Next-Generation JAK Inhibition for Patients With Myelofibrosis

Pacritinib Shows Spleen and Symptom Response in Myelofibrosis With Low Platelets

During a live virtual event, John O. Mascarenhas, MD, discussed the PERSIST-1 and PERSIST-2 trials of pacritinib for patients with myelofibrosis with lower platelet counts.

Targeted OncologyTM: What data from clinical trials supported the FDA’s approval of pacritinib for myelofibrosis?

JOHN O. MASCARENHAS, MD: The PERSIST-1 study [NCT01773187] was an upfront study of pacritinib 400 mg versus best available therapy [BAT]. It excluded ruxolitinib, so no one saw ruxolitinib previously and no one could get ruxolitinib as a best available therapy option. It was purely pacritinib or any other drug in the BAT arm. The primary end point was the classic 35% spleen volume reduction [SVR] at week 24.

The PERSIST-2 study [NCT02055781] may be a more interesting study because it focused in on this unmet need of patients with a platelet count less than 100 x 109/L. It was understood in the development of this drug that you could deliver this drug in the early-phase studies, and there was a lot less myelosuppression. Yet, even in these patients with low platelets at the full dose of the drug, you could get spleen and symptom benefit. This study was meant to tease this out and to confirm that in this patient population with low platelets. They could be ruxolitinib treated or ruxolitinib naive. [Patients] could get pacritinib at 400 mg once daily, 200 twice daily, or BAT, which could also include ruxolitinib. This study was vigorous; it had coprimary end points of SVR 35% and total symptom score improvement of 50%.

Despite the fact that these patients had low platelets and half of them had already seen ruxolitinib, almost half of them were getting ruxolitinib in the BAT arm.1 Then another 20% were getting [observation] because there were not great options for these patients, or they were getting old drugs like hydroxyurea or even trying things like prednisone. This was a very high-risk patient population and at least 40% of them had platelet counts less than 50 x 109/L. This was the first time we ever saw a large study like this in patients, particularly with low platelets that had this profile.

It was quite obvious that pacritinib was effective in this low platelet population. At the 200 mg twice daily dose of pacritinib…from the PERSIST-2 data, particularly in those with baseline platelet counts less than 50 x 109/L, the SVR rate was 29% versus 3% in the BAT arm which could also include ruxolitinib.

Symptoms also got much better with pacritinib, whether it was 400 mg or 200 mg twice daily, compared with ruxolitinib. For the 200 mg dose, the total symptom score response was 32% versus 14% with the BAT option [and 17% with the 400 mg dose of pacritinib].

The individual symptom scores performed better with pacritinib at 200 mg twice daily in the PERSIST-2 study when compared with BAT, so that is now the approved dose. So, across the board, we saw improvements, in some cases very dramatic improvements in individual symptoms with pacritinib 200 mg twice daily in this low-platelet population.

[Patients also] didn’t get a lot of anemia. Particularly, if one looks at those patients who were receiving red blood cell transfusions who received 200 mg twice daily pacritinib, there was lessening of that transfusional need from 1.06 transfusional units per month to 0.74 [in week 12] to 0.67 [in week 24]. So there was a reduction in the need for red blood cell transfusion dependence.

When you look at [percent change in] platelets with the 200 mg twice daily dose, you don’t get the same kinetics of thrombocytopenia that one generally sees with ruxolitinib where you see a more immediate reduction in platelet count, about a median of 40% reduction. Here you get some change throughout the study period, but, in the end, you get that change back to baseline. So you don’t get significant changes in the median platelet count over time.

If you look at all adverse events [AEs] regardless of attribution that [were reported at a rate of] 10% or greater, much like fedratinib, there is a degree of on-target FLT3 inhibitor-related gastrointestinal [GI] toxicity, but it’s rarely high grade and it’s easy to manage. It’s almost always in the first couple of cycles and a little bit of loperamide [Imodium] and/or antiemetics for nausea can easily handle that degree of GI toxicity, which is typically low grade.

Sometimes people ask, “[there are] thrombocytopenic patients in this study and the median platelet count didn’t change, so how is there 32% grade 3 or worse thrombocytopenia?” That is a phenomenon that’s understood to be a consequence of patients who are coming onto trial with these low platelet counts. Even a small change in their platelet count can upgrade their score to grade 3 or grade 4 and then would be captured as a new treatment-emergent AE. These patients had low platelet counts to begin with and frequently would have a change in their platelet count that would up-score their AE profile.


1. Mascarenhas J, Hoffman R, Talpaz M, et al. Pacritinib vs best available therapy, including ruxolitinib, in patients with myelofibrosis: a randomized clinical trial. JAMA Oncol. 2018;4(5):652-659. doi:10.1001/jamaoncol.2017.5818