Ibrutinib-Resistant Patients a Core Challenge in MCL

Partner | Cancer Centers | <b>Moffitt</b>

Bijal D. Shah, MD, discusses the current treatment paradigm of MCL, what therapies are moving through the pipeline at a rapid rate, the potential benefit with CAR T-cell therapy, and pivotal biomarker studies currently being conducted.

Bijal D. Shah, MD

Therapeutic approaches, including chimeric antigen receptor (CAR) T-cell therapy, are being explored in patients with mantle cell lymphoma (MCL) who have demonstrated resistance to the BTK inhibitor ibrutinib (Imbruvica)—a patient population that is worrisome to practitioners, explains Bijal D. Shah, MD.

“The spectra of ibrutinib resistance is the main thing,” said Shah. “If I had 1 thing to say about MCL, ibrutinib resistance is bad for the majority of the patients we see, and it has to be acknowledged. It works wonderfully for some individuals. We see that 20% are going to see these prolonged remissions with ibrutinib and I would never say anything bad about the drug, but the take-home message is when resistance emerges in MCL, it’s really bad.”

OncLive®: What is generating excitement in the field of MCL?

Shah, who specializes in medical oncology at Moffit Cancer Center, discussed the current treatment paradigm of MCL, what therapies are moving through the pipeline at a rapid rate, the potential benefit with CAR T-cell therapy, and pivotal biomarker studies currently being conducted, during an interview at the 2017 OncLive® State of the Science Summit on Hematologic Malignancies. Shah: When I think about MCL, the most complicated patient I see now is the ibrutinib-resistant patient, particularly those who develop ibrutinib resistance early. These individuals tend to have extraordinarily aggressive lymphoma, and these lymphomas tend to be very chemotherapy resistant. I’m going to cite data from Moffitt Cancer Center. For our patients, the median survival when someone progresses on ibrutinib is only about 2 months. About three-fourths of the group who die within 2 months, die within the first 2 or 3 weeks, just to give some insight of what we’re running up against.

We have been profiling patients with MCL to try to understand what drugs work and what do not. We have a sophisticated assay, in which we are finding that almost no drug works, which is frightening. It is what we see clinically, as well. How are we going to succeed in this space? It is no question—when ibrutinib works, it can work wonders. But when it stops working, we have a real problem.

This is where CAR T-cell therapy has a very important role to play. We have treated a number of patients, and the inclusion criteria for patients on the Kite Pharma trial is that they must be ibrutinib-intolerant or resistant. What we are seeing is incredible. I don’t know when Kite Pharma is going to present the data, but I will simply say that we are seeing some truly incredible results and outcomes. Therefore, I am very excited about CAR T cells as they enter the MCL space.

What about other novel therapies on the horizon? I think it’s a done deal: we are going to see venetoclax in MCL. A 75% overall response rate on a phase I trial is unheard of. These are the kind of response rates on ibrutinib that got us excited about ibrutinib. This is not something that can be ignored.

It should be stated that none of the patients were ibrutinib-resistant on that trial. However, to see that magnitude of response tells you, very clearly, that there is a signal here. We are already in the process of generating more data in the space.

Ibrutinib plus venetoclax in MCL [is] something I would pay attention to. Yes, we are going to see more myeloid toxicity with that regimen, but we are significantly prolonging that event-free survival. It tells you the answer: that we have moved on to something important. That is another important agent.

What other combinations are being looked at other than ibrutinib/venetoclax?

Will transplant forever have a role in this paradigm?

We are still trying to understand how to use the ErbB4 inhibitors in this space; that will be another important class of agents in MCL. We will see how that evolves hopefully. Carfilzomib (Kyprolis) is starting to make its way into MCL. The University of Texas MD Anderson Cancer Center has a few clinical trials. I don’t know the status of them, but we are recognizing that this is a drug we need to keep an eye on. Acalabrutinib, the second-generation BTK inhibitor, is also one we need to keep an eye on. It is funny—wherever there is a target, some novel PI3K-delta inhibitors are showing considerable promise. There is a novel one from TG Therapeutics that blocks the Wnt/beta-catenin pathway. It is important in MCL—I will say that. The ability to block that pathway may yield more durable benefits than we see in PI3K-delta inhibitors, such as idelalisib (Zydelig). Those are all promising. The next-generation CD20 antibodies are something to keep an eye on. I have never been a big believer in them; that said, particularly for those who are developing rituximab resistance, there may really be an opportunity for some of these novel CD20 agents in this space.Absolutely not. I am not a believer in autologous transplant for all patients with MCL. There are a group that it benefits; ironically, that group is patients with lower-risk MCL. This is the same group who might benefit, regardless of what you give them. It is not to say that it is wrong to do. We don’t have a good threshold for when to stop rituximab maintenance, where patients get rituximab until progression. Some patients are still getting it every 2 months, even 5 years out from their initial therapy.

