In an interview with <em>Targeted Oncology</em>, Gregorz S. Nowakowski, MD explained the details of the ECOG-ARIN 1412 trial and how the positive results related to patients with DLBCL will further improve outcomes for patients with all lymphoma types.
Gregorz S. Nowakowski, MD
The phase II ECOG-ACRIN 1412 trial showed improvement in progression-free survival (PFS) for newly diagnosed patients with diffuse large B-cell lymphoma (DLBCL) with the addition of lenalidomide (Revlimid) to R-CHOP (rituximab [Rituxan], cyclophosphamide, doxorubicin, vincristine, and prednisone), based on data reported during the International Conference on Malignant Lymphoma (ICML) 2019 .1
As a single-agent, PD-L1 inhibitor lenalidomide has shown activity in patients with relapsed or refractory (R/R) DLBCL, achieving a 30% response rate. However, patient responses were not durable with lenalidomide alone, according to study investigator, Gregorz S. Nowakowski, MD, hematologist/oncologist at Mayo Clinic.
Recent clinical trials have shown that lenalidomide plus R-CHOP (R2CHOP) is an effective combination for prolonging the duration of response in these patients. In the ROBUST trial (NCT02285062), the median PFS rate was 6.2 months.2This study helped provide the rationale for the study of R2CHOP for improvement of outcomes in patients with DLBCL.
In ECOG-ACRIN 1412 trial, patients with stage II bulky-IV disease, IPI ≥ 2, ECOG PS ≤ 2, and measurable disease were accrued regardless of the cell of origin. Patients were randomized 1:1 to receive either lenalidomide at 25 mg plus standard R-CHOP for a duration of 10 days or the R-CHOP alone for 6 cycles. Tissue biopsies were extracted to monitor progression of disease and outcomes in patients with activated-B-cell (ABC) DLBCL.1
The primary endpoint of the trial was PFS, and it had 2 secondary endpoints as well, including overall response rate (ORR), and overall survival.
Patients showed a better response to R2CHOP with an ORR of 97% and a complete response rate of 72%, compared to 92% and 67% for R-CHOP21. Additionally, using R2CHOP resulted in a 33% reduction in risk of disease progression or death compared to R-CHOP21. Grade 3 toxicities associated with R-CHOP21 were as expected and considered manageable.1
Based on these data, R2CHOP significantly improved PFS in patients with newly diagnosed DLBCL. To further improve outcomes for patients with DLBCL, Nowakowski believes R2CHOP can serve as a backbone treatment for DLBCL, upon which other agents, like anti-CD19 monoclonal antibodies, can be added.
In an interview withTargeted Oncology, Nowakowski explained the details of the ECOG-ARIN 1412 trial and how the positive results related to patients with DLBCL will further improve outcomes for patients with all lymphoma types.
TARGETED ONCOLOGY: Can you explain the rationale for conducting the phase II ECOG-ARIN 1412 trial?
Nowakowski: Lenalidomide is active in patients with R/R DLCBL. There are multiple studies which show that as a single-agent, it has a response rate of about 30%. Unfortunately, this response does not last long. There's good rationale for introducing lenalidomide to the upfront regimen with R-CHOP to improve the outcome of frontline therapy in DLBCL.
There were actually 2 phase II, single-arm studies conducted, 1 by the Mayo Clinic and the other by our Italian colleagues. Those 2 studies have shown that the addition of lenalidomide to upfront chemotherapy with R-CHOP had very promising efficacy. When we look at the efficacy retrospectively, based on molecular analysis, it appeared to be particularly active in the ABC subtype of DLBCL, which is typically associated with a worse prognosis. Those two studies provided a very strong rationale for introducing lenalidomide to upfront therapy in DLBCL.
TARGETED ONCOLOGY: Can you explain the efficacy and safety data from this trial?
Nowakowski: This study was designed to maximize the synergy of lenalidomide with R-CHOP. We used a shorter duration of lenalidomide, day 1 through 10 at 25mg, to allow the rest of the cycle for hematologic recovery. The whole design was based on the assumption that it would allow maximal efficacy of R-CHOP while building on it.
In the study, we also put out a great effort to accrue patients early on, particularly those with rapidly progressive disease or sick patients. This has led to the study design in which we took all patients, regardless of the molecular subtype of DLBCL based on local pathology, and those patients were randomized 1 to 1 to receive R-CHOP or lenalidomide plus R-CHOP. Those patients were followed for the outcome as we usually do in these types of studies.
The primary endpoint of this study was PFS. The study actually met its primary endpoint, so it has reduced the risk of progression or death of patients in the experimental arm, who received the addition of lenalidomide, by 34%. If you look at the secondary endpoint, there was also a trend towards improved response rate, and we also saw a trend toward improved survival, although the study was not powered to access any of those.
If we look specifically at the patients with ABC subtype, which is a subset of patients we're recruiting in this study, there was also a strong trend of benefit in this subtype. However, a small number of patients with ABC DLBCL produce definite conclusions, based on significant statistical value.
The toxicity profile was manageable. We saw more hematologic toxicity, as you would expect with the addition of lenalidomide. However, there was no increase in treatment-related mortality. The regimen was very well-tolerated and safe.
TARGETED ONCOLOGY: What are next steps with this research?
