Thank you to all of you. This has been a rich and informative discussion. Before we conclude, I’d like to get final thoughts from each of you. John, let’s start with you.

This is a really exciting time. There’s been an unprecedented opportunity to have access to novel therapies in the relapsed setting—CAR [chimeric antigen receptor] T cell, tafasitamab-lenalidomide, selinexor. Most of us have been through a couple of decades without access to any new therapies in this space. It’s a challenge for oncologists around the world to learn how to use these drugs. It’s tough to get experience with all of these. But staying on top of all the data and being able to use these drugs effectively is important when treating patients with large B-cell lymphoma. There is a lot of hope now in this disease. 

I second that thought from John. There’s a lot of excitement, a lot of novel agents. We just touched on bispecifics. There are many others in the pipeline, including novel targets like TAK-981. There are many other agents rolling out. I hope that in next several years, as Kami mentioned, we will be able to find out which patients will benefit from the treatment—a more personalized approach. Large B-cell lymphoma is a heterogeneous disease. We will probably have to define those characteristics of poor prognostic markers or responses to the treatment based on the patient characteristics or disease characteristics.

A lot of work needs to be done. Also, there are a lot of learning curves. Managing adverse effects of a treatment is the best way to actually get the most out of a treatment. If you don’t handle the adverse effect well, the treatment will fail. A humongous task going forward will be educating the community. But there is a lot of opportunity as well as excitement. I hope that down the line the outcomes of patients with large B-cell lymphoma will get better and better.

As I mentioned, this is an exciting time. We’ve made remarkable progress, and there are a lot of promising agents out there. As we move forward and focus on efficacious agents, it’s also important that we look at safety, access, and cost of these agents. These are barriers that we face, even with the effective therapies that we have.

That’s definitely important. Loretta, I’ll give you the last word.

One thing that I can take away from 2020 is that science can really solve some major problems. When you remove the red tape, you can get things done very efficiently. One thing we’ve learned in the last year is that by leveraging real-world data, you can address some really critical questions and even utilize some of that data to influence some of the decisions made by the FDA. This provides us some opportunity, moving forward, to be more creative with how we conduct trials so we can get answers quicker and potentially shape the treatment landscape more efficiently.

Important words. Thank you. To our viewing audience, we hope you found this 

® discussion to be useful and informative.

Videos

Evolving Therapies in R/R DLBCL: Concluding Remarks

Kami, let’s ask you to look into your crystal ball. In solid tumors, the checkpoint inhibitors, the immunotherapies are often the mainstay of treatment for many diseases. Today R-CHOP [rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone], as we started the conversation discussing, is really our most common therapy for patients with diffuse large B-cell lymphoma. But we do have exciting data with these other forms of immunotherapy. Do you foresee a day when we’re abandoning chemotherapy? If that is possible, what would it take for us to get to that point?

The first thing to recognize is the progress that we have made in the last handful of years in relapsed/refractory diffuse large B-cell lymphoma. Previously, if you were going to get to autotransplant, you didn’t have curable disease and you didn’t have effective outcomes. Now we have CAR [chimeric antigen receptor] T-cell therapy for the patients who can’t get to autotransplant or relapsed autotransplant, and we have bendamustine and rituximab with polatuzumab-tafasitamab-lenalidomide, which have all improved outcomes in patients who otherwise would get palliative therapy.

Moving forward, when you look at the role of CAR T, I do think that this is going to be used earlier in therapy in a large number of patients. There will be select patients who relapse late and are going to benefit from chemotherapy and transplant. But as Loretta mentioned, a lot more are going to be getting CAR T at first relapse. And for those high-risk patients—whether we define them as not achieving an early response to CHOP [cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone]–based chemotherapy by either a PET [positron emission tomography] scan or circulating tumor DNA—I think we will be moving on to CAR-T earlier.

In regard to CAR T as primary therapy, based on the way it’s given now, it’s hard to imagine using it as frontline therapy. Just considering the logistics of giving this therapy, patients with aggressive large B-cell lymphoma can’t wait 3 to 4 to 6 weeks to get their first line of therapy, right? So that’s saying that there’s not going to be a role, especially in the frontline setting. We’re a far way off from that.

I do think about things like Jason Westin’s SMART START study that looked at the role of limited therapy and nonchemotherapy in select patients leading in. Then they used a normal course of chemotherapy. In select patients, they were able to reduce the amount of chemotherapy, potentially not giving it in some patients. I think that’s in the future. 

We’ve come to recognize that large B-cell lymphoma is very heterogeneous, and these patients cannot all receive 1 treatment. We’ve looked at therapies directed at these different subsets of patients. At the same time, when you look at the therapies that have been approved—CAR T, bendamustine and rituximab with polatuzumab-tafasitamab-lenalidomide—these are active across different subsets of patients. So looking at incorporating some of these treatments into the management of patients and using them as combinations may be beneficial to these patients.

It sounds as if there is a lot of opportunity out there, and it is a problem. I guess it’s something to strive for.

R/R DLBCL: A Glance at Using Novel Therapies Moving Forward

We’re going to switch gears and talk about new therapies that are in the pipeline but not yet FDA approved. To me, 1 of the most exciting classes of agents are these bispecific antibodies, Loretta. There are multiple options making their way through the development process. Epcoritamab was recently published on, and others as well. Are you working with these agents? What’s your take?

At ASH [American Society of Hematology Annual Meeting] this year, particularly, there was a lot of excitement surrounding these bispecific antibodies that are targeting tumor antigens, specifically CD20, and then engaging the T cells with CD3 binding. As you mentioned, a number of these were presented. They are being studied in third-line space for large B-cell lymphoma in addition to some other B-cell lymphoma subtypes.

The efficacy looks quite promising with these as single agents. There have been attempts to try to mitigate some of the cytokine-release syndrome [CRS], which appears to be a class effect. In my experience, it’s not the same cytokine-release syndrome we see with CAR [chimeric antigen receptor] T-cell therapy. This is earlier in onset but shorter in duration, and it usually doesn’t progress to severe situations. It is usually grade 1 or 2. It looks a lot more like an infusion reaction, and for that we’ve had many years of experience with rituximab-based approaches. Epcoritamab is a subcutaneous administration, and you have this blunted absorption that leads to less potential cytokine-release syndrome.

There have been strategies with glofitamab, such as debulking patients with obinutuzumab a week before introducing the agent. All of them have looked at a ramp-up dosing schedule where you start with a very low dose at day 1, dose a little higher at day 8, and then finally give the full dose by day 15. With that strategy we’ve seen that you can mitigate some of the CRS. That will lend itself to outpatient administration because most of these studies thus far required admission and monitoring for the first 24 to 72 hours. To me, that is a nonstarter if we’re trying to get this out of the academic centers and into the community setting. I think we can accomplish this with some of these strategies.

The other potential limitation is this ramp-up. For these aggressive-behaving large B-cell lymphoma patients, sometimes you can’t get them to full dose before they progress. So I think these are very promising. We’re still learning how to administer them, but I think they’re going to have a foothold in the treatment of B-cell lymphomas.

Amit, do you think this is going to replace CAR T-cell therapy, or be a challenger against CAR T?

This will definitely challenge CAR T. As Loretta mentioned, most of the studies are single-agent trials. We are not seeing combinations yet. We are not seeing a combination with chemotherapy yet. We have seen that they’re all rapidly escalating in the frontline setting in combination with CHOP [cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone], and they are randomizing them head-to-head with R-CHOP [rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone]. There’s a lot of excitement. In the next 5 years, I hope that we will have a regimen that can beat R-CHOP [rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone] in the frontline setting. Definitely, this will compete with CAR T. As Loretta mentioned, this will surely be a game changer for community physicians. 

One other aspect that I want to mention is that as we have more and more CAR T patients, post–CAR T relapses are very common. Especially at academic centers, we are seeing post–CAR T relapses become an emerging area, in regard to unmet needs. In 2019 at ASH, we had a plenary session where we reviewed data on mosunetuzumab. This drug showed great responses in those patients who progressed after CAR T. So it’s going to challenge using CAR T, but at the same time, for those patients who are progressing through CAR T, this will be a great option.

Yeah. I confess that it wasn’t obvious to me that a bispecific antibody was going to work well after CAR T, but I was impressed and happy to see those data that you mentioned. 

John, let’s get your take on this. Even if the results weren’t quite as good as they could be with a CAR T-cell therapy, if you were able to apply it to many more people, the ultimate number of patients you’re helping is going to be greater. But there are obstacles to doing this in the community. Are you participating in these trials, or do you foresee being able to administer it in your clinic or your partners’ clinics?

That’s a great question, and I have several comments. Looking at the data, do they really approximate what you get with CAR T-cell therapy? Many of these are dose-escalation phase 1 studies. From what was initially reported at ASH, they would report some of the higher-dose levels of these products now. The overall response rate is in the ballpark of two-thirds for most of these, and the CR [complete response] rates are in the ballpark of 30% to 50%. They’re treating 10 to 15 patients at these target full doses with those kinds of numbers. If those data are confirmed with more patients treated and with longer follow-up, this could potentially not be a threat to CAR-T but could be another alternative to CAR-T, especially for the folks who might be more frail, might have a harder time traveling from a community practice to a tertiary center to receive CAR T-cell therapy.

As was mentioned before, it’s not always necessarily a direct threat. Amit talked about combination studies with chemotherapy. This might be more of a threat to rituximab, if it gets moved earlier in the course of disease in large B-cell lymphoma. And CAR T-cell therapy is still there. I don’t think CAR T-cell therapy is going away. I don’t think this is going to replace CAR T-cell therapy. Certainly, you can imagine these being used in different settings, and the same with CAR T. You can imagine both of the therapies being used earlier.

Can community doctors do this? I think the answer is yes, probably. I have a Mosun [mosunetuzumab] trial available, although I have been restricted in my ability to treat large B-cell lymphoma patients. Earlier, the Mosun trials required hospitalization for large B-cell lymphoma because of concerns about CNS [central nervous system] toxicity, driving restrictions and things like that. But as was mentioned, there’s a lot of work going on in terms of trying to mitigate some of the CRS and some of the neurotoxicity with the ramp-up dosing. There is even talk about steroid use and other things that might be able to mitigate that risk and allow these treatments to be delivered in the outpatient setting. Even if it does require a hospitalization or 2 for a couple of days, it’s not to the extent of what has to be done with axi-cel [axicabtagene ciloleucel]. I do foresee that this will be deliverable by community practitioners and will have an impact on patients.

R/R DLBCL: Novel Therapies in the Pipeline

Unfortunately, not all patients are candidates for CAR [chimeric antigen receptor] T-cell therapy. Maybe this is because of comorbidities, maybe it is because of access issues. Amit, can you walk us through some of the considerations that go into selecting a therapy for a patient who’s not a candidate for CAR T-cell therapy?

As I mentioned, identifying the patient for any therapy is important, including CAR T. If we decide that the patient is not eligible for CAR T treatment, and if they have failed 1 or 2 lines of therapy, we luckily have many therapies approved. Of course, we are to be in alignment with the label so that we can get the insurance approval.

As we discussed, BR [bendamustine, rituximab]–polatuzumab is an option, tafasitamab-lenalidomide is an option, and selinexor is an option. Tafasitamab-lenalidomide is approved in the second-line setting. Those patients are not eligible for transplant or aggressive therapy. Tafasitamab-lenalidomide is a good combination. As I mentioned, the management of lenalidomide and dose of lenalidomide is critical in that setting. I’ve used also polatuzumab with Rituxan alone, rather than adding bendamustine. But if you are planning for CAR T down the line and have a collection plan, I would not use a bendamustine-based therapy. If the patient is not eligible, that’s a great regimen to use. I’ve seen great responses with BR [bendamustine, rituximab]–polatuzumab.

We touched on selinexor a little. Managing the toxicity profile of selinexor is key. As I mentioned, I used the drug in a patient with CNS [central nervous system] involvement and had successful outcomes. That could be a niche in that area.

With all these agents, we have to be very careful with their adverse effects, especially bendamustine. As we all know, these patients may have received a couple of lines of therapy before. Bendamustine has been very lymphotoxic. Maybe the patient is at high risk of infections, especially in the COVID-19 [coronavirus disease 2019] era. We were talking about CAR T and long-term adverse effects of CAR T—hypogammaglobulinemia and high risk of infections. That’s something that we should be mindful of when managing these patients.

John, 1 thing we didn’t hear Amit talk about was the time from last therapy. For instance, the patient who is primary refractory. Does that weigh in on your decisions as to which therapy you might use for a patient? For instance, another chemotherapy vs an immunotherapy? What would sway your vote on what to give a patient in the primary-refractory setting? 

We all know those patients have the worst outcomes in this disease. Historically, I have used standard-of-care chemotherapy—second-line salvage therapy with ICE [ifosfamide, carboplatin, etoposide] or DHAP [dexamethasone, cytarabine, cisplatin] or something—and I try to get them lined up for autotransplant. Should they be receiving a targeted agent like tafasitamab-lenalidomide instead? That’s a really good question. I don’t know. I don’t have data to answer that question. It certainly seems that it would be a reasonable thing to try. 

That said, if you have a patient lined up for an autotransplant, we really don’t have super data about tafasitamab-lenalidomide in that patient population. I would probably do what I’ve been doing, knowing that the outcomes are suboptimal. Those are the patients I would line up for CAR T-cell therapy sooner rather than later.

Loretta, let’s get your thoughts on this. For the people who have been referred to you and are primary refractory and not going to make it to CAR T-cell therapy, what would you use as treatment? Are you going back to cytotoxic chemotherapy?

Traditionally, that’s what I’ve done. I’m holding out optimism that the randomized studies are going to be positive and that I’ll have CAR T in the second-line setting. But what will still happen is I’ll probably cool them off with a platinum-based approach while they’re undergoing manufacturing. I don’t know that my practice is going to change dramatically just yet. I have used tafasitamab-lenalidomide in that setting with the idea that if I switch the mechanism of action, particularly in patients who are over 70 years of age, I may be able to change them from this poor trajectory. 

I have limited experience, and I want Kami’s input. They excluded a lot of these patients from that study. I’ve not had good success in taking a patient who is rapidly progressing off CHOP [cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone] and salvaging them with tafasitamab-lenalidomide, but that is a potential option and is something I intend to explore further. We need good therapies in that space, and it’s traditionally been an area that drug companies have not gone after. I’m glad that studies like L-MIND and the randomized phase 3 studies are opening up that second-line space. I still think that’s an area where we don’t have a great strategy, particularly for chemotherapy-refractory patients in the frontline setting.

Though the primary-refractory patients were excluded from L-MIND, there were a handful of patients who were treated on the study before the inclusion changed. When you look at the data, there were better overall responses in the nonrefractory or nonprimary-refractory patients, but duration of response wasn’t great. This has been approved only since July, but I have been using the treatment in this exact situation, and I use it if they don’t respond to R-CHOP [rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone].

I’ve had a few patients who have done really well. The biggest thing is getting lenalidomide approved. Once you get it approved, there’s almost always a co-pay that nobody can afford, and you still have to get assistance. The biggest thing is trying to turn that around fast enough that I can start them on treatment. When I have been able to do that, I’ve actually had reasonably good luck. I have done a short course of treatment in some patients to get them there if I’m close but not quite there, and occasionally I have had to use a cycle of chemotherapy. 

Treating Primary Refractory DLBCL

I want to dive deeper into the discussion about bridging therapy. We hear that the people who are the worst off probably get bridging therapy because you can’t hold them. Is there 1 therapy that you use over another because you think, 1, they’re going to respond to it; or 2, it’s not going to cause problems for you down the line in terms of adverse events; it’s going to be better? Or is every patient just different?

There are a few ways to look at it. First, do they need bridging therapy if they have symptomatic disease? If they have rapidly progressing disease, they’re probably going to need it. If it’s something you can hold off on, it’s better not to do it.

Looking at when to give it, how do you define bridging therapy? Do you define it as before cell collection, like getting them to the actual CAR [chimeric antigen receptor] T consult, or the apheresis of the cells? Or do you define it from when they collect the cells to being able to give it? 

