Immuno-Oncology Resource Center

Immuno-Oncology Resource Center


Healio Interview

Disclosures: Gilham reports employment by Celyad Oncology.
February 04, 2021
6 min read

Trial to evaluate combination of CAR-T and pembrolizumab for metastatic colorectal cancer


Healio Interview

Disclosures: Gilham reports employment by Celyad Oncology.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Although colorectal cancer mortality has declined steadily since the 1940s, the long-term prognosis for patients with late-stage metastatic disease has remained relatively poor.

The 5-year survival rate for patients with metastatic colorectal cancer stands at 14%, according to American Cancer Society data.

The KEYNOTE-B79 trial is a phase 1b study to evaluate CYAD-101 in combination with pembrolizumab for patients with late-stage metastatic colorectal cancer.

Researchers hope to address the unmet treatment needs of these patients by combining chimeric antigen receptor T-cell therapy with an immune checkpoint inhibitor.

Celyad Oncology has collaborated with Merck to conduct the KEYNOTE-B79 clinical trial, a phase 1b study to evaluate CYAD-101 — an investigational non-gene edited allogeneic CAR T-cell therapy — in combination with the anti-PD-1 therapy pembrolizumab (Keytruda) for patients with late-stage metastatic colorectal cancer.

Healio spoke with David Gilham, PhD, chief scientific officer of Celyad Oncology, about the combinational strategy and pending clinical trial.

Question: Why did you decide to target colorectal cancer with a CAR T-cell therapy?

David Gilham, PhD
David Gilham

The rationale is based on the broad targeting specificity of the receptor that we are using in our CAR-T. The receptor, NKG2D, binds eight different ligands. These ligands stretch across a wide range of tumors, but they are expressed in nearly every patient with colorectal cancer that we have investigated. Therefore, from a target perspective, the approach makes sense.

Also, preclinical work carried out with our receptors showed that our CAR-T cells could generate a specific antitumor response in the solid tumor environment and in particular the receptor target — not just tumor cells, but also other nonneoplastic cells within the tumor environment, such as regulatory T cells and myeloid-derived suppressor cells. The underlying hypothesis is that the CAR-T targets the tumor and other elements within the tumor that potentially release the patient’s endogenous immune response.

Our approach makes sense in colorectal cancer because there are few other options for this very large patient population. Given the target specificity, the mode of action that was shown in preclinical models, and our progress through the autologous CAR-T trials and now moving into the allogeneic context, some of those mechanisms appear to be true and underlie the clinical responses that we have seen in this patient population.

Q: What is the rationale for combining CAR-T and pembrolizumab for patients with colorectal cancer?

Our CAR-T cells target not just human cells, but also other cells in the immune environment. The underlying hypothesis is that this can release the patient’s own immune response. We have some early snippets of information in our CYAD-101 trial, where we’ve given our CAR T cells along with a full regimen of preconditioning chemotherapy. We recently presented data showing three patients with evidence of clinical activity also demonstrated an increase in the T-cell immune repertoire, whereas a patient with progressive disease did not show such an increase. This suggests that not only is there a direct antitumor activity, but that the patient’s immune system may well be [activated] where these tumors are typically cold, or not immune responsive. The hypothesis behind the work with our colleagues at Merck is that if our cells are releasing this endogenous immune response, then can we further enhance the antitumor activity by combining with pembrolizumab. The question is, if that patient’s immune response is activated, can we prolong the immune activation by using this combination in patients with microsatellite-stable disease? We think our CAR T cells will have an antitumor effect but also release some aspects to the patient’s immune system, and by combining with pembrolizumab we can strengthen that activity and drive a more durable response.

Q: How does your approach overcome the challenges associated with treating solid tumors?

The challenges of solid tumors come in two main flavors, and one is the tumor tissue itself. The tumor is hidden away from the usual immune compartments, so our T cells generally live in the blood circulation within secondary lymphoid organs, such as the spleen and lymph nodes. To get into a solid tumor, the T cell has to traverse its normal environment and move into the tumor microenvironment. Tumors are very good at trying to blunt that entry. They have a number of mechanisms that are solely designed to blunt immune activity.

