William Wierda, MD, PhD: You mentioned PI3K inhibitors. Let’s touch briefly on PI3K inhibitors and their utility. Dr Shadman, what are your thoughts on PI3K inhibitors, and what has your experience been with them? How do they fit into what we’re doing today with therapeutic strategies in CLL [chronic lymphocytic leukemia]?

Mazyar Shadman, MD, MPH: When we think about the novel agents that we have for CLL, the 3 big groups are basically BTK [Bruton tyrosine kinase] inhibitors—including the next generation, second generation, and now third generation—BCL2 inhibitors, and PI3K inhibitors. PI3K inhibitors are in the 2 large studies for CLL therapy. As John Allan just mentioned, they’re usually not the first or second choice most of the time.

Based on the initial experience, using these agents in the frontline setting usually comes with more toxicity. These are treatments we think about using in the relapse setting. You have 2 approved drugs for CLL, of course: idelalisib and duvelisib. Idelalisib was the first drug approved, and it’s a PI3K-delta inhibitor. A randomized study combined idelalisib with rituximab and the control arm was placebo and rituximab.

The study showed PFS [progression-free survival] and OS [overall survival] benefit based on a relatively short follow-up. I believe the PFS was 5 to 6 months for the control arm and not reached for idelalisib-rituximab. There was an OS benefit. The study population included patients with normal kidney function, significant myelosuppression, and patients with coexisting issues. Recently last year the long-term follow-up was published in the JCO [Journal of Clinical Oncology]. It sounds as if they looked overall for patients who received the combination therapy idelalisib-rituximab.

The median PFS was around 20 months. The take-home point from this paper is the toxicity data. One of the factors we consider when deciding whether to use PI3K inhibitors is the toxicity—namely, colitis, pneumonitis, and transaminitis. As I mentioned, these toxicities seem more common in patients who were previously untreated, maybe because their immune system is more intact.

When we looked at the long-term follow-up data, the longer exposure to idelalisib, the adverse effects like diarrhea, colitis, and pneumonitis become more common and more prevalent. For example, diarrhea grade 3 or more is in the range of 16% when you continue to take idelalisib, and grade 2 diarrhea is in the range of 17%. When you think about grade 2 diarrhea as having 4 to 6 stools per day, that is significant.

The same is true with colitis. For those of us who deal with colitis as a result of taking idelalisib, in my experience the patients who I treated early on treat graft-vs-host disease with the same type of approach: high-dose steroids and tapering. Some patients had to go back on the drug because at the time, there were not many other options. The drug had significant toxicities: pneumonitis, at 6% of grade 3 or more with longer use. We usually see hepatitis early on after starting therapy.

Usually hepatitis doesn’t come back, and the longer follow-up confirms that. Duvelisib is the other drug. It’s a PI3K-delta and PI3K-gamma inhibitor. Duvelisib is also approved as monotherapy for CLL based on the head-to-head randomized trial comparing duvelisib with ofatumumab. The median PFS was 13 months, 13.3 months vs 10 months in favor of duvelisib led to its approval.

I’m not sure how much duvelisib is used. To know more about the true rate of toxicity, we need to use the drug beyond the initial clinical trial. We need to get more real-world data, but it sounds like diarrhea and colitis are the common toxicities. These are recorded separately in the paper, so it’s hard to know the true rate. But to me, the drug seems to have the same toxicity profile with hepatitis occurring a little less. It’s not recorded.

The incidence of hepatitis was less than 10%. But again, we need to use duvelisib more, and the drug has more indications as being used in T-cell lymphomas, for example, more recently. We may get more toxicity information from different histologies. PI3K inhibitors as a group are extremely effective, but if we have to use them, we need to be mindful of the toxicity, mainly infections and autoimmune adverse effects, and prepare prophylaxis that needs to be used. There are strategies to make PI3K inhibitors better tolerated.

At ASCO [American Society of Clinical Oncology Annual Meeting], we have some strategies—maybe we should call them next-generation PI3K inhibitors—in terms of scheduling. We give PI3K inhibitors less frequently after an initial phase. In that approach we may have a PI3K inhibitor that has a different toxicity profile that may be available for CLL and other lymphomas. For umbralisib there are many studies in combination with a CD20 monoclonal antibody and other combinations. We have some work to do from a toxicity standpoint, but in patients who I had on these drugs, and even patients who are high risk, it is extremely effective medication as long as you control the adverse effects.

I see PI3K inhibitors as a third-line treatment in patients who either progress on both BTK inhibitors and venetoclax, or for some reason they can’t tolerate BTK inhibitors. There may be a future in combining PI3K inhibitors with other drugs. It sounds as if the future involves a venetoclax-based therapy. If a BTK inhibitor for any reason is not a good partner, maybe a safe PI3K inhibitor could be used.

Drug-drug interactions are significant with PI3K inhibitors and venetoclax. But there are studies that have shown safety for duvelisib and the newer generations of PI3K inhibitors.

Transcript Edited for Clarity