Managing Patients With HR+ MBC on PI3K Therapy

Joyce O’Shaughnessy, MD: I’d love to hear from everybody what your personal experience has been with alpelisib. We also want to touch on toxicity so we can weave together a little bit of your personal experience efficacy-wise—which patients, where you’ve seen things not work and work, toxicity. Gabe, you could start? What have you seen so far?

Gabriel Hortobagyi, MD, MACP, FASCO: Clearly, the PI3 kinase inhibitors and alpelisib in particular have more adverse events associated with them than the CDK4/6 inhibitors or endocrine therapy alone. Virtually all patients have some adverse events, and about three-quarters of them have hyperglycemia, which is arguably the most common adverse event. Unfortunately, this requires attention because you need to monitor this carefully from the beginning. You need to be proactive in controlling hyperglycemia. Many of us oncologists have fallen a little away from our endocrine days, but we need to recover those skills.

The second is diarrhea. Diarrhea is fairly common, including some grade 3 diarrhea. The same way we deal with diarrhea with abemaciclib and ribociclib, we need to deal with this proactively. Rash is another important adverse effect associated with alpelisib. It can be maculopapular. It can be acne form. Some of it is grade 3 or 4. It can be pruritic. It can be painful. It has to be addressed, both symptomatically and to reduce the severity of it. Otherwise, your patients will quit. Then there are the usual adverse effects of fatigue, asthenia, loss of appetite, and some weight loss.

This drug is in a different category from the CDK4/6 inhibitors. It’s closer to everolimus, which also has its own safety issues, although I find that alpelisib is somewhat more toxic than everolimus. It’s a good drug, but it’s a drug that you need to use with a very aggressive supportive care plan, an education plan, for your patient. If you do that, most of them will take the drug and many will benefit from it. In the SOLAR-1 trial, 25% of patients discontinued prematurely alpelisib. That’s an important figure to keep in mind.

Now that we know what the adverse effects are and the probability of adverse effects and how to deal with them—and there are very clear guidelines as to how to deal with them—we can make alpelisib a tolerable drug, an effective drug. As you said at the beginning, the drug that you don’t take can’t help you.

Joyce O’Shaughnessy, MD: Thanks, Gabe. Thanks for setting the stage for us. How about you, Sarah? What’s your experience been? Any caveats on the toxicity management?

Sarah Sammons, MD: I’ve become more familiar and more comfortable with using alpelisib in the clinic. We’re lucky enough to have an excellent clinical pharmacist who helps us manage the toxicity. A week prior to starting any patient on alpelisib, we start everyone on 500 mg of metformin XR regardless of their hemoglobin A1C. We can peel that off if they don’t have issues with their hyperglycemia, but we have a fairly hardy population in North Carolina. We’ve had success doing that, so we start everyone on metformin XR a week before.

That’s being looked at in clinical trials. We also have 2 trials here [at Duke University School of Medicine] in patients on alpelisib with hyperglycemia resistant to metformin, looking at glucagon inhibitor and a GLP1 analog. There’s a lot going on in that space. We also start patients on Zyrtec [cetirizine] 7 days prior to starting alpelisib. I start them on Zyrtec twice a day; most of my colleagues start them on Zyrtec daily. Seven days prior to starting, we’re starting metformin and Zyrtec on all patients. We’re monitoring them based on guidelines with clinical pharmacy and nursing. I’ve had success that way. It is a good drug if you can manage the toxicity.

I do just want to put 1 caveat [out there]. One of my biggest success stories as an oncologist is that I had a patient in visceral crisis with dual PI3 kinase mutations, with leukemia-like blood counts, and she’s now a year and a half into alpelisib-fulvestrant NED [no evidence of disease].

Joyce O’Shaughnessy, MD: She had a packed marrow?

Sarah Sammons, MD: Packed marrow, lobular breast cancer, dual PI3 kinase mutations. I gave her chemotherapy for a little while, but she was having terrible toxicity. She turned it around with alpelisib and fulvestrant because of those. Data have shown that, preclinically, it looks like dual PI3 kinase mutations are explicitly sensitive to the drug.

Joyce O’Shaughnessy, MD: Thanks. I love those stories because it’s a terrifying situation to be in for the patient and for us. How about you, Neil? What’s your experience been?

Neil M. Iyengar, MD: I echo a lot of what’s been said. This is a drug that to patients is worth the effort that you put in. It does take some effort on the patient’s part. But as we’ve heard, it’s a very efficacious drug. One thing to remember is that the hyperglycemia is an on-target effect of PI3 kinase inhibition. This is expected, and this essentially is telling us that it’s working.

What we’ve seen in preclinical models paradoxically is that hyperglycemia and resultant hyperinsulinemia can provide a positive feedback loop that then reduces the antitumor efficacy of PI3 kinase inhibition. Some of these ongoing trials Sarah mentioned. We have a trial testing a ketogenic diet vs a low-carbohydrate diet vs an SGLT2 inhibitor. These are going to be important as we think about hyperglycemia management strategies. I’m optimistic that there are all these ongoing trials, so that in the near future we won’t have 1 optimal hyperglycemia management strategy. We’ll have multiple hyperglycemia strategies that the patient can choose from. The future of the use of this drug is optimistic.

With regard to the rash, it’s the same thing. This is something we can prophylax for. We also have a clinical trial testing and antieosinophilic agent. This underscoresthe point that the drug is ultimately a good drug and certainly a good enough drug. There’s a lot of interest and effort being put into managing these adverse effects—some of them on target—so that ultimately we can wield this powerful tool in the clinic in a way that’s also palatable to patients.

Transcript edited for clarity.