McKinney Compares Therapeutic Agents in Later-Line DLBCL

Matthew S. McKinney, MD

Cellular Therapy Specialist Hematologic Oncologist Hematologist (Malignant)
Duke Cancer Center/Duke Health

Durham, NC

Targeted Oncology^TM: What are the data for polatuzumab vedotin (Polivy) in DLBCL regarding potential treatment for this patient?

MCKINNEY: Polatuzumab is a CD79b-targeted antibody-drug conjugate, so it delivers a very potent spindle toxin to CD79b-positive large B cells, or lymphoma cells.¹ The FDA approval was achieved in June 2019 for polatuzumab plus BR [bendamustine (Bendeka), rituximab] based upon the results of a phase 2 study, GO29365 [NCT02257567].²

Most of this discussion is going to focus around the DLBCL cohort, and basically that was 80 patients. It’s a phase 2 design, so it wasn’t fully stratified in a phase 3, or powered to be a phase 3 trial.

It was bendamustine plus rituximab vs polatuzumab vedotin plus BR, cycles given every 3 weeks.2 The way this is given is a very similar schedule to giving standard BR. You give the bendamustine [90 mg/m² IV on days 2 and 3 of cycle 1 and then days 1 and 2 of subsequent cycles], and then rituximab [375 mg/m² on day 1 of each cycle], and polatuzumab [1.8 mg/kg IV on day 2 of cycle 1 and day 1 of subsequent cycles]. The primary end point was ORR [overall response rate], done at the end of 6 cycles of that treatment. In this phase 2 study, it was relapsed/refractory patients. They screened out patients with more than grade 1 neuropathy because polatuzumab vedotin can be associated with worsening neuropathy, and patients who had certain criteria around transformed disease, recent transplant, some of those things.

In terms of the patient characteristics, these were mainly older patients, relatively high risk by IPI, and late stage.² The median range of treatments was 2, so this was predominantly a third-line and beyond population. In terms of the response in this patient population, I think BR had been used in the palliative setting for a long time, and those results are what you’d expect, something like a 15% to 20% ORR...a little bit below what you’d deem an acceptable active ther- apy in third-line DLBCL. But when polatuzumab was added, that resulted in a 40% CR rate, with a very small percentage of patients having PR. Most of the responses were CR in the study. Again, this is also time-limited therapy, so it’s 6 cycles every 3 weeks; relatively easy to dose and give in a schedule patients with lymphoma are used to because we’ve been giving bendamustine for indolent lymphomas for many years now.

Progression-free survival [PFS] by IRC [independent review committee], as well as overall survival [OS]—it’s tricky to interpret the median numbers for PFS and the OS, because in this trial the CR was a little bit below 50%.² Estimating the median numbers may underestimate the fact that there may be durable responders that benefit; they’re right below the median. But clearly the addition of polatuzumab to BR resulted in significant increases in the median PFS, at 9.5 months vs 3.7 months. In this study, [there was an] OS of about a year vs [an OS] measured in months with just BR; again, thought to be a palliative measure, historically. There [have] been updates to these data, so consistent PFS and OS benefits... remain now, with more than 3 years of follow-up.

I think the biggest thing that we’ve experienced with dosing pola-BR in terms of adverse events [AEs] is hematologic toxicity.^1,2 Most of the time that’s neutropenia, so the rates of grade 3 to 4 neutropenia are significantly higher than they are with BR [alone]. I tend to start most of these patients—unless they have robust blood counts at the start of therapy—on growth factor support. The rate of either dose reduction, delays, or discontinuations in the phase 2 experience from AEs was about one-third in the pola-BR arm, whereas the BR alone was a less toxic therapy. Clearly the polatuzumab-based drug conjugate is adding efficacy at the risk of toxicity.

Please describe the safety concerns for polatuzumab in DLBCL.

One interesting thing I’ve experienced is that you certainly see the neutropenia, and sometimes the thrombocytopenia, but febrile neutropenia didn’t seem very much changed.^1,2 I think reviewing these data again drive that home. I don’t know the last time that I’ve seen a patient come to our inpatient service with febrile neutropenia from pola-BR. It is relatively unusual. But these patients are at risk of infections, and often you must delay therapy, or a patient gets 3 or 4 cycles in and you must stop therapy because they don’t have enough neutrophils to continue.