Nowakowski: It's very interesting because, at the same meeting, there was a study which used a similar design just for patents with ABC DLBCL, adding a lower dose of lenalidomide for more days. This study was actually negative. It was a larger study which did not show the same signal of efficacy that we saw in the phase II study.
I think because we had those two discordant studies, we need to rethink how we can move forward. We looked deeply and tried to analyze what the differences are between the studies and how to best move forward. One way could be the development of new emits or cell modifiers, which could be potent in a similar way to lenalidomide, and there will be room for future studies because we now understand better. Maybe with some of the factors that contributed to the larger study being negative, we could design this study much better.
The other option, which I think could be even more exciting, is taking an additional step forward and using this R2CHOP backbone to add additional agents, which could have synergy with lenalidomide. From this randomized phase II study, we see this very clear signal of activity of lenalidomide added to R-CHOP. There are a number of agents that can synergize with lenalidomide, including some of the monoclonal antibodies like anti-CD19 and others. This includes some of the other targeted agents which can be combined with lenalidomide using this already known, safe, and effective backbone of R2CHOP. Using this doublet of lenalidomide added to R-CHOP, we could possibly raise the bar higher and show the difference more easily, to ultimately cure patients, which is our goal.
TARGETED ONCOLOGY: What are your thoughts on role of genomic analysis in lymphoma?
Nowakowski: I think genomic analyses in lymphoma play a major role. We are currently working on genomic analyses of the 2 trials we just presented, the ECOG trial and also the larger, more robust trial. There will be a wealth of information coming from those trails, but there's also a wealth of information on genetic characterization coming from other trials as well.
I think the biggest barrier right now is that we had shown in the ECOG trial that taking those patients early on who are progressing very rapidly, it is very critical to enroll those patients early on for the success of those trials. Unfortunately, in the current state, all this molecular analysis takes time. Sometimes, it takes quite a significant amount of time to get the results. Also, the platforms are not completely validated yet as well. [This means] they're not necessarily ready for clinical use.
Liquid biopsies have the potential for being more rapid than tissue analysis in this setting, but I'm sure this technology will develop, and we'll be able to use it in real-time. If we can't use it in real-time, maybe we can start the first cycle of treatment without assigning patients based on molecular profile, but, later on, the agent could be added based on the molecular profile.
TARGETED ONCOLOGY: What are the most significant unanswered questions in frontline treatment for patients with lymphoma?
Nowakowski: The biggest mystery of them all is why do we cure 60% of the patients? It's the same disease; it's DLBCL. Patients are getting the same chemotherapy with R-CHOP, and 60% of them will be cured, and 40% will not. Those 40% will also relapse very quickly, and unfortunately, they'll die from the disease.
We've tried over the years to develop different clinical prognostic factors to identify those patients. They do work where we can identify them to some degree, but it's not a perfect system. The same goes for molecular markers. In the ROBUST study, we conducted an analysis to identify those ABC patients that have worse outcomes, but in prospective trials, the differentiation in the outcome between GCB [germinal B-cell like] and ABC has not been as good as in retrospective studies, and it's also not the perfect marker. We still cure a significant number of patients with ABC, even just with simple R-CHOP.
I hope that this molecular analysis will bring us to the next level. If we can identify the 60% of patients that are cured with R-CHOP, we are good. R-CHOP is sufficient for those patients. Who knows, maybe it is even too much to give it for 6 cycles, so maybe we can make it even shorter in patients with low-risk disease, as we have already done in certain studies.
On the other hand, if we can identify those patients who we know will relapse early-on, there are a number of things we can do. We can think about adding new agents, consolidation therapy early on, immunotherapy or non-chemotherapy combinations, immediately after debulking chemotherapy or even upfront, as discussed in the meeting.
We have plenty of ideas on what to do, but unfortunately, we don't have a perfect test as of yet to identify those patients. Although those molecular tests are looking promising, the platforms still have to develop enough to be able to provide us results in real-time to move the field forward. We also have to be somewhat cautious because 10 years ago, there was huge excitement at ICML about the separation of ABC and GCB, and everyone thought we had it. Then, if you validated it in prospective studies, the difference was less and less, as well as more difficult to build on. That's why we have to make sure that those molecular markers or clusters as we call them are also validated in prospective studies. It is possible that what we've seen in retrospective cohorts in those studies may not be fully validated in prospective studies. Some of them may not be prognostic anymore. Some of them may not have the same discrimination power between the patients who are cured or not cured in DLBCL.
TARGETED ONCOLOGY: What is your key takeaway from ICML 2019?
Nowakowski: It's been a great meeting with highlights of multiple developments in lymphoma. If I was to pick one [key takeaway from this meeting], it would be that field is moving faster than ever. It's a very exciting time for lymphoma patients and lymphoma investigators because we have multiple effective therapies moving quickly to patients. Typically, they are not traditional chemotherapy, but they're non-chemotherapy combinations. There's a lot of progress in immunotherapy and CAR T-cell therapy. We also see a lot focus on improving patients' quality of life because that is also a very important aspect. Sometimes, removing some of the unnecessary or harsh therapy and replacing it with more targeted agents has a major impact on a patient's life. All of those developments are extremely exciting.
I think it's going to take some time before we can actually figure out how to sequence it and how to best build on it in the future. Overall, it's better to have more options than few options.