If you’re thinking about giving them something to be able to get to the collection, you probably want to avoid things like bendamustine or BR [bendamustine, rituximab]–polatuzumab, so maybe use polatuzumab alone. Whereas if giving something after they’ve already collected those cells, maybe a cycle of something like that is OK.

There is a role for radiation here. If you have 1 area that’s more problematic or symptomatic than others, it seems that in non–GC [germinal center] patients, I’ll think about using ibrutinib or something like that as a bridging therapy if possible. This depends on whether it’s before or after collection.

You brought up ibrutinib. Loretta, there were data this year at ASH [American Society of Hematology Annual Meeting] suggesting that ibrutinib might be helpful in certain situations, and we’ve seen the data previously reviewed in other disease states, like CLL [chronic lymphocytic leukemia]. It’s not that easy to get for patients with large B-cell lymphoma, but are you using ibrutinib routinely or in select patients?

Not routinely. I agree with Kami’s comments. All of us are trying to figure out ways to optimize using the therapy. We know that T-cell fitness may lead to better efficacy. We also saw glimpses of that with the ZUMA-12 trial, where you see more robust expansion if utilizing the drug in patients who’d had only 2 cycles of chemotherapy. As you mentioned, that area under the curve didn’t translate into worse safety profiles. So any strategy that might impact T-cell fitness is worth exploring. Where we struggle with interpreting the data that currently exist is none of it is randomized. We don’t know how much of an impact there is. I agree that I’ve been underwhelmed when trying to get it and stabilize a patient with large B-cell lymphoma. But it’s still worth exploring.

Bridging Therapy in R/R DLBCL

As I was listening to you, I was almost depressed. The 2 of you remarked on all the obstacles associated with putting a patient on CAR T-cell therapy. But we have to remember that prior to the introduction of CAR [chimeric antigen receptor] T-cell therapy, all these patients would die. Now we have a therapy that has long-lasting duration of remission and cure, and 40% of patients are eligible for these trials.

Loretta, that makes you wonder, right? If it’s so great a therapy in patients who have relapsed after an autologous transplant or are refractory-ineligible—they never make it to a transplant—why are we transplanting people? Maybe we should just be going directly to CAR T-cell therapy.

We have 3 randomized studies that are asking that same question. There’s strong enough rationale, at least in the third-line setting, to go head-to-head with transplant. It is defining the population that’s most likely to fail transplant, and those patients progressed within 12 months of frontline rituximab and CHOP [cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone]–like treatment, so I have a lot of optimism that those studies will be positive. If they are, then transplant will probably disappear.

I was also disheartened by all the frustrations that were voiced, but I agree that all statements are absolutely correct. The systems are not designed well for CAR T, but that provides 2 opportunities. One is to revisit how we do things. We’re not going to be able to address the insurance approval. You have to have a champion on both sides, and I oftentimes service that champion on this side. I am more than happy to push a patient through the system if I know about them. This has forced us to create these relationships where we share cell phone numbers and emails with folks on the front line, because it’s much more likely to go quicker if you know someone on the inside who can push that patient through. Otherwise, there’s no reason to even continue to pursue CAR T, hearing what we just heard. We should be focusing all our effort on trying to find more effective, more readily available therapy. 

 That’s a great point, and I would emphasize the importance of both, right? We have this technique, CAR T, that’s going to save lives. We have to do everything we can to maximize the utilization of the technique, as we’ve outlined.

But at the same time, those barriers are still there and it’s hard to imagine snapping your fingers and having somebody go from my clinic to a tertiary center the next day. And so at the same time we need to broaden the scope of the availability of CAR T-cell therapy. We need to deliver it more safely. We need to expand where it can be delivered. And we need to find other therapies that can achieve the same degree of success, that don’t require so much effort to make to happen. 

 Maybe we should also be refining our understanding of who’s most likely to benefit and who’s likely to get the adverse events associated with that treatment.

Kami, do we know anything about who’s likely to get cytokine-release syndrome [CRS]? Who’s likely to get neurotoxicity? Does tumor bulk matter? Do prior therapies matter? How can we improve outcomes, or at least our patient selection?

We know that patients with rapidly progressing disease, high LDH [lactate dehydrogenase] levels and high CRP [C-reactive protein] are more likely to have increased toxicity. Looking at the role of early referral, as everybody has mentioned, is key. Identifying patients who will benefit is key. When I look at a patient who is not a candidate for transplant, but I still think they’re a candidate for CAR T, I’m going to refer them early for CAR T. Even though I don’t think they’re a candidate for transplant, I say, “Oh, I’m giving them something.” And then I look at the role of bridging therapy, which is something that I don’t think everybody should receive. We need to decide who is going to benefit from bridging therapy and what that bridging therapy is.

 Do you have to have active disease, Kami, to get CAR T-cell therapy? I’ve been in the fortunate situation a few times where you didn’t think a patient is going to respond to anything. But I collected the CAR T cells, gave them bridging therapy, and all the disease went away. Should I hold that therapy? There’s a notion that you need a target to go after. There are some data on this from some of the trials. It’s not a common situation, but what are your thoughts?

We have had some patients who have at least by PET [positron emission tomography] scan gone into a surprised CR [complete response] and went ahead with CAR T-cell therapy. Whether they’re still in remission because they were going to be in remission with or without that, I guess I can’t really speak to.

R/R DLBCL: Transplants in the Era of CAR-T 

John, as a community oncologist, what are the barriers you experience related to getting patients treated with CAR [chimeric antigen receptor] T cells? We know from the centers that provide them, some of the challenges they have, but can you comment on your experience and the obstacles to actually getting someone treated—very practical real-world issues?

Yeah, there are a lot of them, and this is really important. As was mentioned before, a lot of patients who could be benefiting from this therapy in real life probably don’t get access to it and don’t receive it, which is too bad because it really does potentially offer better outcomes. 

What are those barriers? One of them is that the disease can move too quickly. It takes a while for me, as community doctor, to see the CT scan or PET [positron emission tomography] scan that shows a recurrence, meet with the patient, discuss the possibility of CAR T-cell therapy among others, and put in a referral. The referral has to get sent by a staff member to the receiving tertiary care center. A patient has to wait for an appointment to get in to see the specialist. The patient may have to travel. I live in a big city, but many patients live 2, 3, 4 hours away from big cities. That’s another barrier. The pandemic is not making travel any easier. That’s another barrier that has occurred.

The other thing I’ve seen is insurance authorization barriers. I saw this especially early on, when CAR T first became available. “Yeah, we have this product available, but it’s going to take a few weeks to get the insurance authorization before we can even more forward and do apheresis.” That’s going to take time and will delay treatment, and many people may not have the luxury of waiting.

The high cost of the treatment is also a barrier. The high costs and physician and hospital concerns about getting reimbursed vs losing their shirts is a barrier. The need for inpatient treatment is a barrier. Some patients, especially older ones who may not be candidates for stem cell transplant, may not want to spend a couple of weeks in the hospital. That has been a barrier for some of my patients. I also think patients’ concerns about potential toxicities and safety can pose a barrier. Some folks really don’t want to go through therapy based on some of those risks.

There are a lot of barriers, and I’m probably forgetting many. A lot of things get in the way of patients seeking out and successfully receiving these treatments. The take-home point for practitioners in the community is to really make your referrals early. Your patient doesn’t have to be ready for apheresis. If you’re waiting that long to make your referral for CAR T, you’re waiting too long. You’ve got to recognize that there are going to be barriers that will delay the apheresis by many weeks. You want to get your patient in the hands of a specialist well before that, so they have time to get all those ducks in a row before the treatment begins.

That’s really great advice. Amit, John just spoke on behalf of the referring end. Perhaps you can talk on behalf of the receiving end? What obstacles do you see? If you could reduce some of these obstacles, what would you suggest?

I totally agree with John. Earlier referral is key. If the patient is referred late, we are struggling to get all the ducks in a row. One of the bigger challenges on the receiving side, even after insurance approval, which doesn’t happen right away, is that most of the time I do peer-to-peer talks and provide all the evidence and articles and whatnot. That gives us an approval.

