Aziz Nazha, MD, discusses the different agents being explored for patients who have progressed on ruxolitinib, and why this is a significant unmet need. He also highlights the role of genetics in patients with MF and how a better understanding can help guide treatment decisions.
Aziz Nazha, MD
There is no doubt that ruxolitinib (Jakafi) has changed the treatment landscape for patients with myelofibrosis (MF); however, the responses to this agent do not last forever, and many patients will relapse or become refractory to the JAK1/2 inhibitor, explained Aziz Nazha, MD.
Much of the current research in the field is now focused on combination trials pairing ruxolitinib with other agents, trying to improve upon responses seen with the single agent, with results from several of these trials presented during the 2018 ASH Annual Meeting this past December.
Notably, phase II findings were presented for the combination of ruxolitinib plus azacitidine in patients with myelofibrosis and <20% blasts, which showed a 73% overall response rate (ORR) among the 52 patients on the study. Sixty-five percent of the patients had a palpable spleen reduction >50% at any time on study, and 58% had a spleen reduction by >50% at week 24.
Additionally, initial results from a phase II study of ruxolitinib plus thalidomide (Thalomid) were presented during the meeting, which showed an 80% ORR among 5 of 18 accrued patients with myelofibrosis. An additional 3 patients had improvement in anemia and symptom response and 1 patient met the criteria for a spleen response.
In an interview withTargeted Oncology, Nazha, assistant professor, Department of Medicine, Cleveland Clinic School of Medicine, discussed the different agents being explored for patients who have progressed on ruxolitinib, and why this is a significant unmet need. He also highlights the role of genetics in patients with MF and how a better understanding can help guide treatment decisions.
Targeted Oncology: What are some of the recent updates we have seen in the treatment landscape for myelofibrosis?
Nazha:The majority of the clinical trial updates we saw at ASH [fell] into combination trials with ruxolitinib plus another drug or novel drugs for patients who typically fail ruxolitinib or [where] ruxolitinib failed them. There have been a few oral [abstracts], some of which that might be of interest are phase II clinical trials with ruxolitinib plus azacitidine. That is out of the MD Anderson Cancer Center and showed an ORR around 73% with spleen response rate about 60% or so. The combination was tolerated, at least as written in the protocol, but it does have side effects. The challenge with combining therapy with ruxolitinib is that any time you combine a therapy, the side effects increase. It has some side effects, but it does have some activity.
Another trial that was also presented was ruxolitinib plus thalidomide. This enrolled about 25 patients, but it was also an early report, at least in the abstract form. They reported about 5 patients that completed 6 cycles of treatment, and about 80% of them had some sort of response to their spleen, but also toxicity was observed when you combine the 2 drugs.
There was also a dose-escalation study of ruxolitinib and interferon. It is often thought to be difficult to combine ruxolitinib with interferon because the side effects of interferon on top of ruxolitinib make it very difficult. In this study, they did a dose-escalation starting with a lower dose of interferon, going upward. They reported on a 5-cohort study. It was still early, but what they saw was that the side effects were tolerated and the activity was minimal, too. So, we have to wait for those studies to mature a little bit more.
There are a couple of other interesting studies in general. Imetelstat is a targeted RNA human telomerase that inhibits the telomerase enzymatic activity. There was a phase II study for patients who have intermediate-2 or high-risk [disease] by DIPSS [score] or patients who relapse after ruxolitinib. They reported that about one-third of the patients responded, and about 10% of the patients had spleen response, so it wasn’t really great responses, but a few patients did respond. It seems to be that the toxicity profile is manageable.
Another compound that was of interest is PRM-151. This is a recombinant human pentraxin-2 molecule. This study, mainly done at MD Anderson, really showed that about one-third of the patients will have improvement in fibrosis. The promise of this drug is it can reverse fibrosis inside the bone marrow. The activity regarding spleen size was a little bit modest, so it wasn’t really great, but about one-third of the patients have somehow improved their fibrosis, whether they have grade 3 and improved to grade 2 or have grade 2 and improved to grade 1. The challenge with this is when you grade fibrosis, there is subjectivity of the pathologist who grades the fibrosis. So, whether it’s truly that the drug is converting or reversing the fibrosis or it’s just changes you see by bias from the pathologist, that is not really clear so far.
There have been other trials presented in early stages, but these are the abstracts where there was either some mature or promising data.
Are there any other ongoing trials right now that are encouraging in this landscape?
There are multiple studies, and if you look at the studies in general, again you have a pocket of ruxolitinib plus something else, trying to improve the response to ruxolitinib. The challenge with that is that although many in vitro studies show synergy and that both drugs work in vitro and mouse models, when applied to humans, you don’t have that same activity and the toxicities go up.
