DR EVANS: I was surprised, I think, pleasantly. I think looking at some of the early data on obinutuzumab, at least in indolent lymphoma and some other lymphomas, I was questioning, How more active is it than rituximab? Yes, it’s “likely” engineered for enhanced ADCC and decreased reliance on complement-mediated cell death, but this was an ambitious study from the get-go. It was a randomized study, a large randomized study, from the Germans, called the CLL11 study, randomizing to the standard of chlorambucil versus 2 experimental arms. One was GA101 plus chlorambucil or obinutuzumab plus chlorambucil, and the second arm being rituximab plus chlorambucil.
And what was presented at ASCO was kind of the first stage. They had Stage IA and Stage IB, where they compared the combination of the CD20 antibody with chlorambucil to chlorambucil itself.
DR LOVE: And when you talk about stage, it’s stage of the presentation of the data kind of thing.
DR EVENS: Exactly. And I should also make sure to frame that this was untreated CLL with comorbidities, and to me, this was almost kind of a truly real-case scenario.
When you look at a lot of the CLL studies, especially for untreated, the median age in many of the studies is in the mid-50s, where we know that’s not the median age for most patients with CLL. It’s close to age 70.
DR LOVE: Can you comment on what the difference is between the way this agent and rituximab work?
DR EVENS: So literally it is this glycoengineered gycosolation where it results in increased affinity binding to FC gamma and really results in much enhanced, at least preclinically in vitro, ADCC. And much lower dependence on complement-directed cell death.
DR LOVE: Just, again, a word in terms of the mechanism, Jon, for the nonaficionados, if you think about rituximab, ofatumumab and obinutuzumab, how would you explain to a fellow what the difference is in terms of how they work?
DR FRIEDBERG: I think obinutuzumab being a type II antibody, those antibodies have been associated with direct cytotoxicity far more than the type I antibodies. So — and it’s the only type II antibody except for tositumomab, which is part of the iodine 131-tositumomab regimen that we’re using in practice. So, although ofatumumab is another second- or third-generation anti-CD20 that has been manipulated to enhance cytotoxicity, the type II nature of the GA101 antibody should, at least in the laboratory, suggest that there’s increased chance of direct cytotoxicity as well.
So, if a patient is resistant to ADCC because of lack of effector cells or previous therapy or maybe the presence of a disease like CLL, there’s thought that a type II antibody may result in direct cytotoxicity.
DR LOVE: So maybe you can isolate what you think are the most interesting graphics they presented.
DR EVENS: So I think I would say 2. One would be the end-of-treatment response rates, and in the GA101 or obinutuzumab, the overall response rate, 76%, with a CR rate of 22%. And they’re not to the second stage of the comparison, where they’re going to compare GA101/chlorambucil versus rituximab/chlorambucil, so it’s a little bit of apples to oranges. But the rituximab/chlorambucil rates were 66% and 8%. And then moreover, looking at MRD, it was a fairly impressive reduction in MRD in terms of the obinutuzumab. In peripheral blood, 30% were converted to MRD-negative, 17% in the bone marrow. Conversely, with rituximab, it was 2% and 3%, respectively.
The PFS was pretty impressive for obinutuzumab, median PFS of 23 months and a 1-year PFS of 84%. Comparatively, for the rituximab/chlorambucil, it was 16 months and a 1-year PFS of 63%.
DR LOVE: Julie?
DR VOSE: I think it is very interesting, especially the MRD information. I think it’s a hint that perhaps the antibody does have some potency that the other ones don’t. And we should definitely look at that. And I’ll be anxious to see the final results.
DR LOVE: So Chris, what are your thoughts? Do you think a couple of years from now, at least in CLL, we might be using obinutuzumab instead of rituximab?
DR FLOWERS: I think that’s possible.
As Julie mentioned, the MRD data are somewhat impressive, to see that a single-agent antibody with chlorambucil is producing MRD in patients with CLL. That’s a little bit surprising, even in the small numbers of patients that are becoming MRD-negative. So I think these were very promising early results, and we’ll need to see what the final results in Stage II show in the comparison versus rituximab, to see whether obinutuzumab is actually better.
DR LOVE: Mike, I’m curious about your thoughts about this and, also, the fact that it uses this chlorambucil comparator. It’s older patients. If the trial actually plays out to show an advantage for obinutuzumab over rituximab, do you think it will be open to criticism because of the chlorambucil elderly thing?
DR WILLIAMS: Just to take the second point first, I mean, there’s been criticism of chlorambucil as being the control arm in all of these new agents that are being tested. And it’s probably not a real difficult thing to beat. Chlorambucil does have activity, and so in our view it’s something that we use in elderly, frail patients. It’s convenient. Many patients, although you don’t get a lot of deep responses, you can often control the disease pretty well. So — but that said, I don’t know that it’s ultimately going to be the agent with which these new molecules and immunotherapeutics are going to be combined.
