Risk-Based Therapy for DLBCL

Transcript:Krishna V. Komanduri, MD: We’ve had a great discussion today, and the last topic of discussion will be on emerging concepts for aggressive lymphomas. And the first thing that I think that we’ve learned about aggressive lymphomas is that they’re really not the same, that there are 2 major clinical classifications. There are probably many more than that, but 2 major clinical classifications of aggressive lymphomas. I was wondering, Leo, if you could talk to us about the risk assessment of DLBCL (diffuse large B-cell lymphoma) patients in 2016, beyond, again, clinical factors, but really thinking about the cell of origin and how that should shape our thinking about biology.

Leo I. Gordon, MD: You can have 2 patients with large cell lymphoma, and you try and predict how people are going to do. They can look the same clinically, they can look the same under the microscope, and for some reason, one has a relapse within a few months and one is cured—and they’re being treated with the same thing. So, we’ve evolved a number of ways to try and discern that. One is what you mentioned, the clinical features, and those are fairly obvious: more older, more disease, more symptoms. That’s the International Prognostic Index. And then we tried to change that a little bit. The LDH level turns out to be, I think, a very important prognostic factor. We did a study years ago comparing CHOP and m-BACOD, and looked later at LDH. And LDH greater than 3 times normal had an outcome that was similar to what we saw in pancreatic cancer. It was just a very powerful prognostic factor, and that’s where the NCCN International Prognostic Index ultimately came from.

But even more than that, as we dig deeper, Louis Staudt and Linda Wilson and their group began looking at genomics. They identified 2 major groups: the germinal center-derived diffuse large B-cell lymphoma and the activated B-cell, or non-germinal center, diffuse large B-cell lymphoma, with different pathways of activation and different genes that are mutated that are important. And that becomes a fairly important predictor. We know that the germinal center group does maybe a little bit better than the activated B-cell group, and we begin to think about different therapies for each of them, in a little bit unexpectedly as we dig even deeper. We’re looking now at mutations of c-MYC for that gene. You have activation of a variety of different factors, which makes the disease more aggressive. But, of interest, most of the c-MYC—activated, mutated patients occur in the germinal center group. It’s a little bit of a paradox in a way.

And we also have begun recognizing that you can look for the protein c-MYC and also BCL2. If they’re both mutated, then that’s a so-called “double hit lymphoma.” But if we look at the protein expression bind, immunohistochemistry, we think of those as double expressers, and those tend to fall into the activated B-cell category. And we’re not sure yet what to do with those. Ultimately, we think we should be moving toward treating patients based on the cell of origin—germinal center versus activated B-cell—and even further, whether they’re c-MYC-mutated, BCL2-mutated or not.

The problem we have is we don’t know how to do that yet. We haven’t identified a treatment that’s specific for one versus the other, but we’re beginning to do that. We’re beginning to look at agents that might be more active in the pathway of activated B-cell, and maybe more activated in the germinal center group, and that’s something maybe John wants to discuss.

John P. Leonard, MD: Well, I think there are a few different approaches to this challenge. One is, as you alluded to Leo, the drug or the intervention. And then it’s the technique to identify that subset, whether it’s immunohistochemistry. The NanoString group has one approach. There’s obviously the more basic laboratory research-oriented approaches, as well, that have all, to varying degrees, been used in clinical trials to try to get after this issue. On the flip side, we have several drugs that have been used in this strategy—the first being bortezomib, the second being lenalidomide, the third being ibrutinib—all of which have gone into randomized trials looking at, one way or another, identifying these populations and looking to see if adding the drug in that population makes a difference.

So, I would say that the jury is still out on a couple of these. One challenge is that the steps that lead to selecting a patient for a trial, meaning you see the patient, maybe it takes you some time to get their outside slides if you’re a second opinion. The process of then performing whatever central assays that need to be done that might need to be sent out for the purposes of the trial, those steps lead to a more favorable patient population, and that’s something we observed in the negative bortezomib trial.

There’s an abstract at this meeting. Matt Maurer and the Mayo Group looked at the patients who went from biopsy to starting treatment within 2 weeks versus the patients that went from biopsy to starting treatment after 2 weeks and showed that that was a very important prognostic indicator. And the patients that waited had better prognostic features and actually had better outcomes. So, just that simple step of having a patient wait, where, on the flip side, patients that are able to wait have more favorable outcomes largely, presumably, because they’re walking in and they can wait. They have better disease, so to speak. So that’s going to be a key factor in all of this. And it may be or has already been, for some of them, a factor in the outcomes of these trials. Certainly, with bortezomib, our data were consistent with that.

We have studies that looked at ibrutinib in this sort-of fashion, identifying patients by cell of origin and then randomizing the non-germinal center type, as identified by immunohistochemistry, to be able to then test ibrutinib in a randomized fashion more specifically in that population. And we have data with lenalidomide doing that as well. The ibrutinib and lenalidomide studies are ongoing in various forms. Greg Nowakowski looked at the R2-CHOP, or lenalidomide-rituximab CHOP, and had some data in a variety of different analyses—largely retrospective comparison—suggesting that the addition of lenalidomide could overcome the less favorable outcome of the non-germinal center type, but, of course, that needs to be studied prospectively.

So, I think it’s a great idea. I’m hopeful it will work. I’m a little bit cautious that it might not work because of these practical issues, but, again, we’ll have to see. As of today, I don’t think we should be treating patients on this basis; meaning getting your pathologist, our pathologist, give us cell of origin by immunohistochemistry. But generally speaking, we don’t treat patients on that basis. And I think, as Leo alluded to—the key factors that, in my practice, are beyond clinical factors of the ipilimumab—how much that changes therapy today is even debatable. Clinical factors like does the patient have mediastinal involvement? Do they have testicular involvement? A bone marrow involvement that might tweak my treatment a little bit or again looking at the double-hit group. Those are the decision-making processes, rather than cell of origin, at this point in time.

Transcript Edited for Clarity