Andre Goy, MD, MS: Beyond the IPI [International Prognostic Index], the context of the patient—the well-being, the age, the fitness, the comorbidities—is something that we very often exclude at the clinical trial and then that’s part the problem…for something that has a huge impact. It has an impact on the dose intensity, and then we know that if you have less than 85% dose intensity, in some retrospective studies, even with a good IPI you don’t do that great. Having this dose are important factors that we don’t take into account probably enough. What do you think?
Nathan H. Fowler, MD: This is clearly important. If you look at patients, a significant number of patients who you see in clinic that have newly diagnosed large cell lymphoma are going to be elderly. They may have co-existing heart disease or kidney problems. As we all know one of the most effective drugs that we use with CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone], can affect the heart. If you think about patients who present, especially patients who can’t tolerate typical R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone], traditionally their outcomes have been worse. In my own practice I generally substitute things like etoposide—something like CEOP [cyclophosphamide, etoposide, vincristine, prednisone] instead of CHOP. The outcomes look similar, but they’re still not quite as good as we see with R-CHOP. This is definitely a population of patients for whom there’s an unmet need, patients for whom you can’t do standard therapy.
Andre Goy, MD, MS: We talked about molecular features, we talked about comorbidities, we talked about the context, the well-being of the patient. We talked about the staging for the IPI. There’s a number of other factors that because we have tried for many years to go beyond R-CHOP. So far, we have not really broken the ceiling. You have done a great job trying to do this, and you can comment on this. I think there were very interesting phase II data from the Mayo Clinic looking at R-CHOP versus R-CHOP in combination with lenalidomide or R-squared CHOP [lenalidomide, rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone], particularly in the ABC [activated B-cell] subtype, and then trying to show that it would negate the difference of the negative impact of the ABC subtype. So where do we stand with that?
Grzegorz S. Nowakowski, MD: There are actually a couple of studies done in this regard, very promising single arm phase II studies from both Mayo Clinic and Italy, improving the outcome of patients with the ABC or non-GCB [non—germinal center B-cell] subtype of DLBCL [diffuse large B-cell lymphoma]. There was also an Intergroup randomized study, E1412, which randomized patients to the lenalidomide R-CHOP versus R-CHOP alone. It was a randomized phase II study that used lenalidomide of 25 mg days 1 through 10, and this study actually did show improvement in progression-free survival in patients treated with R-squared CHOP.
Actually for everyone—not only patients with the ABC subtype, although the trend was the strongest for the ABC patients. What was disappointing is that the ROBUST trial, which was a large phase III study just enrolling patients with the molecularly defined ABC subtype, failed to improve the outcome of patients with addition of lenalidomide versus placebo. There are major differences between the randomized phase II study done in ECOG [Eastern Cooperative Oncology Group] and the ROBUST study. The obvious one was the dose of the drug. ROBUST was using only 15 mg, although for 14 days. But we know that most of the data from previous relapsed/refractory setting were coming with 25 mg. So this could have played the role in discordant results of the study.
The other issue is the patient selection, although ROBUST was selecting patients with the ABC subtype molecularly defined. In a sense this was the state of the art, most modern study and using the most modern assay you could do to identify these patients, this actually did result in selection bias of patients with lower-risk disease. It’s something that was difficult to appreciate just by molecular subtyping or even looking at IPI. But we all know that in the clinic, we see 2 types of patients. Some patients are coming with some lymphadenopathy, maybe some elevation of LDH [lactate dehydrogenase], stage IV. But nevertheless, they just don’t progress so quickly, so there is time to do the biopsy, maybe send it to central pathology review, and molecular subtyping, then you know the patient will qualify for the study. That takes several weeks on occasion. Then there are patients who are just sick. You see them Friday afternoon, and they just need to go right to the hospital.
Andre Goy, MD, MS: They don’t make it to clinical trials.
Grzegorz S. Nowakowski, MD: They don’t make it to clinical trials. What was different in the ROBUST and ECOG studies was that ECOG was accepting patients without a molecular marker. This was done retrospectively. The ROBUST did not, which resulted in very different median times from diagnosis to treatment. It was 20 days in E1412 and 31 days in ROBUST. But the median time, I’m not giving the full story, because if you look actually between the patients who were treated within the first 2 weeks, this was around 40% in the E1412 study, and very few patients in ROBUST at all.
What we worry about sometimes are those molecular markers, and when you look at something promising for selection of the patients, this what we all want to do. But if you make the studies very complex, they’re becoming not real world, you’re selecting a population that outperforms the control arm. Then even with an active agent it is very difficult to show the difference.
Andre Goy, MD, MS: It is a challenge. There have been repeated trials where the phase II was very promising, and then the phase III setting the selection process.... I would say that for our audience, the 6 cycles of R-CHOP for basic large cell lymphoma still remains. We’ll go into how we look at the response that might have an impact on maybe what kind of modify....
Dr Maddocks, I think there is 1 subtype situation where I would argue that we should probably not give just R-CHOP, in a patient who actually has double-hit or maybe double-expression lymphoma. Do you want to comment what your experience is in this and recommendation?
Kami Maddocks, MD: I would agree. I think the R-CHOP is the standard for most patients, but in a patient with double-hit lymphoma, they have a very poor prognosis with standard 6 cycles of R-CHOP. There are a lot of retrospective data with more intensive therapies, including dose-adjusted R-EPOCH [rituximab, etoposide, prednisone, vincristine, cyclophosphamide and doxorubicin hydrochloride] regimen, R-CODOX-M/IVAC [rituximab, cyclophosphamide, vincristine, doxorubicin, and high-dose methotrexate, alternating with ifosfamide, etoposide, and cytarabine] regimen, or hyper-CVAD [cyclophosphamide, vincristine, doxorubicin, and dexamethasone] regimen. And there are phase II data using the dose-adjusted R-EPOCH regimen that shows that probably while they don’t do as well, their 2-year progression-free survival is much better than what we initially expected. I would say that the patients with double-hit lymphoma for sure should have more intensive chemotherapy, and I would typically treat them with dose-adjusted R-EPOCH, and they do have a higher rate of CNS [central nervous system] progression or relapse, and we usually give some form of CNS prophylaxis in those.
I don’t think in the double expressor it’s as clear that more intensive therapy helps. The don’t do as well with R-CHOP, and we don’t have great data that they do necessarily better with dose-adjusted R-EPOCH. There was in addition to the R-CHOP lenalidomide study the R-CHOP ibrutinib study, and the non-ABC subtype that showed at least an exploratory analysis that in younger patients there may be benefit in the non-GC with the R-CHOP plus ibrutinib, and in the patients with the double expressor with R-CHOP plus ibrutinib.
Andre Goy, MD, MS: When we talk about double-hit, and I think for our audience the double-hit definition is obviously we’re requiring the FISH [fluorescence in situ hybridization test], typically BCL2 and MYC, sometimes BCL6 and MYC; the triple-hit that we sometimes refer to doesn’t do really differently. But what is interesting is there was a paper published that makes a lot of sense where sometimes we don’t find it in FISH obviously, but the patient has a signature that is consistent with a double-hit signature, and these patients do very poorly as well. This is something very important.
But when we see a patient, we all know from the clinic, they say, “How am I going to do?” I think all our answer is that we need to say, “Well, how will you respond first to chemotherapy?” That brings us to the next stage, the response to chemotherapy.
Transcript Edited for Clarity