Interview of Jeff Sharman, MD

DR SHARMAN: This study was really no surprise whatsoever. It’s putting a Ferrari against a Volkswagen with regards to the clinical efficacy in the front-line CLL. I think this puts the final nail in the coffin for front-line single-agent chlorambucil, at least from a US perspective.

From other work I’ve done, we’ve looked at patterns of care across the United States. And front-line, single-agent chlorambucil is really only about 4% to 5% of the US population. And I suspect that a lot of patients may have gone on this trial with the aspiration of being randomized to the ibrutinib arm, or at least thinking, at crossover, they could gain access to it. So I think it would be hard to rustle up 270 patients willing to take chlorambucil front line.

But, nonetheless, they were able to get the study done. And not surprisingly, the efficacy differences are some of the most enormous hazard ratios you’ll see, with 84%-86% reduction in the risk of progression-free survival, a similar reduction in the rates of death. Dramatic improvements in overall survival, lymph node response rates and so forth.

DR LOVE: There are other studies that kind of remind me a little bit of this; for example, the obinutuzumab study involved chlorambucil. And in a way, sometimes I think it’s not like exactly against the placebo, but it’s really, to me — I know there’s a regulatory reason to do something like that, but I guess in a way it’s really setting up an indirect comparison between what they saw in this trial versus what you might expect with other first-line options. And of course that’s been a question out on the table anyhow. So, what do you think this means?

DR SHARMAN: There’re 2 things I would answer out of your statement. I mean, to me, there lacked clinical equipoise for this study. I don’t think anybody really thought you could flip a coin and predict a winner here. I think it was relatively slam-dunk, but you do have the regulatory requirements for how to get labels. And those two are in tension with one another. And I don’t know that we, as an oncology community, have really settled how to answer those questions.

The more clinically relevant question, aside from this trial, is that now there are 2 treatment options for patients with treatment-naïve CLL and comorbidities — the older population, which really does constitute the majority of patients seen in community practice. You’ll have ibrutinib as an option once the FDA clears this from a regulatory perspective, but you also have the obinutuzumab/chlorambucil.

And how you pick between those 2, I think, is going to be a little bit challenging. They’re both good options. They both have their own unique profile. One of the things that have been seen in other publications is that the likelihood of discontinuation of ibrutinib is directly proportional to patient age. For every decade above age 65 you have a doubling of treatment discontinuation for reasons other than progression. So I think that ibrutinib is more difficult to tolerate in the older population, whether that’s atrial fibrillation, arthralgias/myalgias, bruising/bleeding.

DR LOVE: That's interesting. So, then, in a second, I’ll ask you about younger patients. But in terms of older patients, one of the issues is if you use something like obinutuzumab/chlorambucil, what’s the likelihood that you’re going to have a significant time off treatment?

DR SHARMAN: I think that comes back to really the start of where we discussed here a little bit, which is how do you pick between these two? I think that what we’re seeing, even in this ibrutinib-versus-chlorambucil study, there’s a 13% treatment discontinuation during this study. Now these are highly motivated clinical trial patients. And my rule of thumb is that for every 1 patient who stops there’s 1 or 2 who are contemplating stopping. So you might have a third of patients, maybe even more, who are struggling enough with ibrutinib, on it, that maybe it’s not as easy to take long term as we might have considered.

And I’ve certainly had these discussions where patients look at me and say, “So, how long do I take this?” And I say, “You take these indefinitely. Think of this as a blood pressure medication.” And it is very different from how we conceptualize a lot of our oncology drugs, where we give some sort of pulse therapy and then observe.

With obinutuzumab, the median progression-free survival on that combination is something like 2 and a half years. So I think this is a temporary question because there are a number of other studies looking to replace obinutuzumab/chlorambucil with some of the novel agents, whether that’s a BTK inhibitor or a Bcl-2 inhibitor. I think once those are done, chlorambucil will be officially gone. It’s gone currently as a single agent, but hopefully soon gone as a combination partner as well.

