Myelofibrosis

Myelofibrosis with myeloid metaplasia, also known as agnogenic myeloid metaplasia, chronic idiopathic myelofibrosis, and primary myelofibrosis,^[1] was first described in 1879 and is currently classified as a myeloproliferative disease caused by the growth and proliferation of an abnormal bone marrow stem cell, resulting in the replacement of the bone marrow with fibrous connective tissue. An eponym for the disease is Heuck-Assmann disease, or Assmann's Disease.

Additional recommended knowledge

Guide to balance cleaning: 8 simple steps

Daily Visual Balance Check

Weighing the right way

Presentation

The bone marrow is replaced by collagen fibrosis, impairing the patient's ability to generate new blood cells resulting in a progressive anemia. It is usually reactive following other myeloproliferative disorders, such as polycythemia rubra vera or essential thrombocytosis. Extramedullary haematopoeisis occurs as the haemopoetic cells migrate away from the bone marrow, to the liver and spleen. Patients often have hepatosplenomegaly and poikilocytosis. The mean survival is 5 years and causes of death include infection, bleeding, organ failure, portal hypertension, and leukemic transformation.

Diagnosis

Diagnosis is based upon:

1. Normochromic normocytic anaemia
2. Red cell poikilocytosis on blood film (tear drop RBCs)
3. JAK 2 mutation on Val 617 Phe locus in 50%
4. Raised lactate dehydrogenase
5. Raised neutrophil alkaline phosphatase score
6. Bone marrow biopsy may show increased cellularity and fibrosis

Treatment

Treatment of chronic idiopathic myelofibrosis (primary myelofibrosis, agnogenic myeloid metaplasia) Ayalew Tefferi, MD

INTRODUCTION — Agnogenic myeloid metaplasia (AMM, called chronic idiopathic myelofibrosis (CIMF) in the WHO classification and primary myelofibrosis by the International Working Group on Myelofibrosis Research and Treatment), is one of the chronic myeloproliferative diseases . In addition to AMM, the group includes essential thrombocythemia (ET), and polycythemia vera (PV), both of which may undergo delayed disease transformation into a fibrotic state called postthrombocythemic myeloid metaplasia (PTM) and postpolycythemic myeloid metaplasia (PPM), respectively. The conversion rates after 10 to 20 years of disease are less than 5 percent for ET and approximately 10 to 20 percent for PV. AMM, PPM, and PTM are all referred to as myelofibrosis with myeloid metaplasia (MMM).

The primary disease process in AMM is a clonal hematopoietic stem cell disorder which results in chronic myeloproliferation and atypical megakaryocytic hyperplasia. The secondary process of bone marrow fibrosis (BMF) is the result of nonclonal fibroblastic proliferation and hyperactivity induced by growth factors abnormally shed from clonally expanded megakaryocytes. BMF is the hallmark of AMM and contributes to the impaired hematopoiesis that leads to severe anemia. In addition to BMF and anemia, patients with AMM have marked degrees of splenomegaly, extramedullary hematopoiesis, and severe constitutional symptoms. Similar to the situation with other stem cell diseases, the only treatment modality with curative potential in AMM involves allogeneic hematopoietic cell transplantation [1,2]. All other current treatment strategies are palliative [3,4].

This topic will review the treatment of AMM [5]. An overview of the myeloproliferative disorders, as well as discussions of pathogenetic mechanisms, clinical manifestations, and diagnosis of AMM are discussed separately. (See "Overview of myeloproliferative disorders" and see "Clinical manifestations and diagnosis of chronic idiopathic myelofibrosis (primary myelofibrosis, agnogenic myeloid metaplasia)" and see "Pathogenetic mechanisms in chronic idiopathic myelofibrosis (primary myelofibrosis, agnogenic myeloid metaplasia)").

PROGNOSIS — Most patients with AMM present with anemia, marked splenomegaly, early satiety, and hypercatabolic symptoms including severe fatigue, low grade fever, night sweats, and weight loss [3]. During their clinical course, most patients experience massive hepatosplenomagaly along with progressive anemia requiring frequent red blood cell transfusions. Portal hypertension may occur due to either marked splenomegaly or intrahepatic obstruction. In some patients, extramedullary hematopoiesis (EMH) develops around the spinal column, causing cord compression, or in the pleural and peritoneal cavity, causing pleural effusion or ascites, respectively.

In an epidemiologic study of patients within Olmsted county, the three-year survival rate was 52 percent [6]. Similar results have been noted by others. In one series, risk factors for decreased survival included advanced age (>60 years), hepatomegaly, weight loss, anemia (hemoglobin <10 gm/dL), leukocytosis (white blood cell count [WBC] >30,000/microL), leukopenia (WBC <4000/microL), circulating blasts (>2 percent), thrombocytopenia (platelet count <150,000/microL) and abnormal karyotype [7]. In contrast, splenomegaly and the degree of bone marrow fibrosis did not affect survival.

A simplified prognostic scoring system was devised, based upon the following two findings [7]: Hemoglobin <10 g/dL White blood cell count <4000/microL or >30,000/microL

Median survivals were 93, 26, and 13 months for those with none (low risk), one (intermediate risk), or both (high risk) of these adverse prognostic risk factors, respectively.

In a Mayo Clinic cohort of 160 patients with AMM <60 years of age, we devised a prognostic scoring system based on the following three adverse prognostic findings [8]: Hemoglobin <10 g/dL White blood cell count <4000/microL or >30,000/microL Platelet count <100,000/microL

Median survivals were 155, 69, and 24 months for those with zero, one, or 2 of these adverse features.

Several other studies have reported similar prognostic parameters in AMM [9-12]. Some reports have identified other predictors for worse survival including constitutional symptoms [13], clonal cytogenetic abnormalities [14-16], decreased erythropoiesis or red cell aplasia [12], increased plasma levels of soluble interleukin (IL)-2 receptor [17], and advanced bone marrow fibrosis [18]. In one series of 47 patients, for example, 32 percent had clonal abnormalities; these patients had a much shorter median survival than those with normal chromosomes (30 months versus not reached at six years) [14].

Response criteria — In addition to serving as prognostic parameters, changes in these disease-related features (eg, anemia, splenomegaly, constitutional symptoms, WBC and platelet counts) may be used to evaluate response to treatment [19,20].

Patients of transplant age — The prognostic variables in the above studies were determined in all patients, most of whom were over the age of 55 and not candidates for allogeneic hematopoietic cell transplantation (allo-HCT). This issue was addressed in a report of 123 patients with AMM under the age of 55, who might be candidates for allo-HCT [21]. This subgroup, representing approximately 25 percent of all patients with AMM, had a median survival of 11 years, three years longer than the low-risk group noted above [7]. Three factors predicted poor survival in these younger patients: Hemoglobin <10 g/dL Constitutional symptoms (fever, sweats, weight loss) Circulating blasts >1 percent

Median survival was 15 and 3 years for those with no or one adverse factor versus two or more adverse factors, respectively.