One rationale behind CAR T-cell therapy is that it can avoid some of those tumor-evasion mechanisms. One of those mechanisms is that the tumor can hide from natural T cells just by downregulating some key components that would normally be on the surface of the tumor cell and allow the T cell to recognize it. Effectively, the tumors can become invisible. The idea behind a CAR-T approach is that the receptor that we engineer onto the surface can target proteins that are not downregulated on the tumor cell. This reduces the barrier of invisibility that the tumor cells can have against the natural immune environment. We hypothesize that our T cells can target tumors through these other proteins, but what differentiates the NKG2D CAR we are using is that it targets multiple different cells within the tumor environment. Tumor cells may represent just a minority of cells within a tumor. Most of the tumor could be made up of nonneoplastic supporting cells. But these cells are stressed, and as a result they express the targets that our NKG2D receptor can target.

Using this receptor allows us to target multiple elements of the tumor. We think this provides an advantage in terms of looking at solid tumors and perhaps gives us a stepwise improvement over strategies that focus on a very specific tumor antigen. We think the breadth of targeting allows this approach to have more success in the solid tumor environment.

Q: Can you describe the planned KEYNOTE-B79 trial?

The trial is not open yet. We just announced the collaboration with Merck, and we will continue to work with them to refine the protocol and move this into clinical testing. The patient population will be basically the same population we have treated with our standalone CAR T-cell therapy — CYAD-101. The trial will include patients with late-stage metastatic colorectal cancer who have received three previous lines of therapy and are resistant or refractory at that stage. Unfortunately, this group of patients has very limited treatment options.

We are focusing on patients with microsatellite-stable disease. This means that they tend to have a very stable genome, without high levels of neoantigens. This is important because pembrolizumab and other anti-PD-1 therapies have been shown to be active in the microsatellite-instable population. Unfortunately, that is only 5% of the population with colorectal cancer, so the microsatellite-stable patient population that we are targeting with our colleagues at Merck is 95% of the colorectal cancer population. That puts into context the scale of the challenge we are facing.

Q: What will the trial evaluate?

This is a phase 1b trial, and all phase 1 trials focus on safety first and foremost. We understand the safety profile of our CYAD-101 CAR T-cell therapy from our initial trials, and the safety profile of pembrolizumab is extremely well-known. We do not anticipate toxicities from combining both elements in this treatment strategy.

Apart from safety, the main objective is to see evidence of clinical response. We have previously reported results with CYAD-101 showing evidence of disease control and an approximate 13% objective response rate in patients with metastatic colorectal cancer. We are hoping that we maintain that level of response. The important question is whether we can combine pembrolizumab to extend the durability of both the CAR T cells and the immune system. We have reported PFS in our small trial of around 3.9 months. We hope that will be extended by the inclusion of pembrolizumab.

Q: When do you expect to report preliminary results of the trial?

There are several details to work out before the protocol gets opened. It is in the very early stages, but we expect have some significant data to report by the end of 2021.

Q: Is there anything else you would like to mention?

This is the first non-gene edited allogeneic CAR T-cell therapy. Having worked in the field previously from the autologous angle, I am excited we are developing this off-the-shelf approach that really should be easier and more attractive to my clinical colleagues. The autologous approach is fantastic and is showing great results in hematologic malignancies but there is a time lapse — the vein-to-vein time is the challenge. Conversely, we have really focused on a treatment that is off-the-shelf-ready so that when the patient is recruited for the trial, they should be able to receive our investigational CAR T-cell therapy very quickly. I believe the ability to shorten this time lapse is critical for the future success of these therapies.


American Cancer Society. Colorectal Cancer Facts & Figures 2020-2022. Available at: Accessed January 27, 2021.
Prenen H, et al. J Clin Oncol. 2020;doi:10.1200/JCO.2020.38.15_suppl.3032

For more information:

David Gilham, PhD, can be reached at