And so for a patient who wants to do some sort of therapy and then not do anything for 2 to 3 years, I think that the obinutuzumab might be a good choice. In my own mind, I’m using patients’ FISH status and their IGHV mutation analysis status, kind of to help me parse that apart. Those patients who have IGHV unmutated are typically going to have a more kinetically active disease. And so for such a patient I might favor ibrutinib over obinutuzumab/chlorambucil, or certainly if they have a high-risk marker such as a 17p deletion, or maybe even high-titer 11q deletion, that might be a patient I would really steer towards ibrutinib over obinutuzumab.

DR LOVE: Yes. It’s interesting this study looking at pharmacovigilance: 9% of these patients were on anticoagulants, a third of them on antiplatelet medications and 4% were on medications thought to decrease ibrutinib efficacy. Does this represent a typical population, do you think?

DR SHARMAN: I think this is a typical population and, if anything, perhaps even more healthy than a lot of the patients we see in community practice. These are patients seen in an academic medical center. And we know from some of our registry data that the typical community practice patient is almost a full 10 years older than an academic medical center patient, with diminished creatinine clearance and more medical comorbidities as a result. So this may even underestimate, to some degree, how likely somebody is to be on concomitant medications that could affect ibrutinib.

DR LOVE: So even though we’re not going over any data related to younger patients, I’m curious what your thoughts are, because I don't know how much different the efficacy is in terms of the clinical usage of ibrutinib for CLL. Right now, if you just — putting aside, let’s say, economic issues, but just looking at risks and benefits, how do you decide between, for example, FCR and ibrutinib? Let’s say if you could use ibrutinib in a non-del(17p).

DR SHARMAN: Right. This clinical trial is going to lead to a label change. And it’ll be very interesting to see how the FDA regards the front-line label for ibrutinib, if they restrict it to patients above age 65 or patients with comorbidities or if it’s going to be a broadly open label. And I think that what’s going to affect some of the authorization access to the drugs is exactly how that label looks.

But if we step back and just say, “You can choose amongst anything you want. How would you pick?” again, this is where I think that molecular characteristics are incredibly important. We know that there have been 2 recent publications, one from the German CLL8 study, which was FC versus FCR, and then the MD Anderson group reporting long-term follow-up of their FCR experience.

I think that there’s enough data to say that we can cure CLL in a number of patients up front. I know that that’s a controversial statement, but you see a plateau on the curve that begins around 7 years, and if you haven’t progressed following FCR at 7 years, chances are you’re not going to progress if you take those 2 things.

Now you want to know before you treat somebody if you’re going to be in that group or not. And it looks like those patients who have IGHV-mutated CLL and either trisomy 12, deletion 13q or possibly the normal cytogenetics, those patients have 7-year progression-free survivals. And in the case of trisomy 12s, of 100%. So if I’ve got a young patient, good medical status, good medical fitness, who’s got an IGHV-mutated trisomy 12, I’m going to treat them as aggressively as I can because I think there’s a chance for cure. And I don’t think you’re going to get that chance in the second-line setting.

I will tell you it’s a difficult conversation with patients who oftentimes come preloaded, because they’re web-positive — they might be coming to you saying, “I want ibrutinib.” And I’m saying to them, “Well, okay, but here’s the chance. I might have a chance of curing you if I do this.”

Now I don’t think we’re curing anybody with IGHV-unmutated CLL. And so that’s where I think the discussion becomes more challenging between immunochemotherapy and ibrutinib. I think that that marker may be the dividing line between where I would choose a chemoimmunotherapy versus a tyrosine kinase inhibitor.

DR LOVE: Fascinating.

We also are starting to see data in other BTK inhibitors. And John Byrd presented some data, and then it got published, on ACP-196. What do we know about it and how does it compare to ibrutinib in CLL?

DR SHARMAN: There’s a fascinating story about how ibrutinib came to be. It was when Craig Venter was doing the human genome project. He thought he could simultaneously make drugs. And so he started developing some drugs that — I don't know how far he got in clinical development. But one of them was actually PCI-32765, which subsequently became ibrutinib.

