Wikipedysta:DrPZ/brudnopis/007

http://www.mayoclinicproceedings.com/content/80/1/75.long

Mayo Clin Proc. January 2005;80(1):75-83

Blood Eosinophilia: A New Paradigm in Disease Classification, Diagnosis, and Treatment

AYALEW TEFFERI, MD

Acquired blood eosinophilia is considered either a primary or a secondary phenomenon. Causes of secondary (ie, reactive) eosinophilia include tissue-invasive parasitosis, allergic or inflammatory conditions, and malignancies in which eosinophils are not considered part of the neoplastic process. Primary eosinophilia is classified operationally into 2 categories: clonal and idiopathic. Clonal eosinophilia stipulates the presence of either cytogenetic evidence or bone marrow histological evidence of an otherwise classified hematologic malignancy such as acute leukemia or a chronic myeloid disorder. Idiopathic eosinophilia is a diagnosis of exclusion (ie, not secondary or clonal). Hypereosinophilic syndrome is a subcategory of idiopathic eosinophilia; diagnosis requires documentation of both sustained eosinophilia (absolute eosinophil count ≥1500 cells/μL for at least 6 months) and target organ damage (eg, involvement of the heart, lung, skin, or nerve tissue). Genetic mutations involving the platelet-derived growth factor receptor genes (PDGFR-α and PDGFR-β) have been pathogenetically linked to clonal eosinophilia, and their presence predicts treatment response to imatinib. Accordingly, cytogenetic and/or molecular investigations for the presence of an imatinibsensitive molecular target should accompany current evaluation for primary eosinophilia. In the absence of such a drug target, specific treatment is dictated by the underlying hematologic malignancy in cases of clonal eosinophilia; however, the initial treatment of choice for symptomatic patients with hypereosinophilic syndrome is prednisone and/or interferon alfa.

Eosinophils are derived from hematopoietic stem cells that are committed initially to the myeloid and subsequently to the basophil-eosinophil granulocyte lineage. As depicted in Figure 1, the mature eosinophil, when examined by standard blood-staining techniques, displays a bilobed nucleus and an abundant cytoplasm filled with reddish-orange granules. The material in these granules includes cationic proteins (major basic protein, eosinophilic cationic protein, eosinophil-derived neurotoxin, eosinophil peroxidase), cytokines (interleukins [ILs], tumor necrosis factor), and lipid mediators (leukotriene C4). Interleukin 5 is considered the major eosinophil growth and survival factor, whereas chemokines (eotaxin, platelet-activating factor, RANTES [regulated on activation, normal T expressed and secreted]) and endothelial adhesion molecules (integrins, vascular cell adhesion molecules) contribute to eosinophil trafficking. Major basic protein, eosinophil cationic protein, and eosinophil-derived neurotoxin are the primary mediators of eosinophil-associated toxicity to microbes (parasites, protozoa, bacteria, viruses) and human tissue (myocarditis, pneumonitis, dermatitis, neuropathy, vasculitis). The lung and gastrointestinal systems constitute the main residence for eosinophils. Blood eosinophilia (absolute eosinophil count [AEC] ≥600 cells/μL) is the usual initial clue for the presence of an eosinophilic disorder. The degree of blood eosinophilia, in the absence of active treatment, can be categorized into mild (AEC 600-1500 cells/μL), moderate (AEC 1500-5000 cells/μL), or severe (AEC >5000 cells/μL). Target organ damage is unusual with mild eosinophilia, but its occurrence in association with moderate to severe eosinophilia does not appear to depend on the specific cause of eosinophilia (ie, in addition to the wellestablished association of target organ damage with primary eosinophilia, it also has been reported to occur in both familial and secondary eosinophilia).