You also have the other pocket where drugs are developed for after ruxolitinib has stopped working. There are 2 drugs that seem to be promising now. One of them is fedratinib, a JAK2 inhibitor, that has been around for some time. There has been there clinical trials JAKARTA-1 and JAKARTA-2 where they tried, mainly in JAKARTA-2, to use fedratinib in patients who stopped responding or failed on ruxolitinib. Really, the study showed that about 40% of the patients had improvement in their spleen size, which is a good response for a patient cohort that really has no other options. Even in patients with platelets less than 100,000, their responses were about 40%. The trial, however, was stopped because there was concern from the FDA that the drug was causing Wernicke’s encephalopathy, and the trial was put on hold, so the drug died for some time. After longer follow-up and subsequent follow-up with the FDA, it turns out that these were over reported and were not really true. What happened is that Celgene bought the drug from the company, and now they are conducting a trial called the FREEDOM trial [NCT0375518]. Again, this is fedratinib in patients who failed or are not responding anymore to ruxolitinib. If the trial is positive, this will actually grant fedratinib an FDA approval. That’s more realistic for fedratinib. It just started in late 2018, so it’s something that will take a few years to see results and see whether the drug will be approved or not
The other drug is pacritinib, which is also a JAK2 inhibitor. What the company tried to target [with pacritinib] is the patients that have lower platelets who are typically difficult to target, because pacritinib can still work even if the platelets are less than 100,000. Pacritinib has been in clinical trials, PERSIST-1 and PERSIST-2. PERSIST-2 is a trial for patients with primary MF who have failed ruxolitinib and received either pacritinib 400 mg daily or 200 mg twice a day compared to best available therapy. Really, pacritinib actually showed activity in those patients. Although it can still cause thrombocytopenia, it showed activity in patients who had thrombocytopenia, and it did not affect their platelets that much. It seems the activity really didn’t matter whether the platelets were 50,000 or more than 50,000.
In 2017, the FDA put a halt on the PERSIST-2 trial. The reason was because they were some that thought the drug was [associated with] thrombotic events that were not reported preoperatively, but after further review from the FDA, they took the lift off. Clinical trials for pacritinib are still ongoing, especially with the twice daily doses. If the results we saw in the earlier trials still stand, it could also be interesting to see whether this drug will make it to approval or not.
What are the most pressing challenges or unmet needs you still see in this space?
I guess the challenge now is that we have ruxolitinib now, and it has really changed the landscape for myelofibrosis. However, the responses to ruxolitinib don’t last forever. If you look at the COMFORT-I and COMFORT-II trials, which led to the approval of ruxolitinib, about 50% of the patients will discontinue ruxolitinib at the 3-year mark and about 75% of them will be off ruxolitinib around 5 years after treatment.
There are some hematologic toxicities [associated with] using ruxolitinib in myelofibrosis, mainly anemia and thrombocytopenia, that will lead to discontinuation at a certain time point. What has been reported in the trial is [a rate of] about 25% to 35%, but the number is challenging, of course, as the numbers could be higher if community oncologists are using the drug because they tend not to [alter] the doses [up and down]. It depends how you manage those toxicities. However, there is one-quarter of patients who are able to tolerate ruxolitinib hematologic side effects, mainly anemia and thrombocytopenia.
The other side is, there are some patients with MF with platelets less than 50,000; it is a challenge to treat those patients because if you follow the ruxolitinib package insert, you should not start ruxolitinib with platelets lower than 50,000. Although, we have done that in practice where we start lower doses and you see a little bit of activity, but you can still start in a lower dose and then [increase].
You have this patient population that fails ruxolitinib, or ruxolitinib failed them, and they either responded and then at some point ruxolitinib stopped workingthe spleen started enlarging and they began having constitutional symptoms—or, they developed hematologic side effects that led to their removal from ruxolitinib. That is one pocket that has a huge unmet need.
The other pockets are the patient populations of MF that have lower platelets and anemia, and it becomes very difficult to put them on ruxolitinib. That is when it presents as a problem.
What are we learning about the genetics of MF and how can we use this information to guide treatment?
There have been several studies looking at the genetic landscape of MF. Traditionally, you have what we call canonical mutations; these areJAK2V617F mutations. This occurs in about 50% to 60% of patients with primary MF. You have calreticulin(CALR)mutations that occur in about 25% to 30% of patients, and you haveMPLmutations, which occur in about 5% to 10% of patients.
You have about 10% to 15% of patients who don’t have these mutations, which they call triple-negative with theJAK,CALR, andMPL, but they have other mutations. We have studies looking at other mutations, specifically in the triple-negative patients. Most commonly in the triple-negative patients, more than 8% of them will carry a mutation of eitherASXL1,EZH2,SRSF2,IDH1,orIDH2. When those happen, some of the studies, though small, show that this could affect the overall survival (OS) of the patient. Also, the number of those abnormalities could affect the OS of these patients.
Traditionally, the models that have been used in the clinic to risk stratify patients with MF, like IPSS, the dynamic IPSS (DIPSS), or DIPPS-Plus. All these models use clinical variables, such as white blood cell count, hemoglobin platelets, age, constitutional symptoms, blast percentage, and cytogenetics. But recently, there has been an addition of the molecular data to the scoring system with a new scoring system: MIPSS70 and MIPSS70-Plus. What the authors showed was that in these 2 cohorts from Mayo Clinic and an Italian cohort of about 150 patients with MF, when the [molecular data] were incorporated, if you have one of those non-canonical mutations or 2 or more these mutations, that can have a significant impact on OS.
There have also been some studies trying to look at whether the molecular profile of the patient can predict response to ruxolitinib. These studies are a little small and not replicated in an independent cohort, but generally speaking, the studies have shown that if you have aJAK2mutation allelic burden of about 50%, you have a higher chance of spleen reduction. One study showed that if the patient has aCALRmutation with any mutation inASXL1, their response to momelotinib, one of the JAK2 inhibitors, was higher in terms of shrinkage of the spleen. Some studies have shown that those non-canonical mutations, if you have 2 or less, the response is better compared to if you have 3 or more.
That’s generally speaking, it’s still a work in progress. We have adopted MIPSS70 in terms of prognostication, but in terms of identification for therapy, we have not really put that much weight on it, at least in clinical practice, just in research, since we only have ruxolitinib.