And when he recognized that he shouldn’t be developing drugs, he sold it as an afterthought for a really inexpensive cost. And while we were doing work on SYK inhibitors we were seeing signs of efficacy. And SYK — S-Y-K — resides immediately adjacent to BTK.

And so there were a number of things going on at the time that they had to throw a drug into the clinic pretty quickly. And ibrutinib had never really gone through the full medicinal chemistry workup that a lot of drugs do before they go into the clinic. So there are some liabilities of ibrutinib. It’s not as pure a BTK inhibitor. It has a number of off-target activities. It’s in a difficult formulation, so you have to give 3 pills together. The dose of 480 mg represents some limitations of oral absorption and so forth.

This looked at a very similar population of the original ibrutinib CLL patients; median prior therapies, 3. And when this abstract was actually submitted they had not even seen a single progression on this drug. At the time of eventual presentation there were 2 progressions. But this drug really looks to be a potential improvement upon prior BTK inhibitors. And I think it’s going to present a big challenge to ibrutinib once it gets available.

Now, some of the liabilities: It’s a twice-a-day drug as opposed to a once-a-day drug. And that’s actually selected because BTK resynthesis rates probably enable some BTK to be present at 24 hours. These BTK inhibitors are covalent, so you take them, you swallow them, they are only in your blood for maybe an hour or two and then they’re gone. But in the meantime they covalently bond to BTK and totally inactivate the enzyme. But by 24 hours you actually start to see BTK coming back. And so by switching to a twice-daily dosage you can eliminate that resynthesis problem and enable more complete BTK inhibition.

DR LOVE: So you would think that a major issue here would be tolerability or toxicity. I mean, it was only 61 patients, but I see here no major hemorrhage, atrial fibrillation, tumor lysis, et cetera. What about clinically? I don't know. Have you used the drug? Do you see this bruising that you see with ibrutinib?

DR SHARMAN: I have quite a few patients on this medication. I’ve been very impressed. I have not seen bruising or bleeding. It is reported with the medication but at rates that appear to be less. They have preclinical information looking at thrombus model and so forth, that they think they have less anticoagulant activity. And I think that bears out with my clinical experience with the drug. The arthralgias seem to be less. Actually, I haven’t even seen arthralgias with the drug. I think that there’s some low frequency of arthralgias reported. But it does appear to limit some of those side effects that we see commonly with ibrutinib.

DR LOVE: Atrial fibrillation?

DR SHARMAN: I don’t think it’s been seen.

DR LOVE: No. It’s not in this study. It seems like it’d be tough to figure out a way to prove that it’s more effective than ibrutinib. It might be easier to show that it’s less toxic. I don't know. Do these data suggest greater efficacy? Is there any way you could even think about that?

DR SHARMAN: If you do a cross-trial comparison — and you know the limitations of doing all that — I mean, this is really just looking at the data and waving our hands. The fact that there had been no progressions on this at a time point where, on the ibrutinib study you had seen a handful, you might consider that this could be more effective. But, Neil, they’re really getting very bold with this drug. They’re running a head-to-head study against ibrutinib.

DR LOVE: Wow!

DR SHARMAN: And I think that represents a degree of confidence that they will have at least equivalent activity with lower toxicity. But if they show higher efficacy on that study, that’s going to be really quite an earthquake in the field.

DR LOVE: I’m going to guess, but maybe there have been anecdotal cases that we don’t know very much about, response to ibrutinib after ACP-196 and vice versa?

DR SHARMAN: I’m not aware of any.

DR LOVE: So this certainly is going to be a very interesting story for us to follow and see what happens.

Let’s move on and talk a little bit about some of the papers, other papers that came out in CLL. And I want to ask you about several papers. We’ve heard a lot about venetoclax. And you would think, it seems like it ought to be available — should be available soon, I would hope. But can you talk about the data that was presented there in CLL?

DR SHARMAN: I would anticipate that this drug may very well be approved by the time this education thing goes out to your audience. I would anticipate approval probably still within the first half of 2016. And I anticipate that the label will be predicated upon the 17p-deletion population.