DR LOVE: Do you think that people would then generalize it to other chemotherapeutic agents?
DR WILLIAMS: Right. So add it to bendamustine? I think that would be a logical thing for folks to do. And I suspect they’ll be ahead of the studies. I would comment on the — just comparing rituximab with obinutuzumab, you — I mean, you heard about differences in mechanisms of action. And if you think about just single-agent activity of rituximab in CLL, it’s pretty modest. I mean, in SLL/CLL. And whether that’s because of lower CD20 density, less effective effector cells like NKT cells, I’m not sure. So, for reasons that Jonathan mentioned, this molecule may, in fact, be an incremental, but maybe a fairly significant incremental improvement over what we’ve had to date with our anti-CD20s for CLL.
DR LOVE: Andy?
DR EVENS: The one other thing on tolerability, I think it’s obviously important to look at, that I think the good news is: It is similarly tolerable to rituximab. The one note, there was fairly high incidence, I thought, of Grade 3/4 infusion-related reaction at cycle 1. Twenty-one percent had a Grade 3 or 4 infusional reaction. That was, I guess, the bad news.
The good news is, somewhat like rituximab, the body acclimates and it dramatically drops even a week later, because this was day 1, 8, 15 dosing. For cycle 1. So, even by day 8, the overall incidence dropped from 69% to 3% in terms of infusional reactions. And there were no Grade 3/4. So obviously, still caution on that first dose that you’re giving of this agent for untreated patients.
DR LOVE: John?
DR FRIEDBERG: The dose of the antibody is higher. CLL is a disease where some previous studies, even with rituximab, suggested that there might be a dose response. And, in fact, when we give FCR chemotherapy, the dose of rituximab is higher than we would give in lymphoma. So there are certain people who might argue that one of the reasons why you’re seeing benefit with the GA101 here is you’re simply giving more antibody than you are with rituximab.
DR LOVE: That's fascinating.
DR FRIEDBERG: So Mike may have insights on that, too. He thinks a lot about antibodies. I don't know. One of the questions I was going to ask you, as a follow-up to what Neil was getting at, if this study turns out to be positive and the progression-free survival benefit is there, I think the question on the table that’s literally a million- or billion-dollar question is, should GA101 replace rituximab in BR and FCR? And although the rituximab doesn’t show significant single-agent activity in CLL, it’s in at least the FCR space for overall survival, which seems to be improved when you add rituximab.
The mechanism may, therefore, be different when it’s synergizing with active chemotherapy. And do you think that a type II antibody might — some of its benefit might be less in that setting where you have such active chemotherapy? Or do you think that you might build even more?
DR WILLIAMS: Yes. I mean, it’s a great question, and I don’t know the answer, but I’m happy to answer it. So I think there will be additional benefit when it’s used in combination. I don’t see any reason to think that you would lose any of that novel effect of this antibody when it’s in combination with a cytotoxic. So I think it will likely — if the Phase III studies bear it out, I think it will become probably the combination antibody of choice, if the Phase III data support that and support the safety of it.
Just the other counter to the issue about using higher doses of antibody for CLL/SLL is that we did a small pilot study at our place — and I won’t go over the background biology of what led to it, but we found that frequent low doses of rituximab could be very effective in CLL. So we did thrice-weekly 20 mg/m2 of rituximab in relapsed/refractory CLL, including 17p-deleted patients, and showed very high response rates in those patients. So we did that for 4 weeks, thrice weekly, and actually have done a very small pilot with Adrian Wiestner at NIH, looking at subQ rituximab. And there’s a French study now looking at frequent low-dose subQ, just like alemtuzumab, which uses a much smaller dose of antibody with thrice-weekly subQ dosing, can be used. So that’s also being tested now, and maybe that’s a way to maintain people or consolidate a response based on MRD.
So we always think about giving more antibody as a way to get the end we want, but there may be other strategies as well.
DR LOVE: Really curious to hear what the pharmacologic basis of that is. Don’t they have very long half-lives?
DR WILLIAMS: Right. So the basis of this was that they do have long half-lives. But what was found with CLL — and we found this sort of surprisingly — we were looking at complement activation and complement depletion. So we were giving standard-dose rituximab to CLL patients and then getting blood samples at 30 minutes, 60 minutes, et cetera. And what we found is that there were dramatic compartment shifts where somebody who had a white count of 100,000, by 1 hour had a white count of 10,000. And then over the next several hours to couple of days, their white count came back up, but the cells that were circulating were CD20-negative. And it wasn’t because the CD20 was being masked, it was being removed by a process that we called “shaving reaction.”
So the RE cells were pulling off the immune complexes. CD20-negative cells would circulate, but then over the next several days they would regenerate their CD20. So by giving frequent small doses, you avoided overwhelming the system and saturating the RE system and the effector cells. So, that’s the proposed mechanism. And that’s been shown for — we showed that for low-grade lymphomas. We showed it in tissue as well as circulating cells.