TREATMENT — At present, allogeneic hematopoietic cell transplantation (allo-HCT) constitutes the only treatment modality with a curative potential in AMM. Other modalities that have been successfully employed include androgens, chemotherapy, hydroxyurea, anagrelide, splenectomy, radiation therapy to the spleen, and thalidomide [22,23].

Allogeneic HCT — Allo-HCT has traditionally been limited to patients below the age of 60 years and to those who have an HLA-identical sibling donor, although there has been increasing use of both matched unrelated and mismatched related donors. (See "Donor selection for hematopoietic cell transplantation").

In addition to limited availability of donors, other problems associated with allo-HCT include a high transplant-related mortality due to complications such as acute and chronic graft-versus-host disease (GVHD). On the other hand, responses to donor lymphocyte infusions in patients relapsing after allo-HCT suggest the presence of a graft-versus- myelofibrosis effect in this disorder [23,24]. (See "Clinical manifestations and diagnosis of acute graft-versus-host disease" and see "Clinical manifestations and diagnosis of chronic graft-versus-host disease").

Most reports of allo-HCT in AMM have included small numbers of patients [1,2,25-28]. Furthermore, the median age was usually about 40; this represents a small subset of patients, since only 17 percent of those with AMM present before the age of 50 [9].

A study from the French Society of Bone Marrow Transplantation described 12 patients with AMM (mean age 40 years), 10 of whom had been splenectomized [1]. All but one of the donors was an HLA-identical sibling. Engraftment occurred in 11, grade II-IV acute GVHD in 10, and extensive chronic GVHD in four. At a median follow-up of 25 months, estimated 4-year overall and event-free survivals were 71 and 59 percent, respectively.

A report from Seattle described allo-HCT in 56 patients (mean age 43, range: 10 to 66 years) with AMM or the spent phase of polycythemia vera or essential thrombocythemia [2]. Estimated three-year overall survival was 58 percent for the entire group and 76 percent for those conditioned with targeted dose busulfan plus cyclophosphamide. The 100-day treatment related mortality was 14 percent.

A third study included 55 patients with AMM (median age 42 years) who underwent allo-HCT, 90 percent of whom engrafted [29]. Overall survival at five years was 47 percent; 22 patients (40 percent) achieved a complete histohematologic remission (ie, complete hematologic remission plus disappearance of marrow fibrosis) . Five- year survivals of 83, 43, and 31 percent were seen in the low, intermediate, and high risk groups, as defined above (see "Prognosis" above) [7]. Predictors of survival at five years were hemoglobin >10 g/dL and the absence of osteomyelosclerosis; predictors of failure were abnormal pretransplant karyotype, absence of grade II to IV acute GVHD, and age >45 at transplant [30].

It has been concluded that young patients with two or more adverse features (ie, hemoglobin <10 g/dL, constitutional symptoms, isolated cytogenetic abnormality, or blasts >1 percent) [21,29] should be considered for HCT shortly after diagnosis. For low-risk patients, who might live 10 to 15 years with supportive treatment [21], but might have a transplant-related mortality of at least 8 percent, the answer is not yet clear. Until further information is available, a suggested strategy is to offer transplantation to the latter group when an adverse risk factor appears [23,31].

Although the above study noted an estimated five-year overall survival post-transplant of 14 percent for patients 45 years of age [29,30], a preliminary report from Seattle indicated three- and five-year disease-free survival rates of 63 and 50 percent, respectively, for patients in that age range [32].

Initial studies raised a concern that engraftment would be delayed in the severely fibrotic marrow. However, the above observations suggest that engraftment is not a problem in most patients with MMM. This is in keeping with a study that compared the outcome after HCT in 33 patients with severe myelofibrosis to 33 matched controls without myelofibrosis; there was no significant difference between the two groups in engraftment parameters [33]. Similarly, preliminary information from Seattle indicated no significant advantage or disadvantage of splenectomy on posttransplantation outcome [34].

  Nonmyeloablative allogeneic HCT — A number of studies have reported on the successful use of allogeneic HCT following reduced-intensity conditioning in patients with AMM [23,24,35-38]. In one series of 21 patients, treatment-related mortality was zero and 16 percent at 100 days and one year, respectively [37]. Complete histopathological remission was achieved in 75 percent, with estimated 3-year overall and disease-free survivals of 84 percent.

Autologous HCT — Autologous HCT using peripheral blood stem cell rescue following myeloablative single agent chemotherapy with busulfan has been attempted in 21 patients with advanced MMM (ie, anemia, thrombocytopenia, neutropenia, symptomatic splenomegaly, and/or inadequately controlled disease-related symptoms) [39]. There were 6 deaths, 3 of non-relapse causes. At a median follow-up of 390 days, estimated two-year actuarial survival was 61 percent. Responses were noted in 10 of 17 anemic patients and 4 of 8 patients with thrombocytopenia, while symptomatic splenomegaly improved in 7 of 10 patients. Further investigation of this approach appears warranted.

Androgens — Patients who are not candidates for allo-HCT are treated palliatively. The combination of an androgen preparation (fluoxymesterone, Halotestin, 10 mg PO twice per day) and a corticosteroid (prednisone 30 mg/day PO) improves anemia in some patients. After one month of therapy, treatment with fluoxymesterone is continued in responders and the corticosteroid is tapered.

All patients treated with androgens should have periodic monitoring of liver function tests and men should be screened for prostate cancer (digital rectal examination and measurement of serum prostate specific antigen) before initiating therapy. The virilizing side effects of this treatment program should be emphasized in advance to female patients.

In our experience, approximately one-third of the patients respond to androgens, although rates varying from 29 to 57 percent have been reported [40,41]. In one series of 23 patients, the response rate was 92 percent in those with normal chromosomes compared to 22 percent in those with chromosomal abnormalities [41]. Nonresponders were more likely to have severely compromised hematopoiesis, as evidenced by the presence of thrombocytopenia, erythroid suppression on ferrokinetic studies, and lack of activity on bone marrow scans.

In a controlled study of 24 patients with chronic myeloproliferative disease (more than one-half of whom had myelofibrosis), a good response to fluoxymesterone, 30 mg/day, was seen in 4 of 14 (29 percent) compared to no good responses in the control arm [40]. As in the above-noted study, non-responders were more likely than responders to have reduced effective erythropoiesis.

