Susan M. O’Brien, MD: Although the targeted agents are very effective agents and are not myelosuppressive, which is a major advantage, they do all have their own specific toxicities. So, for example, with ibrutinib, we see bleeding. It’s generally mild, predominantly ecchymosis, but occasionally you can see rare bleeding. And it also can cause atrial fibrillation, which, right now, looks to be in about 7% of the patients. One of the problems with atrial fibrillation, other than the event itself, is that it sometimes requires anticoagulation, which will then potentially increase the risk for bleeding. Other agents, such as idelalisib, have their own toxicities, including pneumonitis, late colitis, and transaminitis. And then venetoclax, which was recently approved for the 17p subset, has a high risk for tumor lysis in certain patients with high lymphocyte counts or very bulky disease. So, they each have their own pattern of side effects.
For example, if I had a patient who’s on anticoagulation, I might be concerned about giving them ibrutinib, particularly if they’re an older patient where I’m worried about them falling or something like that. Maybe I would choose idelalisib in that situation. Whereas, let’s say I had a patient with bad underlying lung disease—for example, COPD—I would be worried about that patient developing pneumonitis since their lung function is already decreased. And I would try and avoid that drug. So, to some extent, we can use the toxicity profiles of the novel agents to try and choose the best treatment for the patient.
Capturing long-term data from the clinical trials with novel agents is very important. We have unanswered questions. For example, will the complete response rates go up over time? Because, remember, the way the drugs are dosed now, for the most part, patients stay on them indefinitely. Another question would be, would we see late side effects that we didn’t see in the short duration from which the trials are usually reported? So, it’s very common to look for late side effects. It’s very important that, in the clinical trials, we continue to follow those patients.
The reality is that clinical trials are limited to a small fraction of the patients who will then be receiving the drug once it’s approved and goes out into the real world. It will be important to have reporting from physicians in the community if they see side effects, or events which we may not know about from the limited follow-up that we have with the clinical trials.
The good news is that on the phase II trial, now we have follow-up probably going out to 4 to 5 years—which it’s not 10, but it’s not bad. And we haven’t seen any real late unexpected toxicities with ibrutinib. But, we still need to follow those patients longer. And with the other novel agents, since they’ve been approved more recently, we still have to get more long-term data on them also.
One of the questions is, do you really need to dose patients long term with some of the novel agents? That’s the way that they’ve been given up until now. We start ibrutinib, we start idelalisib, etc, venetoclax, and we just keep going. Is that really necessary? Obviously, patients don’t really want to be on pills for an extended period of time, years at a time. And, the cost to the healthcare system is quite large. So, there are now a number of trials which are beginning to look at stopping therapy. And I think those are very important trials, both from a patient comfort point of view and a pharmacoeconomic point of view, in terms of trying to be able to stop therapy.
For example, a patient can have a relatively mild side effect, let’s say with ibrutinib, such as arthralgia. And we’re not talking grade 3 to 4. But, the ability to tolerate that side effect, if you know that you’re going to be on a drug for years, is a whole different ball game. Because, even a mild side effect that you anticipate experiencing for years to come is an unattractive idea. And so, it is very important to be looking into trials where we could potentially stop therapy. And I think that will markedly increase patient satisfaction with these drugs, which is already high, but will be even higher.
Patient satisfaction clearly impacts treatment. If a patient is not feeling well, they’re not going to take the drug. It’s very easy for them not to take oral drugs. And, as I mentioned, if they have mild side effects, but they see those side effects going on indefinitely, that’s a deterrent to taking the drug. And we know that no matter how few side effects a drug has, patient adherence is just not great; patients being humans. So, anything we do to improve their satisfaction is very important. And we notice that now that, for example, ibrutinib is out in the real world. There are some presentations that have been made suggesting that as many of 20% of patients are coming off for side effects. That’s much higher than what we saw in the clinical trials. But, if you think about it, when we started the clinical trials with ibrutinib, those were people that were really desperate. They were highly refractory, they’d failed all the treatments that were out there, and now they were on this pill that was generally well tolerated; plus, it was giving them amazing efficacy. And so, they weren’t going to be complaining too much is the bottom line.
We know that clinical trial patients are highly motivated. We go out in the real world and patients are not necessarily very refractory. They could be getting treated at first relapse, and so there are other options for them. And everything is a risk-benefit. If I know I have other options and I don’t like this toxicity, then it’s very likely I’m not going to be continuing the drug. So, patient satisfaction is huge, and that’s particularly true the more options we have for them. Because, then, if they’re not happy on one drug, they have the option of getting off that drug, whereas maybe the original trial patients had very few options and they were much more dedicated to staying on the drug.