One of the things, clinically, that’s disappointing is I think that a lot of docs don’t recognize that FISH changes quite a bit in CLL over time. There’s this notion that 17p is a rare phenomenon, maybe only 5% of patients. Well, that’s true in the front-line setting. But when you go through multiple lines of therapy you select for resistant disease. And so 17p deletion in a variety of studies, anywhere from 1 out of 4 to 1 out of 3 patients in the relapsed/refractory setting, so it’s really common.

But we’ve put out data that shows that docs are only repeating FISH about half the time in patients with relapsed/refractory disease, even less once they’re out 2-3 lines of therapy. So if you’re not looking for it, you’re going to miss it. And this drug will probably initially be limited to 17p, but I anticipate its label will broaden out over time.

DR LOVE: And I think the FDA fast track thing, or whatever, is in 17p. And I’m just kind of curious right now, you wake up tomorrow and find the drug’s approved for 17p; are you going to use this or ibrutinib?

DR SHARMAN: This drug is perhaps the most effective drug in CLL from an efficacy perspective, but therein lies its biggest liability. It works too quickly. I was working with this drug early in Phase I at a time when 2 deaths occurred at other study centers from tumor lysis. One of those may have been explainable from the way the protocol was written. But another death that occurred happened in a patient after a single dose of the drug — took a single 50-mg dose and went into fatal tumor lysis 8 hours later.

When that happened, they reformulated this ramp-up of starting with 20 mg and then doing that for a week and then going up. And in doing that, that reconfigured dosing schedule has been implemented across all these studies. And to their credit, they have not seen any clinical tumor lysis since making those changes.

My biggest concern, though, is that they are still seeing laboratory tumor lysis syndrome. And something like a potassium that goes from 3.5 to 4.0 could actually be a very ominous finding in a patient, if that happens after 6 to 8 hours. And my biggest concern is that docs who use this drug may not totally capture just how quickly it can work. And so physicians need to be really vigilant using this medication — these patients have to be observed very quickly.

Now, you tell me, wake up tomorrow, the drug’s approved. You’ve got a 17p-deleted patient. I think that the label is likely to have some suggestion about consideration of hospitalization for patients who are receiving this drug. One of the things they’ve done is they’ve looked at trying to characterize the risk of individual patients. And so, if you have multiple lymph nodes in excess of 5 centimeters or a lymphocyte count in excess of 25,000, those confer higher risk than if you don't have those. And similarly, if you have a creatinine clearance less than 80 mL per minute, which frankly is most of our patients in the relapsed/refractory setting, that also confers higher risk.

And just operationally, these protocols have all used labs that are 6 to 8 hours post-first dose. And so just operationally, thinking about doing this as an outpatient, if I have a patient come to clinic, take the pill, and then have to get uric acid and phosphorus and magnesium and potassium 6 to 8 hours later, there’re potentially operational challenges in doing that. And so within our own network, I’ve looked at a number of different treatment centers, and some can’t even measure uric acid on site. So they would have to ship that out and get that result back in a timely way.

So I will probably be admitting many of my patients to the hospital to do this. And I think that that’s going to present a barrier in a lot of cases. And so I think that once it’s available, we’ll have to go through all the insurance authorization to get hospital administration — have to get formulary authorization for it. I think there’s going to be a lot of operational barriers to the broad utilization of this up front.

DR LOVE: Where do you see this agent heading, all these papers here with various combinations, anti-CD20, obinutuzumab, rituximab? I don't know. Have they looked at it combined with ibrutinib or idelalisib?

DR SHARMAN: John Byrd is running a study of ibrutinib/obinutuzumab and venetoclax, and nothing’s been reported there. But that’s triplet therapy, if you think about it. I think the sum theme of all of these abstracts from ASH is looking at maybe an alternative way to mitigate the risk of tumor lysis by debulking. And so if that’s giving 2 cycles of bendamustine/rituximab first, or giving obinutuzumab first or something to reduce that lymphocyte burden prior to the initiation of venetoclax — and to me, that makes a whole lot of sense, that that would be the strategy moving forward.

