Obe-cel Shows Superior Efficacy Over Standard of Care in Matched Analysis for R/R B-ALL

Treatment with obecabtagene autoleucelView drug details (obe-cel), a CD19 (search)Search target-directed CAR T-cell therapy with an intermediate-affinity binding design, demonstrated significantly improved efficacy outcomes compared with standard of care (SoC) non-CAR T-cell therapies in adult patients with relapsed or refractory B-cell acute lymphoblastic leukemia (search)Search disease (R/R B-ALL), according to a propensity score-matched analysis comparing data from the FELIX trial against external control arms derived from historical trials.

The analysis, which used patient-level data from the Phase Ib/II FELIX study and ten global historical trials, found that obe-cel produced a significantly higher overall remission rate (ORR) and longer overall survival (OS) and event-free survival (EFS) compared with matched patients receiving SoC therapies, primarily blinatumomabView drug details.

Superior remission rates across both analysis populations

In the intent-to-treat (ITT) analysis, 107 of 112 FELIX patients were matched to an external control arm (ECA1). The ORR with obe-cel was 67.3% compared with 51.4% for SoC (odds ratio [OR] 1.9; 95% CI 1.1–3.4; p = 0.0257). In the modified ITT (mITT) analysis, which included 84 of 94 infused patients matched to ECA2, the ORR reached 79.8% with obe-cel versus 54.8% with SoC (OR 3.3; 95% CI 1.6–6.5; p = 0.0009).

The improvement in outcomes was observed despite 16.8% of patients in the ITT population not receiving obe-cel, with 9.3% receiving only bridging therapy and 7.5% receiving neither obe-cel nor bridging therapy.

Overall survival benefit demonstrated

With a median follow-up of 20.3 months for infused patients in FELIX, obe-cel showed a significant OS advantage when censoring for hematopoietic stem cell transplant (HSCT). In the ITT analysis, median OS was 15.1 months with obe-cel versus 7.0 months with SoC (HR 0.53; 95% CI 0.36–0.79; p = 0.0015). Without HSCT censoring, a trend toward longer OS was observed: median 13.9 months versus 7.8 months (HR 0.70; 95% CI 0.50–0.99; p = 0.0430).

In the mITT analysis, median OS with HSCT censoring was 15.5 months for obe-cel compared with 10.2 months for SoC (HR 0.52; 95% CI 0.32–0.82; p = 0.0046). At 18 months, OS rates with HSCT censoring were 47.1% in the mITT group versus 25.4% in the ECA.

Event-free survival nearly quadrupled

EFS outcomes consistently favored obe-cel across both analysis populations. In the ITT analysis, median EFS with HSCT censoring was 9.8 months with obe-cel versus 2.5 months with SoC (HR 0.47; 95% CI 0.33–0.67; p < 0.0001). Without censoring, median EFS was 8.2 months versus 2.5 months (HR 0.45; 95% CI 0.32–0.63; p < 0.0001).

In the mITT analysis, median EFS with HSCT censoring was 9.1 months versus 2.8 months (HR 0.43; 95% CI 0.29–0.66; p < 0.0001), and without censoring was 9.0 months versus 2.8 months (HR 0.45; 95% CI 0.32–0.66; p < 0.0001).

Comparable safety profile

Almost all patients experienced treatment-emergent adverse events (100% with obe-cel in the mITT group and 97.6% in the ECA). The incidence of grade ≥3 treatment-emergent adverse events was similar between groups (81.0% vs 86.9%). Rates of grade ≥3 cytokine release syndrome within three months of treatment initiation were identical at 3.6% for both obe-cel and SoC.

Grade ≥3 adverse events within the nervous system disorders or psychiatric disorders system organ classes occurred in 10.7% of obe-cel patients versus 9.5% of SoC patients. The proportion of patients with infections within three months was comparable at 56.0% for obe-cel versus 59.5% for SoC; when COVID-19 infections were excluded (FELIX was conducted during the pandemic while historical trials occurred between 2014 and 2017), infection rates were 51.2% versus 59.5%.

Fewer early deaths within three months of treatment initiation occurred with obe-cel compared with SoC (14.3% vs 19.0%), and a lower proportion of patients experienced grade 5 events overall (42.9% vs 60.7%).

Study design and limitations

The prospectively designed, non-interventional study used propensity score matching to balance baseline characteristics between FELIX patients and external controls, including age, sex, ECOG performance status, bone marrow blast percentage, number of prior lines of therapy, presence of extramedullary disease, prior allogeneic HSCT, and refractoriness to last prior line of therapy.

Among ECA patients, 80.4% in ECA1 and 84.5% in ECA2 received blinatumomabView drug details, while standard chemotherapy and inotuzumab ozogamicinView drug details accounted for the remainder. More patients in the ECAs underwent allogeneic HSCT compared with FELIX patients (32.7% vs 15.0% in ITT; 32.1% vs 17.9% in mITT), which may have impacted OS results.

The authors acknowledged several limitations, including reliance on historical data sources that can introduce selection bias and confounding, limited numbers of matched patients with Philadelphia chromosome-positive B-ALL in the ECA cohorts, and the inability to conduct comparisons between obe-cel and individual treatments within the ECAs due to data collaboration agreements. Additionally, approximately 20% of patients in the mITT group had bone marrow blasts below 5% after lymphodepletion, which could have confounded survival comparisons.

The analysis was strengthened by the propensity score-based approach, which balanced prognostic factors between groups, and by a dataset larger than similar studies such as SCHOLAR-3. The outcome analysis remained blinded until after the external control arms were confirmed to be well-balanced with the FELIX study across all baseline characteristics examined.