Amid continuing excitement over the potential for PD-1 pathway immune checkpoint blockade strategies in anticancer therapies, research presented at the 2014 American Society of Hematology (ASH) Annual Meeting helped established a foundation for the use of anti- PD-1/PD-L1 agents in hematologic malignancies.
Philippe Armand, MD, MPH
Amid continuing excitement over the potential for PD-1 pathway immune checkpoint blockade strategies in anticancer therapies, research presented at the 2014 American Society of Hematology (ASH) Annual Meeting in San Francisco in December helped established a foundation for the use of anti- PD-1/PD-L1 agents in hematologic malignancies. Philippe Armand, MD, MPH, of Harvard Medical School, and Alexander Lesokhin, MD, of Memorial Sloan Kettering Cancer Center, are among the leading researchers in the field. They discussed their recent findings and other key questions in separate interviews with OncologyLive.
OncLive: Until now, success with PD-1 pathwaytargeting agents was seemingly limited to solid tumors. What has driven the more recent successes in hematologic malignancies?
Armand: Because of the success in the solid tumor world, many people were interested in bringing the drugs into hematologic malignancies. In most diseases, it’s a little bit of a shot in the dark, although there is a fair amount of preclinical evidence for at least the presence of PD-L1 on some of the malignant cells in some of the tumors. The one disease that is different in this respect is classical Hodgkin lymphoma (cHL) because there’s already very strong science to show that this disease is different from the others. cHL very often has a genetic abnormality that amplifies the genetic material on the short arm of the 9th chromosome and this results in increased expression of the PD-1 ligands, specifically PD-L1 and PD-L2. It’s something that hasn’t been shown before for any solid tumor and very few hematologic malignancies, which implies that HL may have a different behavior under PD-1 blockade than the other set of diseases.
Can you briefly discuss the significance of research presented at ASH?
Armand: There were two studies that were presented on PD-1 blockade in cHL. That’s where the results are the most promising. The studies were similar, the treatments were similar; one treated 23 patients and one treated 31 but analyzed 29 patients. The overall results in terms of response rates and apparent duration of response were also similar. In the nivolumab study,1 which treated 23 patients, the overall response rate was 87% with 13% complete remission. In the pembrolizumab study,2 the overall response rate was 65% with 21% complete remission. It seems from relatively early data that the responses may be durable and there are patients who are over 1 year in response already. So both studies really supported a strong activity of these drugs and they were both done in populations of very advanced patients, most of whom had already failed prior transplant and prior brentuximab. So these were patients who were very heavily pretreated who had these phenomenal responses.
A third stud was on non-Hodgkin lymphoma and multiple myeloma using nivolumab, and there the results were interesting but a little bit less definitive. There were some histologies like large cell lymphoma, follicular lymphoma (FL), and T-cell lymphoma where some of the patients responded, but the response rates were more in the 30% to 40% range in a very small number of patients. That is certainly a strong signal, but there were a small number of patients and the responses seem less durable, so more work is needed to understand the biology that underlies this phenomenon.
Lesokhin: I think the activity [in HL] is remarkable. I participated in the nivolumab study.1 I’ve treated patients with this agent who have relapsed and refractory HL and I saw patients with very advanced disease with multiple prior therapies—up to 14 prior lines of therapy—all of whom responded and are doing great with very little toxicity. I think it’s absolutely amazing. It’s not a home run, it’s a grand slam. You know—the ninth inning of the World Series with two outs, no runners on base, or bases loaded—whatever analogy you’d like to use.
Do you think that these agents will ultimately prove useful in all hematologic malignancies or does the data so far suggest that it is likely to be limited to certain types?
Armand: If you take the results at face value, there’s a very clear difference between HL and everything else, so in terms of the next obvious steps in development, the place where these drugs are most likely to have a significant impact is classical HL. I don’t think there is much doubt that these agents are going to have a place in the treatment of these patients because they are such strong results—granted preliminary, but strong nonetheless. For the other diseases, it depends whether the signals are confirmed, whether we can find how to identify responders, or how to combine or position the drug to get more responses. In a disease like multiple myeloma, there were no real responses but there was some stable disease; the bar is even higher in terms of figuring out if this strategy is valid. It doesn’t mean it isn’t, it just means it’s going to take more work to figure out how use PD-1 blockade in that context.
