Ruxolitinib for Relapsed/Refractory or HU-Intolerant Polycythemia Vera

Harry Erba, MD, PhD:The approval of ruxolitinib for patients with intolerance to hydroxyurea or inadequate response to hydroxyurea gives patients and their physicians another viable treatment option that really wasn’t there before. As a way of exemplifying this and pointing out this glaring hole in our armamentarium, consider the RESPONSE Trial. In the RESPONSE Trial that led to the approval of ruxolitinib for the patients with hydroxyurea intolerance or resistance PV (polycythemia vera), patients were randomized between ruxolitinib and best available therapy, if they were already found to be resistant to or intolerant to hydroxyurea.

So, the 50% of patients who were found to be resistant or intolerant to hydroxyurea who got randomized to best available therapy, what did they get? Sixty percent got hydroxyurea. Why? Because the other options are either ineffective, expensive, toxic, or not approved in the United States. For example, other options include: anagrelide, which is very good for lowering the platelet count. You do get a dilutional anemia, but not really specific for polycythemia vera. Interferon is quite effective in this disease, but not commonly used because of the cost, the route of administration, and the toxicities. Pipobroman, which is a drug that’s only available in Europe, and then the IMiDs (immunomodulatory drugs), such as lenalidomide or thalidomide, have been used in polycythemia vera and other myeloproliferative neoplasms, such as myelofibrosis. There are very few other treatment options. With the availability of ruxolitinib now, we have an effective drug that can be added to that list that really needed an addition.

The RESPONSE Trial was a randomized phase III open-label study looking at the activity or potential benefit of ruxolitinib in a very well defined population of patients with polycythemia vera. These patients with polycythemia vera were defined according to a modified set of response criteria from the European Leukemia Network. These included: the inability to maintain the hematocrit less than 45% without phlebotomy—so patients had to remain phlebotomy-dependent to go on this study—leukocytosis, thrombocytosis not being maintained, or symptomatic splenomegaly.

It’s important to recognize that these were modified criteria to get on the study, because what was not considered a criterion was resistance, in terms of persistent symptoms, which I would maintain is a very important reason for considering alternative therapies in our patients with polycythemia vera. Intractable pruritus can drive our patients crazy—all day itching. Hydroxyurea has not been, in my experience, a very effective drug for that—very few symptomatic reliefs are—or, if they’re having drenching night sweats or other constitutional symptoms. Those reasons for defining resistance to hydroxyurea were not considered entry criteria for the RESPONSE study. So, it’s really hematologic nonresponse, hematocrit over 45% requiring phlebotomy, uncontrolled white blood cell count, platelet count, or uncontrolled splenomegaly.

So, that was the population. There were over 200 patients in this study, and they were randomly assigned to ruxolitinib, 10 mg twice daily or best available therapy. In this study, there was a run-in period where all of the patients were brought to the same starting gate. For a month prior to the randomization, patients were allowed to get phlebotomies, and their dose of hydroxyurea can be altered in order to get the hematocrit less than 45% without phlebotomy at day 1 of the randomization. In the study, 110 patients were randomized to ruxolitinib, and 112 patients were randomized to best available therapy. As I mentioned, 60% of these patients, on best available therapy, remained on hydroxyurea.

The primary endpoint of the study was a composite endpoint. It was both hematocrit control, less than 45% without phlebotomy and a greater than 35% reduction in the splenic volume by MRI, or CT if MRI wasn’t possible. And, this primary endpoint was obtained at 32 weeks. So, it was a composite primary endpoint of the study. There were other exploratory or secondary objectives, such as rate of complete hematologic remission, defined as complete normalization of the white cell, platelet count, and the hematocrit less than 45% without phlebotomy. The duration of these responses was a key secondary endpoint. And, they collected data on symptoms. They collected data on survival. They collected data on thromboembolic events as well. But the primary endpoint was this composite one.

The results of the RESPONSE Trial demonstrate a benefit of ruxolitinib in these patients with intolerance to or resistance to hydroxyurea in PV. In terms of the composite endpoint, 60% of patients achieved a hematologic response with ruxolitinib, so the hematocrit was less than 45% without phlebotomy, versus 20% with best available therapy. The other part of the composite was the splenic size, and 40% achieved at least a 35% reduction in splenic size with ruxolitinib compared with 1% with best available therapy.

And so, when you take both of those together, the composite endpoint was achieved in 24% of patients who received ruxolitinib and 1% of patients on best available therapy. And, what was seen in a follow-up publication from the study group in Haematologica was at 80 weeks, 90% of patients maintained the components of the hematologic response (the hematocrit was less than 45%), and 43 out of 44 patients who had achieved splenic volume reduction and maintained the splenic volume reduction at 80 weeks.

Also, 50% of patients had a greater than 50% reduction in their symptom scores compared to 5% with best available therapy. So, symptoms improved, as we might have expected, from the myelofibrosis studies, COMFORT-1 and COMFORT-2. Those features are not included in the label because this was an open-label study. Since it’s not placebo-controlled, it’s hard to discern the actual benefit of the study drug on the symptoms. So, the data from the study strongly suggest that there is a benefit for these patients who are intolerant or resistant to hydroxyurea.

Finally, the one thing that we worry about, in terms of our goals of therapy, is prevention of thromboembolic disease. The RESPONSE Study looked at 32 weeks, and the number of events is going to be very, very low in a 32-week period. However, they did tabulate the number of thromboembolic events that occurred during the first period of time, roughly about a year of the study. And, there was one event with ruxolitinib and six events in the best available therapy. Now, I don’t know that these were statistically significant differences, but it’s reassuring to me that there was not a high risk of thromboembolic events with ruxolitinib.

A Patient with Disease Progression on Hydroxyurea

Case 1: