Transcript: William G. Wierda, MD, PhD: Let’s talk a little bit about, we have BTK [Bruton tyrosine kinase] inhibitor as a target, we have BCL2. We also have PI3-kinase as a target, and we have a couple of drugs now. Recently a second drug was approved called duvelisib that targets PI3-kinase gamma/delta. So I wonder, Matt, you’ve done a lot of work with duvelisib. Maybe you can comment on its approval, the patient population, and about where you think it fits in how we manage our patients.

Matthew S. Davids, MD, MMSc: Sure. As you probably know, we already have an approved delta-specific PI3-kinase inhibitor, idelalisib which was approved in combination with rituximab. And early on, even before the approval of idelalisib, what was called at the time IPI-145, later duvelisib, was being developed. And they launched a large registrational study called DUO, a randomized phase III trial, international study that randomized patients to either duvelisib as monotherapy given continuously or to a 6-month course of ofatumumab, the other CD20 antibody that we haven’t talked much about, which at the time was a reasonable comparator.

And just recently, the results of the DUO study have reported out showing a very significant progression-free survival benefit for the patients who received duvelisib. They looked specifically at a population that had 2 or more prior therapies, and they seemed to have particular benefit with a progression-free survival median in the range of about 15 months. So that led to the recent FDA approval for duvelisib in that population.

At this ASH [American Society of Hematology] meeting, I presented a poster with the results of the crossover portion of that study, so these were the patients who were on ofatumumab who progressed. We had 90 patients who crossed over to receive duvelisib, and essentially the results looked quite similar. About three-quarters of the patients had a nice response and progression-free survival in the range of about 15 months. So duvelisib is another nice option that we have now. It’s approved as a monotherapy as opposed to in combination with rituximab. The toxicity profile overall looks fairly similar to idelalisib, maybe a little bit less of the transaminitis that we’ve seen with the older drug. And so I think it’s an important new option to consider for patients with CLL [chronic lymphocytic leukemia]. As we move more toward a world where ibrutinib and venetoclax are being used early in the course of the disease, often in combination in these clinical trials, we worry that patients, when they progress, may need new options. And I think PI3-kinase as a class is a good option to consider there.

Susan M. O’Brien, MD: My own experience with it is, I would say that it has the same toxicity spectrum as idelalisib. So low incidence of pneumonitis, perhaps I agree, less transaminitis and colitis. So it’s the same spectrum. My own perception is that it is less frequent than what we see with idelalisib, and whether that’s just a function of smaller patient numbers or that’s going to hold up over time, I think remains to be seen.

William G. Wierda, MD, PhD: There’s another PI3-kinase inhibitor that’s not yet approved, TGR-1202. Anthony, I’d be interested in you summarizing your experience with TGR-1202. I know you have an abstract here that you presented with TGR-1202 in a combination for patients with relapsed CLL and Richter transformation.

Anthony R. Mato, MD: Sure. Probably the best summary of the activity and profile of TGR-1202, or umbralisib, was presented by Matt a couple of years ago. He presented an integrative analysis. I think it was about 347 patients, something like that.

Susan M. O’Brien, MD: About 347?

Anthony R. Mato, MD: I just remember that number. Is that right or no?

Matthew S. Davids, MD, MMSc: Yes, I think that’s right.

Anthony R. Mato, MD: OK. Where he looked at this large patient population and really focused in on the AE [adverse event] profile across a broad spectrum of trials. I think it was probably lymphoma and CLL, and some were monotherapy, some were combination, and some were triple. So it was a diverse group of patients. But the take-home from it was that although the immune-mediated toxicities that are classically associated with idelalisib were still present, they were present to much less of an extent, suggesting that maybe this drug is a little bit different in terms of its ability to control disease, but at the same time not have a really significant immune-mediated toxicity profile.

Fast forward to this ASH; we presented data looking at the combination of umbralisib with ublituximab, which is a glycoengineered anti-CD20, with pembrolizumab. And the rationale here was that we would have, that there is strong preclinical data to suggest that checkpoint inhibitors should work in CLL and transformed disease. But the clinical data haven’t been the best so far for monotherapies. So maybe having that little bit of an immune adjutant with the PI3K delta inhibitor and also control of disease might actually help the checkpoint inhibitor work a little bit better.

And so we looked at patients with relapsed/refractory disease. We saw about a 90% response rate, although not a lot of CRs [complete responses], only one CR, and then we focused in on 5 patients with Richter; 2 of the 5 had a CR, 1 too early, and 2 had progression of disease. I think probably the most interesting thing from the Richter cohort was that these weren’t transient responses that we saw. One was about 10-plus months, 1 was between 15 and 16 months, and these were patients who were so heavily pretreated, there was nothing left to think about, including transplant or CAR [chimeric antigen receptor] T [T-cell therapy] for these patients. So I think maybe a distinct profile in terms of AE, and there are studies ongoing that will make this drug available for patients with relapsed or frontline needs.

Susan M. O’Brien, MD: Do we have an idea why? We’ve always thought all along that the toxicities are on target, to some extent, because of inhibition of T-regs [regulatory T cells] and allowing a cytotoxic T-cell population to emerge. Do you have any thoughts on why the toxicity profile appears to be so different?

Anthony R. Mato, MD: Well, I think that all PI3K delta inhibitors inhibit T-regs to differing degrees, and that there have been preclinical studies looking at idelalisib, duvelisib, and umbralisib at doses that are probably higher than we would normally expose patients to that suggest that maybe the T-regs are maintained with the umbralisib-treated patients as compared to the other PI3K inhibitors. And maybe having the CD8-positive T-cell activation in combination with T-reg decline is what is resulting in the toxicities that we see with the other 2 drugs. I think the 1 piece of information that came out of the study we presented, and it was only 10 patients, was that we did look at T-regs pretreatment, after umbralisib treatment, and then after the triple treatment, and saw that the numbers were unchanged, essentially. So that may be part of the explanation.

Alexey V. Danilov, MD, PhD: That mirrors some of the preclinical data where when this drug is compared to duvelisib and then idelalisib, there is less effect on T-regs and Th17 [T-helper 17] cells. And it’s not clear why that is. Maybe it has to do with that second target, the CK1 epsilon, and we just don’t know.

William G. Wierda, MD, PhD: Nitin Jain [MD] from our institution presented an update of a trial that we’ve been doing for Richter patients with ibrutinib combined with nivolumab. And we’ve seen responses among patients with Richter transformation with not a lot of toxicity. We don’t see the same toxicity profile in the CLL patients with checkpoint inhibitors as seen in the solid tumor patients. It was intriguing to me the tolerability of the combination that you were presenting, and I was impressed by the fact that it was well tolerated, and we didn’t really see any signals of terrible colitis or pneumonitis...

Anthony R. Mato, MD: I was very happy to see that, too. I thought Nitin’s data were excellent. One of the big strengths of his data set is that it’s a larger Richter patient population. I forget the exact number. It was between 20 and 30 patients, and there they were seeing pretty significant activity. And he, in terms of the practice strategy, had been able to bridge a fair number of patients to go on to stem cell transplantation, which I think remains a very important and viable option for patients with Richter given their overall prognosis is quite poor.

Transcript Edited for Clarity