DR LOVE: When is the optimal time to introduce a JAK2 inhibitor, and is there a rationale for treating asymptomatic disease?

DR VERSTOVSEK: The COMFORT-II study compared ruxolitinib to “best available therapy,” which in most cases was hydroxyurea. No benefit was reported with best available therapy, so one could argue that the correct way to care for patients who are symptomatic and/or have an enlarged spleen is to start with a JAK2 inhibitor. Administering a JAK2 inhibitor to patients with early-stage disease who are asymptomatic seems reasonable, but we don’t have data to substantiate that.

Many patients with myelofibrosis are older and retired, and after treatment with ruxolitinib their condition improves so much that they can perform activities they have missed for years. Ruxolitinib controls the symptoms of the disease and prolongs survival, but it is not curative. The duration of the benefits of ruxolitinib is variable. The signs and symptoms will come back, at which point one can try different options. Patients feel so much better on the agent, they can consider a bone marrow transplant to attain cure. We don’t know whether patients with myelofibrosis can have their disease controlled indefinitely with ruxolitinib. We may be able to slowly discontinue therapy over time.

A Phase I/II study initiated in the summer of 2007 at MD Anderson focused on 107 patients with myelofibrosis who received ruxolitinib. After a median follow-up of about 3 years, approximately 50% of the patients are still enjoying the benefit of the drug. So far, the maintenance of therapy on the ruxolitinib arm in the COMFORT-I study mirrors the MD Anderson study. So I would expect that approximately half of the patients would still be receiving ruxolitinib after 3 years of follow-up.

DR KANTARJIAN: The survival benefit observed with ruxolitinib in the COMFORT-I trial is impressive and suggests that early treatment can save the lives of many patients. That’s why we observe the survival benefit even though many patients crossed over from placebo to ruxolitinib.

When the data were initially published there was a belief that perhaps the responses to ruxolitinib were not durable. The Mayo Clinic updated the results from their Phase I study and showed that most of the patients were taken off the study early on. We updated our studies from MD Anderson, and we found that most of our patients continued on therapy. So one of the issues is whether part of that difference is related to different study groups or whether it was related to the beliefs or interventions of the physicians who may have taken the patients off therapy earlier and those who have continued the treatment despite the fact that the disease was stable or maybe progressing slightly.

For community practice, one of the general suggestions is, if you have a patient who has responded to ruxolitinib with significant shrinkage of the spleen and then some of the splenomegaly is recurring, do not discontinue ruxolitinib unless the patient loses all benefit from the treatment. Ultimately, now that the COMFORT-I and COMFORT-II trials are showing a survival benefit and, in general, an improvement in the condition of the patients, these data suggest that JAK2 inhibitors should continue in development and that perhaps better JAK2 inhibitors or possibly combinations of JAK2 inhibitors and other treatments could benefit patients further.

DR CORTES: It is important to acknowledge that, at least from our knowledge today, we don’t know fully what the JAK inhibitors can and cannot do. We know these agents can improve splenomegaly and symptoms significantly. So clearly the patients who would be good candidates for ruxolitinib are those who have an enlarged spleen. Those are the patients who would benefit the most.

The question is, what about a patient who does not have an enlarged spleen? In the COMFORT-I study we had a few patients who did not have an enlarged spleen or had undergone splenectomy, and they did benefit significantly with respect to their symptoms. Those are the important clinical issues.

So for a patient with an early diagnosis of myelofibrosis who has minimal symptoms and no enlarged spleen, we wouldn’t administer ruxolitinib because we do not know whether it cures the disease, whether it prevents the development of further fibrosis or whether it reverses the myelofibrosis. In those settings, we don’t know that the JAK inhibitors offer any benefit.

In our experience, more than 50% of patients are still receiving ruxolitinib more than 3 years after we started using these drugs. So most of the patients who respond can maintain their responses. This has been a little controversial. Other investigators have reported that most patients eventually discontinue therapy by the third year. In our experience, patients can experience sustained responses for a long time, and the discontinuation rate for adverse events was low. That observation was confirmed by the COMFORT-I study, in which the discontinuation rate for adverse events was 11% for patients receiving ruxolitinib or placebo.

DR CERVANTES: Myelofibrosis evolves with time, and it typically occurs in older patients who are vulnerable to developing complications from which they may die. We know that if we can improve not only the symptoms but also the general condition of the patients, this decreases the risk of developing complications. This would probably be the main reason explaining the survival prolongation.

We don’t know if starting ruxolitinib earlier confers a benefit, but I believe we will know shortly whether it will be beneficial to administer ruxolitinib to patients who are beginning to have even mild symptoms from the splenomegaly or, of course, constitutional symptoms.

DR TALPAZ: When you perform a clinical study, you want definitive parameters for the FDA. The FDA usually requires survival parameters. Patients with intermediate- and high-risk myelofibrosis have a limited survival, between 2 and 5 years. That was the reason for choosing these patients for the studies. Ruxolitinib will almost certainly be more active in the earlier stages of the disease, so there is no real reason not to administer it earlier. Studies are being conducted for patients with much earlier-stage disease — low risk and intermediate-1 — because in these patients you can get much, much longer remissions.

DR LOVE: How often do you see a patient who has symptoms, yet the risk is lower than intermediate-2?

DR TALPAZ: That happens. Patients with intermediate-2 myelofibrosis must be older than 65 years. Younger patients with symptoms will have disease that is classified as being at intermediate-1 or low risk. A 50-year-old patient who is highly symptomatic with weight loss and a huge spleen but without anemia and with a white blood cell count not yet reaching 25,000 will be classified as having low or intermediate-1 risk disease, but this patient is as sick as can be. In other words, the IPSS doesn’t characterize well how sick the patient is.

The question is, what is the primary objective? If the primary objective is deeper shrinkage of the spleen — and that is, for me, a primary objective — then I try to raise the dose as high as I can. If the spleen shrinks dramatically, it’s a worthwhile accomplishment. So my goal for most of the patients is to reach a dose of 15 mg BID of ruxolitinib, and even more, up to 20 mg BID. As long as the patient doesn’t have life-threatening thrombocytopenia or life-threatening neutropenia, I will continue.

The update of COMFORT-I, which is the randomized study comparing ruxolitinib to placebo with a crossover, has now reached 3 years of follow-up. It’s important to emphasize that 59% of the patients experienced splenic volume reduction of 35% or more, which is the criterion for a good response. Fifty percent of these patients who experienced this good response continued the response for 2 years or more, which means the notion that the response to ruxolitinib is trivial or short-lived is not necessarily correct.

The second point to emphasize is that some degree of response continues in 90% of the patients. Some degree of spleen size reduction continues beyond 2 years, which indicates that this is not a transient, short-term, symptomatic response only.

DR GOTLIB: I would have no problem trying to escalate the dose of ruxolitinib in a patient with symptomatic improvement, but I would monitor the patient closely. If the patient were obtaining some benefit from ruxolitinib but truly believed more benefit were needed, then you could increase the dose slowly and carefully and be ready to pull back if needed.