However, I would be very apprehensive and would not endorse and would strictly avoid, myself, the co-administration of venetoclax with anything else simultaneously. If you were to just have somebody come into clinic, take rituximab and venetoclax, I’d worry that the tumor lysis there is unmeasurable. Fortunately that’s not the label, but that’s — when we think about how things get used out there in the real world, there’s a lot of creativity, if you will, of how people use things. And so I think that the proper role of this drug is going to be after some debulking has occurred, but that’s not going to be the label up front. So that’s what studies are going to have to answer for us moving forward.

DR LOVE: Any specific comments on this late-breaking abstract in 17p, particularly in terms of what they saw in terms of MRD? And do you think that somewhere in here is the potential cure for CLL?

DR SHARMAN: Yes. This study highlights the really unbelievable efficacy of this drug. What we’re seeing is in a highly relapsed/refractory, 17p-positive patient, where nothing works, we’re getting overall response rates of 80%. And amongst responders, you’re getting MRD negativity in 20% of those. When patients have 17p, those sort of numbers are virtually impossible. We’ve never seen them with anything else. So the efficacy here is unprecedented.

And if we’re potentially curing patients in the up-front setting with FCR, the right genomic patients, I think that this drug will be an important means by which we get to a cure in patients with CLL. But how that’s either combined or sequenced, or where that fits, is yet to be seen.

DR LOVE: Interesting. We were talking before about obinutuzumab. And of course a lot of people are asking the question of obinutuzumab with things other than chlorambucil, as you were talking about earlier. Abstract 493 reported on safety and efficacy of obinutuzumab with bendamustine in untreated patients. And this was a subgroup analysis. This is the so-called GREEN study that I think we’ve seen data on before. Anything that came out of that that you thought was important?

DR SHARMAN: There’re actually several findings in this study that I think are really interesting and curious. This was a nonrandomized study of obinutuzumab where patients in the German CLL network were given obinutuzumab, and then the investigator could pick the corresponding chemotherapy to go with it. So it wasn’t randomized. It was really kind of multiple Phase II studies all in parallel with each other.

And the findings in this, I think, were really quite interesting. We know from the chlorambucil/obinutuzumab that not only was there a progression-free survival benefit, there was overall survival benefit when combined with chlorambucil. But it also appeared considerably more effective than rituximab for progression-free survival. Now some people have said that’s dosing and scheduling. I think they’re qualitatively very different antibodies. And so it’s my belief that obinutuzumab is more effective than rituximab.

This study combined it with bendamustine, which is really one of the most commonly used front-line regimens in CLL in the United States. And there were a couple of things that I would highlight. The first is that neutropenia looks to be a little bit worse. This study did not mandate growth factor support early on, but we’ve been doing, actually, a parallel study to this in our own network and heavily encouraged pegfilgrastim growth factor support. Whether that will ameliorate the neutropenia we have to see when we see the data.

But interestingly, 10% rate of tumor lysis was pretty high for a front-line study. And similarly, if you look at minimal residual disease in this study, it was pretty darn impressive, with almost 60% of patients MRD-negative, with a great number of missing data points. It might have been considerably higher had they had all the data points. I think this is a very active regimen.

There’s no Phase III study looking at this, although we will have another Phase II study. I think it’ll be interesting to see if this can get compendia listed on that, and, if it does, whether or not that gets utilized. I would feel comfortable doing that, if it got compendia listing for the combination, because it is a very active regimen.

DR LOVE: Yes. It kind of gets into that question that goes all across oncology, which is, how much evidence does it take to get a drug approved and get a drug used? If you just look at the clinical data that you have right now, if you could use bendamustine/obinutuzumab, would you use it? Or do you use it?

DR SHARMAN: We’ve been doing that study for a while. We’ve closed accrual to it, and so we’re really in the process of analyzing the data. As of right now, one of the findings in this study was actually a surprising number of deaths in the front-line setting. So I think it needs to be done with caution. And it really does beg for a Phase III study. That’s not going to happen. Let’s just say that it gets compendia listed. I would be considering doing that in the older patients who have good comorbidity status and favorable genetic changes, where I think I can possibly get them 4 to 7 years of remission with this combination.

