Dr Andre Goy on Predictive Biomarker Use for CAR T-Cell Therapy Response in Patients With Large B-Cell Lymphoma

Andre Goy, MD, MS, chairman and executive director of the John Theurer Cancer Center, speaks on molecular features of diffuse large B-cell lymphoma and how this explains differences in patient response for chemotherapy and chimeric antigen receptor (CAR) T-cell therapy.

A majority of clinical prognostic features associated with poor response to chemotherapy in patients with diffuse large B-cell lymphoma did not seem to apply to chimeric antigen receptor (CAR) T-cell therapy, in which response to CAR T was shown to depend more on T-cell fitness and prior lines of therapy, said Andre Goy, MD, MS, chairman and executive director of the John Theurer Cancer Center.

Goy was co-author of the abstract presented at the 63rd Annual American Society of Hematology Meeting and Exposition, titled, “Impact of molecular features of diffuse large B-cell lymphoma on treatment outcomes with anti-CD19 chimeric antigen receptor (CAR) T-cell therapy.”


In abstract 165, the varied responses to CAR T-cell therapy based on predictive biomarkers show that some patients will have a strong response to CAR T-cell therapy after a poor response. How is this information useful to both clinicians and payers?

I started alluding to this when we talked about how to make a patient benefit more from axi-cel [axicabtagene ciloleucel] in relapsed/refractory large B-cell lymphoma, and that alludes to the fact that we can predict a signature that the patient will do well.

I will just summarize, as I mentioned before, is that the standard clinical prognostic features such as primary refractory, multiple lines of therapy, and more importantly, highly proliferative disease—K7 (keratin 7), double hit, double-expressor lymphoma, TP53–all of these very poor prognostic features for chemotherapy–activated B-cell (ABC) worse than germinal center B-cell (GCB)—all of these features are not really applying to CAR T-cell.

The abstract we’re referring to, 165, was a multicenter retrospective analysis of a patient with large B-cell lymphoma who had received CAR T-cell, looking at comprehensive signatures to see if we can identify markers. So, as I mentioned, the standard prognostic features did not apply in CAR T, so patients still responded.

Patients who had highly expressive, mutated BCL2 seemed to have a worse outcome, and while MYD88, for example, had a better outcome. This is really starting to scratch the surface of these biomarkers, and the take home message is that, again, the prognostic features that are very well established in chemo immunotherapy do not apply typically in CAR T-cell therapy.

Looking at the microenvironment we had some hint or so from the study that, obviously which we know already from studying the mechanism of resistance of CAR T-cells, microenvironment is going to be very important and T-cell fitness. And as I mentioned, that's why it's important to bring CAR T-cell therapy earlier in the journey of the patient with less prior lines of therapy.

Can you discuss the findings showing that predictive biomarkers of response to traditional chemoimmunotherapy and cellular immunotherapy are distinct?

The typical prognostic factors of chemo immunotherapy, such as again, ABC worse than GCB, K7, double hit, double expressor, high-grade lymphoma, TP53, etc. All of these are markers of poor outcome for chemo immunotherapy. They do not seem to apply with cellular immunotherapy, which makes sense. This is absolutely not the same mechanism.

What matters likely more, again, is the T-cell fitness and making sure that we bring this therapy before patients have received multiple lines of therapy. And the number of lines of therapy is not because a disease becomes more refractory to CAR T necessarily, it’s because T cells get exposed to more lines of therapy and therefore we have less expansion and less functional T cells when it comes to use of the CAR T-cell therapy and manufacturing those T cells.

Related Videos
Related Content
© 2023 MJH Life Sciences
All rights reserved.