Etanercept, a soluble tumor necrosis factor receptor, palliates constitutional symptoms in patients with myelofibrosis with myeloid metaplasia: results of a pilot study: Presented at the 43rd annual American Society of Hematology meeting, Orlando, Florida, December 2001.

Myelofibrosis with myeloid metaplasia (MMM) is an aggressive clonal myeloproliferative disorder characterized by peripheral blood cytopenias, reactive marrow fibrosis, and extramedullary hematopoiesis.1 Constitutional symptoms such as night sweats, low-grade fevers, fatigue, and cachexia are common and debilitating in MMM, but currently available treatments for the disease have little effect on these symptoms.

Several cytokines are implicated in the pathogenesis of marrow fibrosis.2 These cytokines include transforming growth factor β, basic fibroblast growth factor, vascular endothelial growth factor, platelet factor 4, calmodulin, and tumor necrosis factor (TNF).1 TNF directly inhibits hematopoiesis,3,4 stimulates fibroblast proliferation,5 and is a key mediator of fever and cachexia.6,7 Anti-TNF approaches are currently being evaluated in cancer-related cachexia and in a wide range of disorders associated with fibrosis.

We hypothesized that anti-TNF therapy might be effective in MMM. We report an open-label pilot study of etanercept, a dimeric soluble recombinant form of the extracellular domain of human p75 TNF receptor fused to the Fc fragment of human immunoglobulin G1, in patients with MMM complicated by anemia and/or constitutional symptoms.

We recruited patients with MMM from the hematology practice at the Mayo Clinic in Rochester, MN. All patients gave informed consent before study enrollment. To be eligible, patients had biopsy-confirmed MMM (a senior pathologist, C.-Y.L., reviewed all marrow samples) and did not have the bcr-abltranslocation. We excluded patients with pre-existing infections or poor liver or kidney function (direct bilirubin level higher than 2 times the upper limit of normal, aspartate aminotransferase level higher than 5 times normal, or creatinine level higher than 2.0 mg/dL). All other therapies for MMM, including hematopoietic growth factor support, were stopped at least 4 weeks before trial initiation.

The institutional review board of the Mayo Clinic approved the study protocol. Patients self-administered 25 mg etanercept (Enbrel; Immunex, Seattle, WA) subcutaneously twice weekly. Patients responding to therapy at 12 weeks had the option to continue the drug for an additional 12 weeks. All patients completing 24 weeks of therapy underwent a follow-up marrow biopsy.

There are no standard response criteria for clinical trials in MMM. The minimal responses we defined as necessary to qualify as a clinically meaningful improvement in this trial are listed in Table 1.

At the start of the trial and every 4 weeks thereafter, we asked patients about the following symptoms: presence or absence and degree of night sweats and fevers, daily mean fatigue level using a 10-point numerical analog scale, and unintentional weight loss. We weighed patients at enrollment and monthly thereafter. Spleen size was determined by physical examination.

We enrolled 22 patients in this study. Of these, 2 patients were excluded shortly after enrollment: one patient with a pre-existing but undiagnosed dental abscess, and a second patient traveling to Mexico who forgot to take the study medication with him. These patients took less than 2 weeks of study medication. Of the 20 evaluable patients (aged 45-82, 15 men), 18 had anemia, 8 had thrombocytopenia, and 13 had palpable splenomegaly at the time of trial enrollment (3 had previously undergone splenectomy) (Table 2). With regard to constitutional symptoms, 8 patients had drenching night sweats, 7 complained of unintentional weight loss in the previous 6 months (ranging from 5 to 23 kg, median of 7 kg), 1 had persistent unexplained low-grade fevers, and all 20 patients complained of excessive fatigue.

Objective responses were seen in 4 patients (Table 2). There were 3 hemoglobin responses (1 increase in hemoglobin by 3 g/dL and achievement of transfusion independence, 1 stable increase in hemoglobin by 1 g/dL, and 1 decrease in red cell transfusion requirements by > 50%), 1 platelet response (increase from 16 000 to 182 000/μL), and 1 spleen response (decrease from 10 cm to 2 cm below the costal margin on shallow inspiration). However, 1 patient had marked worsening in his blood counts (hemoglobin dropped from 8.2 g/dL to 6.3 g/dL, absolute neutrophil count from 2630 to 640/μL, and platelet count from 166 000 to 136 000/μL). This patient was the only person who discontinued the drug because of toxicity, and the cytopenias resolved promptly.

Overall, a constitutional improvement of one type or another was seen in 12 of 20 patients. Improvements were seen in 6 (75%) of 8 patients with drenching night sweats (5 of the 6 enjoyed complete cessation), 7 (100%) of 7 patients with unintentional weight loss (2 simply stopped losing weight and 5 gained weight with range from 1.2 to 5 kg), and 10 (50%) of 20 patients with fatigue. Four patients who did not complain of weight loss at the beginning of the trial gained weight during the study (range, 1.5-4.5 kg). We observed no improvement in the 1 patient with unexplained fevers.

In addition to the 1 patient with pancytopenia, toxicity attributable to the drug included mild injection-site reactions in 4 patients, development of new low-titer antinuclear antibody in 2 patients, minor infections (upper respiratory infections and gastroenteritis) in 2 patients, and myalgias, abdominal cramping, and mild sedation on the day of injection in 1 patient each. No serious infections were observed. One patient who did not have night sweats at the beginning of the study developed them while on the study.

Nine responding patients elected to continue therapy for the full 24 weeks and underwent a follow-up marrow biopsy. There were no changes in the degree of marrow fibrosis. Minor alterations in overall marrow cellularity were seen, but these alterations were not consistent and sampling error cannot be excluded.

In this preliminary study, etanercept at a dose of 25 mg subcutaneously twice weekly was well tolerated and relieved troubling constitutional symptoms in some patients with MMM. We observed several objective and clinically significant responses in peripheral blood counts and spleen size, but caution is warranted because one patient developed pancytopenia. We did not observe a change in marrow fibrosis in this study. This finding may be because etanercept has no effect on marrow fibrosis or because 24 weeks of anti-TNF therapy may not be long enough to effect a change in marrow fibrosis.

This study is limited by its relatively small sample size (in part due to the rarity of MMM), the subjective nature of constitutional end points, and the lack of blinding of patients and outcome assessors. Therefore, these results will need to be confirmed in a larger, randomized study using placebo or active controls and with appropriate blinding of patients and outcome assessors. In light of these preliminary results, similar studies of the palliative role of other anti-TNF agents (eg, infliximab) should be conducted.

If more definitive studies confirm this study's promising results, the widespread use of etanercept in MMM may be limited by the need for parenteral administration and by its high cost. The wholesale pharmacy cost of 8 doses (25 mg) of etanercept in our institution is US$888, and facility and administration charges increase the patient price to more than US$1600 per month. In fact, several patients in this trial who wanted to continue etanercept beyond the completion of this study could not pay for the drug and had to discontinue it. A financial assistance program is available from the manufacturer.

In conclusion, etanercept may be safe and effective in palliating constitutional symptoms in some patients with MMM. The most severe toxicity observed was reversible pancytopenia. Patients with MMM and bothersome hypercatabolic symptoms may benefit from a trial of this medication.

Etanercept was provided free of charge to the patients for the duration of the study by the Immunex Corporation of Seattle, WA. The drug manufacturer did not have access to or control of study data and did not participate in the preparation of this report. We thank Dr Victor Montori for his helpful comments on the manuscript.