Patients who fail to respond to one androgenic preparation may respond to another, such as oxymetholone or one of the injectable testosterone preparations. (See "Testosterone treatment of male hypogonadism", section on Testosterone preparations). Responses may not occur until after at least three months of treatment.

Supportive therapy for anemia — Anemic patients not responding to androgen therapy are usually supported by periodic red cell transfusions. Secondary hemosiderosis from chronic blood transfusion may be palliated by the use of chelation therapy. (See "Iron overload syndromes other than hereditary hemochromatosis", section on Chelation therapy).

Other modalities have been tried with variable success: Erythropoietin or darbepoietin have generally not been successful in alleviating the anemia associated with AMM [42-44], although others have reported responses, most often in patients not requiring transfusional support and/or those with inappropriately low serum erythropoietin levels [45-47]. Some patients respond to danazol (200 to 800 mg/day), both in our experience and that of others [48,49]. Approximately 25 to 50 percent of anemic patients experience a reduction in transfusion requirement after splenectomy (see "Splenectomy" belowsee "Splenectomy" below) [50-52]. Coombs-positive autoimmune hemolytic anemia has been described in AMM; it usually responds to prednisone at a dose of 40 mg/m2 per day PO.

Immune mechanisms also may impair erythropoiesis. Cyclosporine has been employed in occasional patients with AMM, severe anemia, and immune defects who do not respond to prednisone. One series evaluated ten patients with MMM and steroid-resistant anemia; eight showed evidence of immune defects [53]. Three responded to a six month course of cyclosporine (5 mg/kg PO twice per day).

Chemotherapy — Chemotherapy in AMM is used to diminish the degree of hepatosplenomegaly with attendant improvement in ascites, pain and/or cytopenias, to relieve constitutional symptoms such as fever and weight loss, or reduce symptomatic thrombocytosis. In the Mayo Clinic series of patients with AMM receiving chemotherapy, a significant reduction in splenic size with relief of pressure symptoms occurred in 70 percent of patients [54]. Responses lasted a median of 4.5 months and only 16 percent of patients obtained sustained symptomatic relief. Toxicity was common and often necessitated cessation of treatment.

Busulfan (Busulfex, Myleran; starting dose 2 to 4 mg/day PO) and other alkylating agents have been used in the past [54]. However, patients with AMM are unusually sensitive to these agents, which may result in prolonged and severe cytopenias even after therapy has been discontinued.

An Italian study employed the alkylating agent melphalan (Alkeran) in reduced doses (initial dose 2.5 mg PO three times/week increasing to a maximum dose of 2.5 mg/day) in 104 patients with AMM and splenomegaly, transfusion-dependent anemia, leucocytosis, and/or thrombocytosis [55]. All eligible patients were pretreated with danazol (200 mg/day) or prednisone (0.25 mg/kg per day) for at least two months. Responses were noted in 66 of 99 evaluable patients at a median time of 6.7 months. Splenic size, leukocytosis, and thrombocytosis were normalized in 23, 86, and 93 percent, respectively. Anemia improved in 12 of 20 severely anemic patients not requiring transfusion; of the 16 patients requiring transfusion, six became transfusion independent. Hematologic toxicity was common, but was reversible in all patients by discontinuing or reducing the treatment. However, the potential leukemogenicity of this agent has tempered our enthusiasm for its use in this disorder.

  Hydroxyurea — Because of its demonstrated utility in the other chronic myeloproliferative disorders (eg, chronic myelogenous leukemia, PV, and ET), hydroxyurea has become the agent of choice for such treatment [56]. Hydroxyurea may result in a reduction in spleen size and/or control of thrombocytosis and leukocytosis in some patients with AMM [56-59]. In one series of 59 patients with a chronic myeloproliferative disease (10 with AMM) and thrombocytosis, chronic therapy with hydroxyurea reduced the platelet count to less than 500,000/microL in six of the ten with AMM (and over 80 percent of those with PV or ET) [56]. In another report by the same authors, hydroxyurea also reduced the degree of bone marrow fibrosis in these patients [57].

Although the suggested starting dose for hydroxyurea in PV and ET is 15 mg/kg per day PO (average adult daily dose of 500 mg twice per day PO), the initial dose in advanced AMM should be much lower, given the presence of cytopenias in many patients. A starting oral dose of 500 to 1000 mg every other day, with dose modifications depending upon results of frequent blood counts and the patient's clinical status, would seem to be the most prudent approach under these circumstances.

Splenectomy — Splenomegaly, often massive and occupying a large portion of the abdomen, is present in virtually all patients with AMM. Surgical removal of such spleens is considered for patients who have symptomatic splenomegaly as manifested by mechanical discomfort, recurrent episodes of splenic infarction, transfusion-dependent anemia, refractory thrombocytopenia, hypercatabolic symptoms, or portal hypertension [50-52,60]. (See "Clinical manifestations and diagnosis of chronic idiopathic myelofibrosis (primary myelofibrosis, agnogenic myeloid metaplasia)", section on Splenomegaly).

Preoperative laboratory evidence of low-grade disseminated intravascular coagulation (positive soluble fibrin monomer and plasma D-dimer >500 microg/L) may increase the risk of perioperative bleeding, and it is recommended that the operation be postponed until these abnormalities are corrected. At experienced centers, the mortality rate from the procedure should be less than 10 percent, provided that the operation is conducted by an experienced senior surgeon, who pays meticulous attention to achieving hemostasis.

Given the complications that may be associated with removing a massively enlarged spleen (see below), an initial report described success with subtotal splenectomy in three patients [61]. More experience is required before this approach can be recommended.

Two large reported series analyzed the indications for and efficacy of splenectomy in AMM in 321 published cases and 314 patients at a single institution, respectively [50,51]. The following findings were noted: The main indications for the procedure were transfusion-dependent anemia (25 percent) and symptomatic splenomegaly (49 percent) [51] The frequency of improvement in one series was 97 percent for painful splenomegaly, 83 percent for portal hypertension, 70 percent for anemia, and 56 percent for thrombocytopenia [50]. However, long-term improvements were less common in the second series, with respective values of 49, 40, 50, and 30 percent, respectively [51]. The operative mortality after splenectomy was 7.5 to 9 percent, rising to 26 percent after three months; operative morbidity was 31 to 46 percent [50]. Patients at greatest risk for perioperative death were those with spleen weight exceeding 2000 g or a platelet count less than 70,000/microL [50]. In the second series, operative mortality was in the range of 5.5 to 7.6 percent, with a perioperative morbidity of 25 percent [51]. Postsplenectomy thrombocytosis of >450,000 and >1,000,000/microL occurred in 29 and 5 percent of patients, respectively, and was significantly associated with postoperative thrombosis and decreased survival [51]. There was no evidence for an effect of splenectomy on survival, the median values for which were 13 and 19 months after surgery in the two series [50,51]. The main causes of death not related to surgery were infection, cardiac or thrombotic events, bleeding, and leukemic transformation (which occurred in 11 to 16 percent of patients).

