Obecabtagene autoleucel (obe-cel; Aucatzyl) showcased a beneficial therapeutic effect in adult patients with relapsed/refractory B-cell acute lymphoblastic leukemia (B-ALL) irrespective of disease burden at the time of screening or lymphodepletion (LD), according to results from a post-hoc analysis of the phase 1b/2 FELIX trial (NCT04404660).
The data, which were presented at the 2026 EHA Congress, showed that at a median follow-up of 32.8 months (range, 19.9-52.8), the median event-free survival (EFS) was not reached in patients in morphologic remission, defined as less than 5% bone marrow (BM) blasts at screening and/or LD, compared with 9.0 months (95% CI, 5.1-12.3) in those with no morphologic remission, defined as 5% or more BM blasts at either time point.
Moreover, the median overall survival (OS) was also not reached in the morphologic remission groups vs 13.5 months (95% CI, 10.2-16.8) in those without remission at either time point. For the overall infused population (n = 127), the median EFS was 11.9 months (95% CI, 8.0-not estimable [NE]), and the median OS was 17.1 months (95% CI, 12.9-28.8).
Lower Disease Burden Linked to Better Outcomes With Obe-Cel
- Obecabtagene autoleucel demonstrated durable efficacy in relapsed/refractory B-ALL regardless of disease burden, with the strongest outcomes observed in patients who achieved morphologic remission before lymphodepletion.
- Patients who entered lymphodepletion with less than 5% bone marrow blasts experienced prolonged event-free and overall survival, with median outcomes not yet reached after nearly 3 years of follow-up.
- No grade 3 or higher CRS or ICANS occurred in patients in morphologic remission at lymphodepletion, highlighting the potential for improved safety and expanded outpatient treatment strategies.
“Broadly speaking, obe-cel definitely has a beneficial therapeutic effect in these adult patients with B-ALL, regardless of their morphologic remission status at screening or LD, but we have to acknowledge the fact that these patients in morphologic remission seem to experience the greatest benefit, both in terms of efficacy and…safety,” presenting author Claire Roddie, MD, PhD, FRCPath, said.
“What is really more compelling also is the fact that no grade 3 cytokine release syndrome [CRS] or immune effector cell–associated neurotoxicity syndrome [ICANS] is observed in those patients in morphologic remission at LD,” she added. “This opens up the whole question of whether this can be an ambulatory approach for these patients. This is an exploratory post hoc analysis, but we believe it supports obe-cel use in patients with low bone marrow blast percentage before their CAR-T therapy, and that includes patients in morphologic remission.”
Roddie is an associate professor of hematology at the University College London (UCL) Cancer Institute and a consultant hematologist at UCL Hospitals NHS Foundation Trust in London, United Kingdom.
Why does morphologic disease burden matter for obe-cel in relapsed/refractory B-ALL?
Obe-cel is an autologous, fast-off-rate 4-1BBζ CD19-directed CAR T-cell therapy approved for adults with R/R B-ALL.2 The decision was supported by previously reported FELIX data, which showed that 42% (95% CI, 29%-54%) of evaluable patients (n = 65) achieved a complete remission (CR) within 3 months. The median duration of CR achieved within 3 months was 14.1 months (95% CI, 6.1-not reached). With longer follow-up, approximately 38% of patients remained in remission at 3 years without further therapy, including stem cell transplant (SCT).3,4
Prior FELIX analyses showed that patients with less than 5% BM blasts before CAR T-cell infusion experienced better efficacy and safety outcomes than those with 5% or more BM blasts, raising the question of how bridging therapy and morphologic response before LD influence long-term results.3
“The EFS for these patients was just under 40% at 3 years of follow-up, and that’s without further therapy,” Roddie said.1 “That, as a finding, is pretty unprecedented in the CAR T-cell field in adults.”
How was the FELIX post hoc analysis of morphologic remission designed?
FELIX enrolled adult patients with R/R B-ALL who were at least 18 years of age and used tumor burden–guided dosing of obe-cel. The target dose for the agent was 410 × 10⁶ CAR T cells, and this was split across day 1 and day 10, with a 20% BM blast threshold. The primary end point was overall remission rate (CR/CRi) per independent radiology review committee, with EFS, OS, and safety as secondary end points. Bone marrow was assessed at screening and again at LD, with bridging therapy administered between the two assessments at investigator discretion.
The post hoc analysis stratified the 127 infused patients by morphologic remission status (BM blasts < 5%: YES vs ≥ 5%: NO) at each time point, yielding four groups: YES/YES (in remission at both; n = 10), YES/NO (remission at screening but not LD; n = 9), NO/YES (no remission at screening but remission at LD; n = 26), and NO/NO (no remission at either; n = 82). Overall, 19 patients (15.0%) were in morphologic remission at screening, and 36 (28.3%) were in remission at LD.
“The majority of patients have frank morphologic disease at the point of screening; that's 108 patients in total. This had changed a little bit by the point of LD. So at this point, there were more patients in morphologic remission, 36 patients in total, and there were still the majority of patients at frank morphologic relapse, and that is 91 patients,” Roddie noted. “What we're interested in is this interim period. What happened to those patients who were in remission at the point of screening, and those patients with frank disease at the point of screening? What happened to them? How did their fate change through bridging?”
What should be known about the baseline characteristics by disease status at screening vs LD?
Roddie noted that investigators wanted to ensure that the patient groups were matched, so they examined their baseline characteristics. They observed that the patients with morphologic disease at the time of screening, and particularly those who had ongoing morphologic disease at LD, appeared to be more refractory to the last prior line of therapy, and they were less exposed to previous stem cell transplant.
