Andre Goy, MD, MS: For our audience, everyone is familiar with, as you mentioned, ABC [activated B-cell] and GCB [germinal center B-cell], but we know this is much more complex, both in the molecular typing of the tumor cells, but also the microenvironment that plays a factor and then describing different subsets. I would like Nathan to tell me what you’re thinking. If you can tell our audience, after Grzeg did an overview of the fact that the signature is getting more and more complex, what in 2020 do we need to actually look at when we see a patient to make sure that we customize a therapy, because that’s our goal?
Nathan H. Fowler, MD: The bad news is, at least in 2020, we don’t have a good way to prognosticate these types of tumors. As Grzeg mentioned, a lot of the studies are using next-generation sequencing. I think many of us, even in academic centers—outside of maybe checking for MYC, BCL2, and maybe BCL6—many people are not getting the level of sequencing that is going to be required to separate patients into these different subgroups. There are several groups looking at other ways to potentially prognosticate patients…looking at some other profiles of these large cell lymphomas at diagnosis. But the unfortunate truth is that in 2020, outside of maybe double hit, which I think we’re going to talk about, I don’t see that most of us are really stratifying patients in the different treatments based upon their molecular portrait, at least yet.
Andre Goy, MD, MS: I would say that looking at ABC and GCB, the routine is probably not done enough. It matters because when we look at patients who have ABC and very high Ki-67, very often they’re going to be a double expression—we will talk about double expression in a minute. I think that for our audience you’re right, we don’t have a clear understanding of what’s the minimum testing, but I think it’s important that we go be beyond just calling it large B-cell lymphoma and definitely have at least a Ki-67, ABC, GCB, I think p53 is underestimated probably as well, EBV [Epstein Barr virus] in the elderly population. But as a lymphoma expert, if you see a patient coming, they have a diagnosis, Peter, of large-cell lymphoma, what routinely at Weill Cornell Medicine, at your institution, would you request to try to customize your treatment?
Peter Martin, MD: I think Nathan’s point is a good one in that in this era, it’s hard to necessarily modify therapies based on some of the data that we’ve looked at coming out, like molecular signatures. Nonetheless routinely our pathologists, who follow the World Health Organization…report, origin based on immunohistochemistry, which is GCB, non-GC. It’s different than the original gene expression profiling. We routinely also do immunohistochemistry for MYC and BCL2, which as you mentioned are more commonly expressed in the ABC subtype. Again, we do not necessarily right now modify therapy based on that information. But BCL2 expression will be eligibility for an upcoming national cooperative group trial that will look at the addition of venetoclax to standard therapies. I think it’s worthwhile for all of us to start incorporating that into standard pathology reports, partly because its eligibility for a national trial but also because if that trial is positive, then we will need to able to do that. There is also FISH [fluorescence in situ hybridization] testing as well for MYC and BCL2.
Interesting, I was at a meeting recently and it’s clear that every institution is doing it, but we’re not all doing it routinely in all of the same patients. We know that it’s more common, for example, in the germinal center subtype. It’s probably more cost-effective to only do it in that group. Of course, if you do that, we’ll miss some of them in that non-GC subtype.
Andre Goy, MD, MS: For the audience I think, particularly at ASH [the American Society of Hematology 2019 annual meeting], there are a few abstracts looking at the next level of complexity. As Grzeg mentioned, as we move forward we’ll be able to hopefully define a meaningful small signature that we’re going to use in a practice. Talking about practice, we should talk about other prognostic markers that we take into consideration when we see newly diagnosed large-cell lymphoma.
Dr Chavez, can you mention your prognostic factors. We’re talking about the IPI [International Prognostic Index], or the NCCN [National Comprehensive Cancer Network] IPI, or the revised-IPI. What’s your experience, and what’s your recommendation?
Julio Chavez, MD: I think the use of the score assessments are very important to risk stratify patients and actually to explain to patients their prognosis to the therapy that they will receive. This is because this is a common question in the clinic about how long they’re going to be in remission, or “Is this going to cure me?” The NCCN IPI are also useful tools to determine the overall survival and progression-free survival of patients. They’re informative, and they are easy to do. It’s done in a clinic routinely, and you can do that in front of the patient once you have all the information. But I agree, additional information like immunohistochemistry now in the GCB, non-GCB will give prognostic information to the patient as well as the FISH status, which actually will help to determine the treatment.
Andre Goy, MD, MS: But the IPI does not really help us decide on the type of treatment. You can have a low-risk IPI and have a bad biology, right? Grzeg, you guys at the Mayo Clinic worked on EFS24 [event-free survival at 24 months], and this is interesting. Also, do you want to mention the next prognostic model for our audience?
Grzegorz S. Nowakowski, MD: Yes, the EFS24 is a very important prognostic model for overall survival. What we have seen is that patients who are relapsing within the first 24 months have significantly shorter overall survival, as you would expect. What’s interesting about it is that patients who are surviving without disease beyond 24 months have actually overall survival that is close to the normal population.
Andre Goy, MD, MS: That’s very reassuring for our patients, there’s a model that you have defined that predicted, the EFS24, but it does not necessarily help us. We’ll go into this in a minute on how to customize a treatment, because we know that large cell lymphoma typically does well with R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone], but the failures occur early, and 80% of the failures occur in the first 18 months. Those patients are very challenging….
Transcript Edited for Clarity