CD34 Selection Process Evolves into Practice-Changing Approaches in Hematologic Cancers

In Partnership With:

Miguel-Angel Perales, MD, discusses the science behind Orca-T and other approaches targeted to improve GVHD relapse-free survival in select patients with hematologic cancers.

Miguel-Angel Perales, MD

Miguel-Angel Perales, MD

Research efforts with T-cell depletion, specifically with CD34 selection, continue to be refined in the treatment of patients with hematologic cancers, including acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), and myelodysplastic syndromes (MDS). 

A revolutionary approach is Orca-T, a cellular therapy product that is currently entering a randomized, multicenter, phase 3 trial. If the study is successful, it could pave the path for an FDA approval of Orca-T, which combines donor-derived stem cells and regulatory T cells, along with a small, measured dose of conventional T cells to improve outcomes for patients undergoing stem cell transplantation.

In single-arm, phase 1b/2 data presented at the 2021 ASH Annual Meeting, the treatment significantly reduced acute and chronic graft-vs-host disease (GVHD) while improving GVHD relapse-free survival (RFS) and time to engraftment. Orca-T also showed a trend toward improved overall survival (OS) vs standard of care (SOC) in patients with serious hematologic malignancies.

In an interview with OncLive®, Miguel-Angel Perales, MD, a medical oncologist and chief of the Adult Bone Marrow Transplant Service at Memorial Sloan Kettering (MSK) Cancer Center, discussed the science behind Orca-T and other approaches targeted to improve GVHD relapse-free survival in select patients with hematologic cancers.

OncLive®: Could you highlight the technology behind Orca-T and some of the groundbreaking research on the road that led to it?

Perales: We at MSK have been working on the use of T-cell depletion for over 2 decades. When I speak of T-cell depletion, what I'm really talking about is ex vivo T-cell depletion or in vitro, where we take the graft from the donor, and then positively select out the CD34 cells that will then be infused to the patient. That is known as CD34-selected grafts. We have clinically shown that, in patients with acute leukemia and myelodysplastic syndrome, we can perform a myeloablative transplant with a matched donor that results in a significant reduction in acute and chronic GVHD, but not a significant or obvious increase in relapse, which has always been one of the concerns.

One way to try and "fix" the problem of CD34 selection is to improve immune reconstitution. That is where the work that the group at Stanford has done over the years [comes in], first in preclinical mouse models in Dr. Robert Negrin's lab, and then taking that forward clinically is really where they have made a big impact.

What they figured out is taking the same platform that we have used, the CD34 selection platform, and adding back T cells. It's not as simple as just adding T cells, because once you start adding T cells, you run into the risk of GVHD. What they figured out is you don't just add what we call effector T cells, or Tcons. These are the cells that can help fight infections and may actually help fight the underlying disease.

However, you need to balance that with a population of T cells that are important in regulating immune responses. In the case of an allogeneic transplant, that means regulating GVHD, and those are regulatory T cells, or Tregs, that have been recognized now for many years. Alexander Rudensky, MD, who is at MSK, is one of the pioneers in the Treg field.

How do the efforts with T-cell depletion and PROGRESS II (NCT02345850) fit into Orca-T?

[At MSK,] we did several retrospective analyses comparing our data with data from The University of Texas MD Anderson Cancer Center in acute lymphoblastic leukemia, acute myeloid leukemia, and myelodysplastic syndrome in 3 different papers, where the patients on the MD Anderson paper were treated with an ablative transplant—matched donor but modifying graft—and we did CD34 selection. All 3 papers showed the same thing: similar OS, no difference in relapse, but a significant reduction in acute and chronic GVHD.

Therefore, the question is: If we see less acute and chronic GVHD, why are we not seeing better OS? All these studies were retrospective. We went on to do an additional study comparing our data to European registry data. The Europeans use quite a bit of in vivo T-cell depletion with anti-thymocyte globulin [ATG], more so than in the United States. We compared the MSK data with CD34 selection with the European data for patients with AML with ATG, and again, [there was] no difference in OS. There was maybe a slightly increased survival in the MSK cohort, but it wasn't statistically significantly different. We only saw less chronic GVHD with CD34 selection than with ATG.

