SAN DIEGO—The JAK2/FLT3 inhibitor pacritinib significantly reduces spleen volume and symptoms in patients with myelofibrosis and low platelet counts,
compared to best available therapy (BAT), according to results of the PERSIST-2 trial.
In this phase 3 trial, BAT included the JAK1/2 inhibitor ruxolitinib. And pacritinib demonstrated benefits over BAT despite a truncated trial.
The US Food and Drug Administration (FDA) placed PERSIST-2 on clinical hold in February 2016 due to concerns over interim survival results, bleeding, and cardiovascular events.
Patients randomized at least 22 weeks prior to the clinical hold contributed data to the week 24 endpoint, the results of which were presented at the 2016 ASH Annual Meeting.
John Mascarenhas, MD, of Icahn School of Medicine at Mount Sinai in New York, New York, delivered the results as a late-breaking abstract (LBA-5).
Ruxolitinib is FDA-approved to treat myelofibrosis, but it is associated with dose-limiting cytopenias and is not indicated for patients with platelet counts less than 50,000/μL.
The earlier PERSIST-1 trial demonstrated sustained spleen volume reduction (SVR) and symptom control with pacritinib therapy regardless of baseline platelet count, but it did not include ruxolitinib in the BAT arm.
PERSIST-2 study design
Patients were randomized on a 1:1:1 basis to pacritinib at 400 mg once daily (PAC QD), pacritinib at 200 mg twice daily (PAC BID), or BAT, including ruxolitinib.
Patients had to have primary or secondary myelofibrosis and platelet counts of 100,000/μL or less. Patients were allowed to have had prior JAK2 inhibitors.
Crossover from BAT was allowed after progression at any time or at week 24.
The study had 2 primary endpoints: percent of patients achieving a 35% or greater SVR and the percent of patients achieving a 50% or more reduction in total symptom score (TSS) by MPN-SAF TSS 2.0.
The primary study objective was the efficacy of pooled QD and BID pacritinib compared to BAT. The secondary objective was the efficacy of QD or BD separately compared to BAT.
Patient demographics
The study randomized 311 patients, and 221 were included in the intent-to-treat efficacy population.
The efficacy population consisted of all patients randomized prior to September 7, 2015, which allowed their data to be included in the week 24 endpoint analysis prior to the clinical hold.
The PAC QD arm consisted of 75 patients with a median age of 69 (range, 39-85). Seventy-one percent were 65 or older, and about half were male.
About three-quarters had an ECOG performance status of 0-1, 61% had primary myelofibrosis, 21% had primary polycythemia vera (PPV), and 17% had primary essential thrombocythemia (PET). About half were DIPSS Intermediate-2 risk, 80% were JAK2V617F positive, and 51% had a platelet count less than 50,000/μL.
The PAC BID arm consisted of 74 patients with a median age of 67 (range, 39-85). Sixty-two percent were 65 or older, and 65% were male.
Eighty-eight percent had an ECOG performance status of 0-1, 74% had primary myelofibrosis, 19% had PPV, and 7% had PET. About half were DIPSS Intermediate-2 risk, 80% were JAK2V617F positive, and 42% had a platelet count less than 50,000/μL.
The BAT arm consisted of 72 patients with a median age of 69 (range, 32-83). Seventy-one percent were 65 or older, and about half were male.
Three-quarters had an ECOG performance status of 0-1, 60% had primary myelofibrosis, 22% had PPV, and 18% had PET. About half were DIPSS Intermediate-2 risk, 71% were JAK2V617F positive, and 44% had a platelet count less than 50,000/µL.
Prior ruxolitinib therapy was consistent across the arms—41.3% (PAC QD), 42% (PAC BID), and 46% (BAT).
“Now, it’s important to note,” Dr Mascarenhas said, “that the most common BAT was ruxolitinib, 45%, and hydroxyurea, 19%. And 19% of BAT patients actually had no treatment, watch and wait. This highlights the fact that this is an area where there is really no other viable therapeutic option for these patients.”
