Role of Tazemetostat and Selecting Appropriate Therapies in R/R FL

Alexey Danilov, MD, PhD: As you said, this field is becoming more complicated with the introduction of additional agents like tazemetostat, which was recently approved, as we discussed earlier. Tazemetostat is essentially an epigenetic-directed therapy. It’s an EZH2 inhibitor, which is a methyl transferase, so that gets pretty complicated. As we know, some of the tumors are EZH2 mutant, and others are EZH2 wild type. But it turns out that tazemetostat can have activity in EZH2 wild-type follicular lymphoma.

The data that came in this space is from a phase 2 study of tazemetostat in patients with follicular lymphoma. It enrolled 99 patients; 45 had mutated EZH2, and 54 had wild-type EZH2. The drug is administered orally at 800 mg twice a day. In that study, the response rate in EZH2-mutant disease was 70%. That’s a high response rate in patients with relapsed/refractory follicular lymphoma who receive 2 or more prior therapies. In patients with wild-type EZH2 the response is half that: 35%. What’s interesting is that PFS [progression-free survival] was not that different in the end. PFS was 14 vs 11 months. Clearly patients who have EZH2-mutant disease benefit more. The drug was well tolerated. For example, high-grade neutropenia was seen in only 3% of patients, and discontinuation due to adverse events was under 5% to 10%, which is different from what we see with PI3K inhibitors. Also, in unselected patient population, EZH2 wild types, the efficacy is not as high. How do we choose between PI3K inhibitors and EZH2 inhibitor tazemetostat in patients who ready for their third line? In my practice, I don’t know the EZH2 starters on most patients. Is it different in your practice?

Tycel Jovelle Phillips, MD: No. This is around the time where you start evaluating for EZH2 mutations, once you start getting in third-line and later. For the most part I agree. I don’t know the status of my patients when they hit the third-line setting. That makes choosing between delta inhibitors and tazemetostat challenging because delta inhibitors in those who lack a EZH2 mutation would be much more beneficial than tazemetostat and give a much more durable response. One key point from the study is that the duration of response they reported was more misleading in the simple fact that EZH2 nonmutants enrolled a lot faster than a mutant population. They just had much longer follow-up. The mutant population data are still mature, and it won’t be known for quite a bit of time.

For me, if a patient is fit enough, and I don’t have any concerns about them having a big issue with tolerance with a delta inhibitor, then I tend to choose a delta inhibitor in a third-line setting unless I know that the patient has an EZH2 mutation. If they have an EZH2 mutation, then tazemetostat would be a better option based on tolerability and likely efficacy in their situation. But if their EZH2-mutation status is not known and I can’t test it, or if I know that the patient isn’t a mutant, then I’m looking for a delta inhibitor. The delta inhibitor of choice will depend on the availability of clinical trials in this situation and patient comfortability and dependability with taking an oral agent. Rather, they can come back to the clinic weekly. It would be as needed for copanlisib to give it because of its IV [intravenous] formulation and recommended schedule.

Alexey Danilov, MD, PhD: I take a similar approach. I admit that I don’t have extensive experience with tazemetostat. The reason is that, unfortunately, often we don’t know the EZH2-mutational status and EZH2 wild-type disease, so the efficacy may be lower than I’d like. I have experience with all the PI3K inhibitors that have been mentioned in follicular lymphoma and other diseases. If it comes to an oral therapy and I need therapy, the illuminated effects of oral continuous PI3K delta inhibitors are concerning. In that sense and the mutant dosing with IV [intravenous] copanlisib, sometimes it becomes effective if it’s acceptable to pay a patient.

Many of our patients refill oral therapies where weekly IV infusion visits may not always be feasible if patients live far. However, others are totally unopposed to weekly IV infusion. These agents are compared head-to-head, which I don’t necessarily see happening in the near future. It will always be a choice based on multiple factors and patient comorbidities if they have fully controlled diabetes. Copanlisib is certainly not for them.

You need to check blood sugars before you administer the agent. Ideally you wouldn’t give copanlisib with nonfasting blood sugars above 200 mg/dL or fasting blood sugars above 160 mg/dL. On the other hand, if the patient already has an autoimmune disease, then the use of PI3K delta inhibitors is difficult. It becomes a multifactorial choice between tazemetostat and PI3K inhibitors. It will be nice to see some longer follow-up data on EZH2-mutant disease. Hopefully, as this agent becomes more widespread, we’ll see community testing for EZH2 mutations on patients with follicular lymphoma. We certainly try to do it when we have enough material using next-generation sequencing and 1 FFPE [formalin-fixed, paraffin-embedded] slides, but often you must make a decision before that information is available.

Transcript Edited for Clarity