One thing I’ve learned during the last year is that approval is not sufficient. The cost of the therapy is so high, and somehow that contract is not under the existing contract with the institution. This means we have to go into a separate agreement with the insurance company, what we call a case-based agreement. The higher-ups in the institution get involved with the insurance company. So that is an additional step to be sure that the therapy is covered and the cost of the therapy will be paid to the institution. Without that, the institution will not move forward with using that almost $0.5 million treatment in that setting.

Once that is done, which also takes an additional week to 10 days, that’s when the scheduling starts. I’ll tell you that the leukapheresis and collection of cells, even though it seems very easy, still requires a sync between the institution and the company, or the commercial side of that. We have to have a slot on their side as well as ours. That’s an additional delay.

I totally agree with John. COVID-19 [coronavirus disease 2019] has changed many things. Even if the patient needs a line, they need to have a COVID test and need to be negative within 3 days. The team doing this may be very busy because of the pandemic. You may not get an appointment right away. These are all barriers. Apart from that, remember that these patients are progressing very fast. So you’re also struggling to control their disease.

There are so many challenges, especially in the era of COVID-19. If I have 1 successful CAR T treatment, I’ll sleep better.

CAR T for R/R DLBCL: Improving Access and Utilization

John, I’d like to talk a little about the safety of CAR [chimeric antigen receptor] T-cell therapies. The safety is getting better regarding the products we have as well as the next-generation therapies that are going through clinical trials. We see at least 2 major adverse events associated with CAR T-cell therapies: cytokine release syndrome [CRS] and neurotoxicities. Is this getting better? If it is, why is it getting better from when they were originally approved?

I haven’t delivered these treatments. Not being a CAR T-cell doctor, I can look at the literature and what people are saying and reporting. To me it looks like the incidence of cytokine release syndrome is lower with liso-cel [lisocabtagene maraleucel] compared with the others. Tisa-cel [tisagenlecleucel] is in between. The timing is different. CRS occurs a bit later after liso-cel [lisocabtagene maraleucel]. That potentially makes that particular product capable of being delivered in the outpatient setting. Patients could get the treatment and then wait several days. If they have a problem, they can get admitted later.

That’s exactly what the study that Loretta mentioned earlier showed at ASH [American Society of Hematology Annual Meeting]. This could be done safely, at a slightly wider number of centers and type of centers, compared with in the past. The newer product, liso-cel [lisocabtagene maraleucel], looks to have a lower toxicity level. That’s what other people who have used all the products have indicated to me. Whether it’s the 4-1BB or the 1:1 ratio of CD4 to CD8 T cells, or the patient population, those may be factors in regard to the differences.

 Kami, do you think we’ve made progress in the adverse-event profiles of the products that we have? Some of the trials have suggested that later on in the trials the adverse-event incidence rates went down. In your hands, at your center, do you feel like we’re in a better place than we were a year or 2 years ago in terms of managing adverse events? If so, why is that true?

Probably for a couple of reasons. One reason is the ability to recognize the CRS and the neurotoxicity earlier. Then the treatment approaches: when and how to use steroids, tocilizumab, and intrathecal chemotherapy, even with neurotoxicity. But also recognizing when it’s occurring and then the treatment for it and instituting that treatment earlier on has helped improve the toxicity profiles as we’ve gotten more experience.

 Does that cut into efficacy if you’re intervening early with steroids? When CAR T cells were first being developed, there was this great fear of giving people steroids. There was concern that you’d be killing the T cells, killing the efficacy. Do you think that affects efficacy?

We’ve had a little bit of retrospective data on this. I know Loretta probably can speak to it best, because the data comes out of her institution The University of Texas MD Anderson Cancer Center, but there was some suggestion that maybe there could be a little change. Overall, it probably doesn’t largely affect them.

I agree. I think the message from our retrospective study is that you have to be careful about the dose and duration of therapy. We’ve gotten much more comfortable intervening with corticosteroids earlier on in the course of toxicity. You should still be mindful of having patients on huge doses for prolonged periods of time. That’s probably not a good thing.

CAR T for R/R DLBCL: Safety/Tolerability

Amit, there are some other differences. We talked about the time to give the therapy once you’ve acquired the T cells. There is some variability. It appears that the vein-to-vein time is quicker for some products. Regarding the patients, are there differences that change this vein-to-vein time? Are T cells easier to produce with 1 product or another, or 1 patient vs another?

We still need more data. There are reports. In my practice, there are very few chances for failure to produce a product. That has been published on. Failure ranges from 1% to 2%. However, it also depends on what kind of treatment they received before. As you mentioned, if you have used more lymphotoxic regimens before the leukapheresis because a patient is progressing very fast, there is a chance that they may not be able to produce or collect enough T cells. 

The other factor is that most of the centers that are certified to treat with CAR [chimeric antigen receptor] T are also part of the expanded-access program. What I mean is that most of these studies have shown that cell viability of more than 80% would successfully hold up to the results and outcomes as published. However, some patients with even lower viability, like 70%, also held responses. By default, the way FDA has approved the therapy option, cell viability needs to be 80% or more. Those patients who don’t meet that criteria or other specifications are enrolled in a different protocol and still have significant or almost similar responses. That will also be helpful for patients who are not able to have viability cutoff for FDA.

John, in your mind what is a reasonable amount of time, regarding the struggle of acquiring the CAR T cells? What is a reasonable amount of time to wait for these CAR T cells to come back and get infused? These patients are often very sick and have very aggressive disease. How do you manage them?

Some of these patients have very rapidly progressing disease. The reasonable amount of time depends on the patient. Some patients can’t wait the 2½ to 3 weeks that it generally takes to manufacture the product, so they need this bridging therapy sooner—something to control the disease before they get treated. For others, they may have time. You catch the relapse relatively early, and they’re not in dire straits at the time you’ve picked this up, so 3, 4, 5 weeks may not be a problem. It depends on the individual, regarding how much time is reasonable.

Understanding the CAR-T Process: Collection Through Infusion 

Loretta, you’ve published in the area of real-world data, and there are multiple publications looking at these different products. One of the arguments that came out when ZUMA-1 was published with axi-cel [axicabtagene ciloleucel] was, “You can never re-create those patient populations.” There was such a bias stating that the results of people who went on the trial wouldn’t reflect what’s going to happen when this becomes commercially available. Were those skeptics? Is this validated with the real-world data, or is it better than that?

I always pause and recognize that it’s not the real world, right? If we were looking at the real world, we’d look at every large B-cell lymphoma patient out there, not just the ones who are treated at The University of Texas MD Anderson Cancer Center and other centers where we collaborated with folks who had experience on ZUMA-1. If you take patients who would not have qualified for ZUMA-1 for various reasons and look to see what their outcomes are, across the board we have shown that you can reproduce the findings of ZUMA-1 and JULIET. We’re all eager to see if we can reproduce the findings of TRANSCEND-NHL-001 once we collect real-world experience at the centers that have experience.

There were a number of patients who got standard-of-care products who never would have qualified for those prospective studies because they were too frail, too sick. That is reassuring to me. Despite the population being skewed in that direction, we don’t see differences in safety particularly. We can reproduce the efficacy. I do think that provides much more ammunition to that argument that this is just a boutique therapy that’s only going to be offered in select centers. I think it should be offered to every relapsed/refractory large B-cell lymphoma patient. Where we’ve struggled is to get more access to patients. We’re just looking at the folks who get through our door. There’s a huge portion who never get there, and part of it is logistics and the perception around the toxicity.

You asked this question to Kami. In my experience, there is clearly a difference in the toxicity profile, at least in the first 30 days. There might be financial reasons why a center may choose 1 construct over the other. How much of that translates into efficacy? Probably very little. It’s more in terms of when you expect that fever to occur, can you give it outpatient vs inpatient? But we have shown that there will be differences in the patient population, depending on the construct, because of that safety perception. But the efficacy looks to be just like what we saw in these prospective studies.

You brought up some of the barriers regarding why not as many patients get treated with this as should. Some of that has to do with access, financial barriers, social barriers, and such. There are also issues with delivering the therapy once you’ve acquired the cells, Loretta. Some products take longer to give than others. How does that factor into your decisions about which product to give? Some data suggest that bridging therapy is not your friend, right? It doesn’t necessarily improve outcomes. It may be necessary, but it doesn’t necessarily improve outcomes. How does that factor into things?