Ultimately, with all these retrospective data, you must [address] the big question with a randomized phase 3 trial. We did that and we just published that paper in the Journal of Clinical Oncology late last year, and that was the BMT CTN Protocol 1301 trial, or PROGRESS II. That trial was done by myself, Marcelo C. Pasquini, MD, from [the Medical College of Wisconsin] and Leo Luznik, MD from Johns Hopkins [Medical Center]. We compared 2 different approaches: CD34 selection with a PTCy graft with standard-of-care tacrolimus/methotrexate with a bone marrow graft.

The study was of more than 300 patients, all receiving ablative transplant up to the age of 65 years and with a matched donor. The primary end point was a composite endpoint of chronic GVHD RFS, meaning the patient is alive, they haven't relapsed, and they don't have chronic GVHD. We know that chronic GVHD is associated with a significant impact on quality of life. If you explain that to a patient, they get that, "I'm alive, I haven't relapsed, and I should be having decent quality of life."

Unfortunately, this was a negative trial, meaning the primary end point was the same across the 2 experimental arms and the control arm. In fact, OS was decreased in the CD34-selection [group], which is certainly not what I had hoped for, since we had spent many years developing this at MSK, and it was also somewhat surprising given the other retrospective work that we had done. [One] explanation has to do with the fact that when you go to a single-center, high-volume center with experience in a specific approach, and you go to multicenter trials—and we had [more] study centers involved—that there is a learning curve, and that may certainly be part of it.

One thing we recognized as well, is that the field has evolved and things have changed and some of the approaches we use, [such as] the T-cell depletion, particularly the conditioning regimen, and the use of ATG have not changed. We have data now to show that it's possible that the dose of ATG needs to be modified. That may be what we call personalized dosing of ATG, rather than sort of using a standard dose [on everyone]. That is one of the directions we are going in.

[A challenge is that we see] a significant number of patients who have viral infections, and the treatment for the infection can lead to complications and the infection itself can lead to complications. That is probably what is contributing to increased mortality [in the PROGRESS II trial].

This still leaves us with the issue of: What is the problem with CD34 selection? Is it an approach that should be abandoned? Or, is there room for development? This is where the Orca-T story comes in. This is because the main problem that we have with T-cell depletion is not relapse, and it's not GVHD, but it's the complications that come from delayed immune recovery.

How are researchers trying to alleviate this obstacle of delayed immune recovery?

The results that the Stanford team has shown and have now been confirmed, not just in their single-center data, but now also in multicenter data. Both of which have been updated at the 2021 ASH Annual Meeting and the results are quite encouraging. They showed that the transplant approach is safe, and [the patients] don't have an increase in GVHD. In fact, their levels of GVHD were similar to what we have reported historically CD34 selection alone.

The big difference is where they look at immune reconstitution, and that there is definitely an advantage over the CD34-selection platform. I really see the future of CD34 selection as being a platform for additional T-cell populations to be added. I wasn't completely enthusiastic about the results of PROGRESS II, especially since I spent a decade of my life working on that. However, I do believe that there is still a role for T-cell depletion, and I think Orca-T is one of the examples and there are others.

[A question facing us is] how can we make T-cell depletion better? There are several advantages to the approach. One is the reduction of both acute and chronic GVHD. The other is the fact that we can treat all the patients with ablative transplant. We know from randomized data from PROGRESS II, that patients with AML undergoing ablative transplant results in better disease control than reduced-intensity transplants.

However, one of the limitations of the older patient is that they are too frail and maybe have comorbidities that they then won't tolerate the need of transplant.