Are any types of biomarker studies being done?

For my patients who have a complete remission after frontline therapy, I tell them, “It’s up to you.” Transplant certainly can provide a durable benefit for some patients and it’s something to think about. If you don’t want the transplant, we are seeing incredible responses with maintenance rituximab. We will have a more definitive answer from an Alliance study, where we are going to be formally asking the question: rituximab or transplant, independent of frontline therapy? That will be really cool. At Moffitt Cancer Center, we want to understand drug response, particularly in relapsed/refractory ibrutinib-resistant cases, where we are desperate to find a hit. Essentially, we know the outcomes are very poor. What we do at Moffitt Cancer Center is we will layer MCL on stroma. These are stromal cells that function very similarly to patient stroma.

We have done all of these analyses. Importantly, they can sustain the MCL survival for well over 1 week; in some cases, 14 days or longer. We can sustain them for 14 days in culture primary cells, so we can begin to answer questions about what drugs work. Now, asking these questions has traditionally been answered by MTT assays, in which you basically take cell survival at a given time point. It is a little challenging to distinguish stroma from the mantle cell, and you ask, broadly, what is the IC50 value? At what concentration do you kill out the cells?

We didn’t like that concept. Part of the reason is because I find that IC50 in a 48- to 72-hour time point out to be very misleading. With the model that we use, we actually have been able to capture cell motion as a surrogate of viability. We can do a live, in vivo image capture over a period of 5 to 14 days. But, we can do these extended in vivo capture imaging in the setting of a drug. Why is that relevant? We capture something that others may or may not see with their traditional assays.

It gives us a better signal. Whatever assay you use, we can turn around on a layer the response data to the sequencing data. Now, we don’t have to guess where we see a variant of unknown significant FGFR14. We can say, “Okay, we see this variant and we see response to a drug that emits FGFR. We have a hit. It is not a guessing game anymore—or less of a guessing game.”

People are using patient-derived xenograft (PDX) models in much the same way. For us, the PDX approach was cumbersome. It takes a long time to generate and it’s expensive—to put it bluntly—and it’s hard to take data from a PDX model and apply it to a patient sitting in front of you. It takes too long.

Importantly, what we have found is that it seems to replicate what we’re seeing in the PDX models—at least the early data. We will publish on this once we have more data. The early data show a pretty good correlation. It shouldn’t be that surprising, because we have been using this assay in myeloma for 2 or 3 years now.

What about other biomarkers?

As a biomarker, this is a something to watch. We are not the only ones doing these kinds of tests, but we have been able to generate with clinical data. It is pretty remarkable and that is something I am very excited about. I am doing it at Moffitt Cancer Center, so I am very biased—but it’s something I am excited about. You heard me allude to sequencing. Sequencing is going to play a bigger role in MCL. What you are seeing emerge from some of the trials in Europe, now that we have some mutational data, is that, “Wait a minute, transplant doesn’t benefit patients with TP53 mutations, which can occur in a good 25% or so of the MCLs that we see.” It is not going to work as well in those with CDK, 2A, and 2B mutations. What you see us doing as we peel back the mutation data is, unfortunately, they work well in low-risk disease. Asking chemotherapy to work in the same space, in a disease that really biologically is defined by impaired DNA-damage response, is asking a lot.

Therefore, sequencing is going to play a big role in helping us understand who to give what therapies to. Clinically, right now, when I see a TP53 MCL, my first go-to [drug] is lenalidomide (Revlimid). I am not going to try chemotherapy in that patient. I already know what I am going to see at the end of the day.

The next big biomarker to talk about in line with sequencing is going to be the clonal minimal residual disease (MRD) approach to assessing depth of remission. We are now seeing pathologic complete response results emerge from large studies I have seen in the last 3 years. There is no question that when we see MRD emerge, it predicts for relapse down the line. You can prolong the time before they relapse. In the maintenance rituximab trials, it was 18 months.

We talk about rituximab maintenance not having an endpoint in sight, but maybe we do have an endpoint. Perhaps MRD-negativity is an endpoint. As long as we are vigilant and we do MRD-testing 6 months later, we can watch these individuals without having them come back every 2 months for another intravenous administration of rituximab.