  Complications — Splenectomy is not a trivial procedure in AMM and is associated with both immediate postoperative and longer term complications. Postsurgical complications include intraabdominal bleeding, subphrenic abscess, sepsis, extreme thrombocytosis that may be associated with thrombosis (eg, stroke, pulmonary embolus, portal vein thrombosis), and accelerated hepatomegaly in 16 to 24 percent due to worsening hepatic myeloid metaplasia and/or loss of splenic sequestration of immature myeloid precursors [51,52]. The thrombocytosis and hepatomegaly may be transiently controlled with plateletpheresis, hydroxyurea (starting dose 500 mg PO three times per day) or cladribine (a two-hour intravenous administration of 0.14 mg/kg per day for five days) [62].

As noted above, splenectomy is often performed in an attempt to reduce extremely high transfusion requirements associated with splenic sequestration. The question has been raised concerning the potential effect of splenectomy in removing an important site of (extramedullary) erythropoiesis, thereby worsening transfusion requirements. The following information may be helpful in making this decision: Massively enlarged spleens, even with significant degrees of erythropoiesis, also exhibit significant degrees of splenic sequestration and ineffective erythropoiesis. The transfusion requirement for an adult with no effective erythropoiesis (as in severe aplastic anemia or pure red cell aplasia) is usually two to three units of packed red cells every two weeks. Thus, a chronic transfusion requirement in a nonbleeding patient with AMM greater than three units of red cells every two weeks indicates that the spleen is not an "effective" source of red cell production.

Another concern that has been raised is the potential facilitation of leukemic transformation following splenectomy [63]. The concern of these investigators was based on a retrospective series of splenectomized patients compared to a "control" group of patients who had not undergone splenectomy. In our experience, it is impossible to case-match patients who are in need of splenectomy. However, splenic histopathology may help in determining the course of the disease post-splenectomy [64]. Thus, the presence of microscopic splenic infarcts, a pattern of immature granulocyte predominance, or the detection of an abnormal splenic karyotype were found to be significantly associated with decreased survival post-splenectomy.

Radiation therapy

  Splenic irradiation — Splenic irradiation (SI) usually provides a transient (three to six months) benefit and can be appropriate for patients who are poor surgical candidates. In our series of 23 patients with MMM, eight of whom received multiple courses of SI, 94 percent of courses resulted in an objective reduction in splenic size and symptomatic relief; the median duration of response was six months [65]. The median dose of SI per course was 2.8 Gy, administered in a median of 7.5 fractions. Significant cytopenia occurred in 32 percent of courses, and life-threatening pancytopenia after a single course of SI occurred in six patients (26 percent), resulting in fatal sepsis or hemorrhage in three. Nine patients underwent subsequent splenectomy with a perioperative mortality of 11 percent; three of these required reexploration for postoperative bleeding.

Other centers have used total SI doses as low as 0.15 Gy and as high as 65 Gy, with generally excellent results, especially if SI is started in small fractions of 0.25 to 0.50 Gy given two to three times per week, with modifications as dictated by the clinical situation and frequently obtained blood counts. The range of regimens can be illustrated by the following two reports: In a retrospective study of 14 patients with MMM treated with SI for symptoms due to splenomegaly, doses ranged from 7 to 24 Gy [66]. Relief of symptoms was achieved in all patients, with 30 to 70 percent reduction in splenic size. Side effects were mild and did not require interruption of treatment, although four patients developed severe anemia. In another series, 17 patients with CML or MMM were treated with SI to a total dose of 0.15 to 16 Gy, with fractions given only two to three times per week [59]. Fourteen of 19 courses given for splenic pain produced a significant subjective relief, while 17 of 26 courses given for splenomegaly obtained at least 50 percent reduction in splenic size.

  Other sites of disease — Radiation therapy is extremely successful in the management of symptomatic foci of extramedullary hematopoiesis (EMH), including the spinal cord, peritoneal and pleural cavities, focal areas of bone pain due to granulocytic sarcoma (chloroma) or periostitis, and whole-lung treatment for pulmonary hypertension. (See "Clinical features, diagnosis, and prognosis of acute myeloid leukemia", section on Extramedullary presentations/Granulocytic sarcoma).

Specific treatment programs that have been successfully employed include [67-70]: In four patients with paraspinal or intraspinal EMH, the median dose of radiation was 1 Gy (range: 1 to 10 Gy), delivered in one to five fractions. Two patients with pulmonary and/or pleural EMH requiring treatment received 1.0 to 1.5 Gy delivered in one to 10 fractions. Administration of fractional doses of radiation (0.25 Gy/day, with rotation into all four abdominal quadrants, to a total dose of 5 to 10 Gy) is extremely effective in the treatment of ascites due to peritoneal implants of EMH [69]. Low-dose radiation to the liver (median dose per course: 1.50 Gy; range: 0.50 to 10 Gy) has been given for symptomatic hepatomegaly in 14 patients with advanced MMM [70]. Such treatment has been found to be myelosuppressive, providing only temporary (median three months), and mainly subjective, relief.

Anagrelide — Anagrelide is an oral agent that has a specific platelet-lowering effect in humans, and is capable of controlling thrombocytosis associated with PV and ET [71]. In one study, 17 patients with MMM (including patients with post-polycythemia and post-thrombocythemia myelofibrosis) were treated with anagrelide (starting and median maintenance doses of 1 mg/day PO) for a median period of 2 years [72]. Platelet counts were decreased in 13, unchanged in one, and increased in three patients. No clinical benefit was obtained in any patient in terms of anemia, transfusion requirement, or reduction in hepatosplenomegaly. Bone marrow megakaryocyte numbers were increased after six months of treatment; no patient had a significant change in either bone marrow fibrosis or osteosclerosis.

Interferon alfa — Interferon alfa (IFNa) induces both hematologic and cytogenetic remissions in chronic-phase CML, and has shown utility in ET and PV. IFNa has multiple effects on reducing proliferation of fibroblasts and bone marrow progenitor cells in vitro, suggesting that it might be useful in patients with MMM.

The efficacy of IFNa has been best evaluated in a series 54 patients with myeloproliferative disease (PV, ET, and MMM) who were treated with initial subcutaneous doses of IFNa of five million IU/day; the following benefits were noted [73]: Control of thrombocythemia was seen in 24 of 24 patients with this problem Control of hyperleukocytosis was seen in 14 of 14 patients At least a 10 percent reduction in splenic size was seen in 26 of 39 patients with splenomegaly.