AEC = absolute eosinophil count CEL = chronic eosinophilic leukemia CML = chronic myeloid leukemia CMML = chronic myelomonocytic leukemia FGFR1 = fibroblast growth factor receptor 1 FISH = fluorescence in situ hybridization HES = hypereosinophilic syndrome IL = interleukin MPD = myeloproliferative disorder PDGFR = platelet-derived growth factor receptor SM = systemic mastocytosis SM-eos = SM associated with prominent blood eosinophilia

GENERAL CLASSIFICATION OF EOSINOPHILIC DISORDERS

Blood eosinophilia can be classified as either familial or acquired (Table 1). Acquired eosinophilia is classified further into a primary or a secondary process, depending on whether eosinophils are considered integral to the underlying disease. Infectious causes and noninfectious causes of secondary eosinophilia (Table 1) are prevalent in underdeveloped and developed countries, respectively. Primary eosinophilia occurs primarily in males and is considered clonal in the presence of either cytogenetic evidence or bone marrow histological evidence of an otherwise defined neoplastic hematologic disorder. Otherwise, a working diagnosis of idiopathic eosinophilia is made, and in the presence of both sustained eosinophilia (AEC ≥1500 cells/μL for at least 6 months) and target organ damage (eg, skin, heart, lung, nerve tissue), the process is subclassified as hypereosinophilic syndrome (HES).

PATHOGENESIS AND CLINICAL MANIFESTATIONS FAMILIAL EOSINOPHILIA

Familial eosinophilia is rare, and its genetic basis, at least in some cases, may be similar to that of clonal eosinophilia. In such cases, the chromosomal region 5q31-33 has been identified as the “hot zone,” although it may not involve mutations of some of the resident genes responsible for the synthesis of known eosinophil growth factors including IL-3, IL-5, and granulocyte-macrophage colonystimulating factor. Interestingly, one recently reported familial case involved a translocation-associated mutation of the platelet-derived growth factor receptor gene PDGFRB, which is known to reside in the 5q31-33 chromosomal region and to be pathogenetically linked to some cases of clonal eosinophilia.21,23 In general, familial eosinophilia is an extremely rare autosomal dominant disorder that is characterized by a stable eosinophil count and, compared with HES, a lesser degree of eosinophil activation as well as a more benign clinical course.

ACQUIRED EOSINOPHILIA

Secondary Eosinophilia. The most common cause of secondary eosinophilia is tissue-invasive parasitosis including schistosomiasis, visceral toxocariasis, strongyloidiasis, filariasis, ancylostomiasis, fascioliasis, trichinellosis, and paragonimiasis (Table 1). In general, parasites that are isolated in either the intestinal lumen (Cestoda, Ascaris) or an intact cyst (Echinococcus granulosus) do not cause blood eosinophilia unless they are introduced systemically through tissue invasion or cyst disruption. In addition to helminths, some (Toxoplasma gondii, Dientamoeba fragilis, Isospora belli) but not other (Giardia lamblia, Entamoeba histolytica) protozoan infections may induce blood eosinophilia. Rare reports of eosinophiliaassociated bacterial (borrelia) or viral (human immunodeficiency virus) infections have been published. Noninfectious causes of secondary eosinophilia include drugs (sulfa derivatives, gold compounds, carbamazepine, myeloid growth factors, purine nucleoside analogues), toxins (associated with eosinophilia-myalgia syndrome, toxic oil syndrome, toxic shock syndrome), allergic disorders (asthma, hay fever, atopic dermatitis, allergic bronchopulmonary aspergillosis), idiopathic/autoimmune inflammatory conditions (granulomatous and/or systemic vasculitis, Churg-Strauss syndrome, Wegener granulomatosis, Kimura disease, angiolymphoid hyperplasia with eosinophilia, bullous pemphigoid and other cutaneous disorders, eosinophilic fasciitis, systemic sclerosis or scleroderma, polyarteritis, sarcoidosis, inflammatory bowel disease), malignancies in which eosinophils are not considered part of the neoplastic clone (Hodgkin and non-Hodgkin lymphoma, metastatic cancer), and endocrinopathies including Addison disease (Table 1). Secondary eosinophilia in inflammatory and malignant conditions is mediated by tissue-derived or tumor-derived eosinophilogenic cytokines.