There was some other nonhematologic toxicity like diarrhea or nausea.^1,2 I think fatigue is sometimes a little bit more with the pola-BR vs BR alone. Usually, you don’t see too much peripheral neuropathy. In my practice, we haven’t seen too much severe peripheral neuropathy get uncovered with adding the polatuzumab vedotin. I think some of that is probably [because] we get more limited by the blood counts, and so you don’t unmask as much neuropathy.

What is the reasoning for using tafasitamab (Monjuvi) plus lenalidomide (Revlimid) in DLBCL?

Tafasitamab with lenalidomide is another novel agent combination now approved for DLBCL in the relapsed/refractory setting.^3,4 The rationale for this combination was the fact that, potentially, the lenalidomide combination could be leveraged with this compound MOR208, or tafasitamab. Tafasitamab is a CD19-directed monoclonal antibody. It has very specific modifications to encourage certain cell-related antibody-dependent cytotoxicity mechanisms, and also to encourage lymphoma cell death. CD19 is also a very robust tumor antigen that can be targeted in large cell lymphoma. Usually it’s pretty invariable, so CD20 you can often lose with targeting with CD20 antibodies. Usually, CD19 is a bit more consistent across the subtypes and treatment patterns of large cell lymphomas.

Tafasitamab also had encouraging single-agent activity, and relatively few [AEs] in some early-phase studies.⁴ The [use] of lenalidomide, which has already been used in mantle cell lymphoma, follicular lymphoma, marginal zone lymphoma, with FDA approvals reflective of that; the idea was that maybe you could leverage some of the effects with T-cell and natural killer–cell activation and get a synergistic drug combination. The phase 2 L-MIND trial [NCT02399085] data set led to the FDA accelerated approval in 2020.⁵

What were the design and patient characteristics of L-MIND?

This was a study done in relapsed/refractory DLBCL and had a bit different patient population than the phase 2 study with pola-BR. This was a patient population that could get as few as 1 prior regimen and was supposed to be a transplant or high-dose therapy–ineligible population. And then, there were some patients with primary refractory disease, and some of the higher-risk features that were excluded on the trial.

The FDA approval is the same [dosing] schedule as was done in the trial.^4,5 It’s a little bit tricky scheduling for the tafasitamab. The first week there’s...a loading dose at day 4, and then it goes to weekly, and then it goes to every other week treatments, basically done on an every 4 week lenalidomide-based cycle, with the maximal dose being the usual 25 mg/kg a day, given days 1 to 21 of a 28-day cycle. The trial dictated that patients could get a year of lenalidomide, and then patients who basically had clinical benefit or response to that combination, you could continue every other week tafasitamab, basically until [disease] progression. This is also not a time-limited therapy. The primary end point of the trial, similar to other trials in the relapsed/refractory setting that led to approval, is ORR.⁵

[Let’s review] the baseline patient characteristics. Compare these to the data for pola-BR. This was generally an older patient population, although there were some younger patients at 41 years.⁵ I’m not sure what makes a 41-year-old not a transplant candidate in the eyes of the trial, but [those are] the data. Most of these were more advanced stage.

I think the other key thing to remember in this study: It was basically a second-line relapsed/refractory DLBCL trial. The median lines of therapy [were] 2, whereas the other data sets were 3 or beyond. This is a less heavily pretreated population. That should be considered when thinking about the patient population you want to treat with this therapy. That also obviously drove what the FDA labeling is because it’s also approved as a second-line therapy in a transplant-ineligible population.⁴

What were the results for patients in the L-MIND study?
In terms of response rate, this is best response after almost several years of follow-up.⁵ Similar to pola-BR, there was a 40% CR rate. Additionally, in the L-MIND study with tafasitamab plus lenalidomide, there was an additional patient population that had both PR as well as stable disease for a long period. Very unusual to get progressive disease in terms of the best response; usually, patients get some clinical benefit somewhere in there. So, potentially, different kinetics from pola-BR. But again, it’s continuous scheduling with both agents for at least a year, and then indefinite tafasitamab, although I can tell you that some of the current studies are looking at alternative scheduling around the tafasitamab and different time points.

