Efficacy and Toxicity of Sorafenib in Advanced DTC
Transcript:R. Michael Tuttle, MD: Let’s talk about the data just a little bit. I want to ask you to talk about the phase III sorafenib trial, and Eric, maybe talk about the lenvatinib trial, because we’re now sort of getting down into the weeds of these trials. Give me a high level view of your sorafenib trial.
Marcia S. Brose, MD, PhD: The way the DECISION trial was structured, it was the first phase III trial that went forward, and it was based on phase II data that was done at our center and others that showed that there’s activity. So, now we had to have a placebo-controlled trial. Because sorafenib was available and people would get it off label, it was impossible to do an overall survival as a primary endpoint because people were going to get the drug one way or the other. We actually included in the trial that patients were randomized to placebo or sorafenib. They were allowed to cross over at the time of progression, so virtually all patients eventually got sorafenib. And, then, the primary endpoint was progression-free survival. There’s been pretty good data to say that progression-free survival can be a surrogate. Now, there had been a few cases historically where the FDA said that’s not been true, and so there was a concern about that. But, I would say in the TKIs, it looks like PFS is a good predictor for overall survival. That said, we can’t prove it in a study that has crossover.
Then, patients were using these criteria of progression within 14 months, RECIST criteria disease. So, they have to have, at minimum, a 1 cm nodule—which I agree, it’s a very low tumor burden. But a lot of patients had more than that, of course. And, then, they were treated and they had to have RAI refractory based on having no uptake, as Naifa said, or they had uptake and grew anyway, or they had just had multiple rounds and now they were over 600 millicuries. We know the data says that no one is going to be cured virtually after 600 millicuries. Now, again, that doesn’t mean that there isn’t the rare cancer patient who might. You have a reason, because they got such long periods, that you would go past 600. But these are sort of, I’d say, strong guidelines. There could be exceptions, right? So, you take those patients and then they were studied for progression-free survival with a central review. The scans were done every 2 months, and basically what the data showed on the placebo arm, without any intervention, there are two things we know. They progress within 5.8 months. That was extended to 10.8 months with the use of sorafenib. But, of interest, if you look at the data, those placebo patients were not asymptomatic. Over 30% of them developed symptoms while they were on. So, again, all of the pre-thinking in the world saying, ‘Oh, this is an asymptomatic patient, they’re going to be fine,’ remembering that 30% of those patients are going to develop symptoms.
Again, I’d say that that speaks to the criteria being very good to pick patients who will be the people who are bad actors. And, the effect of sorafenib was a positive trial with extension of progression-free survival from 5.8 months to 10.8 months. The only caveat that’s off the actual design of the study is the fact that, in practice, when patients progress... So, say they had a 40% response and then they grow back 20%, their volume of disease is still less than when they started. And if it’s one nodule that’s grown, or one new nodule that you can observe (because it’s not in a bad spot, or maybe it’s in a place where you could do surgery or external beam), or they have a new bone metastasis, now we can irradiate it. But, all of the rest of their disease, their 100 other little spots are fine. In practice, we will continue to treat those patients. In fact, that was allowed in the study. It was at the time of progression, patients based on clinical decision, the doctor was allowed to continue to treat. That was part of the design, which I think was nice because it really is realistic with what we actually do in clinic.
R. Michael Tuttle, MD: So, Frank, there definitely is a benefit. We’ve heard the progression-free survival. You’ve heard about the side effects on sorafenib and these drugs. What’s the reality of that side of the coin?
Frank Worden, MD: So, these drugs are benign, and I think we have to take that into consideration. Nothing is ever perfect. I do agree with Marcia. We do have a tendency to want to start treatment, sooner than later, because once you get symptoms, it gets harder to treat. And oftentimes, the symptoms don’t reverse. They can, but they don’t. When you treat these patients, you have to watch for various things. Hypertension is a big issue. Part of the work-up, too, we need to bring up is that I get head CTs on all these patients prior to start therapy.