As we get more experience, we’ll get much better at this. It still boils down to patient selection. And not surprising to me, anytime you have to utilize bridging therapy it’s because the patient has a much worse prognosis, much worse disease status. And so you don’t have the luxury of waiting during that 3- to 4-week window while you’re undergoing manufacturing. The other aspect is we have to be stewards of our practices. This is super expensive and requires a huge demand of health care resource utilization. There will be differences in terms of which construct you might select based on where you practice as well.

That forces us to revisit how we do this, in terms of not only inpatient vs outpatient but also our networks. Can you broaden this out? We saw data at ASH [American Society of Hematology Annual Meeting] looking at centers that are not traditional university settings but for big groups like Sarah Cannon, like US Oncology. It can safely be done in those settings as well. That just provides reassurance to me that as we get more experience, we’re probably getting better at the delivery of this therapy.

At the end of the day, what matters most to patients is, are you going to have successful manufacturing, and are you going to be able to live long enough to get this therapy? That falls on us, as physicians, to do a better job of selecting patients. As more treatment options become available, not every patient is a CAR [chimeric antigen receptor] T patient. Some of these other therapies, if they’re just as effective, might take up some of that space.

Treating R/R DLBCL With CAR-T: Current Barriers 

Kami, there are now 2 FDA-approved products for relapsed large B-cell lymphoma, and there’s another 1 that’s going through the process, hopefully to get approved. The trials are different, in regard to how patients went on them, but there are some very real differences in the constructs of these therapies. Do you think they’re different? Do you use different therapies for different patients? Will you use one CAR T-cell therapy for 1 patient vs another? Can you discuss some of the differences and similarities among these?

Yes. The general structure is similar. They differ in their combinations of the transmembrane and costimulatory domain. All 3 are a bit different in that regard. When you look at the trial populations, they all included relapsed/refractory diffuse large B-cell transformed lymphoma, high-grade lymphoma. Axi-cel [axicabtagene ciloleucel] also included primary mediastinal patients, as I believe Amit Mehta mentioned. Liso-cel [lisocabtagene maraleucel] included primary mediastinal patients and grade 3B follicular lymphoma patients.

The axi-cel trial did not allow bridging therapy, whereas the tisa-cel [tisagenlecleucel] and liso-cel [lisocabtagene maraleucel] trials did allow bridging therapy. So there’s a little bit of a difference there.

Of the 2 approved products, axi-cel [axicabtagene ciloleucel] is typically given as an inpatient therapy, whereas tisa-cel [tisagenlecleucel] can be given as an outpatient therapy.

In regard to the liso-cel [lisocabtagene maraleucel] product, when you look at the general patient population they didn’t seem to treat as robust of a patient population. Almost 40% or a little higher of the population was over age 65, whereas with the other 2, about a quarter of patients were over age 65. So it’s an older population. The cutoffs for both, as Loretta mentioned, from renal function and cardiac function were lower than in the other trials. There were differences in the patient populations, as you mentioned.

Only axi-cel [axicabtagene ciloleucel] and tisa-cel [tisagenlecleucel] are approved. At our institution, it seems that we tend to use it in the older population.

Are you standardly using just 1, or do you use both?

OK. We’re going to get Loretta’s take on the real-world data in a second, but do you think the CD28 construct has a different adverse-event profile compared with a 4-1BB product, or are those just the differences we’re seeing in the trial—different time, different patient population—and maybe that explains it? What is your experience treating people with a commercial product off-trial?

It varies. Based on the patient population and the patient you’re treating, there’s probably a difference. There’s some difference in the product but also the patient’s disease status and underlying comorbidities, at least from what we’ve seen.

The major question that comes up a lot, at least in the marketing wars between trying to differentiate 1 from another: Does persistence matter? What’s your take on this? Some of them have a different area under the curve, for lack of a better term. Do you think that is a difference, or are the results the results, and it doesn’t really matter how you got them?

I think there is some benefit to the persistence.

CAR-T Treatment Options for R/R DLBCL

Amit, I want to get back to you. Let’s switch gears and focus on CAR [chimeric antigen receptor] T cells. In which patients are you using CAR T cells on a routine basis?

That’s a great question. Both of the products that are approved and the 1 that is going to be approved soon are approved in second-line and beyond settings. Patients have to have failed 2 lines of treatment to be considered for CAR T. As we were discussing about frontline treatment of large B-cell lymphoma, those who are at high risk, high IPI [International Prognostic Index] patients, or double hit, double expressors are the patients who tend to relapse early. If I start a treatment in a patient with double-hit lymphoma, they are on my radar to consider for CAR T down the line. But currently, the FDA indication is a failure of 2 lines of treatment.

The only difference between the tisa-cel [tisagenlecleucel] and axi-cel [axicabtagene ciloleucel] is that axi-cel [axicabtagene ciloleucel] allows for primary mediastinal B-cell lymphoma because those patients were part of the study. That’s the only difference in the label. A high-risk patient that I’m starting treatment on is on my radar for the CAR T treatment down the line. 

Also, it is important to know how they are relapsing. We cannot put every lymphoma case in the same bucket. Some patients have a low burden of disease, some have a high burden of disease, some also have CNS [central nervous system] involvement, for some their LDH [lactate dehydrogenase] is high and the velocity of their progression is high. This is important because CAR T is a process, right? You have a collection, and then you have, depending on the product, 2 to 3 weeks before that product will be ready. We have to see, in that perspective, whether the patient can be held, so to speak, during that time. If they explode between the collection and the infusion, they cannot successfully get the CAR T. All those factors are important.

Also, like transplant, it is important that they have social support. For 8 weeks, maybe they cannot drive. All those are factored in when we consider using CAR T-cell therapy. It also depends on the patient’s insurance status, whether they have good social support, and how their financial systems are. Those all factor into decisions regarding use of CAR T.

 Loretta, Amit just nicely outlined a few of the issues, but there are some real issues in regard to delivering CAR T cells to patients, from financial barriers to logistical issues. You hear some people say, “If he couldn’t get a transplant, there’s no way he’s going to get CAR T cells.” Do you find that to be true? What are your thoughts on that?

It’s grossly underutilized because the logistics are challenging. It requires referral, and it requires commitment from the patient and family members. That being said, when you have a patient who you know would be dead and now they’re cured with this treatment—this is a onetime treatment, though I recognize it includes 6 weeks weeks leading up to it and another several months post infusion where you’re monitoring and adjusting things—it is really rewarding when we see it work.

Where I struggle is, can you actually optimize patients who you’re concerned about? I see the discussion often happening in referrals to me when hospice is offered vs CAR T. I don’t think it should be utilized in that setting where it’s a last-ditch effort. The logistics are so challenging, just as Amit mentioned, so you need a patient who has disease stability while you’re undergoing manufacturing, at least for these auto products. But I still think it should be considered, at least at some point in the disease course for any patient with relapsed/refractory large B-cell lymphoma. You cannot utilize the same criteria we apply to identify transplant candidates.

 So you don’t have a litmus test. You’re not requiring every patient to have an echocardiogram and have an EF [ejection fraction] of 50%, or pulmonary function tests, or something like that, is that correct?

You have to use your physician eyeball test that you already have heard about. You have to be able to predict that they’re going to tolerate the therapy if they have severe CRS [cytokine release syndrome] or severe neurotoxicity. You have to predict that they’re going to have the pulmonary, cardiac, renal, and hepatic reserve to get through that. But it’s not the same stringent criteria we utilize to identify who’s going to tolerate high-dose chemotherapy. You don’t have the whole notion of, “They have to have chemotherapy-sensitive disease—step 1. Step 2, make sure they don’t have comorbidities that would predict for bad outcomes.” That part is still very fuzzy to me because we looked at ejection fractions of less than 50%, and we looked at renal insufficiency and liver insufficiency. So far none of those has held up as being a really strong factor associated with either safety or efficacy.

CAR T-Cell Therapy for R/R DLBCL

 John, I want to go back to you and talk about some of the other therapies that have recently been approved for relapsed large B-cell lymphoma. First, let’s talk about polatuzumab. Polatuzumab was approved in combination with bendamustine-rituximab. What has your experience been here, and how are your using this regimen?

