VIDEO: Advancements in DLBCL subtyping

Editor’s note: This is an automatically generated transcript. Please notify editor@healio.com if there are concerns regarding accuracy of the transcription.

The truth is, not all of these things are absolutely clear for prime time, meaning that most of these things don't end up changing how we consider first-line therapy. Some of them are a bit more important in later lines of therapy, but there are some little nuances there, so I'll just point a couple of 'em out. We do have a few subsets of DLBCL that we think of as super high risk, and that probably does affect what frontline therapy should be. These are not real new things, but we have the worst type of DLBCL, we call double hit lymphoma. Really, sometimes people don't even call it DLBCL anymore, we give it a separate name called HGBCL, high-grade B-cell lymphoma.

We still think of it, I think, many of us, as a subtype, and this is diagnosed by FISH and features translocations of 14;18 alongside 8;14. You know, maybe one advance there is, do we need to do FISH on every single patient biopsy? In the old days, I think we said, "Yeah, you probably should." Although you could frequently kind of tell by the lick-thumb test, you know, whether this is gonna have a good chance of being a double-hit lymphoma. Nowadays, we also have a cell line that are few years old now, a MIC antibody for immunohistochemistry, and not every pathologist or hematopathologist has that in their reflex panel, but you can always ask for it. It's a pretty standard test now, and we do it routinely.

And if someone was MIC-positive by immunohistochemistry, you probably would want to do FISH testing to make sure they're not double-hit lymphoma. And if someone's MIC-negative, we don't have an official report on this, but I think it's pretty likely if someone is MIC-negative by immunohistochemistry, as long as it's a good specimen, not a bone marrow specimen that's been decalcified, then if they're MIC-negative, they're probably not double-hit lymphoma. So that is the worst subgroup, still, and they get treated differently. You know, something more than R-CHOP. We don't know the right answer, but R-EPOCH, or modified Magrath, or something more than R-CHOP. You have something just sort of under that but related, instead of double-hit lymphoma, called double-expressor lymphoma.

And that's just what I hinted at, MIC immunochemistry positive, and BCL2 immunochemistry positive, call them double-expressor lymphoma. They have worse outcomes, not nearly as bad as double-hit lymphoma. We don't really know if there's a right answer for how to treat those patients differently. Maybe we sometimes give them more intensive therapy if they're young and healthy and would tolerate the therapy. We could talk about which therapies to give them.

Just in terms of other upfront cell-of-origin classifications and subtyping, this is not a great change, but we have these two classes we call germinal-center B-cells and activated B-cell-like, GCB and ABC. We don't actually do the proper testing to diagnose ABC lymphoma, it would involve fancy transcriptomics. So we just do normal immunohistochemistry.

We have a surrogate called GC versus non-GC, and it mostly refers back to what would've been GCB or ABC. And the prognosis between those is not hugely different in recent studies. But one actionable thing about that is non-GC DLBCL does seem to respond better to certain therapies. The easiest examples are things like Revlimid, lenalidomide, usually in second or third-line therapy. So non-GC, a bit more responsive to Revlimid, maybe to BTK inhibitors. And then interestingly, in mostly post-hoc analyses, is we've seen that non-GC seems to respond better to this antibody drug conjugate polatuzumab, and super-relevant now, 'cause now we have, somewhat recently, more than a year ago, but the POLARIX trial, showing that Pola-R-CHP is somewhat more effective than R-CHOP and that seems to maybe be even truer in post-hoc analysis for the non-GC patients. So there's not a standard piece of advice to say if someone is non-GC, we should give them Pola-R-CHP.

But it is a consideration. And just a couple of other, you know, subtypes and classifications that risk-stratify a bit. There's a rare group of DLBCLs that are CD5 expressing, they have worse outcomes, higher risk for CNS spread. We have another group called mutated p53. If we do next-gen sequencing on patients, we discover these in less than 10% of patients, but still a fair fraction. And even we can use immunochemistry to get a hint of p53 mutations, because P53 over-expressors tend to be mutated. It's not absolute, but there's a correlation there. So all of these double-hit, double expressor, GCB versus non-GC, CD5, p53 mutations or over-expressions, are all somewhat important. The action to take for all of them, not completely clear.