Primary Eosinophilia. Clonal Eosinophilia. A diagnosis of clonal eosinophilia requires the presence of either cytogenetic evidence or bone marrow morphologic evidence for either acute leukemia or a chronic myeloid disorder (Table 1). Hematologic disorders that can be accompanied by clonal eosinophilia include acute myeloid leukemia, acute lymphocytic leukemia, chronic myeloid leukemia (CML), myelodysplastic syndrome, and both classic and atypical cases of myeloproliferative disorders (MPDs). The atypical MPD category includes chronic eosinophilic leukemia (CEL), systemic mastocytosis (SM), and chronic myelomonocytic leukemia (CMML). Recent developments in the molecular characterization of pathogenesis, in a subset of patients with clonal eosinophilia, have identified activating mutations of 3 receptor tyrosine kinase genes: PDGFRA, PDGFRB, and fibroblast growth factor receptor 1 (FGFR1). Within the context of hematologic malignancies, PDGFRA mutations have been connected so far with a subset of patients with SM associated with prominent blood eosinophilia (SM-eos). In general, SM-eos makes up approximately 20% of all cases of SM in adults. Of these cases of SM-eos, approximately half may be associated with a specific PDGFRA mutation caused by an 800-kilobase interstitial deletion involving chromosome 4q12, resulting in juxtaposition of PDGFRA and FIP1L1 genes (FIP1L1-PDGFRA). The fusion gene generates a constitutively activated PDGFRA tyrosine kinase that transforms hematopoietic cells. Other cases of SM-eos are associated with a C-kit mutation. Systemic mastocytosis not associated with eosinophilia does not display the FIP1L1-PDGFRA mutation. Although rare, PDGFRA also may be activated through chromosomal translocations as in t(4;22)(q12;q11). Patients with FIP1L1-PDGFRA+ SM-eos display a “myeloproliferative” clinical phenotype with organomegaly, cytopenia, increased serum B12 levels, and bone marrow hypercellularity with myelofibrosis. As a result, investigators use the terms myeloproliferative-variant HES78 and FIP1L1-PDGFRA+ CEL79 to describe the same disease. However, most, if not all, such patients display bone marrow mast cell infiltration with morphologically and immunophenotypically abnormal mast cells. Regardless, the term FIP1L1-PDGFRA+ eosinophilic disorder could be used to avoid confusion. Patients with FIP1L1-PDGFRA+ eosinophilic disorder also are characterized by a high risk of eosinophilic heart disease and inferior survival compared with those with FIP1L1-PDGFRA–clonal eosinophilia. The PDGFRB gene is located on chromosome 5q33, and its activation through translocation to various partner chromosomes, including t(5;12)(q33;p13), t(5;10)(q33;q21), t(5;7)(q33;q11.2), t(5;14)(q33;q13), t(5;17)(q33;p11), and t(1;5)(q23;q33), produces a distinct MPD that usually is associated with prominent eosinophilia and sometimes with monocytosis. Pathologic diagnosis in previously described cases of PDGFRB+ clonal eosinophilia have included CEL, atypical CML or MPD, and CMML. The gene for FGFR1 is located on chromosome 8p11, and its activation through translocation to different chromosome partners, including t(8;13)(p11;q12), t(8;9)(p11;q33), t(6;8)(q27;p11), and t(8;22)(p11;q22), produces a myeloproliferative syndrome with eosinophilia, lymphadenopathy, and a high incidence of T-cell non-Hodgkin lymphoma with progression to acute myeloid leukemia. Other karyotypic variants involving FGFR1, including t(8;17)(p11;q25), t(8;11)(p11;p15), t(8;12)(p11;q15), and ins(12;8)(p11;p11;p21), have been associated with SM, acute leukemia, or MPD associated with both T-cell lymphoma and marrow eosinophilia.

TABLE 1. Classification and Causes of Blood Eosinophilia* Familial Acquired Secondary Infectious causes Tissue-invasive parasitosis (most common) Bacterial or viral infections (rare) Noninfectious causes Drugs (sulfa, carbamazepine, etc) Toxins (associated with eosinophilia-myalgia syndrome, toxic oil syndrome, etc) Allergy (asthma, atopic dermatitis, etc) Idiopathic/autoimmune inflammatory conditions Vasculitis (Churg-Strauss, Wegener) Kimura disease Eosinophilic fasciitis Systemic sclerosis or scleroderma Polyarteritis and other CTDs Sarcoidosis Inflammatory bowel disease Malignancy (metastatic cancer, Hodgkin lymphoma) Endocrinopathies (Addison disease, etc) Primary Clonal Acute leukemia Acute myeloid leukemia Acute lymphocytic leukemia Chronic myeloid disorder Molecularly defined chronic myeloid disorder Bcr/Abl+ chronic myeloid leukemia PDGFRA-rearranged eosinophilic disorder PDGFRB-rearranged eosinophilic disorder Kit-mutated systemic mastocytosis 8p11 Syndrome Clinicopathologically defined chronic myeloid disorder Myelodysplastic syndrome Myeloproliferative disorder Classic myeloproliferative disorder (polycythemia vera, etc) Atypical myeloproliferative disorder Chronic eosinophilic leukemia Systemic mastocytosis Chronic myelomonocytic leukemia Unclassified myeloproliferative disorder Idiopathic HES Pre-HES