The combination of an ablative transplant and methotrexate and the calcineurin inhibitor is what is too intense for that patient population. We have shown in our data that you can transplant a patient in their 60s and sometimes even into their early 70s if they are otherwise fit with an ablative transplant. That is probably why we see this excellent disease control in our patients with T-cell depletion. It really argues the case that one of the drivers of the effectiveness of the transplant in AML, MDS, and ALL is not so much the immune effects—what we call the graft-vs-leukemia effect—but really the ability to increase the dose intensity. That is what explains the results that we saw compared with the MD Anderson group—that there was no difference in relapse. In the randomized trial, there was no difference in relapse either.

That whole notion that has often been out there in the community, that T-cell depletion is associated with high relapse, has not been borne out by the facts. There are diseases where the GVL effect is critical, particularly in patients with chronic myeloid leukemia. There, we know historically, that if you do T-cell depletion, there is a high risk of relapse.

However, since we hardly do those transplants anymore, because there are so many drug options for those patients, that is less critical to transplant today—where the number one indication for an allograft is AML. The data from PROGRESS II, which showed no difference in relapse, and the randomized data of later vs reduced intensity is what we have to look at in terms of guiding our practice.

With the Orca-T approach, they're going to be able to push the age limit as well. That is why I'm encouraged that this will be similar to what we've done to treat older patients, with what ends up being a better transplant in terms of disease control.

Taking account all the intriguing data that have been observed with Orca-T thus far, what is it about this treatment that could have clinical implications on these patient populations?

The critical issue today is that the most common indication for an allogeneic transplant is AML. Patients with AML tend to be older. As the population ages, the incidence of AML is increasing. Therefore, we are seeing more and more patients referred for transplant for AML. As I mentioned, the best transplant for a patient with AML is an ablative transplant. If you are in a situation where you're limited and can only offer reduced-intensity transplant because of the package—meaning they get tacrolimus, methotrexate, and cyclosporine—then you're never going to be in a situation where you offer what we would consider optimal transplant from a disease perspective.

Using T-cell depletion alone gets around some of those issues in terms of being able to offer an ablative transplant, and [puts them at] low risk of GVHD. However, as I mentioned, the randomized phase 3 trial did not support that as a standard-of-care approach because unfortunately, it wasn't better than standard of care. In fact, there was decrease in survival due to toxicity. If we want to be able to treat the patients that we're seeing, and not just the 25-year-olds, but the actual real-world patients that we see with AML who are typically in their 60s and give them the best chance of a curative treatment, we need to find ways to improve on today's standard. Therefore, reduced-intensity transplant is not the solution. T-cell depletion alone unfortunately, does not appear to be the solution either.

Therefore, a strategy developed by the group at Stanford, now being moved forward by Orca Bio with the Orca-T product, gets around one of the main limitations that we saw with T-cell depletion, which was delayed immune recovery. They have been able to combine populations of T cells that provide that immune recovery that helps fight infections and improves outcomes, while at the same time, not resulting in increased risk of GVHD because of the addition of the Tregs. That is the main advantage of that platform with that combination of 2 populations of T cells, and finding the right source on the backbone of innovative transplant.

Then, the next step is the obvious one, which is to do a randomized phase 3 trial, which is what the company is working on. Because, as I've learned, until you do a phase 3 trial, you don't know what the results are going to be. That will really be the determinant. What has changed in the landscape as well, since we started PROGRESS II almost 10 years ago is, at the time, there was a lot of skepticism around T-cell depletion. We were one of the few centers doing it.

However, as a result of the trial, which involved almost 30 centers in the United States, and including some of the largest centers, more people have become accustomed to the use of CD34 selection for T-cell depletion. I have had many doctors reach out to me and say, "Wow, we had never tried this. But I have this patient who has no GVHD whatsoever, and it's so different from all the other patients in my clinic." If you can make it work, and if the patient survives beyond 1 year without complications, then [that is key].

In fact, a few years ago, we published a landmark analysis that Christina Cho, MD at MSK did. If you take patients who survived 1 year without relapse after T-cell depletion, those patients have an 80% to 90% chance of being alive long-term. They don't have GVHD, so those patients do very well. What we need to do is get to that first year, and I think the Orca-T platform is potentially going to get us [there] by giving that early immune recovery, and getting us over the hump of that delayed immune recovery, because we have patients with the traditional T-cell depletion, who just never recovered their immune system. Although they don't relapse, and they don't have GVHD, unfortunately, their quality of life was compromised by recurrent infections. 