Just to summarize, polatuzumab is an anti-CD79b antibody-drug conjugate linked to MMAE [monomethyl auristatin E], the same drug that’s in brentuximab vedotin. Data supporting its use comes from a randomized phase 2 trial, a relatively small trial of 80 patients, where the addition of polatuzumab to bendamustine and rituximab improved many efficacy outcomes, including complete response rates, progression-free survival, and overall survival. To my knowledge, this is still, since the PARMA study decades ago, the only randomized trial in relapsed large B-cell lymphoma that demonstrates a survival advantage with 1 regimen over another. This was a pretty significant leap forward.

Even bendamustine-rituximab can lead to considerable toxicities in these patients. The polatuzumab–BR [bendamustine-rituximab] combination can cause neutropenia, and polatuzumab is also associated with peripheral neuropathy. 

I have found this to be a useful regimen for a similar patient population relapsing after an autotransplant. I even used it before CAR [chimeric antigen receptor] T-cell therapy was around. And I’ve used it as a second-line therapy in patients not eligible. 

Just to elaborate on that, the FDA approval for this combination is for treatment in the third-line setting or beyond. But the study included second-line patients. In fact, those patients, in a subset analysis, had better outcomes. I’ve used it as either second- or third-line therapy. I have not had any trouble getting it covered by insurance, even in the second-line setting. That’s where I’ve used it.

 Thanks. You talked about using this prior to the introduction of CAR T cells. It’s attractive as a bridging therapy for patients who require it, that were acquired CAR T cells. But I certainly worry about a T-cell–depleting agent like that and giving it prior to acquiring the leukapheresis product. It’s an important new therapy for patients. I’ve been very impressed with its efficacy.

Kami, what about the selinexor? This is an oral agent, and it’s attractive because of that. How are you using that in your practice?

I’ll be honest, I have not used this in my practice. Where I see a role for this is based on it being an oral therapy. In regard to toxicities, we see cytopenia, some GI [gastrointestinal] toxicity, fatigue. It does require some work, as far as the premedications you use. However, it seems tolerable with those premedications. To me this is something that if you have a patient who relapses after CAR T-cell therapy, isn’t eligible for a trial, wants to try something, or maybe if you have an older patient who is not eligible for transplant, is not eligible for CAR T, and wants to try something, it’s an oral therapy. As I said, it has toxicity, but that can be managed. That’s really where I see it being used. I have not personally used it.

 You’re not using it in the second-line setting mostly because the results weren’t as impressive as the results with BR [bendamustine-rituximab]–polatuzumab or other agents, I assume. Is that right?

Yeah. I would say that in the second-line setting, for somebody who’s not going to be eligible for a curative therapy, I’m using tafasitamab-lenalidomide; or prior to that, if I could get access to it, as John said, BR [bendamustine-rituximab]–polatuzumab.

I was just going to add 1 thing, Ian. The trial with selinexor required at least 2 prior lines of therapy. It hasn’t been well studied as a second-line therapy. As was pointed out, the responses are probably less impressive than what we’ve seen with some of the other drugs that we’re mentioning that have become available in the last couple of years for relapsed disease. Moving forward, selinexor may potentially be combinable with chemotherapy or other drugs. That may really be the future of this drug—to combine it with some of the other therapies that are out there to see if they can improve efficacy. There is personal hope for this drug, in regard to having more utility.

One more thing on that. There were a few case reports of CNS [central nervous system] penetration with selinexor. I’ve seen responses in CNS involvement. There are also studies ongoing. So that’s an extra niche for this drug.

 Great, we’ll look forward to the results of that trial that you mentioned.

Novel Therapies for R/R DLBCL

Amit, have you started using this regimen, tafasitamab, lenalidomide, in your practice? Maybe I’ll get a few opinions on this, actually. What patient population are you using the regimen in? What are your thoughts?

This is a very effective regimen, especially for those patients who are not eligible for transplant, as approved based on the L-MIND study. As I was mentioning, in the frontline setting for frail, older patients who are not candidates for aggressive therapy, and similarly in the relapsed setting, those patients are perfect for this combination.

And as Dr Kami Maddocks mentioned, the key with this combination is management of lenalidomide adverse effects. If you manage those, you will be able to successfully implement and get a successful outcome in your patients who are frail. I’ve been using this regimen. As I mentioned, the dose and management of lenalidomide’s adverse effects is key in these patients.

 John, I want to get your opinion on this. I’ve also heard Loretta speak eloquently on this in the past. In regard to sequencing, should we just be using this regimen in older and infirm patients? Some people worry about using a CD19 antibody in patients for whom you’re considering trying to get them to CAR [chimeric antigen receptor] T cells. There are a lot of theoretical reasons, probably less data. Maybe you could walk us through how you’re using this regimen in your practice. Is it similar to Amit, or are you doing things differently?

I’ve used it only once in an older patient who was frail. Her prior therapies included R-CHOP [rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone], or mini-CHOP [cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine, prednisone], and then POLA-BR [polatuzumab vedotin, bendamustine, rituximab]. She had been through both of those regimens and then had another relapse. It came down to, “Well, she’s not an autologous transplant candidate.” We also decided she was probably not a CAR T candidate for various reasons. So we had excluded using curative therapies, as Loretta had alluded to before. You know, “Is this someone we’re going to really try to cure?” We felt she was too frail because of her comorbidities.

That’s an example of where I’ve used it. In this particular case, it didn’t go well. She was probably even too frail for tafasitamab and lenalidomide. She got pretty weak and was tired after just a couple of weeks, and wound up having various unrelated medical complications. But that’s an example of how I would attempt to use it. It’s approved for second-line therapy, but it’s not something you want to use in someone who’s going to go through an autotransplant. 

Is there value in using this regimen prior to CAR T? Is there harm in using this regimen prior to CAR T? Is the anti-CD19 antibody removing CD19? Do you lose CD19 expression, thereby rendering a subsequent CAR T product less effective?

We don’t have great data on that. There are some in vitro data out of the Mayo Clinic suggesting that probably does not happen. There are also a couple of case reports of patients who have been successfully treated with 1 CD19 product followed by another. The best we can tell is it doesn’t seem to be a problem, but I’m not a CAR T-cell doctor. I actually don’t know what they think about this. I suspect they have some theoretical concerns, as you alluded to.

Obviously we have a lot of enthusiasm surrounding the potential cure with CAR T, recognizing it’s a small portion. It’s probably about 40% of patients. There is concern that if you’re introducing a CD19-directed therapy based on what we’ve observed in ALL [acute lymphoblastic leukemia], maybe you’re going to select out for CD19-negative clones.

However, that is not supported by any evidence. It’s theoretical, and we’ve looked at outcomes among patients who received CD19-directed standard-of-care CAR T. It didn’t matter what their CD19 expression level was prior to receipt of that therapy, whether it was determined by IHC [immunohistochemistry] or flow and not even known. But that was also in patients for whom the vast majority had never seen a CD19-directed therapy.

What we really need is more real-world experience, where people have seen these agents and they have been sequenced in various ways. I suspect we’ll have tafasitamab prior to CD19-directed CAR T-cell therapy, and vice versa. 

As John mentioned, in the small amount of data that currently exist, it does not appear that introducing tafasitamab prior to CD19-directed CAR T-cell therapy is going to impact that legacy.

R/R DLBCL: Treating With Tafasitamab-Lenalidomide

All the regimens we’re talking about are relatively toxic, right? Giving some R-ICE [rituximab-ifosfamide, carboplatin, etoposide] is not for everybody. That probably speaks to the need for newer therapies that are less harsh on patients and more tolerable. Loretta, do you have any thoughts on that? There are some new regimens recently approved, but in general, trying to give people a less toxic therapy with fewer adverse events would certainly be important.