[There were] different potential features, or subgroup features, related to response.⁵ I think that many of the features that we think of as higher risk may be less important, or negated, or dealt with equipoise with tafasitamab/lenalidomide, such as high LDH going on study, high IPI, certain cell-of-origin phenotypes. I think you have to take some of these data with a grain of salt. Some of these subgroups were pretty small numbers of patients; 20 patients, pretty big confidence intervals for many of these groups. It’s still not clear until we have more, bigger data sets how some of these features may predict response. Also, as someone who’s done lymphoma genomics work in the past, it also seems like, as others reported, maybe we should look at whether there’s a cell-of-origin type basis to high or low responses with a lenalidomide combination. But nevertheless, this is what was found in the study.

What I do think is also encouraging is the duration of response reported in the L-MIND study.⁵ The key take-home point is that many of these patients in L-MIND did have very durable responses. Some of this may reflect the relative duration patients were on therapy. Again, they were on therapy indefinitely, with at least a CD19-based antibody, and then a year with lenalidomide. So, it’s a little bit different time frame, in terms of treating the patient and duration. Whether that extends either PFS or OS, we don’t know, but it is a lot of infusions, it’s a lot of oral therapy, and that may relate to patient-reported outcomes or having to come back and forth for infusion vs a more time-limited therapy. Nevertheless, a median OS of almost 3 years with this therapy makes it very attractive in many of these patients who may be older; they’re transplant or CAR [chimeric antigen receptor] T ineligible, but able to tolerate a basically outpatient regimen that many community oncologists would be comfortable managing.

What are the safety concerns for tafasitamab in DLBCL?
The other thing to talk about with L-MIND is hematologic AEs.⁵ Most of the AEs on this trial were related to the lenalidomide, and most of those being, as you’d expect, hematologic AEs: neutropenia, anemia, thrombocytopenia. Most of those were manageable, and I can tell you in my practice we’ve used usual guidelines of where we start off the lenalidomide, especially in the older patient. So, if you’re treating someone in their mid-80s, I usually will start at a lower dose because you know you’re perhaps going to have to dose-reduce. And that’s just a lesson from treating either other lymphomas, or multiple myeloma, with lenalidomide. Many may use as much lenalidomide in the myeloma space as I do over on the lymphoma space. Nonhematologic AEs—again, most of this is related to the lenalidomide administration, much of that being fatigue, diarrhea, cough, edema, presumably things like rash, infection, anorexia.

In terms of discontinuations or serious AEs, at least in the L-MIND trial, discontinuations of the combination were unusual, related to AEs.⁵ I think it’s interesting, because there was a high rate of serious AEs but they were not, at least by the investigators, reported to be around the study’s intervention, which seems to be a bit of a disconnect. The other part of the AE analysis is the timing of these. When the patients were on the tafasitamab/lenalidomide combination, that’s when they had the toxicity. With tafasitamab and lenalidomide, you see all the higher-grade neutropenia, hematologic toxicities, as you’d expect. When the lenalidomide comes off, very little toxicity in this monotherapy time frame of tafasitamab. Again, potentially, you won’t see the cytopenias, or gastrointestinal AEs, or infection risk. I wonder if many of these toxicities are residual from the combination therapy that patients were already on.

How does loncastuximab tesirine (Zynlonta) play a role in DLBCL?
Loncastuximab is a novel CD19 antibody-drug conjugate.⁶ It delivers a toxic payload into CD19-positive cells, like tafasitamab, except that it has a novel PBD [pyrrolobenzo-diaz-epine] linker, or DNA crosslinking agent, that gets delivered when the agent hits CD19-positive cells. The technology is very novel in terms of this agent....It wouldn’t cross-react with the spindle toxin that’s in things like brentuximab vedotin [Adcetris] or polatuzumab vedotin; it’s a separate, unique mechanism of action.

So ostensibly, there are patients who may respond to that strategy, whereas they may be resistant to potentially other agents. The FDA approval came in April 2021 in DLBCL after 2 or more lines of therapy.^6,7 There [are] additional indications for transformed lymphoma as well as high-grade B-cell lymphoma.

