Lenalidomide and prednisone for myelofibrosis: Eastern Cooperative Oncology Group (ECOG) phase 2 trial E4903

Myelofibrosis (MF) includes primary MF. MF that has arisen from an antecedent polycythemia vera (PV) or essential thrombocythemia (ET; post-PV/ET MF)¹  leads to anemia and/or splenomegaly. Anemia not only contributes to the substantial symptoms of fatigue associated with MF,²  but is strongly correlated with shorter survival.^3,4 

Pilot studies with thalidomide (with and without prednisone tapers) have shown improvement of anemia and splenomegaly in approximately 20%-60% of subjects yet can lead to problematic neuropathy.^5,6  Lenalidomide, is a first-generation IMiD immune-modulatory drug with pleiotropic cytokine-modulating activity in a large phase 2 trial of 68 subjects.⁷  Lenalidomide, 10 mg/d, resulted in the overall response rates of 22% for anemia, 33% for splenomegaly, and 50% for thrombocytopenia.⁷ 

Based upon the single-agent activity of lenalidomide and the synergistic/incremental benefit a prednisone taper added to thalidomide, a cooperative group phase 2 trial combining lenalidomide with a prednisone taper was undertaken by the Eastern Cooperative Oncology Group (ECOG).

Persons with primary or post-ET/PV myelofibrosis were eligible. All subjects had to have a baseline hemoglobin < 10 g/dL or be red blood cell (RBC)–transfusion dependent (1 transfusion in the 2 months before enrollment). Subjects were required to have discontinued prior therapy for their disease (including corticosteroids) for at least 14 days before enrollment. Adequate hepato-renal function, neutrophils (> 1 × 10⁹/L), and platelets (> 100 × 10⁹/L) were required. Subjects were counseled regarding the potential teratogenicity of lenalidomide and on the use of contraception, and required (for females of childbearing age) to undergo serial pregnancy examinations. This study was approved by the institutional review boards of all participating institutions.

Eligible subjects were begun on lenalidomide, 10 mg/d, by mouth continuous dosing on a 28-day cycle. For the first month, prednisone (30 mg/d by mouth) was given and was decreased to 15 mg/d for the second month and to 15 mg every other day for the third month. Subjects continued lenalidomide only for up to 3 more cycles. Subjects discontinued therapy for progressive disease or unacceptable (in the determination of treating physician or patient) toxicity at any time during the treatment phase.

Response assessments while on the study used the International Working Group for Myelofibrosis Research and Treatment (IWG-MRT) criteria.⁸  Responses most applicable for the analysis of this trial fall under the category of clinical improvements: increased hemoglobin > 2 g/dL above baseline for ≥ 2 months or RBC-transfusion independence for the same interval. Similarly, clinical improvements for splenomegaly require > 50% reduction in the palpable component of the spleen for ≥ 2 months.

We report results with a median follow-up of 2.3 years (range, 1.4-3.1 years) from trial entry (Figure 1). Forty-two subjects were evaluable for response. Six subjects were enrolled but not treated for the following reasons: too high a hemoglobin at entry (n = 2), registration issues (n = 2), thrombocytopenia (n = 1), or use of concurrent therapy (n = 1). Demographics were typical of MF (Table 1).

A median of 6 cycles of therapy was given: 10 subjects received 3 months, and 25 subjects received 6 months of therapy (Figure 1). Premature discontinuation of therapy occurred in 17 subjects because of progressive disease (n = 3), toxicity (n = 7), subject withdrawal (n = 5), death from disease (n = 1), or seeking alternative therapy (n = 1).

Therapy with lenalidomide and prednisone was associated with toxicity in 37 subjects with ≥ 1 grade 3 or 4 toxicity (Table 1). Neutropenia (50% grade 3-4) and thrombocytopenia (17% grade 3-4) were common. Anemia was observed in 20 subjects (grade 3-4); it is difficult to distinguish in these individuals anemia as a toxicity of therapy versus anemia arising from their underlying illness. Thrombotic events were observed in 3 subjects. Nonhematologic toxicities were less frequent but still common (Table 1) with 19 subjects (45%) having at least 1 grade 3-4 event, and 18 having a grade 1-2 event (42%). Two subjects had hyperglycemia, presumably from the prednisone.