I 100% agree. Even before I head down that path of choosing my salvage chemotherapy, I have to make a determination regarding whether I think I’m going to cure the patient—that’s going to entail more intensive therapy—or pursue a palliative approach because the toxicities associated with our traditional regimens have been so great. There’s probably a large number of patients who are having fewer of these intensive approaches because we make the determination they’re not going to tolerate it. There was a huge unmet need in terms of having tolerable but effective second-line strategies. What fills that space is tafasitamab and lenalidomide, because the investigators defined the folks they did not feel were suitable for an intensive approach, such as consolidation with high-dose therapy.

More study needs to be done to help us now that we have more options. We need to learn who the optimal candidates are for these less-intensive approaches. But we have treatment options that we didn’t have a year ago.

Yes, let’s dive down on that a little, Kami. Last summer there was the approval of the tafasitamab-lenalidomide regimen for patients with previously treated diffuse large B-cell lymphoma. You’ve done a lot of work with this regimen. I was wondering if you could help us understand some of the clinical results and the rationale for the trial.

Yes. Tafasitamab is a CD-19 antibody. In a phase 2 study, there were some responses with the single agent in relapsed/refractory diffuse large B-cell lymphoma. The few patients who did respond and achieve a complete response [CR] actually had a fairly long duration of response. That was the rationale for then combining it with lenalidomide.

L-MIND was a phase 2 study that treated 80 patients who, as Loretta mentioned, were considered not eligible for aggressive therapy and autologous stem cell transplant. Patients were treated for a year with combination tafasitamab and lenalidomide. The first 3 months were pretty aggressive. They were treated with weekly infusions, then they were treated bimonthly. After a year, patients were able to go on to maintenance therapy with the antibody tafasitamab alone if they were responding. Overall, the therapy was well tolerated. It’s similar to what you see with lenalidomide in terms of cytopenias, fatigue, rash.

In the phase 2 study there was a 60% overall response rate, with 40% of those patients achieving a complete remission. And in those patients who achieved a CR, the responses were pretty durable. Then in the summer, with longer-term follow-up, we saw that the median duration of response was almost 3 years, which is pretty impressive in this population of patients.

Tafasitamab Combined With Lenalidomide in R/R DLBCL

Amit, let’s turn to you. Can you walk us through the recommended treatment options for second-line therapy? Kami just talked about a trial looking at CAR [chimeric antigen receptor T-cell therapy, but we have transplant in that setting. What is your standard of care?

I think the standard of care is still a consolidated bone marrow transplant for patients who are chemotherapy-sensitive. If you look at the recent data, even for patients who had a partial response, if you pursue the transplant, they still have a progression-free survival at 3 years of about 40%. In those patients, the first question I ask is whether they are a candidate for transplant. I always say that beauty lies in the eyes of the beholder. You cannot make a decision on your own. I always like to have a consultation with my transplant colleagues. If they don’t have any red flags that this patient is absolutely not eligible for transplant, the plan is to consider salvage chemotherapy: R-ICE [rituximab, ifosfamide, carboplatin, etoposide], R-DHAP [rituximab, dexamethasone, cytarabine, cisplatin], or gemcitabine–based treatment. None of them, except for R-ICE [rituximab, ifosfamide, carboplatin, etoposide] and R-DHAP [rituximab, dexamethasone, cytarabine, cisplatin], which were compared head-to-head, turn out to be similar in terms of response rates.

When investigators looked at the GCB [germinal center B-cell] and ABC [activated B-cell] subtype, there was a subgroup analysis where the GCB subtype had a bad response with R-DHAP [rituximab, dexamethasone, cytarabine, cisplatin]. So in some patients, if they can handle cytarabine, that is my favorite approach. Once they get salvage treatment, if they respond, if there is a response on the PET [positron emission tomography] scan, then I consider consolidated autologous stem cell transplantation.

I agree with everybody else. At the academic centers, these are the patients, especially primary-refractory patients, who are an area of need. They may qualify for multiple clinical trials. When ZUMA-7 was up and running, that was a trial for which we would consider these patients; it randomized patients between salvage transplant versus axi-cel [axicabtagene ciloleucel]. There are many others also looking at that approach. If the patient qualifies, we would probably consider enrolling him or her on the trial.

John, Amit just went through a variety of salvage regimens, but the question comes up, not infrequently: Should we be using a CD20 [antibody] at all? If we do, should we be switching to a different anti-CD20? Should we switch from rituximab to obinutuzumab? How are you approaching this? If you decided that you did want to do it, is it even feasible? Can you get that approved?

I haven’t personally made the switch to obinutuzumab or ofatumumab or anything like that. Based on the GOYA study, where obinutuzumab was not better and did add a little toxicity in the frontline setting, it’s not a strategy that I have tried. Back many years ago when I was training, the group I was with at Memorial Sloan Kettering Cancer Center looked at the value of adding rituximab to ICE [ifosfamide, carboplatin, etoposide], and it really didn’t add much benefit. But I think most people probably still do it, myself included. If I have a patient with CD20-positive large B-cell lymphoma who relapses, I’ll generally add rituximab to the salvage regimen, whether it’s ICE [ifosfamide, carboplatin, etoposide] or DHAP [dexamethasone, cytarabine, cisplatin].

That’s been my practice. I don’t know the answer to your question, Ian, about using obinutuzumab or some other anti-CD20 antibody, whether it would be authorized. I don’t know.

R/R DLBCL: Second-Line Treatment Options

Thankfully many patients don’t relapse, but there is a group of people who do. Some have some of these high prognostic risk factors, some do not. Can you talk a little about what you do after frontline therapy? How do you approach that patient?

Fortunately, a good portion of patients do very well. We’re particularly nervous about identifying those who will not do well. And so, prior to initiation of treatment, we’re always trying to risk stratify patients and see if there’s something other than CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] chemotherapy. We utilize PET [positron emission tomography] scans in the midst of therapy, with all of the caveats in terms of the positive predictive value and how important the interim PET is. But we’re always looking for any sign that a patient is going to fail.

A small portion of patients fail in the midst of CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone]-like therapy. It’s a larger portion who fail within the first 12 months of treatment. Another small percentage will have late relapses. My approach to those 3 different scenarios is different. We saw some data emerge at ASH [the American Society of Hematology 2020 annual meeting] utilizing an interim PET in these predefined high-risk patients—those who had double-hit features and high IPI [international prognostic index score] who had an interim PET after 2 cycles of CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] or CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone]-like therapy. If they failed to achieve a Deauville score of 1 to 3, they may be eligible for intervention with something like axi-cel [axicabtagene ciloleucel], or CAR [chimeric antigen receptor] T-cell therapy.

Again, there are caveats to that. These are patients who were treated at centers that had access to CAR T-cell therapy. These are patients who had a Deauville score of 4, and maybe they would have gone on to do well. Nevertheless, it identifies how we might utilize the tools at our disposal to find or identify those high-risk patients for whom we might start down a path separate from more intensification of chemotherapy. But again, I recognize that there is a portion of patients who do very well with R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone], and we’re stuck with surveillance imaging.

I think the other technology that might emerge and might be useful in this setting is cell-free DNA. This could potentially help identify patients who are going to fail outside of our traditional imaging approaches such as PET and CT scans. We saw some differentiating data at ASH 2020 with the adaptive approach not being an effective strategy, but maybe the CAPP-Seq [cancer personalized profiling by deep sequencing] cell-free DNA is a strategy that might be employed. I think it’s too soon to decide what the preferred method is, but I do think there is optimism surrounding that strategy to identify high-risk patients, maybe even before they fail on a PET scan.

I think the other thing that I don’t put as much emphasis on in the relapsed setting is cell of origin, or even double-hit cases. I have not adopted an approach for which I might approach them differently. I just know that if they fail early, they do poorly. If they fail late, they might do better.

 Kami, to follow up on what Loretta was saying, do you find that to be the same? Let’s take 2 equivalent patients, or 2 patients who relapse 9 months after completing R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone]. I think Loretta was saying it really doesn’t matter at that point whether they were a GCB [germinal center B-cell] or non-GCB patient. The approach is going to be relatively the same. Is that what you do at The Ohio State University Wexner Medical Center? What do you do for that patient?

I completely agree with that. The biggest indicator is going to be if they’re primary refractory or have an early relapse. We have been fortunate enough to currently have and had the prior trial looking at sending patients to CAR T. Patients are randomized to standard of care with salvage chemotherapy and autologous stem cell transplant versus early CAR T. I try to get patients on that study because I think this may be a population that benefits from that.