The data supporting loncastuximab’s approval come from the LOTIS-2 trial [NCT03589469].⁸ There were some phase 1 data in the LOTIS-1 trial [NCT02669017], and then it went on to LOTIS-2 with some of the dosing and premedication modifications. The way that LOTIS-2 was done, and the way that loncastuximab is advised to be dosed, is that it’s a 30-minute infusion given once every 3 weeks.^6,8 The first 2 cycles, patients get 150 μg/kg, a weight-based dosing, and they move to half of that dosing with subsequent cycles. LOTIS-2 treated patients on that every 3 weeks—again, very quick 30-minute infusion, very tolerable infusion schedule— for up to a year, and then the patients were followed every 3 months for up to 3 years on the trial. Similar to other relapsed/refractory DLBCL studies, it led to recent approvals. The primary end point was read out as ORR, but there’s been separate analysis around other secondary end points, including duration of response, CR rates, PFS, survival, and then safety measures.

Most of these patients were DLBCL NOS [not otherwise specified].⁸ The study contained a population of patients who had high-grade B-cell lymphomas. In terms of the diagnostic criteria, I assume the patients with double-hit and triple-hit lymphoma were counted under the DLBCL heading, but I don’t know how that was done considering the current World Health Organization [WHO] 2016 criteria. There was a population of patients who had primary mediastinal B-cell lymphoma; again, a small number, and those patients generally didn’t do very well with loncastuximab, so loncastuximab doesn’t have approval in that patient population. This was a third-line and beyond trial; many of these patients were quite heavily pretreated. There were patients who had prior autologous transplant, prior allogeneic transplant, prior CAR T-cell therapy. There were some patients who went on to subsequent therapy after loncastuximab treatment on trial.

Please discuss the efficacy data from LOTIS-2.

I think the way I remember the response rates is 50/50. Half of the patients on LOTIS-2 had a response, with half of those being CR rates, or roughly that.⁸ So, 24% and 24%, respectively.

The median time to first response was pretty quick, roughly 1 or 2 cycles of therapy. Most of the time that CRs were achieved, it was durable. Most of the [patients who] did respond had that response after 2 cycles, although there were some patients who had PR, then converted later to CR after a period of months of therapy. Some patients had ongoing responses. The mean number of loncastuximab cycles that were administered was 4.5; on average, patients get almost 5 cycles before they progress or have to come off therapy.

One of the keys is that, like other therapies in the DLBCL and high-grade lymphoma spaces, in patients who get a good response it tends to be very durable, which it was.⁸ Then it becomes more quickly apparent if a patient’s not going to respond and benefit from loncastuximab. All of that has to be balanced by the AEs and aspects around administration of the medication, however.

There is also some efficacy by subgroup. There seemed to be pretty good responses of roughly half of patients getting a PR or CR, even high risk; either transformed disease, or high-risk cell of origin, or high-grade B-cell lymphomas as well. Many of these patients had past stem cell transplant, past CAR T-cell therapies, and that didn’t seem to impact the ORR to any significant degree in this study.

The median PFS for all comers was what you’d expect in a therapy where the response rate is right around 50%, similar with OS.⁸ I still think OS is probably favorable, comparable with...some of the other [options in the] relapsed/refractory population. Remember, this is a single-agent therapy, single-arm trial.

The other thing that’s been questioned with loncastuximab and other CD19-based novel therapies that are not CAR T cells is the sequencing of these with CAR T. There are data from LOTIS-2 that have now been published as part of that experience. So, in LOTIS-2, there were 15 patients who went on to subsequent CD19-directed CAR T-cell therapy.⁸

The best response rates in that group were around 50%. It was a heterogeneous group; it was different CAR T products. Most were axi-cel [axicabtagene ciloleucel (Yescarta)], but some were tisa-cel [tisagenlecleucel (Kymriah)], and some were other experimental therapies. It’s a bit hard to assess those response rates. Nevertheless, it does seem that patients can progress on one CD19-based therapy, or at least loncastuximab, and then either go on to CAR T— there [are] some other data back and forth, there are some data out there with tafasitamab. Again, nothing ready for prime time. Several patients went on to stem cell transplant, mostly autologous transplant, as consolidation of response to a good response to loncastuximab in the relapsed/refractory setting.⁸ That’s the other option for some of these remarkable responder patients that have been seen.