The median follow-up was 7.3 years and at the time of publication, one-half were still participating at a median follow-up of 3.8 years. Treatment was generally well-tolerated; toxicity caused treatment withdrawal in 7 patients (13 percent). Thirty-nine of the 54 patients maintained the response for a median of 39 weeks after withdrawal of IFNa; repeat courses of therapy in previously responding patients produced similar results.

However, our own, as well as the ECOG, experience with IFNa has not been favorable [74]. In a phase II trial involving 11 patients with previously untreated "hyperproliferative" MMM, weekly doses of up to 15 million units of IFNa were given for as long as one year [75]. Unacceptable drug toxicity was present in seven of the 11 patients, necessitating discontinuation. No beneficial changes in splenic size, reticulin fibrosis, osteosclerosis, or microvessel density were seen in any patient.

Thalidomide — Thalidomide has been evaluated mainly in previously-treated patients with MMM. Beneficial responses, which have been seen in 20 to 40 percent of patients, have included disappearance of constitutional symptoms, reduction in splenic size, improvements in hemoglobin concentration, white blood cell, and platelet counts, as well as transfusion independence [76-80].

In these studies, few patients were able to tolerate doses higher than 400 mg/day because of side effects, which included drowsiness, constipation, fatigue, paresthesias, and neutropenia [81]. In one study of 63 patients, the median maximally-tolerated dose was 100 mg/day and the cumulative drop-out rate due to adverse drug side effects was 51 percent after six months of therapy [79].

In our study, the median duration of treatment was only 16 weeks; preliminary results suggested that low starting doses (ie, 50 mg/day) with more gradual dose escalation may be as effective and more tolerable [78,82]. Adverse hematologic effects were seen in four of 15 patients (eg, extreme thrombocytosis and leukocytosis and pericardial extramedullary hematopoiesis with clinical tamponade).

  Thalidomide and prednisone — In order to improve tolerability of thalidomide, we prospectively treated symptomatic MMM patients with low-dose thalidomide (50 mg/day PO) plus a three-month oral prednisone taper (starting prednisone dose 0.5 mg/kg PO per day) [82,83]. Twenty of the 21 patients were able to complete the 3-month course, with an objective clinical response in 13. Responses included improvement of anemia in 13 patients, red cell transfusion independence in four, platelet count increases in six, and >50 percent decreases in splenic size in four. This dose of thalidomide was better tolerated than those used in previous studies (see "Thalidomide" above).

Leukocytosis and thrombocytosis were observed in 38 and 19 percent of patients, respectively; one episode of deep vein thrombosis was noted, a complication which has been seen with this agent when used in other malignant disorders. Adverse events associated with prednisone were mild and transient. (See "Thrombotic complications following treatment with thalidomide and its analogues").

At a median follow-up of 25 months for the 36 patients treated with thalidomide in our two series, the overall long-term response rate was 28 percent (10 patients) [78,82,83]. Durable treatment responses were seen for anemia and thrombocytopenia, but not for splenomegaly.

  Lenalidomide — A number of cooperative phase II studies are evaluating the utility of the thalidomide analog lenalidomide with or without prednisone in patients with AMM. In one report, overall response rates were 22, 33, and 50 percent for anemia, splenomegaly, and thrombocytopenia, respectively [84]. Grades 3 or 4 adverse events included neutropenia and thrombocytopenia in 31 and 19 percent of patients, respectively.

Etanercept — Use of etanercept (Enbrel, 25 mg SQ twice weekly), a soluble tumor necrosis factor receptor, resulted in improvement of constitutional symptoms (eg, weight loss, night sweats, fatigue, fever) in 12 of 20 evaluable patients with MMM [85]. Objective responses (eg, anemia, thrombocytopenia, spleen size) were noted in four patients. Reversible pancytopenia was noted in one patient, necessitating cessation of this agent. Toxicity was otherwise mild and well tolerated.

In an small phase II trial, the combination of thalidomide, prednisone, and eternacept alleviated symptoms and was well tolerated, but did not appear to be superior to thalidomide/prednisone in terms of therapeutic value for anemia, thrombocytopenia, or splenomegaly [86].

Imatinib mesylate — We have conducted a phase II trial with the tyrosine kinase inhibitor imatinib mesylate in 23 patients with MMM [87]. Toxicity was appreciable, with no discernible clinical benefit. In a second study, 13 of 18 patients treated with this agent showed only minor degrees of clinical or hematologic improvement [88].

RECOMMENDATIONS

Prognosis and treatment overview — The prognosis and need for treatment of agnogenic myeloid metaplasia (AMM) or the fibrotic states following polycythemia vera or essential thrombocythemia (postpolycythemic myeloid metaplasia and postthrombocythemic myeloid metaplasia, respectively) are strongly dependent upon the presence or absence of signs and symptoms of the disease (eg, anemia, painfully enlarged spleen, systemic symptoms, refractory cytopenias, thrombocytosis, portal hypertension, cord compression secondary to foci of extramedullary hematopoiesis). (See "Clinical manifestations and diagnosis of chronic idiopathic myelofibrosis (primary myelofibrosis, agnogenic myeloid metaplasia)"). In the absence of adverse prognostic factors such as anemia and constitutional symptoms, there may be very little need for treatment, with an expected survival in excess of 10 years (see "Prognosis" above). Survival and quality of life can be severely limited in this disorder, with an expected survival as short as one year for the most symptomatic patients (see "Prognosis" above). While allogeneic hematopoietic cell transplantation (HCT) is the only curative treatment for this disorder, it is associated with a high treatment-related mortality, and may not be available for older patients with severe comorbidities and those without a suitable donor. Other than HCT, all of the other available treatment modalities are palliative, with variable degrees of efficacy and treatment-related complications. There are few randomized trials comparing these modalities, preventing our ability to give strong recommendations for selecting one treatment over another.

Hematopoietic cell transplantation For younger patients with two or more adverse features (ie, hemoglobin <10 g/dL, constitutional symptoms, isolated cytogenetic abnormality, or blasts >1 percent) we suggest that the patient be considered for hematopoietic cell transplantation [HCT] shortly after diagnosis (Grade 2B). (See "Allogeneic HCT" above). For low-risk patients, who might live 10 to 15 years with supportive treatment alone, but might have a transplant-related mortality of at least 8 percent, the answer is not yet clear. Until further information is available, we suggest HCT for the latter group when an adverse risk factor appears (Grade 2C).