I think that is best for those patients. Outside of a study, we don’t have a different practice for our standard of care for these patients. Hopefully that will change.

DLBCL: Treating Patients Following Relapse on Frontline Therapy

One of the patient groups that I find most challenging to treat are patients who either have central nervous system [CNS] involvement at the time of diagnosis or those for whom you’re worried about it. They may have risk factors. Some people will use the CNS-IPI [central nervous system international prognostic index] to try to risk stratify patients. Can you tell us about your approach for patients who are either at high risk or already have it? Does that change how you treat patients? Is your work-up different? Perhaps walk us through some of your approaches.

At initial diagnosis to look at CNS risk, I look at double-hit status, double-expressor status, and then use the CNS-IPI, which is the IPI plus adrenal and renal. For patients who are high risk, CNS-IPI 4 to 6, and then double-hit and double expressor, I typically tend to offer CNS prophylaxis. This can be done with both intrathecal or intravenous [IV] methotrexate. If somebody is going to be able to tolerate it, my preference is to use the IV methotrexate. There’s some suggestion that that’s probably a better preventive, although we saw data at ASH [the American Society of Hematology 2020 annual meeting] that kind of questioned the benefit of all prophylaxis. But I usually will use that for 3 cycles. If somebody has CNS and systemic disease, we’ll sometimes use it with the R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone]. So, we will give it on days 1 and 2. We also sometimes will fold it in, giving it on day 15, especially if there are questions of tolerability. Or occasionally, we’ll use it at the end of treatment.

What kind of doses are you using, out of curiosity? Say you gave it on day 15, are you giving 4 g, 6 g?

Definitely a challenging patient population. Loretta, what about the double expressors? Are you doing anything different for that patient population?

I echo all of the comments that we’ve heard. One of the things that we’ve learned from some of these prospective studies that have failed miserably is if you have too many barriers to identify these poor-risk patients, it sways them toward standard of care treatments like R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone]. You don’t have the luxury of time to pursue a trial or something more novel. And so, for the double expressors we have some potential advantages, and you usually can determine that by IHC [immunohistochemistry], and you should have that information when you’re first seeing the patient.

I think we probably do a better job of enriching our prospective studies with the double expressors. As of this point, I’m still using R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone] if I can’t get them on trial, and I have some disappointment or lower expectations in terms of outcomes. I have not pursued anything else. I know there are data suggesting that dose-adjusted EPOCH-R [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, and rituximab] might be beneficial. When we look at the randomized study, it’s not powered to address this particular cohort. It’s disappointing. We don’t really see a potential benefit there.

The other thing I’ll comment on is the Mayo Clinic group did define this time from diagnosis to initiation of treatment as being another prognostic factor. When we start thinking about how we might design those next iterations of studies, if we can shorten that time by either implementing tools that can identify these poor-risk patients that are more real-time and more readily available, or enroll them into more of a standard of care approach. And then at cycle 2, maybe distinguishing with our experimental intervention. I think we’re going to have to revisit how we do studies, or we’re going to continue to have these negative phase 3 trials.

DLBCL: Treatment Considerations for Challenging Populations

®, “The Evolving Treatment Landscape in Relapsed and Refractory Diffuse Large B-Cell Lymphoma.”

I’m Dr Ian Flinn, from the Sarah Cannon Research Institute in Nashville, Tennessee. Joining me today in this discussion are my colleagues: Dr John Burke, from the Rocky Mountain Cancer Centers in Aurora, Colorado; Dr Kami Maddocks, from The Ohio State University Wexner Medical Center in Columbus, Ohio; Dr Amitkumar Mehta, from the University of Alabama Birmingham Comprehensive Cancer Center [O’Neal Comprehensive Cancer Center at UAB]; and Dr Loretta Nastoupil, from The University of Texas MD Anderson Cancer Center in Houston, Texas.

Today we are going to discuss a number of topics pertaining to the use of novel agents in relapsed/refractory diffuse large B-cell lymphoma. We’ll discuss newly available agents, the latest research in the field, including from the ASH [American Society of Hematology] 2020 virtual meeting, and the impact of recent clinical trials on making decisions around treatment selection.

Let’s get started on our first topic. John, perhaps we’ll start with you. R-CHOP [rituximab with cyclophosphamide, doxorubicin, vincristine, and prednisone] is really the mainstay of our therapy for most patients, or for many patients with diffuse large B-cell lymphoma in the frontline setting. But under what circumstances would you use a different regimen, and why would that be?

I would say, Ian, there are probably a few circumstances I can think of when I might use regimens different from conventional R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone]. Examples would include to treat the so-called double-hit or triple-hit lymphomas. For those patients, my practice has generally been to use dose-adjusted EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin] with rituximab. Certainly, other regimens are reasonable as well in that high-risk subgroup. We know that those patients fare poorly with R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone]. We do not have randomized trials demonstrating improvement in outcome with any other regimens. We have a number of retrospective studies that suggest that some of the more aggressive regimens may lead to better outcomes in those patients who don’t do well with R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone]. One of those is the dose-adjusted EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin] regimen. Another is rituximab plus hyper-CVAD [cyclophosphamide, vincristine, doxorubicin, dexamethasone]. The third is rituximab plus CODOX-M/ IVAC [cyclophosphamide, vincristine, doxorubicin, methotrexate/ifosfamide, etoposide, cytarabine.

I find that subgroup to be challenging to really know what to do with. In many areas in our field, we don’t accept results of retrospective analyses as being the gold standard. I don’t think there really is a gold standard. Simply because of relatively unfavorable outcomes with R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone] and the fact that the best we have are these retrospective data, I think one of the more aggressive regimens is reasonable to use. I tend to use dose-adjusted EPOCH-R [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, rituximab] in my practice, which is a community practice. I can administer that treatment as an outpatient procedure without having to hospitalize patients. That’s a scenario for which I’ll use something different from R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone].

Another is in primary mediastinal large B-cell lymphoma. Most studies have suggested that while R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone] is acceptable, the risk of local recurrence may be higher without radiation. Patients with primary mediastinal large B-cell lymphoma are frequently younger women. Radiation may result in some long-term toxicities, including coronary disease and higher risk of breast cancer. The dose-adjusted EPOCH-R [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, rituximab] regimen has been shown to be a highly effective regimen without the use of radiation therapy. And again, in the nonrandomized trial very good results have been seen with that regimen. So I’ll choose the dose-adjusted EPOCH-R [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, rituximab] regimen in that group.

HIV-associated lymphoma might be another scenario for which one may consider using dose-adjusted EPOCH-R [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, rituximab], although R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone] is still acceptable. I don’t treat a lot of HIV-associated large B-cell lymphoma—I just don’t see a lot of it—but I have used dose-adjusted EPOCH-R [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, rituximab] in that group.

John, let me break in here and get a few other people on board here.

Amit, he mentioned several different groups. There are the double-hits, the primary mediastinals, HIV. What about other groups for you? Do you use any of the prognostic factors? We talked a little about FISH [fluorescence in situ hybridization] testing, looking at the double-hit cases, but are there other groups for which you might use something different?

All of the regimens use anthracycline, one of the important chemotherapies. There are 2 groups I can think of: One group is those patients who have a heart-related issue. For instance, if they have CHF [congestive heart failure]. In those patients, we cannot use anthracycline no matter what the risk factors are. In that case, we have to supplement anthracycline, classically we use doxorubicin, with something else.

There are many regimens that have been discussed. One of them is EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin] with a low dose of anthracycline we could use, or supplement gemcitabine in place of anthracycline. The other group that is an area of unmet need, so to speak, in my opinion, is the elderly population. These are patients who cannot handle full-dose CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] or EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin]. Consistently, if you look at the frontline study when they compared head-to-head R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone] with rituximab dose-adjusted EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin], the outcomes were similar, but the toxicity was higher in the EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin] group. So if you have an elderly patient who cannot handle full-dose CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] or dose-adjusted EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin], using mini-CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] or a reduced-dose of CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] has been my practice.

John M. Burke, MD, Rocky Mountain Cancer Centers

Articles