Efficacy
Pacritinib-treated patients had significantly greater spleen reduction from baseline to week 24 than BAT-treated patients, with 18% (QD+BID), 15% (QD), and 22% (BID) achieving 35% or more SVR compared to 3% in the BAT arm.
Pacritinib-treated patients also experienced greater TSS reduction, with 25% (QD+BID), 17% (QD), and 32% (BID) reporting 50% or more reduction in TSS compared to 14% in the BAT arm. However, only the PAC BID arm was significantly different from BAT.
SVR in all subgroups—age, gender, JAK2 mutation status, prior treatment, platelet count, hemoglobin, peripheral blasts, and white blood cell count—demonstrated superiority for pacritinib.
Median changes in individual symptom scores were also better in the pacritinib arms than in the BAT arm in almost every category—tiredness, early satiety, abdominal discomfort, inactivity, night sweats, bone pain, and pain under ribs on the left side.
Pruritus was the only category in which BAT was superior, and that was compared to the QD arm and not the BID arm.
The majority of patients who stopped pacritinib therapy were taken off due to the clinical hold.
There were no significant differences between the groups in overall survival. Hazard ratios for overall survival (95% confidence intervals) were 0.68 (0.30, 1.53) for PAC BID vs BAT, 1.18 (0.57, 2.44) for PAC QD vs BAT, and 0.61 (0.27, 1.35) for PAC BID vs QD.
PAC BID maintained this survival advantage compared with BAT across nearly all demographic and myelofibrosis-associated risk factors.
Patients who were red blood cell transfusion-dependent at baseline experienced a statistically significant decrease in transfusion frequency from baseline to week 24 in both the QD and BID pacritinib arms compared with BAT.
And thrombocytopenia was not a significant factor for patients who were receiving pacritinib and had platelet counts less than 50,000/μL.
Toxicity
The most common treatment-emergent adverse events (AEs) associated with pacritinib were diarrhea, nausea, vomiting, anemia, and thrombocytopenia. They were generally less frequent for BID compared with QD administration.
The most common serious adverse events (SAEs)—occurring in 5% of patients or more in any arms—were anemia (5%, 8%, and 3%), thrombocytopenia (2%, 6%, and 2%), pneumonia (5%, 6%, and 4%), and acute renal failure (5%, 2%, and 2%) in the QD, BID, and BAT arms, respectively.
SAEs of interest included congestive heart failure, atrial fibrillation, cardiac arrest, epistaxis, and subdural hematoma, which occurred in 3% or fewer patients in any arm.
“They [the SAEs] were relatively infrequent, and there was not a clear signal of toxicity,” Dr Mascarenhas said.
Deaths
Deaths in the intent-to-treat evaluable population were censored at the time of full clinical hold.
For the entire enrolled population, 15/104 deaths occurred in the QD arm, 10/107 in the BID arm, and 14/100 in the BAT arm.
After the full clinical hold, 7, 10, and 6 deaths occurred in the QD, BID, and BAT arms, respectively.
Seven of 20 patients died after crossover to pacritinib. Five were due to AEs—3 cardiac, 1 bleeding, and 1 other.
“It’s important to note that progression of disease was the leading cause of death in the PAC BID arm,” Dr Mascarenhas noted. “This is after the patients stopped the drug [when the trial was on clinical hold].”
Conclusions
Despite study truncation, pacritinib (QD+BID) was significantly more effective than BAT for SVR (P=0.001) and trended toward improved TSS (P=0.079).
Pacritinib BID appeared more effective than QD versus BAT for SVR and TSS, and pacritinib BID appeared to have a better benefit/risk profile than BAT.
“This is a well-tolerated drug in many respects,” Dr Mascarenhas said. “This is a patient population that is quite ill, low platelets, poor outcome, and they did pretty well.”
When asked about the future of pacritinib, Dr Mascarenhas said he believes the benefit-to-risk ratio is in favor of the drug.
“Pacritinib offers patients in this vulnerable situation an opportunity for symptom relief, basically spleen and symptoms,” he said. “So I think this is a drug that will hopefully move forward.”
Dr Mascarenhas disclosed research funding from CTI Biopharma, the sponsor of the trial.
