Ruxolitinib Plus Azacitidine Improves Survival, Bone Marrow Fibrosis in Myelofibrosis

Adding azacitidine to ruxolitinib led to an “encouraging” spleen and bone marrow fibrosis response rate in patients with myelofibrosis (MF), compared with previous experience with single-agent ruxolitinib, according to results from an open-label, nonrandomized, phase II trial published in Blood. The combination also was considered safe, with a low rate of treatment discontinuation due to drug-related adverse events (AEs), the authors, led by Lucia Masarova, MD, from the University of Texas MD Anderson Cancer Center, reported.

“Despite the approval of ruxolitinib for the treatment of MF, problems in the management of this disease remain,” corresponding author Naval Daver, MD, also from MD Anderson, told ASH Clinical News. “Certain patients will have disease that doesn’t respond or responds sub-optimally to the JAK1/2 inhibitor, only a small subset of patients experience reversal of fibrosis, and some will develop anemia that can restrict administration of therapeutic doses of ruxolitinib.”

In this prospective trial, researchers evaluated whether combining the JAK1/2 inhibitor with the hypomethylating agent could overcome some of these challenges. They enrolled adults with intermediate- or high-risk primary myelofibrosis (PMF; n=25), post–polycythemia vera myelofibrosis (PPV-MF; n=10), or post–essential thrombocythemia myelofibrosis (PET-MF; n=11). Participants also had an absolute neutrophil count (ANC) ≥1.0×109/L and a platelet count ≥50×109/L.

Patients were excluded from the trial if they had received prior therapy with either study agent or if they had received any therapy for MF within 14 days of starting study therapy, with the exception of hydroxyurea.

To mitigate myelosuppression and high early discontinuation rates observed in prior studies of ruxolitinib-based combinations, the study protocol called for a sequential dosing approach, starting with a three-month run-in phase of single-agent ruxolitinib. Ruxolitinib was given continuously in 28-day cycles, starting at 5, 15, or 20 mg orally twice-daily to patients with platelet counts of 50 to 100×109/L, 100 to 200×109/L, and >200×109/L, respectively.

Azacitidine 25 mg/m2 was introduced on day 1 of cycle 4, then given subcutaneously or intravenously on days 1 through 5 of each cycle. If a patient had proliferative disease or elevated bone marrow blasts (>10%), azacitidine could be introduced earlier or dose could be increased up to 75 mg/m2 at the discretion of the treating physician.

After a median follow-up of 28 months (range = 4-50 months), 33 patients (72%) responded to ruxolitinib-azacitidine treatment that lasted more than 12 weeks (primary endpoint; response defined according to International Working Group for Myelofibrosis Research and Treatment [IWG-MRT] criteria). This included two patients who had a partial remission, and 31 patients who had clinical improvement in MF symptoms.

The median time to initial clinical response was 1.8 months (range = 0.7-19.0 months). Seven of the 33 responders (21%) achieved the primary endpoint following the addition of azacitidine in the fourth treatment cycle, the authors noted. In this group, the median time to response after starting azacitidine was 4.4 months (range = 0.1-16.5 months).

Of the 34 participants who had splenomegaly >5 cm below costal margin at the time of study enrollment, 21 (62%) had clinical improvement in spleen response (>50% reduction in palpable spleen length), per IWG-MRT criteria.

Twenty-five patients (54% of the study population) had a reduced Total Symptom Score (TSS) following treatment, at a median time to response of two months (range = 0.7-16 months). Among the 29 patients with high baseline TSS scores (>20), 17 (59%) experienced a TSS response. This occurred after the introduction of azacitidine in four patients.

Responses were observed regardless of mutational profile, disease severity, presence of anemia, or splenomegaly at baseline, the investigators reported. The ruxolitinib-azacitidine combination also allowed one patient who was transfusion-dependent at baseline to achieve transfusion independence.

Bone marrow (BM) response was measurable in 31 patients (67%) who had sequential BM samples available. Nineteen of these patients experienced an improvement in BM morphology, including improvement in BM reticulin fibrosis at 24 months. “Although these are small numbers, they compare favorably to BM reticulin fibrosis improvement rates of 15.8 percent at 24 months with ruxolitinib alone in a phase I/II study,” the authors wrote.

Thirty-three patients discontinued the study; this included seven patients who continued on the combination off-study and six patients who went off therapy to undergo allogeneic hematopoietic cell transplantation. The remaining 20 patients discontinued the study due to:

  • progression to acute myeloid leukemia or accelerated-phase MF (n=5)
  • myelosuppression (n=4)
  • drug-related toxicities (n=2)
  • uncontrolled proliferation requiring cytotoxic therapy (n=2)
  • death (n=3)
  • financial or travel difficulties (n=3)
  • concurrent medical illness (n=1)

Most participants experienced at least one AE (n=44, 95%), with grade ≥3 AEs observed in 67 percent (n=31) of participants. The most frequent grade ≥3/4 AEs included anemia (n=21; 46%), thrombocytopenia (n=14; 30%), neutropenia (n=11; 14%), and infection (n=5; 11%). In addition, the investigators observed cases of new-onset anemia (n=16; 35%), thrombocytopenia (n=12; 26%), and neutropenia (n=11; 24%).

At 28-month follow-up, the overall survival rate was 67 percent. In a multivariate analysis, factors associated with longer overall survival included:

  • achievement of spleen clinical improvement (hazard ratio [HR] = 5.6; 95% CI 1.84-17.3; p=0.002)
  • non–high-risk disease (HR=5.1; 95% CI 1.7-15; p=0.003)

“Longer follow-up is required to determine whether the improved fibrosis seen with the combination of ruxolitinib and azacitidine in our trial translates to a survival advantage,” said Dr. Daver, “but at this time, we are using this combination as our frontline approach for all patients newly diagnosed with MF who require therapy.”

Limitations of the study include its small number of patients, short follow-up period, the recruitment of patients from a single center, its nonrandomized design, and the lack of a control group receiving monotherapy.

“Although we have robust and contemporary single-agent ruxolitinib controls, these are never sufficient to substitute for a true control arm,” Dr. Daver explained when asked about the study’s limitations. “A frontline randomized study evaluating single-agent vs. combination ruxolitinib in patients with newly diagnosed MF who require therapy would be ideal.”

The authors report financial relationships with Incyte, which also cofunded the study with the MD Anderson Cancer Center.


Masarova L, Verstovsek S, Hidalgo-Lopez JE, et al. A phase II study of ruxolitinib in combination with azacytidine in patients with myelofibrosis. Blood. 2018;132:1664-74.

“From a practical standpoint, the major problems with ruxolitinib monotherapy remain the lack of any effect on MF-related anemia and the lack of any apparent effect on the disease itself, as measured by BM fibrosis.

Perhaps most encouraging from this study was that 19 of 31 evaluable patients (61%) had improvement in marrow morphology, and 13 of 16 evaluable patients had a decrease in the JAK2 allele burden, with a few having a >50 percent reduction. This suggests that this combination might be affecting the disease pathogenesis – at least in some patients.

One limitation of the study is the platelet range at study entry: The range was 125-1,070×109/L, but a large proportion of patients with MF have platelet counts lower than this range. Thus, these patients might not be good candidates for this type of combination therapy.”

Peter Emanuel, MD
CHI St. Vincent
Little Rock, AR