Daniel J. George, MD: When I’m starting a patient on second-line VEGF-targeted therapy, it’s really important that patients get a fresh start. I like to take a couple of weeks’ break, if possible, before starting the second-line agent. It’s not always possible. Some of these patients are symptomatic, some of these patients have disease progression that is more rapid. We don’t feel comfortable with that long a break. So, when that’s the case, I will go right from one agent to another, with at least like a 1-week break, if possible, in order to minimize any kind of drug interactions. But 1 to 2 weeks is what I like to do for a break. And then, counseling patients early on about how to manage that toxicity. I know some of this might sound redundant, a lot of these patients heard it before with their frontline VEGF TKI. But, no 2 drugs are exactly alike.
As I mentioned earlier, cabozantinib might have a little bit more delayed onset of some toxicities. That can give some patients some false confidence, and it’s important to guard against that. I like to see patients back, when I start them on cabozantinib, within 2 to 3 weeks. Not 1 week, because we’re not going to see those changes then, but 2 to 3 weeks. And if I see them back at 2 to 3 weeks, I want to see them back one more time at 2 to 3 weeks.
Once I get a patient out to 4 to 6 weeks, that’s when I start to feel a little bit more comfortable that we’re starting to see the full profile of toxicities. But honestly, the degree of those toxicities might not maximize until you’re all the way to 12 weeks out. And that’s different than almost all the other agents in this class. So, for me, it’s really important for patients to understand that time course.
And then, the toxicities might be a little bit different. Even if patients didn’t have diarrhea, we can see it in this space, and making sure if they didn’t have it before, they understand the need for Imodium. They might be afraid; they might have had constipation before and been afraid to use it. It’s important to make sure these patients realize these are still side effects and toxicities we need to manage aggressively. Following a thyroid function test and managing that is important. THS (thyroid stimulating hormone) elevation is one thing; changes in T4 is another. So, for us, we generally are going to manage around the changes in T4 levels, not so much around mild changes in TSH.
Other toxicities we’re looking at include hypertension. Hypertension is a very positive predictive factor, but it can also be a significant toxicity in this patient population. So, we’re going to manage that and, again, try to stay a little bit ahead of the cure on that for these patients because this is generally continuous therapy. It’s important that we get that managed before it gets too high because we won’t necessarily be giving breaks. But I do use dose interruptions when I see toxicities, particularly any grade 2 toxicityit could be GI, it could be nausea and vomiting, it could be diarrhea, it could be hand-foot syndrome, or it could be hypertension.
These are going to be the things I want to use a dose interruption for because it does 2 things for me. One, it gives me time to intervene, but that’s not as likely to escalate up to grade 3. Two, it also gives us an opportunity then to figure out, do we actually need to dose reduce? Because cabozantinib has a long half-life. It’s about 5 days’ half-life. Even though we’re holding the drug, that toxicity may actually get worse before it gets better. And if I see that, it’s maybe more likely in a population I’m going to consider a dose reduction in. On the other hand, if that toxicity improves very quickly, well, then I know I’m probably at a pretty good dose and I don’t necessarily need to dose reduce. It helps me gauge where I am with this dosing, it helps me control this from getting up to grade 3, and it helps me get on board the additional medicines we need to better manage that in the future. So, I like a dose interruption first.
Case Scenario 1: A 50-year old male with relapse of metastatic RCC