Is there anything else about Orca-T, or this type of approach that we're working towards in oncology, that is important for the oncology community to know about?

We can probably mention a couple of other approaches outside of Orca-T, which are using [a similar] backbone of T-cell depletion. We have a couple of studies here at MSK. One study is looking at the addition of a donor device CAR T cells. This would be the idea that you have patients who have relapsed ALL who would not typically be candidates for an allogeneic transplant because their risk of relapse is too high.

We combine a transplant with CAR T cells from the same donor, which are given early posttransplant, and we have done some very interesting preclinical work showing that if the CAR is activated, somehow that downregulates the GVHD signal.

Therefore, inside the cell, if you push the signal through the CAR, you take away the signals through the native T-cell receptor, which would be driving GVHD. The mouse experiments are vague and sort of explained the biology. Obviously, we now have to do the clinical trial to support that, and that's the trial that is open.

There is another trial that we're doing as well; this is with a French company called Smart Immune. They have developed an interesting approach that we had also worked on, but unfortunately, we're not able to translate to the clinic at MSK. They have worked with an approach called the DL1/DL4 signaling. This is Notch signaling, and if you take CD34 stem cells and stimulate them with certain cytokines, that sort of drives T cells, and with this Notch signaling in the lab, you can develop T cell precursors—baby T cells, if you will.

At least in the preclinical setting, when you give those to the mice, those baby T cells become mature T cells. It's a different way to give an immune system you give a T-cell–depleted transplant. Then, you give these T-cell precursors that don't necessarily need to come from the donor; initially we were going to do it from the donor, but they could even be off the shelf, at least in theory, because those baby T cells are going to be educated in the host.

Therefore, they should go through the normal process of T-cell selection with positive and negative selection such that the body makes sure that we don't develop T cells that fight us because otherwise we'd have autoimmune diseases. We have done a lot of preclinical work on that. That is a slightly different approach from the Orca-T approach where we are using T-cell precursors.

There are various examples of strategies that use the CD34-selected backbone to sort of try and improve transplant outcomes.

The last one I'll mention is a company called Vor BioPharma; what they're doing is also very interesting. They're taking a CD34 selected graft, and they're going to knock out a CD33 on the cells. CD33 is commonly expressed on AML, as well as our normal myeloid cells. [If we treat] a few transplanted patients with AML with a gene knockout of CD33 graft, then those patients can then be treated with agents that target CD33.

There is an agent that's FDA approved for that, gemtuzumab ozogamicin [Mylotarg], which has been around for a long time. One of the challenges of treating AML with CAR T cells is that if you knock out the AML, you also knock out the healthy bone marrow. This [approach] would get around that because the healthy bone marrow is now mutated to not express CD33. That's a different approach using the CD34 selection, where they're getting straight out the disease itself by sort of engineering a new bone marrow that's resistant to the treatment that can target AML. There are a lot of exciting things going on today that will use CD34 selection as a platform, but take it in different directions.

Editor’s Note: Miguel-Angel Perales, MD reports honoraria from OrcaBio and Vor Biopharma.


  1. Hoeg RT, Pavlova A, Gandhi A, et al. Orca-T results in high GVHD-Free and Relapse-Free Survival following myeloablative conditioning for hematological malignancies: results of a single center phase 2 and a multicenter phase 1b study. Blood. 2021;138(suppl_1):98. doi:10.1182/blood-2021-154191

Related Videos
Expert on DLBCL
Andrew Katims, MD, MPH
Tycel Phillips, MD
Expert on ALL
Yonina R. Murciano-Goroff, MD, MSc, DPhil
Yonina R. Murciano-Goroff, MD, MSc, DPhil
Komal Jhaveri, MD, FACP
Michael Andreeff, MD, PhD
Brian C. Ball, MD
Related Content