DR LOVE: What about the use of obinutuzumab alone? I know the last time I talked to you, I was interested to find out that a lot of times you just use it by itself, without chlorambucil or anything else. Are you still doing that? And do you see prolonged remissions just using it alone?

DR SHARMAN: Our group had run a clinical trial of obinutuzumab monotherapy. And we’ve reported that data. It’s been published in Blood. It’s been presented at several congresses. We saw really nice activity with that. Interestingly, that study was a randomized study, but it was a randomized dose-level study where we looked at a 2-g dose versus a 1-g dose. And it did look like patients did better on the 2-g dose, although it wasn’t statistically significant. It was very close.

Again, this is for the patient who is older, has more medical comorbidities, and it’s for the patient who doesn’t want to be taking pills or might have multiple other pills that could be interfering with ibrutinib and so forth. So I am still doing it. My hope is that — right now, we’re actually doing a study in which it’s combined with the ACP-196 molecule. And that’s actually a randomized Phase III study. So this is really kind of a temporary time. And I think that we will have better combination partners with obinutuzumab in the front-line setting quite soon.

DR LOVE: No interview in the field of oncology is complete nowadays without asking about checkpoint inhibitors, although the hematology angle’s come in fairly recently. But there was a paper, 834, at ASH looking at PD-1 blockade with pembrolizumab, which, of course, was approved in a bunch of other cancers, in relapsed/refractory CLL, including Richter transformation. Any thoughts about this paper?

DR SHARMAN: Yes. This paper was a little short on efficacy data, but it did kind of give us a clue to an area of activity that I think is both welcome and surprising, which is the level of efficacy in patients with Richter’s transformation. I’ve had a handful of Richter’s transformations recently, and 2 of my last 3 died within 72 hours of the diagnosis. Richter’s transformation can be unbelievably aggressive and fulminate when it presents. And this is a very challenging patient population to study because frequently they are quite ill. They’ve been beaten up with a lot of different chemotherapy regimens beforehand, and they’re not going to tolerate all that much.

And what this paper showed was that 4 out of 5 patients with Richter’s transformation responded. There was one-plus complete response. There was another one that was near CR. I mean, they used some colorful terms to describe these. But what this data really cries out for is a larger experience of this drug in this disease setting. That is not an easy study to conduct because these patients are quite variable in their presentation, and they’re quite ill, oftentimes, at their presentation, as well.

So I have used this study to justify doing this. Right now, actually, I have a patient who has CLL transformation. He’s in his mid-eighties. He has horrible neuropathy just at baseline. And he actually had a Hodgkin’s transformation from his CLL, which is the less common version of transformation but pathology confirmed at an academic medical center. And so I’m going to start him on pembrolizumab next week.

The challenge in him was he had kind of a bony-predominant disease. And we had a couple of unsuccessful biopsy attempts of it, so a lot of time passed. But I think with a Hodgkin’s transformation, where, in Hodgkin’s disease, these checkpoint inhibitors really look sensational — I mean, it’s a “wow”-type thing. I’m really optimistic for this patient.

DR LOVE: Right. So, a couple of other papers I wanted to ask you about. I guess you could say this was R²/I. It was an Alliance study, a Phase I study of rituximab/lenalidomide and ibrutinib in previously untreated FL.

DR SHARMAN: I think that this study highlights the fact that this is not a good combination. R² alone gives response rates on the order of 90% to 95% with very high CR rates. And in this study I don't know that you could tell a difference in the response rate whatsoever, but what you could see was really intolerable toxicity, with nearly three quarters of patients having rash that was dose limiting or dose interrupting.

And I think that when we start doing combinations of these targeted drugs we see things that we don’t necessarily expect. And why rash? I don't know. But, to me, this triplet — and their conclusions, as well, were that further investigation of this triplet in this setting seems unwarranted. So I do not think that this is going anywhere.