CTD = connective tissue disease HES = hypereosinophilic syndrome PDGFR = platelet-derived growth factor receptor

Idiopathic Eosinophilia. When both clinical and laboratory evaluations do not clearly identify either a secondary or a clonal cause, a working diagnosis of idiopathic eosinophilia is reasonable. For now, HES is considered a subcategory of idiopathic eosinophilia that requires documentation of both target organ damage and an AEC of 1500 cells/μL or greater for at least 6 months. Several review articles have described the clinical manifestations and natural history of HES. These and other case reports have shown that some cases evolve into either acute leukemia or an aggressive disease phenotype that is indistinguishable from an MPD. This information has long suggested that HES may represent a clonal hematologic disorder. This contention was supported by clonality studies that used X-linked DNA analysis.92 In contrast, the presence of clonal TH2 lymphocytes in a subset of patients with “HES” suggests a pathogenetic heterogeneity that may include an IL-5–mediated secondary eosinophilia in the mix. In a study of 60 patients with HES, 16 (27%) had T cells with an aberrant immunophenotype, and T-cell clonality was suggested in half of these 16 patients. As in clonal eosinophilia, more than 90% of patients with HES are males. The reason for male predominance in HES and in clonal eosinophilia is unknown. Clinical manifestations can be either nonspecific (cough, nocturnal sweating, fatigue, anorexia, weight loss, gastrointestinal symptoms) or directly related to an affected target organ including the skin (pruritus, erythematous papules, nodules, urticaria, angioedema, mucosal ulcers, eosinophilic cellulitis or Wells syndrome), heart (mural thrombus, endocardial fibrosis, cardiomyopathy, elevated serum troponin level), nervous system (sensory and motor neuropathy, mononeuritis multiplex, isolated vasculitis of central nervous system, eosinophilic meningitis, transverse myelitis), lung (pulmonary infiltrates, lung nodules, acute respiratory distress syndrome), or gastrointestinal system (gastroenteritis, inflammatory bowel disease, sclerosing cholangitis). Thromboembolic complications, including Budd-Chiari syndrome, are not infrequent in HES, and in isolated reports, HES has been associated with retinal vein occlusions, polymyositis, arthritis, and renal disease.

DIAGNOSIS

A thorough patient history is the most important part of the evaluation for blood eosinophilia, and it should guide the extent and type of laboratory tests performed. The initial focus should be on international travel history and a review of current and recent medications. In the absence of possible drug association, a stool test should be performed in all cases to look for ova and larvae of intestinal worms (ascaris, schistosoma, ancylostoma, cestodes, fasciola, strongyloides). Next, depending on the travel history, the following tests and procedures should be performed: urine sediment tests (schistosoma), blood concentration tests (filaria), serology (strongyloides, filaria, trichinosis, visceral larva migrans, schistosomiasis), sputum examination (paragonimiasis, visceral larva migrans), chest radiography (paragonimiasis, ascariasis), computed tomography (echinococcus, cysticercosis), small bowel biopsy (isosporiasis, strongyloidosis), or muscle biopsy (trichinosis). However, work-up beyond stool examination is recommended only if history dictates. Similarly, a good history and basic laboratory tests should be adequate to address the possibility of a noninfectious cause of secondary eosinophilia, and only in the presence of historical clues should more invasive and costly interventions be pursued. All patients with suspected primary eosinophilia should undergo bone marrow examination with cytogenetics. Also, my colleagues and I encourage use of bone marrow immunohistochemical stains for tryptase and mast cell immunophenotyping (neoplastic but not normal or reactive mast cells express CD25) to test for SM. The following PDGFRB rearrangements currently are detected by conventional cytogenetics: t(5;12)(q33;p13), t(5;10)(q33;q21), t(5;7)(q33;q11.2), t(5;14)(q33;q13), t(5;17)(q33;p11), and t(1;5)(q23;q33). In contrast, standard cytogenetic methods do not detect the usual PDGFRA rearrangement that is a result of an interstitial 4q12 deletion. This must be sought with either a fluorescence in situ hybridization (FISH)–based or a reverse transcriptase polymerase chain reaction–based laboratory technique. At present, we perform 3 laboratory tests (bone marrow biopsy, bone marrow cytogenetics, peripheral blood FISH for FIP1L1-PDGFRA) in all patients with primary eosinophilia because of the important therapeutic implications (Figure 2). Finally, in addition to looking for the cause of eosinophilia, laboratory tests to assess possible eosinophilic tissue damage may be required and include echocardiography, chest radiography, pulmonary function tests, and, in the presence of symptoms, tissue biopsy.

