A discussion on decision factors for switching to and deciding upon a second-line therapy, and an overview of the phase 3 REACH2 study with ruxolitinib for glucocorticoid-refractory acute graft-vs-host disease (GVHD).
Usama Gergis, MD, MBA: The trigger to go on to a second line after steroids is basically progression or no response—so progression after 3 days, no response within a week, or all the other problems that come with steroids.
For those of us who are doing this every day, it’s not that hard to find these patients. After a while we get what Malcolm Gladwell calls “blink and educated gut feel” of who will get better or not.
Corey Cutler, MD, MPH, FRCPC: Until recently, when we didn’t have great data in the steroid-refractory setting, what would go through your mind when deciding the next drug for the patient in front of you who needs a second-line agent? What are the factors you considered when choosing second-line therapy?
Usama Gergis, MD, MBA: I considered the organs affected, and I considered the ease of giving this treatment. I’ll explain. Let’s say I was in New York and I have a patient who comes from Long Island or Westchester and they have skin acute GVHD [graft-vs-host disease] and I want to give them photopheresis. It’s tough for them to come twice a week for photopheresis, so I’ll probably think of something else. If they live on the Upper East Side, then that would be my first choice for skin GVHD. Historically, we have given Crohn disease treatments, like infliximab, for GI [gastrointestinal], lower GI acute GVHD. I am a fan of what we call steroid-sparing agents. These are the topical GI steroids for upper and lower GI GVHD. This is basically what we did when I was at Weill Cornell Medical College. After that, it’s been the flavor of the day.
Corey Cutler, MD, MPH, FRCPC: The other important consideration is the toxicity profile of the compound you’re considering and what adverse effects the subject might already have. For example, we tend to avoid oral agents in patients with very severe gastrointestinal disease. There are certain compounds that cause cytopenias, so you might want to avoid them with patients who have fragile counts. Incorporating patient preference and the toxicity profile of the agent is also very important. But now we can use clinical trial data to try to help guide what agents we’re going to use in second line and beyond.
Earlier in our conversation, you started to allude to the REACH series of trials. You briefly gave us an overview of REACH-1, which was the open-label trial of ruxolitinib in steroid-refractory acute GVHD. I’d like to focus a little more on the phase 3 REACH2, which has more usable data because these were randomized data. We should discuss that trial for a few minutes. Why don’t you tell us about that?
Usama Gergis, MD, MBA: As I alluded to earlier, the REACH-1 trial set the stage for ruxolitinib to be the only FDA-approved treatment for steroid-refractory acute GVHD. Unfortunately, the number was very small. They enrolled 70 patients for safety, and they could evaluate only 40 or 41 for efficacy.
Corey Cutler, MD, MPH, FRCPC: The efficacy in that trial was good. The overall response rate was 55%. That’s better than what we’ve had in the past.
Usama Gergis, MD, MBA: Correct.
Corey Cutler, MD, MPH, FRCPC: In individuals with grade 3 and grade 4 disease, the response rate was only in the low 40s. It’s an advantage. It’s some improvement, but it’s certainly not the panacea we’ve been looking for.
Usama Gergis, MD, MBA: This was rightfully followed by the REACH2 trial, which was a phase 3 randomized trial for steroid-refractory acute GVHD patients. I’ll get into definition and differences from the REACH-1 trial. Patients were randomized to either ruxolitinib 10 mg twice a day or investigator choice. The investigator choices were a large range of choices including ATG [anti-thymocyte globulin], sirolimus, CellCept, and a few others.
Corey Cutler, MD, MPH, FRCPC: It was a European trial. There were some choices in the best-available therapy arm that we would not routinely reach for in North America. That’s worth mentioning at least.
Usama Gergis, MD, MBA: Correct.
Corey Cutler, MD, MPH, FRCPC: It’s also important to note that the investigator had to declare which of these best-available therapies they were going to use prior to randomization to take away some of the bias. The investigators were given a list, and they had to choose from 1 of those. They couldn’t decide on a whim.
Usama Gergis, MD, MBA: Correct. The primary end point was overall response rate at 4 weeks, just like the REACH-1 trial.
It’s worth mentioning that the definitions were a little different in REACH2 than REACH-1, particularly in the steroid-dependent and steroid-resistant group. As we know, whether patients are steroid-refractory or steroid dependent, the outcomes are bad. The outcome is more or less as bad in both of them. The definition of steroid dependence in REACH2 was the need to increase. What happens classically in that group of patients is your patient will have a response to steroids. They cannot stay on 2 mg/kg forever. You start to taper down, and then they get a flare at some point. Once they get a flare, you go back to the 2 mg/kg and you can go in this circle until something bad happens. The definition of that and the REACH2 is the need to increase the steroids to 2 mg/kg or the inability to decrease or taper down below 0.5 mg/kg. That’s important because what happens is—you have a [patient with] grade 3 acute GVHD who is steroid refractory. You start steroids; they do well; you start to taper the steroids. When you are at less than 0.5 mg/kg, they flare. That is considered a steroid dependent, and that’s very important.
These patients were included in the REACH2 trial, and overall response rate at 4 weeks was the primary end point. There were a bunch of secondary end points, but the most important is the durability of the overall response rate. That was defined as 8 weeks afterward. What’s the percentage of these patients who respond at a month who will continue to respond? Those of us who do this every day know that most of the responders will relapse at some point.
The primary end point is overall response at 28 days. The overall response rate was a little better than REACH-1 at 62%. That was divided between CRs [complete responses] at 34% and PR [partial response] at 28%.
This study not only confirmed the phase 2 REACH-1 trial, but it compared this with an array of other agents. As expected, the comparative arm overall response rate was 39%, 19% CR, and 20% PR. So ruxolitinib 34% CR, others 19% CR. Ruxolitinib 27% PR, others 20% PR.
Corey Cutler, MD, MPH, FRCPC: It’s clearly an improvement that was highly statistically significant, but I agree with you that it was not just statistically significant but clinically significant as well. That’s an increase in the response rate by about 50% from 40% to 60%.
We would have all loved to have seen that be very durable. About a third of the ruxolitinib patients lost their response by day 56. The response rate by day 56 fell to only 40%, unfortunately, but that still was almost twice as high as the response rate in the control arm, which fell from 40% by almost half, to about 22%.
It’s not as durable as we would have loved, but more durable than the control arm. It’s a statistically significant and clinically meaningful difference by day 56. This was recently published in the New England Journal by Dr Robert Zeiser, who was the lead author of this study. I agree with you that this confirms and solidifies ruxolitinib as the singular FDA-approved agent in this setting.
Transcript edited for clarity.