GALLIUM Trial: Follicular Lymphoma

John M. Burke, MD:The choice between obinutuzumab and rituximab is an interesting one. In my practice, I choose obinutuzumab for the majority or all of my patients when I’m combining the antibody with chemotherapy. The rationale for that is the GALLIUM study. The GALLIUM study was a randomized phase III trial in which patients were randomly assigned between either rituximab or obinutuzumab. Both antibodies were combined with a chemotherapy partner that was CHOP [cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone], CVP [cyclophosphamide, vincristine, and prednisone], or bendamustine. The results of that study showed the 3-year progression-free survival was a little bit higher with obinutuzumab. At the same time, obinutuzumab caused a little bit more toxicity.

Knowing that that’s the data we have, I think obinutuzumab is a bit more effective as an antibody in treating follicular lymphoma. For that reason, I almost always choose it as the antibody that I use with chemoimmunotherapy in my patients with previously untreated follicular lymphoma. That said, I am probably in the minority of oncologists in the United States in that regard. I’m told that the majority of oncologists choose rituximab, not obinutuzumab as their initial treatment. Why is that? Well, I think 1 reason might be rituximab has been around a long time and oncologists are very comfortable with it. Obinutuzumab does add a little bit of extra toxicity. The cost might be a factor in that perhaps obinutuzumab as the newer agent is a little bit more costly.

Certainly in the upcoming era that we are entering into now with biosimilar versions of rituximab, that cost differential might be more significant than it has been until now. For all those reasons, I think most doctors are still choosing rituximab. I don’t think anyone thinks that’s unreasonable. It’s still the go-to antibody for most people, and it’s still a very reasonable choice. But, for me, I will generally choose obinutuzumab first.

As I mentioned early on, 1 of the prognostic issues in follicular lymphoma is the so-called POD24 or the patients whose follicular lymphoma progresses within 24 months of their initial therapy. That data set was generated from a review of the National LymphoCare Study. Other data sets have found similar findings, namely that patients whose lymphoma relapses fairly soon, whether 1 year or 2 years from the initiation of therapy, have a less favorable prognosis than those whose follicular lymphoma relapses later.

Unfortunately, we do not have the ability to predict in advance with a lot of accuracy who those patients are going to be. It would be nice if when a patient with follicular lymphoma walked in the door, we could say, “This patient has the features that make them likely to relapse within 24 months, and so they have a bad prognosis. Therefore, they need a novel strategy.” We are simply not very good at doing that, and that speaks to the limitations of our prognostic models in follicular lymphoma.

The FLIPI [Follicular Lymphoma International Prognostic Index] score is certainly not a perfect score at predicting who is going to have POD24; neither is the FLIPI2 nor is the so-called m7-FLIPI, which incorporates FLIPI, the ECOG [Eastern Cooperative Oncology Group] performance status, and mutational status of 7 different genes that are known to have an impact on follicular lymphoma prognosis. So, even the m7-FLIPI is not a perfect predictor of who is likely to have an early relapse.

We don’t have a perfect prognostic score to be able to predict how patients are going to do. We do the best we can, but reality is when patients come in the door who need treatment, we treat them, cross our fingers, and hope for the best. Eighty percent of the time, things work out, but about 20% of the time, the patient still has a relapse within a couple years. We now know that many of those early relapsers do have transformation into a more aggressive lymphoma, large-cell lymphoma, but not all of those patients. Some patients just have early relapse of their follicular lymphoma.

At the ASH [American Society of Hematology] Annual Meeting & Exposition in 2019, the results of the GALLIUM trial were analyzed using the m7-FLIPI. So to go back to the m7-FLIPI, it was initially developed using a data set where all the patients were treated with R-CHOP [rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine sulfate, and prednisone], and then it was validated using a data set where the patients were treated with R-CVP [rituximab, cyclophosphamide, vincristine, and prednisone]. That’s how the m7-FLIPI was developed. It has not been tested for its validity in patients treated with obinutuzumab or treated with bendamustine. A group of investigators took the patients from the GALLIUM trial who had been treated with obinutuzumab and bendamustine and looked to see whether the m7-FLIPI was an effective predictor of outcomes in those patients. Interestingly, the answer was no, that the m7-FLIPI was not a very good predictor of outcomes in patients treated with obinutuzumab and bendamustine.

That’s 1 of the limitations of the m7-FLIPI, and it’s part of the reason that the m7-FLIPI is not used in clinical practice. I don’t know any of the doctors with whom I work that send a genetic panel to look at the gene mutations in follicular lymphoma and calculate an m7-FLIPI score on those patients. Right now, the m7-FLIPI is more of a research tool but is not applied that often in clinical practice.

Also in the same study that was presented at ASH, the investigators looked at the role ofEZH2mutations. They found thatEZH2mutations predicted longer survival or better outcomes in patients treated with CHOP or CVP. But it predicted less favorable outcomes in patients treated with bendamustine. The corollary of that is, the thought is for when a patient with follicular lymphoma comes in, should I check for anEZH2mutation? If they have it, should I make CHOP or CVP chemotherapy be my backbone? If they don’t have it, the 80% that don’t have it, should I make bendamustine my chemotherapy of choice? In my opinion, this analysis was a retrospective one. It is not practice defining, and I think it’s more of a hypothesis-generating finding. I personally do not intend to change my practice to start checking forEZH2mutations and picking my chemotherapy partner or chemotherapy backbone based on the presence or absence of anEZH2mutation.

Transcript edited for clarity.