TREATMENT OF PRIMARY EOSINOPHILIA CLONAL EOSINOPHILIA

In general, the underlying hematologic malignancy dictates treatment in clonal eosinophilia. In acute leukemia, for example, standard induction chemotherapy is used. In CML, imatinib is the drug of choice. Imatinib works in CML by inhibiting the leukemogenic bcr-abl gene product, which is a constitutively activated Abl tyrosine kinase. Although relatively selective in its action, imatinib inhibits other receptor tyrosine kinases including PDGFR and Kit. The mechanism of action involves a competitive occupation of an adenosine triphosphate–binding site of the catalytic domain of the kinase by a conformationdependent mechanism.

The in vitro demonstration of imatinib-induced inhibition of both PDGFR- and Kit-associated signal transduction formed the rationale for using the drug in clonal eosinophilia associated with SM (SM-eos). This is because (1) Kit ligand and Kit-associated signaling are key for the growth and development of human mast cells127 and (2) approximately 50% of patients with SM-eos may carry a constitutively activated PDGFRA mutation, as discussed previously. In vitro, imatinib effectively inhibits normal mast cell growth and development128 as well as the growth of human mast cell lines and primary cells that carry certain (Val560Gly) but not other (Asp816Val) c-kit mutations. Consistent with these in vitro observations, imatinib has been shown effective in a subset of patients with SM-eos who carry the FIP1L1-PDGFRA mutation but not the c-kit Asp816Val mutation. Patients with SM-eos and the FIP1L1-PDGFRA mutation experience both molecular and mast cell immunophenotypic remissions with relatively low doses of imatinib (100 mg/d). Other investigators refer to FIP1L1-PDGFRA+ SM-eos as either myeloproliferative-variant HES, FIP1L1-PDGFRA+ CEL, or FIP1L1-PDGFRA+ HES and have reported equally impressive imatinib treatment results. Imatinib therapy produces complete clinical, pathologic, and molecular remissions in clonal eosinophilia connected with PDGFRB mutations that are usually associated with a chronic myeloid disorder that phenotypically resembles CEL, CMML, or atypical MPD.

IDIOPATHIC EOSINOPHILIA INCLUDING HES Most patients with symptomatic HES respond to prednisone alone or in combination with hydroxyurea. Patients with HES refractory to corticosteroids have been reported to respond to cyclosporine, vincristine, interferon alfa, cladribine, or etoposide. My preference during symptomatic disease is to use interferon alfa early either by itself or combined with low doses of prednisone. For patients in whom first-line treatment with corticosteroids, hydroxyurea, and interferon alfa fails, the choice of second-line treatment is currently arbitrary, and any of the aforementioned agents can be used before an aggressive approach such as allogeneic hematopoietic stem cell transplantation is considered seriously. Because of the documentation of imatinib-induced partial remissions in approximately 50% of patients with true HES (ie, FIP1L1-PDGFRA–), it is reasonable to try the specific drug as a second-line therapy. However, it should be noted that patients with true HES are unlikely to respond to low-dose imatinib (100 mg/d), and a higher dose of the drug (400 mg/d) should be used in this instance (Figure 2). In patients with HES refractory to drugs, both myeloablative and nonmyeloablative allogeneic hematopoietic stem cell transplantation have been used successfully and can be considered. Currently, the role of novel drug treatment approaches to HES, including the use of imatinib and monoclonal antibodies to either IL-5151-153 or CD52 (alemtuzumab), is being defined.

