Impact of prior treatment on CAR T-cell outcome in adult ALL

In this issue of Blood Advances, Aldoss et al¹ reported a retrospective analysis of the impact of inotuzumab (InO) therapy on the outcome of brexucabtagene autoleucel (brexucel) in adult acute lymphoblastic leukemia (ALL). They found that InO-exposed patients have inferior outcomes but this is likely to be due to patient/intrinsic disease–related selection factors rather than exposure to the drug per se. They analyzed 189 patients with relapsed/refractory ALL, more than half of whom had exposure to InO, and compared these patients with those who were unexposed. It is clear that InO is commonly used in this difficult-to-treat patient population.

Brexucel, which has CD28 and CD3 zeta costimulatory domains, was US Food and Drug Administration (FDA)–approved in 2022. This cohort, to our knowledge, represents the largest group of patients treated with this cellular therapy. Tisagenlecleucel, a very different 41BB chimeric antigen receptor (CAR) T cell, only available in patients up to the age of 26 years, has a much larger patient experience (with considerably greater follow-up). Earlier studies have indicated that bridging with InO is associated with poor outcomes. A real-world study from the United Kingdom found worse event-free survival (relative risk, 2.21; P = .03) with InO in 125 patients who received tisagenlecleucel.² These data and the presence of plausible biological mechanisms for InO affecting outcome were the reasons for performing the study.

InO, which was FDA approved in 2017, is a CD22 antibody-drug conjugate with proven efficacy in relapsed and highly refractory ALL.³ It has a 70% to 80% complete remission rate in the first or second salvage and is the most effective agent in highly proliferative ALL, often the setting for adults being considered for CAR T-cell therapy. Many ALL clinicians have used it before CAR T cells,⁴ either before leukapheresis or as bridging but there has been significant concern that it could affect the CAR T-cell product (by altering the cells collected at leukapheresis) or affect CAR T-cell outcomes by eliminating normal and malignant B cells required for T-cell function, or interfering with T-cell expansion. Accordingly, the study by the Real-World Outcomes Collaborative of CAR T in Adult ALL group was timely.

The timing of InO administration, of course, could be critical, although the current study did not show this. The median time between preapheresis InO treatment and apheresis was 62 days, and that between the last dose of InO during bridging and CAR T-cell infusion was 20 days for patients who received InO preapheresis and those who received InO as bridging therapy, respectively.

InO-naïve patients had superior event-free survival and progression-free survival to those who were InO-exposed, and unsurprisingly, of the InO-exposed group, responders had superior outcomes to nonresponders. However, the multivariate analysis did not confirm these different outcomes. InO-treated patients were more likely to have active disease and be more heavily pretreated. Indeed, many patients may not have survived to receive brexucel had they not received InO. This study provides some justification for using InO before pheresis where necessary.

There were also no significant effects of InO on short-term safety, either short-term nonrelapse mortality or the incidence of veno-occlusive disease, albeit with limited numbers.

The use of InO during bridging before brexucel infusion is less certain. Only 28 patients received InO as bridging therapy (16 also received it before leukapheresis); this is an insufficient number of patients for us to be confident that it does not affect the outcome. It may positively affect outcome by decreasing tumor load⁵ at the time of CAR T-cell infusion, but its obliteration of the B-cell compartment clearly could affect T-cell function and proliferation. This requires further study, both by prospective clinical and biologic studies of CAR T-cell fitness and proliferation. A randomized study comparing InO bridging with other standard-of-care therapies is possible but would likely require large patient numbers and multinational collaboration. Meanwhile, it would be helpful if the US collaborative group continued to accrue data from more patients with InO bridging. Biologic studies will also be difficult to perform because relevant samples are unlikely to be available from the US-treated patients. Analyses of CAR T-cell proliferation and persistence are not routinely performed; these will need to be performed prospectively. It would be helpful for CAR T-cell registries, such as Real-World Outcomes Collaborative of CAR T in Adult ALL, to collect these data.

In the United Kingdom, all patients undergoing brexucel therapy for ALL have to be approved by the national ALL panel. This group is required to report outcomes regularly and assess the effect of InO as a bridging therapy, a strategy commonly recommended by the panel. However, this method of study is prone to bias and will only have the power to detect major effects on outcomes. Over the last 10 years, we have been fortunate to have blinatumomab, InO, and 2 CD19-directed CAR T cells approved for patients with ALL, and we are still learning how to sequence and combine these novel therapies for maximum benefit in adult ALL.⁶ 

CAR T-cell therapy, despite all its transformative successes, is still in its infancy. There are many unanswered questions but the technology continues to evolve, and as a new therapy, we should study all aspects of the CAR T-cell patient pathway, as well as assiduously follow these patients long term to look for serious adverse events such as the small number of T-cell malignancies that have been recently reported.⁷ 

Conflict-of-interest disclosure: D.I.M. and A.Z.C. have performed consultancy work for Pfizer.