I talked about some of the challenging subgroups. When you take out the primary mediastinal B-cell lymphoma, some of the numbers get even better.⁸ In high-grade B-cell lymphomas in LOTIS-2, there was a high rate of CR, which I think makes this potentially attractive in some of these tough cases. For example, I have a patient I don’t think is an autologous stem cell transplant candidate in the second line for high-grade B-cell lymphoma.

He’s not a great CAR T-cell candidate because of caregiver issues. He can generally only come every few weeks on a Friday for infusions, and so for someone like that, loncastuximab may look very attractive. Again, pretty quick response rates with that therapy, so in many of these patients with pretty aggressive disease, you can get a fairly quick response rate, move on with deepening the response rate and palliating their disease.

What was the toxicity seen with loncastuximab in DLBCL?

There is some discussion to be had with the safety results from LOTIS-2.⁸ There [are] some very specific AEs that were interesting from this study. [There were some interesting] AEs that led to study discontinuation. In the trial, GGTs [γ-glutamyltransferase] were routinely measured, and that was the most frequent thing that led to treatment discontinuation. Most of us, I think, in the United States, do not routinely check GGTs. There’s no label requirement to do that, and so as long as you’re OK with the other AEs, you can go ahead without checking GGTs.⁶ Like other cytotoxic [therapies] in the third-line and beyond setting, you run into a lot of hematologic toxicities.8 Most of the time, those are very manageable.

There are very clear labeling requirements for when to start and stop therapy, when to delay therapy, when to dose reduce, and those sorts of things.⁶ The other thing that came out as a unique AE with loncastuximab in this study was a very small incidence of edema.⁸ Most of that can be managed with a steroid premedication that’s built into the therapy; and also, things like furosemide [Lasix] and spironolactone [Aldactone] seem to work fairly well. The other issue with this agent is that patients can have photosensitivity, and so patients should be instructed to avoid sun exposure, they should cover up, wear broad-brimmed hats, sunscreen, all those sorts of things.

_REFERENCES

<sub>1. Polivy. Prescribing information. Genentech; 2020. Accessed March 30, 2022. https://bit.ly/3JZRr26
2. Sehn LH, Herrera AF, Flowers CR, et al. Polatuzumab vedotin in relapsed or refrac- tory diffuse large B-cell lymphoma. J Clin Oncol. 2020;38(2):155-165. doi:10.1200/ JCO.19.00172</sub>

<sub>3. FDA grants accelerated approval to tafasitamab-cxix for diffuse large B-cell lymphoma. FDA. Updated August 3, 2020. Accessed March 31, 2022. https://bit.ly/3bodb9v
4. Monjuvi. Prescribing information. MorphoSys US Inc; 2020. Accessed March 31, 2022. https://bit.ly/2S38ZW7</sub>

_{5. Salles G, Duell J, González Barca E, et al. Tafasitamab plus lenalidomide in relapsed or refractory diffuse large B-cell lymphoma (L-MIND): a multicentre, prospective, single-arm, phase 2 study. Lancet Oncol. 2020;21(7):978-988. doi:10.1016/S1470-2045(20)30225-4}

_{6. Zylonta. Prescribing information. ADC Therapeutics; 2021. Accessed March 31, 2022. https://bit.ly/3L4tsAs
7.FDA grants accelerated approval to loncastuximab tesirine-lpyl for large B-cell lymphoma. FDA. Updated April 23, 2021. Accessed March 31, 2022. https://bit. ly/3uWWZX9}

_{8. Caimi PF, Ai W, Alderuccio JP, et al. Loncastuximab tesirine in relapsed or refractory diffuse large B-cell lymphoma (LOTIS-2): a multicentre, open-label, single-arm, phase 2 trial. Lancet Oncol. 2021;22(6):790-800. doi:10.1016/S1470-2045(21)00139-X}