FIGURE 2. A diagnostic and treatment algorithm for blood eosinophilia.

FISH = fluorescence in situ hybridization RT-PCR = reverse transcriptase polymerase chain reaction.

Evaluate for secondary causes Drug history Travel history Stool for ova and parasites Other tests if indicated Clonal eosinophilia with a molecular target for imatinib Clonal eosinophilia without a molecular target for imatinib Hypereosinophilic syndrome Hydroxyurea Vincristine Cladribine Allogeneic transplant Imatinib, 100 mg/d Consider treatment option that is optimal to the underlying clonal disorder Interferon alfa with or without prednisone No response No response Imatinib, 400 mg/d Evaluate for clonal eosinophilia Bone marrow examination with tryptase stains Cytogenetic studies FISH or RT-PCR to detect FIP1L1-PDGFRA mutation

CURRENT ALGORITHM FOR THE DIAGNOSIS AND TREATMENT OF EOSINOPHILIC DISORDERS

Figure 2 outlines a proposed algorithm for the diagnosis and treatment of eosinophilic disorders. To begin treatment of a patient with blood eosinophilia, the possibility of secondary eosinophilia must be excluded. Once this is accomplished, I recommend obtaining a set of blood and bone marrow studies. The blood studies should include serum tryptase (an increased level suggests SM), T-cell receptor gene rearrangement analysis (positive test results suggest an underlying clonal T-cell disorder), and serum IL-5 (an elevated level requires careful evaluation of bone marrow studies and T-cell gene rearrangement studies for the presence of a clonal T-cell disease). Bone marrow examination should include cytogenetic studies, tryptase immunostains, and FISH or reverse transcriptase polymerase chain reaction to screen for FIP1L1-PDGFRA. The last-mentioned test also can be performed on peripheral blood.

If this work-up reveals an imatinib-sensitive molecular target (PDGFRA or PDGFRB mutations), then treatment with low-dose imatinib (100 mg/d) is recommended, and a complete and durable remission can be expected. We currently recommend an initial imatinib dosage of 100 mg/d for FIP1L1-PDGFRA+ clonal eosinophilia because this specific dosage, as discussed previously, has been associated with molecular disease remission. It is reasonable to consider treatment even in the absence of symptoms for FIP1L1-PDGFRA+ clonal eosinophilia in hopes of preventing serious cardiovascular complications that may or may not be amenable to imatinib therapy once they occur.

In the absence of an imatinib-sensitive molecular target, the physician first must decide whether treatment is indicated and if so, choose a first-line treatment that is suitable for the underlying clonal process or, in cases of HES, treat with interferon alfa and/or prednisone. Imatinib is unlikely to benefit patients with clonal eosinophilia who do not have a known drug-sensitive mutation. Similarly, although some patients with true HES experience partial remission with imatinib therapy at 400 mg/d,72 the long-term benefit of single-agent imatinib therapy in FIP1L1-PDGFRA– HES is unknown. Finally, we and others have described lifethreatening cardiogenic shock associated with imatinib therapy for HES. Fortunately, this potentially fatal complication responds well to systemic corticosteroid therapy and may be predicted by elevated serum troponin levels before treatment. For patients with either baseline elevation of serum troponin level or echocardiographic evidence for eosinophilic heart disease, we recommend concomitant prednisone therapy (1 mg/kg) during the first week of treatment with imatinib. Prednisone can be tapered during the second week of therapy.

CONCLUSION

The clinical observation that some patients with “HES” responded to treatment with imatinib, which is a small molecule inhibitor of certain receptor tyrosine kinases, led to the recent discovery of the association between blood eosinophilia and a cytogenetically occult genetic mutation (FIP1L1-PDGFRA). Such molecular characterizations may ultimately result in HES becoming a nonentity in the spectrum of primary eosinophilic disorders. More importantly, the identification of disease-causing mutations is an essential step toward the development and application of rational drug therapy. At present, approximately 10% to 20% of patients with primary eosinophilia are known to harbor a molecular target that predicts an excellent response to imatinib therapy. However, a subset of patients with HES without known imatinib targets also has been reported to respond to the drug. These observations raise the exciting prospect of discovering new drug targets in eosinophilic disorders as well as future therapeutic successes with the use of second-generation kinase inhibitors that are more potent than imatinib and have a broader spectrum of activity.