Table 2.
Indirect comparison of tisa-cel (ELARA) vs usual care (ReCORD-FL) clinical outcomes
| ELARA enrolled |
ReCORD-FL (weighted) |
ReCORD-FL ≥ 2014 Subgroup (weighted) |
|
|---|---|---|---|
| N = 97 | N = 99∗ | N = 95† | |
| Response rate (%) | |||
| CRR (95% CI) | 69.1 (59.8-78.3) | 37.3 (26.4-48.3) | 30.5 (13.1-47.8) |
| ORR (95% CI) | 85.6 (78.7-92.5) | 63.6 (52.5-74.7) | 58.8 (39.9-77.7) |
| Difference in CRR (95% CI) | 31.8 (18.1-45.3) | 38.6 (19.3-57.9) | |
| Difference in ORR (95% CI) | 22.0 (9.4-34.5) | 26.8 (6.8-46.7) | |
| PFS or EFS‡ | |||
| Median (95% CI), mo | NR (18.8-NA) | 13.1 (8.1-NR) | 6.3 (2.8, 15.4) |
| 6-mo rate (95% CI), % | 85.3 (78.3-92.3) | 66.5 (55.6-77.3) | 54.8 (35.2, 74.5) |
| 12-mo rate (95% CI), % | 70.5 (61.4,79.7) | 51.9 (40.6, 63.3) | 37.3 (18.8, 55.8) |
| 24-mo rate (95% CI), % | 54.1 (41.2, 66.9) | 42.2 (31.0, 53.5) | 26.1 (11.1, 41.2) |
| HR§ (95% CI) | 0.60 (0.34-0.86) | 0.38 (0.17-0.60) | |
| TNT-D‖ | |||
| Median (95% CI), mo | NR (20.1-NA) | 14.4 (9-NR) | 7.9 (3.9, 15.4) |
| 6-mo rate (95% CI), % | 94.7 (90.2-99.2) | 71.5 (60.9-82.0) | 60.8 (41.1, 80.4) |
| 12-mo rate (95% CI), % | 85.9 (78.8, 92.9) | 53.9 (42.5,65.2) | 39.4 (20.6, 58.1) |
| 24-mo rate (95% CI), % | 68.4 (50.6, 86.2) | 44.2 (32.8, 55.6) | 29.6 (14.0, 45.3) |
| HR§ (95% CI) | 0.31 (0.14-0.49) | 0.20 (0.07-0.34) | |
| OS§ | |||
| Median (95% CI), mo | NR | NR | NR |
| 12-mo rate (95% CI), % | 96.6 (92.9, 100) | 71.7 (61.2, 82.2) | 77.4 (60, 94.8) |
| 24-mo rate (95% CI), % | 87.8 (78.0-97.6) | 64.8 (53.3-76.2) | 75.1 (57.5, 92.7) |
| HR§ (95% C) | 0.20 (0.02-0.38) | 0.30 (0-0.94) |
Weights, such as those applied to our control group (ie, the ReCORD-FL sample), can reduce the precision of statistical estimates such that the analyses behave as though we have a smaller-than-actual sample of controls. The ESS quantifies this reduction of precision in the controls. Thus, the larger the reduction from actual to effective sample size, the more statistical precision that is lost via weighting.
NR, not reached; NA, not applicable.
Sample size after weighting (ie, sum of weights) was 99, and the effective sample size was 95 for the main analysis.
Sample size after weighting (ie, sum of weights) was 95, and the effective sample size was 37 for the subgroup analysis.
PFS or EFS estimation considers new anticancer therapy as a progression event (in the absence of clinician-assessed progression or death before start of a new therapy).
HR was calculated by weighted Cox proportional hazard model for indirect comparison between the ELARA trial and ReCORD-FL study.
TNT-D estimation considers death as an event.