Supportive modalities

In the absence of allogeneic HCT as a therapeutic option, we suggest the use of one or more of the following palliative modalities for the symptomatic patient (Grade 2C). For each of the agents listed below, the symptom(s) for which is agent is most effective is noted in parentheses. Androgens, danazol (anemia) Blood transfusions with or without erythropoietin (anemia) Hydroxyurea (splenomegaly, thrombocytosis, leukocytosis) Alkylating agents (splenomegaly, thrombocytosis, leukocytosis) Thalidomide and prednisone (systemic symptoms, anemia, splenomegaly, refractory cytopenias) Etanercept (systemic symptoms) Splenic irradiation (painful splenomegaly) Radiation therapy (symptomatic areas of extramedullary hematopoiesis, cord compression)

  Splenectomy — Splenectomy can be considered in selected patients with a painfully enlarged spleen, anemia and other refractory cytopenias, and/or severe degrees of portal hypertension. This procedure carries a very high operative mortality and morbidity, and should not be undertaken lightly

Notes

REFERENCES 1. Guardiola, P, Esperou, H, Cazals-Hatem, D, et al. Allogeneic bone marrow transplantation for agnogenic myeloid metaplasia. French Society of Bone Marrow Transplantation. Br J Haematol 1997; 98:1004. 2. Deeg, HJ, Gooley, TA, Flowers, ME, et al. Allogeneic hematopoietic stem cell transplantation for myelofibrosis. Blood 2003; 102:3912. 3. Tefferi, A, Silverstein, MN, Noel, P. Agnogenic myeloid metaplasia. Semin Oncol 1995; 22:327. 4. Hennessy, BT, Thomas, DA, Giles, FJ, et al. New approaches in the treatment of myelofibrosis. Cancer 2005; 103:32. 5. Tefferi, A. Myelofibrosis with myeloid metaplasia. N Engl J Med 2000; 342:1255. 6. Mesa, RA, Silverstein, MN, Jacobsen, SJ, et al. Population-based incidence and survival figures in essential thrombocytemia and agnogenic myeloid metaplasia: an Olmsted County Study, 1976-1995. Am J Hematol 1999; 61:10. 7. Dupriez, B, Morel, P, Demory, JL, et al. Prognostic factors in agnogenic myeloid metaplasia: A report on 195 cases with a new scoring system. Blood 1996; 88:1013. 8. Dingli, D, Schwager, SM, Mesa, RA, et al. Prognosis in transplant-eligible patients with agnogenic myeloid metaplasia: a simple CBC-based scoring system. Cancer 2006; 106:623. 9. Visani, G, Finelli, C, Castelli, U, et al. Myelofibrosis with myeloid metaplasia: Clinical and hematological parameters predicting survival in a series of 133 patients. Br J Haematol 1990; 75:4. 10. Thiele, J, Kvasnicka, HM, Werden, C, et al. Idiopathic primary osteo-myelofibrosis: a clinico-pathological study on 208 patients with special emphasis on evolution of disease features, differentiation from essential thrombocythemia and variables of prognostic impact. Leuk Lymphoma 1996; 22:303. 11. Kvasnicka, HM, Thiele, J, Werden, C, et al. Prognostic factors in idiopathic (primary) osteomyelofibrosis. Cancer 1997; 80:708. 12. Barosi, G, Berzuini, C, Liberato, LN, et al. A prognostic classification of myelofibrosis with myeloid metaplasia. Br J Haematol 1988; 70:397. 13. Varki, A, Lottenberg, R, Griffith, R, Reinhard, E. The syndrome of idiopathic myelofibrosis. A clinicopathologic review with emphasis on the prognostic variables predicting survival. Medicine (Baltimore) 1983; 62:353. 14. Demory, JL, Dupriez, B, Fenaux, P, et al. Cytogenetic studies and their prognostic significance in agnogenic myeloid metaplasia: A report on 47 cases. Blood 1988; 72:855. 15. Reilly, JT, Snowden, JA, Spearing, RL, et al. Cytogenetic abnormalities and their prognostic significance in idiopathic myelofibrosis: A study of 106 cases. Br J Haematol 1997; 98:96. 16. Dingli, D, Schwager, SM, Mesa, RA, et al. Presence of unfavorable cytogenetic abnormalities is the strongest predictor of poor survival in secondary myelofibrosis. Cancer 2006; 106:1985. 17. Wang, JC, Wang, A. Plasma soluble interleukin-2 receptor in patients with primary myelofibrosis. Br J Haematol 1994; 86:380. 18. Chelloul, N, Briere, J, Laval-Jeantet, M, et al. Prognosis of myeloid metaplasia with myelofibrosis. Biomedicine 1976; 24:272. 19. Barosi, G, Bordessoule, D, Briere, J, et al. Response criteria for myelofibrosis with myeloid metaplasia: results of an initiative of the European Myelofibrosis Network (EUMNET). Blood 2005; 106:2849. 20. Tefferi, A, Barosi, G, Mesa, RA, et al. International Working Group (IWG) consensus criteria for treatment response in myelofibrosis with myeloid metaplasia, for the IWG for Myelofibrosis Research and Treatment (IWG-MRT). Blood 2006; 108:1497. 21. Cervantes, F, Barosi, G, Demory, JL, et al. Myelofibrosis with myeloid metaplasia in young individuals: Disease characteristics, prognostic factors and identification of risk groups. Br J Haematol 1998; 102:684. 22. Cervantes, F. Modern management of myelofibrosis. Br J Haematol 2005; 128:583. 23. Arana-Yi, C, Quintas-Cardama, A, Giles, F, et al. Advances in the therapy of chronic idiopathic myelofibrosis. Oncologist 2006; 11:929. 24. Merup, M, Lazarevic, V, Nahi, H, et al. Different outcome of allogeneic transplantation in myelofibrosis using conventional or reduced-intensity conditioning regimens. Br J Haematol 2006; 135:367. 25. Przepiorka, D, Giralt, S, Khouri, I, et al. Allogeneic marrow transplantation for myeloproliferative disorders other than chronic myelogenous leukemia: review of forty cases. Am J Hematol 1998; 57:24. 26. Singhal, S, Powles, R, Treleaven, J, et al. Allogeneic bone marrow transplantation for primary myelofibrosis. Bone Marrow Transplant 1995; 16:743. 27. Creemers, GJ, Lowenberg, B, Hagenbeek, A. Allogeneic bone marrow transplantation for primary myelofibrosis. Br J Haematol 1992; 82:772. 