compared overall survival (OS), adverse events, and healthcare utilization for a cohort of patients with RR LBCL consecutively treated with CAR-T versus a cohort of historical controls treated with standard-of-care therapy prior to CAR-T approval. Methods This is a propensity-weighted retrospective cohort study of patients with RR LBCL treated at Princess Margaret (PM) Cancer Centre. Using linked clinical and administrative datasets, consecutive patients treated with CAR-T (2020-2022) following provincial funding approval were compared to a matched cohort of historical controls (2012-2017). Patients were followed from index date, defined as the date of progression following 2 lines of chemotherapy (2L) in the historical controls and following last therapy (2L or higher) in the CAR-T patients for up to 3-years, with maximum follow-up to March 31, 2023. Stabilized inverse probability treatment weighting (sIPTW) was used to account for confounding between cohorts (age, sex, lactate dehydrogenase, and comorbidities). Kaplan meier curves and IPTW-weighted Cox proportional hazard regression analyses estimated the adjusted hazard ratio (HR) between treatment cohort and OS. A landmark survival analysis of patients alive at 3 months post-index date addressed immortal time bias. Adverse events (AEs) from inpatient/emergency department [ED] diagnoses and healthcare utilization were reported per 1000 person-days at risk. Results A total of 86 CAR-T patients and 150 historical control patients were evaluable for comparison. Variables were balanced after applying the sIPTW (based on standardized difference <0.1); mean age was 56 years and males comprised 61%. Prior treatment included ASCT in 27.6% CAR-T vs 38.4% historical control patients (standardized difference 0.21; p=0.09). Post-2L progression, 58% of historical controls had no further treatment, 32% received intravenous or oral chemotherapy/targeted agents, 6.2% had an autologous stem cell transplant, and 10% received palliative chemotherapy and/or radiotherapy. CAR-T patients received tisagenlecleucel (33.3%) and axicabtagene ciloleucel (66.7%) CAR-T products. The OS probabilities at 1-, 2-, and 3-years were 68% (95% CI 53-79%), 60% (95% CI 44-73%), and 57% (95% CI 39-71%), respectively, in the CAR-T group, and 18% (95% CI 12-25%), 11% (95% CI 7-17%), and 10% (95% CI 5-16%), respectively, in the control group. The IPTW-adjusted HR for all-cause death was 0.22 (95% CI 0.15-0.33), and landmark survival analysis for all-cause death 3-months post-index date to end of follow-up generated a similar HR of 0.28 (CI 0.19-0.44) in the CAR-T group. CAR-T patients had a lower number of days hospitalized (77.73 [95% CI 75.00-80.57] vs 86.11 [95% CI 83.12-89.21] per 1000 person-day; p<0.001), ICU admissions (0.55 [95% CI 0.36-0.84] vs 1.26 [95% CI 0.94-1.69] per 1000 person-day; p=0.001), and ED visits (4.79 [95% CI 4.10-5.61] vs 2.07 [95% CI 1.65-2.60] per 1000 person-day; p<0.001). Additionally, CAR-T patients had lower events per 1000 person-days of: infection (1.44 [95% CI 1.10-1.87] vs 3.02 [95% CI 2.50-3.64] p<0.001), neutropenia (0.65 [95% CI 0.44-0.96] vs 1.79 [95% CI 1.4-2.29]; p<0.001), febrile neutropenia (0.45 [95% CI0.28-0.72] vs 1.47 [95% CI 1.12-1.92]; p<0.001), gastrointestinal toxicity (0.26 [95% CI 0.14-0.49] vs 0.69 [95% CI 0.46-1.02]), and respiratory infections (0.51 [95% CI 0.33-0.79] vs 0.91 [95% CI 0.65-1.29]; p=0.04). Conclusions CAR T-cell therapy produced a significant and sustained survival benefit versus historical standard-of-care management, with fewer hospitalizations and infections. Despite well-described CAR-T toxicities, historical control patients had more AEs, underscoring the lack of other effective salvage treatments. This study describes one of the largest real-world comparisons of patients with RR LBCL receiving CAR T-cell therapy compared to previous standard-of-care therapies and demonstrates its effectiveness amongst a broad cohort of eligible patients, consistent with the results of pivotal trials.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".