Transcript:Andre Goy, MD, MS: I want to mention and build upon what you just said about the Hodgkin data: the response rate is very impressive compared to that of other lymphomas. And, actually, referring to your point about the PD-1 ligand, in Hodgkin, there is overexpression of specific translocations of chromosome 9q24.1, which increases the abundance of PD-L1 and PD-L2. That is very classic in the Hodgkin lymphoma. So, we believe that this is really the reason why Hodgkin lymphoma cells respond to PD-1 inhibitors and Hodgkin disease cells respond very well. One of the interesting questions in that setting is that when you try to look at reasons why patients respond—you mentioned melanoma and the first approval—the mutation load is very important for the response. And the more the cells are mutated, the more immunogenic the cells become. What’s your comment on that?

Krishna V. Komanduri, MD: I think just to talk about what that concept is, we know that various kinds of malignancies can have higher mutational loads. And on the lower end are actually malignancies, like acute myeloid leukemia, which you map along the genome, and you say how many mutations are occurring per a certain stretch of the genome. There have been really elegant studies, that have looked across large numbers of patients with large numbers of malignancies, saying how frequent the various, what we call somatic, mutations are that can lead to changes that can be recognized by T cells. It appears that the frequency of these somatic mutations along the genome in cancer cells does appear to be associated with their ability, at least in solid tumors, to respond to checkpoint-inhibitor therapy.

On one end of that mutational spectrum, with higher somatic mutation load, are lung cancer and melanoma. On the other end of the spectrum is acute myeloid leukemia. It looks like, when we’re talking about the lymphomas, CLL, and myeloma, they’re intermediate between these two extremes. So, I think that we actually need more data about how mutational load drives responses to checkpoint inhibitors. We know more about that in the solid tumor setting than in the hematologic cancers.

Andre Goy, MD, MS: In melanoma, it clearly correlated with the duration of response.

Ian W. Flinn, MD: Fred, let’s get you in on this. They’re B cells, but there are striking differences between Hodgkin’s disease and multiple myeloma. Is it all just the mutational differences or is there a microenvironment effect? Tell us a little bit about that.

Frederick L. Locke, MD: Although the response rates in solid tumors do seem to correspond to some degree to the mutational rate, for hematologic malignancies, it’s not clear that that’s going to hold up. We talked about Hodgkin’s disease and the 9p24.1 chromosomal translocations that are amplifications, that are common, and drive increased expression of PD ligand 1 and PD ligand 2 on the Reed-Sternberg cells. So, that doesn’t have to do with mutational rate; it has to do with expression, likely, of the ligands for these checkpoint antibodies. It’s unclear for other hematologic malignancies whether PD-1 or PD ligand 1 expression is going to be what drives the responses. We know for solid tumors that the expression of PD ligand 1 on the tumor or within the tumor microenvironment can correspond to responses. By immunohistochemistry, you’re looking at the expression of these markers.

However, it doesn’t function to rule patients out for these therapies for solid tumors. Patients can have robust anti-tumor immune responses even if they don’t express PD ligand 1 or PD ligand 2. So, it’s early days for hematologic malignancies, and we really need to do more work. In addition, different hematologic malignancies have different immune infiltrates. Diffuse large B-cell lymphoma (DLBCL) has a homogenous infiltration of large B cells, whereas in Hodgkin’s lymphoma, there’s very few and rare Reed-Sternberg cells, and there’s many different heterogeneous immune infiltrations. It’s likely that heterogeneous immune infiltrate suppresses a natural anti-tumor immune response, and perhaps we can turn that back on.

Ian W. Flinn, MD: Even in follicular lymphoma, right? There’s different infiltrates that have been associated with different prognosis, and maybe there’s a lot to be learned there about what’s going to be the most responsive group and perhaps how to change that.

Frederick L. Locke, MD: Yes.

Andre Goy, MD, MS: There haven’t been that much data yet in follicular lymphoma. As you said, you would expect that it would have good response. So, there’s been a combination with pidilizumab and a combination with rituximab in patients that were not rituximab-refractory. The CR rate was impressive in that setting, more than you would expect maybe, but it’s still early. And I agree with you: it will be interesting to see if the different microenvironments have more macrophage and pro-tumor effect versus anti-tumor effect, more T-cell driven, and if that would affect the response. You can see and feel that at ASCO, it’s a very hot topic; finding biomarkers and better understanding what is the signature that corresponds to response and durability in checkpoint inhibition.

Krishna V. Komanduri, MD: The other comment I would make is that you talked in your introduction about the importance of allogeneic transplantation as really a lesson for how immunotherapy can work. And we know that in the context of allogeneic transplantation, there have been some surprising developments. We started off doing allogeneic transplants for acute myeloid leukemia primarily, but we learned that really with very minimal intensive chemotherapy and the use of non-myeloablative transplantation, we can easily cure follicular lymphoma. On the other hand, if we look at myeloma, the immunologic responses in the allogeneic transplant setting for myeloma clearly are much less. And we depend on intensive chemotherapy. That’s often really been a barrier to greater success. So, I think that even though the principles of how an allogeneic transplant response works against myeloma may be different than how it works against follicular lymphoma, there may be some parallels about immunogenicity. And the success of those tumor responses may be common principles that we have to learn about.

Andre Goy, MD, MS: Well, I think what’s very interesting about this whole aspect of checkpoint inhibition—and we’re going to discuss later on combination, dual inhibition, etc—is the fact that some of these responses are very durable and that you can actually stop the treatment compared to a lot of small molecules that you have to keep giving for years until progression. Here you can have a patient achieve CR very quickly and have very durable response. The first patient that was treated with ipilimumab has been in remission for 10 years, so this is really important.

Transcript Edited for Clarity