28. Benesch, M, Deeg, HJ. Hematopoietic cell transplantation for adult patients with myelodysplastic syndromes and myeloproliferative disorders. Mayo Clin Proc 2003; 78:981. 29. Guardiola, P, Anderson, JE, Bandini, G, et al. Allogeneic stem cell transplantation for agnogenic myeloid metaplasia: A European Group for Blood and Marrow Transplantation, Societe Francaise de Greffe de Moelle, Gruppo Italiano per il Trapianto del Midollo Osseo, and Fred Hutchinson Cancer Research Center Collaborative Study. Blood 1999; 93:2831. 30.  Guardiola, P, Anderson, JE, Gluckman, E. Myelofibrosis with myeloid metaplasia. N Engl J Med 2000; 343:659;. 31.  Maziarz, RT, Mesa, RA, Tefferi, A. Allogeneic stem cell transplantation for chronic myeloproliferative disorders and myelodysplastic syndromes: the question is "when". Mayo Clin Proc 2003; 78:941. 32.  Deeg, HJ, Appelbaum, FR. Stem-cell transplantation for myelofibrosis [letter]. N Engl J Med 2001; 344:775. 33. Soll, E, Massumoto, C, Clift, RA, et al. Relevance of marrow fibrosis in bone marrow transplantation: A retrospective analysis of engraftment. Blood 1995; 86:4667. 34.  Li, Z, Gooley, T, Appelbaum, FR, Deeg, HJ. Splenectomy and hemopoietic stem cell transplantation for myelofibrosis [correspondence]. Blood 2001; 97:2180. 35. Devine, SM, Hoffman, R, Verma, A, et al. Allogeneic blood cell transplantation following reduced-intensity conditioning is effective therapy for older patients with myelofibrosis with myeloid metaplasia. Blood 2002; 99:2255. 36. Hessling, J, Kroger, N, Werner, M, et al. Dose-reduced conditioning regimen followed by allogeneic stem cell transplantation in patients with myelofibrosis with myeloid metaplasia. Br J Haematol 2002; 119:769. 37. Kroger, N, Zabelina, T, Schieder, H, et al. Pilot study of reduced-intensity conditioning followed by allogeneic stem cell transplantation from related and unrelated donors in patients with myelofibrosis. Br J Haematol 2005; 128:690. 38. Rondelli, D, Barosi, G, Bacigalupo, A, et al. Allogeneic hematopoietic stem-cell transplantation with reduced-intensity conditioning in intermediate- or high-risk patients with myelofibrosis with myeloid metaplasia. Blood 2005; 105:4115. 39. Anderson, JE, Tefferi, A, Craig, F, et al. Myeloablation and autologous peripheral blood stem cell rescue results in hematologic and clinical responses in patients with myeloid metaplasia with myelofibrosis. Blood 2001; 98:586. 40. Brubaker, LH, Briere, J, Laszlo, J, et al. Treatment of anemia in myeloproliferative disorders: A randomized study of fluoxymesterone v transfusions only. Arch Intern Med 1982; 142:1533. 41. Besa, EC, Nowell, PC, Geller, NL, et al. Analysis of the androgen response of 23 patients with agnogenic myeloid metaplasia: The value of chromosomal studies in predicting response and survival. Cancer 1982; 49:308. 42.  Tefferi, A, Silverstein, MN. Recombinant human erythropoietin therapy in patients with myelofibrosis with myeloid metaplasia. Br J Haematol 1994; 86:893. 43. Rodriguez, JN, Martino, ML, Dieguez, JC, Prados, D. rHuEpo for the treatment of anemia in myelofibrosis with myeloid metaplasia. Experience in 6 patients and meta-analytical approach. Haematologica 1998; 83:616. 44. Mohr, B, Herrmann, R, Huhn, D. Recombinant human erythropoietin in patients with myelodysplastic syndrome and myelofibrosis. Acta Haematol 1993; 90:65. 45. Aloe Spiriti, M, Latagliata, R, Avvisati, G, et al. Erythropoietin treatment of idiopathic myelofibrosis. Haematologica 1993; 78:371. 46. Cervantes, F, Alvarez-Larran, A, Hernandez-Boluda, JC, et al. Erythropoietin treatment of the anaemia of myelofibrosis with myeloid metaplasia: results in 20 patients and review of the literature. Br J Haematol 2004; 127:399. 47.  Cervantes, F, Alvarez-Larran, A, Hernandez-Boluda, JC, et al. Darbepoietin-alpha for the anaemia of myelofibrosis with myeloid metaplasia. Br J Haematol 2006; 134:184. 48. Levy, V, Bourgarit, A, Delmer, A, et al. Treatment of agnogenic myeloid metaplasia with danazol: A report of four cases. Am J Hematol 1996; 53:239. 49. Cervantes, F, Alvarez-Larran, A, Domingo, A, et al. Efficacy and tolerability of danazol as a treatment for the anaemia of myelofibrosis with myeloid metaplasia: long-term results in 30 patients. Br J Haematol 2005; 129:771. 50. Benbassat, J, Penchas, S, Ligumski, M. Splenectomy in patients with agnogenic myeloid metaplasia: An analysis of 321 published cases. Br J Haematol 1979; 42:207. 51. Mesa, RA, Nagorney, DS, Schwager, S, et al. Palliative goals, patient selection, and perioperative platelet management: outcomes and lessons from 3 decades of splenectomy for myelofibrosis with myeloid metaplasia at the Mayo Clinic. Cancer 2006; 107:361. 52. Barosi, G, Ambrosetti, A, Buratti, A, et al. Splenectomy for patients with myelofibrosis with myeloid metaplasia: Pretreatment variables and outcome prediction. Leukemia 1993; 7:200. 53. Centenara, E, Guarnone, R, Ippoliti, G, Barosi, G. Cyclosporin-A in severe refractory anemia of myelofibrosis with myeloid metaplasia. A preliminary report. Haematologica 1998; 83:622. 54.  Silverstein, MN. Agnogenic Myeloid Metaplasia. Publishing Sciences Group, Acton, MA, 1975. 55. Petti, MC, Latagliata, R, Spadea, T, et al. Melphalan treatment in patients with myelofibrosis with myeloid metaplasia. Br J Haematol 2002; 116:576. 56. Lofvenberg, E, Wahlin, A. Management of polycythaemia vera, essential thrombocythaemia and myelofibrosis with hydroxyurea. Eur J Haematol 1988; 41:375. 57. Lofvenberg, E, Wahlin, A, Roos, G, et al. Reversal of myelofibrosis by hydroxyurea. Eur J Haematol 1990; 44:33. 