Lesokhin: Clearly, there is potential for PD-1 targeting agents, specifically targeting the receptor PD-1, in a number of lymphoid malignancies. The results have been most striking in HL and there is some activity in other lymphomas such as diffuse large B-cell lymphoma (DLBCL) and FL. I think one of the other striking features is that there are no objective responses in multiple myeloma. That is unfortunate, of course, but it remains to be determined whether or not we are, as it were, missing something about the immunology of multiple myeloma, and whether there is a way to target this particular pathway, maybe not at the level of the PD-1 receptor but maybe at the level of the ligand PD-L1, for example, or in combination with other myeloma drugs, to somehow harness the immunologic effect as well as the drug-targeting effect in a combination.
With 8 therapies approved by the US Food and Drug Administration (FDA) for the treatment of patients with advanced renal cell carcinoma (RCC), practitioners are faced with the challenge of selecting the most appropriate therapies for their patients within this crowded therapeutic landscape.
Is combination therapy likely to become a significant focus, and if so, what are the most rational combinations?
Armand: I think combination therapy is likely to be important. In HL, the results are so striking it’s conceivable that these drugs could have a place as single-agent therapy. For the other diseases, that is much less likely and, overall, I suspect that this will end up in combination. Then it’s a very difficult question to know what to combine it with and whether the activity with chemotherapy would be enhanced or suppressed based on the mechanism of action of these drugs. So again, there’s a lot of interesting discussion about what best to combine it with and that’s going to take some trial and error and good science.
Lesokhin: This is a question that’s on everybody’s minds and I think that anything I say is very speculative. I think there are a number of combinations that can be tested; these include combinations of various checkpoint targets, drugs targeting specific checkpoints in combination with PD-1, monoclonal antibodies that are already approved in B-cell lymphoma, for example, in combination with PD-1, targeted therapies, chemotherapies. And what’s the most rational? There’s rationale for many of them to varying degrees.
But I think what we see from the signal in initial PD-1 studies with nivolumab is that there was quite disparate activity; all the rationale you can have doesn’t always equal clinical activity. I think all of these combinations that I’ve mentioned are probably going to be tested in small or moderate- size phase II studies, and I think it’s going to take us a little while to sort out the best use of PD-1 in combination, which is where I think research will wind up going.
What are the most significant unanswered questions relating to the use of PD-1 pathwaytargeting agents in hematologic malignancies?
Lesokhin: I think the main question is: How is it that, when the target is present and T cells are present, there is such significant activity in one disease and such significant absence of activity in a different disease. I am talking about HL versus myeloma as the two sort of extremes in this study. So I think one of the main unanswered questions is: What is the biology of the PD-1/ PD-L1 pathway in the diseases where it doesn’t work as well as we would like? Also presented at the ASH meeting was the negative prognostic implication of having PD-L1 in your myeloma plasma cells, and so clearly there is a role for this pathway in the biology of the disease. There are a lot of questions concerning that specific role that remain unanswered.
Is PD-L1 useful as a biomarker in hematologic malignancies?
Lesokhin: I think similar to the solid tumor experience, they may be helpful but I don’t think they’ll be definitive. We don’t have an answer based on the clinical trials that have been reported and performed so far. Everybody in the HL cohort had strong expression of PD-L1 and PD-L2 based on genetically driven expression in malignant Reed-Sternberg cells. I think in DLBCL the presence of PD-L1 was limited, so was the size of the number of patients where it was tested and the size of the cohort, but there were responses and PD-L1 presence wasn’t so prevalent. I think there are many more questions around that than answers—even more so than one would see in the solid tumor literature.
Does PD-1 blockade play a significant role in the dynamic between graft-versus-tumor (GVT) and graft-versus-host disease (GVHD) after allogeneic stem cell transplant?
Armand: That’s another great question. We don’t have the clinical data. This is another area where people who are involved in transplant want to try the drug, but postallogenic transplant is a completely different playing field. There are some clinical data to suggest that PD-L1 is important in maintaining immune control, and that without PD-L1 it could be that GVHD would be much more prominent, which would likely impact GVT as well. So the question to which everybody wants to know the answer is whether these drugs can be safely administered in the postallogenic context. The science indicates that there is likely risk and benefit, but how this all plays out—and whether the benefit outweighs the risk, whether the risk can be controlled by judicious use—is something that we don’t know yet, and also something that many of us obviously want to know.