They also wanted to determine the features that made patients with morphologic disease at screening more or less likely to respond to bridging therapy. BM disease burden at screening was similar between bridging responders and nonresponders, and patients with persistent disease at LD carried substantial blast burden (median, 40.0% in YES/NO and 65.5% in NO/NO).
“We wondered whether it was a bone marrow disease burden thing, but it transpires that both groups, whether they responded to bridging or not, had sort of similar bone marrow burden of disease at the point of screening,” Roddie said. “[As such,] we cannot say that this necessarily dictated the response to bridging [therapy]. It wasn't that they had lower burden disease that they responded to bridging; we could not make that conclusion.”
How did bridging therapy affect morphologic remission status before obe-cel?
Overall, 92.9% of patients received bridging therapy. Patients with no morphologic remission at screening who achieved remission by LD (NO/YES) were less likely to receive chemotherapy alone (30.8%) and more often received inotuzumab ozogamicin (Besponsa)–based, tyrosine kinase inhibitor–based, or other novel-agent regimens, whereas 74.4% of patients who did not respond to bridging (NO/NO) received chemotherapy alone.
“Maybe this is telling us something, that this should be the direction of travel for our patients at the point of screening…we should be considering novel agents and not chemotherapy approaches,” Roddie noted.
What were the efficacy outcomes by morphologic remission status with obe-cel?
With approximately 3 years of follow-up, more than 45% of patients in morphologic remission at screening and/or LD remained in ongoing remission without salvage therapy or SCT (50.0% in YES/YES, 55.6% in YES/NO, and 46.2% in NO/YES), compared with 19.5% in the NO/NO group. Treatment failure or relapse occurred in 51.2% of NO/NO patients vs 10.0%, 22.2%, and 26.9% in the YES/YES, YES/NO, and NO/YES groups, respectively.
“[In patients with] ongoing remission without salvage therapies or stem cell transplant, we can see in those patients, even those high-risk patients with burdensome disease at screening, if you get that burden down, their outlook is very similar to what we see in those patients with morphologic remission at the point of screening,” Roddie said. “So we can, to some extent, level the playing field. Obviously, in contrast, we can see our patients who didn't respond to that bridging therapy have done much less well on the study.”
The median EFS and OS were not reached in the YES/YES, YES/NO, and NO/YES groups, whereas the NO/NO group had a median EFS of 9.0 months (95% CI, 5.1-12.3) and a median OS of 13.5 months (95% CI, 10.2-16.8). On Kaplan-Meier analysis, effective bridging shifted the EFS and OS curves of high-burden patients who responded to bridging (NO/YES) toward those of patients in remission at screening.
“We perhaps have the power to…shift the risk category of our patients into a more favorable category through the careful use of appropriate bridging therapy, even in high-burden disease,” Roddie said.
What did the safety analysis show with obe-cel by disease burden?
No grade 3 or higher cytokine release syndrome (CRS) or immune effector cell–associated neurotoxicity syndrome (ICANS) occurred in patients in morphologic remission at LD (YES/YES and NO/YES groups). Grade 3 or higher CRS was reported only in the NO/NO group (3.7%; n = 3), and grade 3 or higher ICANS occurred in the YES/NO (11.1%; n = 1) and NO/NO (9.8%; n = 8) groups. Any-grade CRS occurred in 20.0% of YES/YES, 66.7% of YES/NO, 57.7% of NO/YES, and 78.0% of NO/NO patients, and any-grade ICANS occurred in 10.0%, 22.2%, 7.7%, and 29.3% of patients, respectively. Treatment-emergent adverse effects of any grade were reported across all four groups.
“No grade 3 CRS or ICANS [were] observed in those patients in morphologic remission at LD,” Roddie said. “This opens up the whole question of whether this can be an ambulatory approach for these patients.”
Roddie concluded that the exploratory analysis supports obe-cel use in patients with low BM blast percentage before CAR T-cell therapy, including those in morphologic remission, and that bridging therapy—particularly with novel agents—may improve outcomes in patients with high disease burden at screening while helping maintain low disease burden at LD.
Disclosures: Roddie reported serving in a consulting or advisory role for and receiving honoraria from Autolus Therapeutics, Bristol-Myers Squibb, and Gilead Sciences; and serving in a consulting or advisory role for Johnson & Johnson and Miltenyi Biomedicine. The FELIX study was sponsored by Autolus Therapeutics.
References
- Roddie C, Park JH, Sandhu KS, et al. Improving efficacy and safety outcomes in patients with relapsed/refractory B-cell acute lymphoblastic leukemia in morphologic remission treated with obecabtagene autoleucel. Presented at: 2026 European Hematology Association Congress; June 11-14, 2026; Stockholm, Sweden. Abstract S113.
- FDA approves obecabtagene autoleucel for adults with relapsed or refractory B-cell precursor acute lymphoblastic leukemia. FDA. November 8, 2024. Accessed June 13, 2026. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-obecabtagene-autoleucel-adults-relapsed-or-refractory-b-cell-precursor-acute
- Roddie C, Sandhu KS, Tholouli E, et al. Obecabtagene autoleucel in adults with B-cell acute lymphoblastic leukemia. N Engl J Med. 2024;391(23):2219-2230. doi:10.1056/NEJMoa2406526
- Park JH, Roddie C, Tholouli E, et al. Can chimeric antigen receptor T-cell therapy be a definitive treatment for adult relapsed/refractory B-cell acute lymphoblastic leukemia without stem cell transplant? Long-term findings and predictors of sustained remission for obecabtagene autoleucel. Presented at: 2025 Society of Hematologic Oncology Annual Meeting; September 3-6, 2025; Houston, TX. Abstract ALL-712.