58. Manoharan, A. Management of myelofibrosis with intermittent hydroxyurea. Br J Haematol 1991; 77:252. 59. Wagner, H Jr, McKeough, PG, Desforges, J, et al. Splenic irradiation in the treatment of patients with chronic myelogenous leukemia or myelofibrosis with myeloid metaplasia. Results of daily and intermittent fractionation with and without concomitant hydroxyurea. Cancer 1986; 58:1204. 60. Lafaye, F, Rain, JD, Clot, P, et al. Risks and benefits of splenectomy in myelofibrosis: An analysis of 39 cases. Nouv Rev Fr Hematol 1994; 36:359. 61. Petroianu, A. Subtotal splenectomy for treatment of patients with myelofibrosis and myeloid metaplasia. Int Surg 1996; 81:177. 62. Tefferi, A, Silverstein, MN, Li, CY. 2-Chlorodeoxyadenosine treatment after splenectomy in patients who have myelofibrosis with myeloid metaplasia. Br J Haematol 1997; 99:352. 63. Barosi, G, Ambrosetti, A, Centra, A, et al. Splenectomy and risk of blast transformation in myelofibrosis with myeloid metaplasia. Italian Cooperative Study Group on Myeloid with Myeloid Metaplasia. Blood 1998; 91:3630. 64. Mesa, RA, Li, CY, Schroeder, G, Tefferi, A. Clinical correlates of splenic histopathology and splenic karyotype in myelofibrosis with myeloid metaplasia. Blood 2001; 97:3665. 65. Elliott, MA, Chen, MG, Silverstein, MN, Tefferi, A. Splenic irradiation for symptomatic splenomegaly associated with myelofibrosis with myeloid metaplasia. Br J Haematol 1998; 103:505. 66. Sciascia, R, Mazza, P, Tondi, L. [Splenic irradiation in the treatment of idiopathic myelofibrosis. Retrospective study of 14 cases]. Radiol Med (Torino) 1988; 76:87. 67. Steensma, DP, Hook, CC, Stafford, SL, Tefferi, A. Low-dose, single-fraction, whole-lung radiotherapy for pulmonary hypertension associated with myelofibrosis with myeloid metaplasia. Br J Haematol 2002; 118:813. 68. Koch, CA, Li, CY, Mesa, R, Tefferi, A. Nonhepatosplenic extramedullary hematopoiesis: Associated diseases, pathology, clinical course, and treatment. Mayo Clin Proc 2003; 78:1223. 69.  Hoffman, R. Agnogenic myeloid metaplasia. In: Hematology. Basic Principles and Practice, 3d ed, Hoffman, R, Benz EJ Jr, Shattil, SJ, et al (Eds), Churchill Livingstone, New York, 2000, chap. 63. 70. Tefferi, A, Jimenez, T, Gray, LA, et al. Radiation therapy for symptomatic hepatomegaly in myelofibrosis with myeloid metaplasia. Eur J Haematol 2001; 66:37. 71. Tefferi, A, Silverstein, MN, Petitt, RM, et al. Anagrelide as a new platelet-lowering agent in essential thrombocythemia: mechanism of actin, efficacy, toxicity, current indications. Semin Thromb Hemost 1997; 23:379. 72. Yoon, S-Y, Li, C-Y, Mesa, RA, Tefferi, A. Bone marrow effects of anagrelide therapy in patients with myelofibrosis with myeloid metaplasia. Br J Haematol 1999; 106:682. 73.  Gilbert, HS. Long-term treatment of myeloproliferative disease with interferon-alpha-2b: Feasibility and efficacy. Cancer 1998; 83:1205. 74. Radin, AI, Kim, HT, Grant, BW, et al. Phase II study of alpha2 interferon in the treatment of the chronic myeloproliferative disorders (E5487): a trial of the Eastern Cooperative Oncology Group. Cancer 2003; 98:100. 75.  Tefferi, A, Elliot, MA, Yoon, SY, et al. Clinical and bone marrow effects of interferon alfa therapy in myelofibrosis with myeloid metaplasia. Blood 2001; 97:1896. 76. Giovanni, B, Michelle, E, Letizia, C, et al. Thalidomide in myelofibrosis with myeloid metaplasia: a pooled-analysis of individual patient data from five studies. Leuk Lymphoma 2002; 43:2301. 77. Canepa, L, Ballerini, F, Varaldo, R, Quintino, S. Thalidomide in agnogenic and secondary myelofibrosis. Br J Haematol 2001; 115:313. 78. Elliott, MA, Mesa, RA, Li, CY, et al. Thalidomide treatment in myelofibrosis with myeloid metaplasia. Br J Haematol 2002; 117:288. 79. Marchetti, M, Barosi, G, Balestri, F, et al. Low-dose thalidomide ameliorates cytopenias and splenomegaly in myelofibrosis with myeloid metaplasia: a phase II trial. J Clin Oncol 2004; 22:424. 80. Thomas, DA, Giles, FJ, Albitar, M, et al. Thalidomide therapy for myelofibrosis with myeloid metaplasia. Cancer 2006; 106:1974. 81. Merup, M, Kutti, J, Birgergard, G, et al. Negligible clinical effects of thalidomide in patients with myelofibrosis with myeloid metaplasia. Med Oncol 2002; 19:79. 82. Mesa, RA, Elliott, MA, Schroeder, G, Tefferi, A. Durable responses to thalidomide-based drug therapy for myelofibrosis with myeloid metaplasia. Mayo Clin Proc 2004; 79:883. 83. Mesa, RA, Steensma, DP, Pardanani, A, et al. A phase 2 trial of combination low-dose thalidomide and prednisone for the treatment of myelofibrosis with myeloid metaplasia. Blood 2003; 101:2534. 84. Tefferi, A, Cortes, J, Verstovsek, S, et al. Lenalidomide therapy in myelofibrosis with myeloid metaplasia. Blood 2006; 108:1158. 85. Steensma, DP, Mesa, RA, Li, CY, et al. Etanercept, a soluble tumor necrosis factor receptor, palliates constitutional symptoms in patients with myelofibrosis with myeloid metaplasia: results of a pilot study. Blood 2002; 99:2252. 86.  Mesa, RA, Steensma, DP, Li, CY, et al. Phase II study of the combination of low-dose thalidomide, prednisone, and etanercept (PET regimen) in the treatment of anemia, splenomegaly, and constitutional symptoms associated with myelofibrosis with myeloid metaplasia (MMM) (abstract). Blood 2005; 106:724a. 87. Tefferi, A, Mesa, RA, Gray, LA, et al. Phase 2 trial of imatinib mesylate in myelofibrosis with myeloid metaplasia. Blood 2002; 99:3854. 88. Cortes, J, Giles, F, O'Brien, S, Thomas, D. Results of imatinib mesylate therapy in patients with refractory or recurrent acute myeloid leukemia, high-risk myelodysplastic syndrome, and myeloproliferative disorders. Cancer 2003; 97:2760.