Bilal Abid MD, MS, FACP, MRCP, FRCP, Texas Tech University Health Sciences Center, Lubbock, Texas, shared results from a retrospective study from MD Anderson examining the impact of cell of origin on outcomes after CAR T-cell therapy among patients with R/R diffuse large B-cell lymphoma (DLBCL) at the 2025 American Society of Hematology (ASH) Annual Meeting & Exposition in Orlando, Florida.

Among patients treated with 3 CAR T-cell products, those with germinal center B-cell (GCB) subtype achieved significantly better progression-free and overall survival than non-GCB counterparts, particularly in the third-line or later setting. 

Dr Abid concluded, “For now, it does allow an opportunity to tailor CAR construct according to patients’ disease, cell origin subtype, at least to begin with, paving the way for prospective studies, larger registry-based analysis and further mechanistic studies will be conducted by our group and others based on our findings.”

Transcript:

Thank you for having me. My name is Bilal Abid. I represent the Texas Tech University Health Science Center and UMC Cancer Center in Lubbock, which is the western half of the state Texas. Here I presented on behalf of my investigators from MD Anderson Cancer Center. We looked at outcomes of CAR T-cell recipients in patients with large B-cell lymphoma who underwent 3 constructs of CAR T-cell therapy and defined it based on putative cell of origin.

The premise of the study is that cell of origin, putative cell origin, GCB, germinal center B, or activated B-cell, non- GCB, confers prognostic implications on the outcomes of large B-cell lymphoma and have also been shown to confer an impact on response to treatments. Now, this specific impact of cell origin and its impact on CAR T-cell therapy outcome has not been looked at in any capacity.

With this premise, we designed this single-center, retrospective study at MD Anderson Cancer Center, including nearly 350 patients who met the inclusion criteria. We included both cell of origin subtypes, which was the main effect, and we further stratified them by line of therapy because we know that second line, we have 2 CAR approvals commercially available, and in third line DLBCL, we have 3 CARs available. Then progression-free survival (PFS) was the primary end point, and we did COX models for multivariable analysis to determine what factors are driving impact when we kind of look at that impact of cell of origin on CAR-T outcomes.

Results were actually very interesting. We saw 3 main things I'll share. The first we saw that CARs in GCB subtype have a significantly superior progression-free survival as well as overall survival as compared to their non-GCB counterparts. That was our key takeaway. Second is that when you do CARs in the second-line setting, which means first relapse as compared to third-line setting and beyond, the outcomes are superior, the survival outcomes are superior, and that's something which is incongruent with the existing data.

Now, when we look at cell origin stratified by line of therapy, we saw that in second-line setting the survival outcomes were comparable across the 2 cell origin subtypes. However, in the third-line and beyond setting, GCB derived a significant survival benefit with CARs as compared to non-GCB. These are main key 3 findings. Based on this analysis, certainly this is retrospective, there are limitations. This is single center, although we control for as many patients CAR-T disease related factors as we possibly could in a retrospective study.

This study does set the premise for larger prospective validation. For now, it does allow an opportunity to tailor CAR construct according to patients’ disease, cell origin subtype, at least to begin with paving the way for prospective studies. Larger registry-based analysis and further mechanistic studies will be conducted by our group and others based on our findings. Thank you for having me.

Source:

Abid MB, Feng L, Ayers A, et al. Impact of putative cell of origin on CD19+CAR-T outcomes in patients with large B-cell lymphoma. Dec 6-9, 2025; Orlando, FL. Abstract: 5500