R. Michael Tuttle, MD: We do, too.
Frank Worden, MD: In the lenvatinib study, in particular, there were bleeding events. And I know, Eric, you and I had a discussion about this in the past, so we need to watch that, as well, and then radiate them prior to that. And with sorafenib, the hand-foot syndrome can be pretty substantial. Now, I know Marcia and I share a similar philosophy, in that I don’t necessarily back down. We hold the drug, allow them to go ahead and recover. And, we treat them with IV fluids if they’re dehydrated. We give them nonsteroidals, keratolytic agents—such as urea, or salicylic acids—and various creams and hand lotions. And, then, once they recover, we start them back and we find that we don’t really see the side effects recurring. We just keep plowing ahead because I’m afraid to completely drop the dose or to decrease the dose because I don’t think they’re going to get the same benefit. So, we take that all into consideration. And, then, we have them come back regularly. When we start these patients on therapy, it’s not like you say, ‘here’s a 30-day supply’ like you do for a lot of other medications. You see them back weekly. Here’s my nurse practitioner’s number, and here’s an appointment to the infusion room in case you’re dehydrated. That’s the key. So, the side-effect profile can be very manageable, but, you have to be proactive.
R. Michael Tuttle, MD: So, talk a little bit about the proactive. You guys had to build an infrastructure to make this happen with nurses and nurse oncologists, that you guys take for granted, because you have that as part of the system. In your endocrine clinic, how did you make this happen?
Naifa Busaidy, MD, FACP, FACE: Yes. From a quality improvement standpoint, for doing the right thing for patient safety, we wanted standards, so that we’re all managing these adverse events early, being proactive about it and remembering to look for it. So, we used various triggers. We made a list of the most common and less common adverse events that you want to specifically ask the patient about, and made an adverse-event chart for that, similar to what one would do in a clinical trial, but in a manageable way for everyday questions. And the nurse would go in there and ask the patient. When the patient is initially starting, the nurse actually makes phone calls to the patient to check how they’re doing, what’s the blood pressure log, how’s the hand-foot, because, just as Frank mentioned, the earlier on you get in this on the game, the better the patient is going to be able to tolerate the therapy, and then, hopefully, have the longer progression-free survivals. And, then, we distributed that, and everybody in their chart needed to have performance status, comorbidities, adverse events listed. Then in their final plan say, why are they choosing this therapy versus another or a clinical trial. And, we went back into the quality improvement and looked at all the charts to be sure that we are doing these things, because this does have a lot of implications for patient safety. And it works very well. It’s now become our standard of care.
R. Michael Tuttle, MD: There’s a skill to do these.
Naifa Busaidy, MD, FACP, FACE: There are two other things I just want to add to what Marcia and Frank said so eloquently. It’s very easy to figure out the patients. There’s a beautiful chart that you always include in a lot of your lectures, where on the Y axis is mortality and on the X axis is time. And, if you look, there’s different cohorts of patients with this metastatic refractory thyroid cancer. It’s very easy to figure out the patients who have that acute curve in the line, where their mortality is high in a short period of time versus those that have the metastatic disease and have that long time with low mortality. So, these are patients that don’t need the systemic therapy, whereas these do. And it’s what do we do with these patients, in between, taking all the guidelines that you just stated in terms of figuring out when to treat them, but remembering to do that cross-sectional imaging, remembering to do it early, and start doing it more often. If it’s somebody that’s even growing by millimeters, it’s easy to have them come back in 3 months and look, or 2 months if you’re worried about them rather than just once a year or twice a year. So, just remembering that trajectory of disease and there’s a different numbers of patients.
And, the other thing is on the sorafenib trial, you all looked—or I think Schlumberger looked—at a cohort of the patients. If they have smaller burden of disease and their target lesions were less than 15 mm in total, those patients may have a different benefit from that systemic therapy. So, it’s not a one-size-fits-all. It is time to progression, burden of disease, symptoms. And, when you come to symptoms, physicians don’t do a very good job at figuring out if patients are symptomatic or not.
Marcia S. Brose, MD, PhD: Well, when you look at the cohort, I’ll give you the data from that. This is a post hoc analysis, it wasn’t preplanned, but it did seem that if you took people, for whom the largest lesion was under 1.5 centimeters, they didn’t get a benefit. Again, the same way I use that 600 millicuries cut-off where I won’t, in general, treat people beyond that, but there are exceptions. For the most part, if their biggest lesion is less than 1.5 mm, now the exception is the miliary patient who doesn’t have a single nodule over 3 mm, but they have symptoms, or they just have a large burden of disease. The other interesting subset, we did look at symptomatic versus non-symptomatic and both groups had a benefit. So, symptoms, really, in my opinion, should not be used as a criteria. Of course, if they’re symptomatic, you’re going to treat, but the lack of symptoms should not be used as a criteria to not treat.
Transcript Edited for Clarity