Among 42 subjects eligible for response assessment, the best confirmed response (by IWG-MRT criteria) observed was clinical improvement in 10, stable disease in 29, and progressive disease in 3. Improvement in anemia was seen in 8 subjects. Grouping the 4 subjects who were not receiving RBC transfusions, median hemoglobin increase was 2.75 g/dL (range, 1.6-3.8 g/dL). Median time from randomization to initial response was 2.0 months (range, 0.1-5 months). Major reductions in spleen size occurred in 4 subjects (10%). Median time from randomization to response is 4.2 months (range, 0.5-9.0 months). Serial bone marrow analyses were available in a subset of subjects. These showed no substantial decrease in fibrosis, hypercellularity, or angiogenesis

Currently, 27 subjects completed planned therapy, 7 stopped early because of toxicity, 5 withdrew by choice (largely logistical), 6 had disease progression, and 2 stopped because of comorbidities. Twenty-three of 42 subjects are alive, 18 died, and 1 refused follow-up. Among 10 responders, 3 relapsed and 1 died without relapse. One- to 2-year event-free survival (EFS) is 82% (95% confidence interval [CI], 70%-94%) and 3-year EFS is 61% (41%-81%). Median EFS among responders is 36.6 months.

Lenalidomide achieves a modest response rate (< 20%) in persons with MF and anemia with the potential to decrease splenomegaly. However, tolerability of lenalidomide is problematic. A significant proportion of subjects have bone marrow suppression and or nonhematologic toxicities. In addition, the risk of thrombosis was substantial with 3 events despite aspirin (81 mg/d) prophylaxis.

The first issue is whether there is a role for lenalidomide in MF given the potential toxicity and modest activity. Three trials (including this report)^7,9  indicate lenalidomide is active. Options for anemia in MF are limited, thalidomide is active but neuropathy typically prevents prolonged use. Given limited options, using lenalidomide seems reasonable but is best considered for subjects without neutropenia or thrombocytopenia. Presence of a del5(q) is associated with a more robust response¹⁰  to lenalidomide perhaps mirroring the benefits seen with myelodysplastic syndrome.¹¹ 

Is there any benefit from adding prednisone to lenalidomide? Our anemia response rate was 19% and the splenomegaly response rate was 10%. Our prior trial of lenalidomide without prednisone reported 22% for anemia and 33% for splenomegaly.⁷  Comparing different trials and trial designs is often difficult or inappropriate. However, adding prednisone to lenalidomide did not appear to increase responses. This contrasts with a recent report of lenalidomide and prednisone with a 30% anemia response (42% for splenomegaly response).⁹  Similar rates of bone marrow suppression (58% neutropenia grade 3 or 4) were reported. Lower response rates in this trial may reflect the multicenter nature of our trial compared with the single institution experience reported previously.⁹ 

Pomalidomide is a second-generation IMiD immunomodulatory compound significantly more potent than thalidomide, but with no neuropathy or sedation.¹²  We reported an adaptive phase 2 randomized trial of pomalidomide with a 20%-30% anemia response in a low-dose cohort without substantial bone marrow suppression or neuropathy.¹³  A similar anemia response rate was seen at low dose in a phase 1/2 study.¹⁴  A phase 3 trial of pomalidomide for anemia of myeloproliferative neoplasm (MPN)–associated myelofibrosis is planned.

In this large, rapidly accruing trial we demonstrated that a great demand for MF trials exists and the cooperative group mechanism worked well for an MPN trial. The results suggest lenalidomide as a modestly active, but frequently toxic, therapy to consider in a subset of MF patients particularly those with del5(q). Future use may be supplanted by pomalidomide, or histone deacetylase, inhibitors.

The authors acknowledge the investigators and study assistants of the ECOG for their support and participation in this trial. In addition, the authors acknowledge the ECOG Leukemia Committee for their support in the design and conduct of this trial.

National Institutes of Health

Contribution: R.A.M. L.D.C., J.R, M.L., E.P., A.T., and M.S.T. designed and coordinated the clinical trial; R.A.M. and X.Y. analyzed the data from the clinical trial; C.Y.L. reviewed the serial bone marrow biopsies obtained during this trial; and all authors contributed to and approve the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Ruben A. Mesa, MD, Professor of Medicine, Division of Hematology and Oncology, Mayo Clinic, 13400 E Shea Blvd, Scottsdale, AZ 85259; e-mail: mesa.ruben@mayo.edu.