- Email: [email protected]
Severe Congenital Neutropenia
Severe Congenital Neutropenia Karl Welte,a Cornelia Zeidler,a and David C. Daleb Severe congenital neutropenia (CN) includes a variety of hematologic disorders characterized by severe neutropenia, with absolute neutrophil counts (ANC) below 0.5 ⴛ 109/L, and associated with severe systemic bacterial infections from early infancy. One subtype of CN, Kostmann syndrome, is an autosomal recessive disorder, characterized histopathologically by early-stage maturation arrest of myeloid differentiation. CN with similar clinical features occurs as an autosomal dominant disorder and many sporadic cases also have been reported. This genetic heterogeneity suggests that several pathophysiological mechanisms may lead to this common clinical phenotype. Recent studies on the genetic bases of CN have detected inherited or spontaneous point mutations in the neutrophil elastase gene (ELA 2) in about 60% to 80% of patients and, less commonly, mutations in other genes. Acquisition of additional genetic defects during the course of the disease, for example, granulocyte colony-stimulating factor (G-CSF) receptor gene mutations and cytogenetic aberrations, indicates an underlying genetic instability as a common feature for all congenital neutropenia subtypes. Data on more than 600 patients with CN collected by the Severe Chronic Neutropenia International Registry (SCNIR) demonstrate that, regardless of the particular CN subtype, more than 95% of these patients respond to recombinant human (rHu)G-CSF with ANCs that can be maintained above 1.0 ⴛ 109/L. Adverse events include mild splenomegaly, osteoporosis, and malignant transformation into myelodysplasia (MDS)/leukemia. If and how G-CSF treatment impacts on these adverse events is not fully understood. In recent analyses the influence of the G-CSF dose required to achieve neutrophil response (ANC >1,000/L) in the risk of developing acute myeloid leukemia (AML) has been reported. Hematopoietic stem cell transplantation (HSCT) is still the only treatment available for patients who are refractory to G-CSF treatment. Semin Hematol 43:189-195 © 2006 Elsevier Inc. All rights reserved.
I
n 1956 the Swedish physician Rolf Kostmann described an autosomal recessive hematologic disorder with severe neutropenia with an absolute neutrophil count (ANC) less than 0.2 ⫻ 109/L and early onset of severe bacterial infections.1,2 Most children died because of these infections despite antibiotic treatment. The underlying genetic defect of Kostmann syndrome is largely unknown, but genetic analyses in autosomal dominant and sporadic cases of severe congenital neutropenia (CN) indicate that 60% of cases are attributable to mutations in the gene for neutrophil elastase (ELA 2). Current knowledge on the pathophysiology of severe CN implicates that the clinical diagnosis of CN includes a heterogeneous group of disorders following different patterns of inheritance (Table 1).
aDepartment
of Pediatric Hematology/Oncology, Medical School Hannover, Hannover, Germany. bDepartment of Medicine, University of Washington, Seattle, WA. Address correspondence to Karl Welte, MD, PhD, Department for Pediatric Hematology/Oncology, Medizinische Hochschule Hannover, Carl-Neuberg-Str. 1, 30625 Hannover, Germany. E-mail: [email protected]
0037-1963/06/$-see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1053/j.seminhematol.2006.04.004
Bone marrow transplantation (BMT) was previously the only curative treatment option for patients with human leukocyte antigen (HLA)-compatible donors3. The availability of recombinant human granulocyte colony-stimulating factor (rHuG-CSF) in 19874 dramatically changed both the prognosis of CN and the quality of life of patients with CN.5,6 Since the establishment of the Severe Chronic Neutropenia International Registry (SCNIR) in 1994, data on 611 patients with CN have been collected to monitor the clinical course, treatment, and disease outcomes. More than 90% of patients in clinical trials responded to rHuG-CSF treatment with an increase in ANCs to greater than 1.0 ⫻ 109/L. All responding patients required significantly fewer antibiotics and fewer days of hospitalization.6-10 Hematopoietic stem cell transplantation (HSCT) remains the only currently available treatment for patients who are refractory to rHuG-CSF treatment and continue to have severe and life-threatening bacterial infections and when the disease has transformed into myelodysplasia (MDS)/leukemia.11 189
K. Welte, C. Zeidler, and D.C. Dale
190 Table 1 Genetic Aberrations Causing Congenital Neutropenia Disorder Congenital neutropenia with ELA 2 mutations* Congenital neutropenia with GFI-1 mutation WHIM syndrome Shwachman-Diamond syndrome Glycogen storage disease, type Ib
Genetic Defect
Recessive Dominant (R-CN) (D-CN)
Neutropenia Plus:
ELA2
ⴚ
ⴙ
Preleukemic syndrome
GFI-1
ⴚ
ⴙ
B-/T-cell deficiency
ⴚ ⴙ ⴙ
ⴙ ⴚ ⴚ
Myelokathexis, IgG deficiency, warts Exocrine pancreas insufficiency Hypoglycemia, lactic acidosis
X-linked X-linked
ⴚ
X-linked
ⴚ
IgG, IgA, IgE deficiency Dilatative cardiomyopathy, skeletal myopathy, short stature, Monocytopenia, platelets normal
ⴙ
ⴚ
CXCR4 SBDS Glucose-6-phosphateTranslocase Hyper IgM CD40-L Barth syndrome (3-methylglutaconic Taz 1 aciduria) Congenital neutropenia with WASP WASP mutation Hermansky-Pudlak syndrome AP3B1 Congenital neutropenia with p14 (MAPBPIP) mutation Griscelli syndrome Chediak-Higashi syndrome
P14/MAPBPIP
ⴙ
ⴚ
Rab27a LYST (CHS1)
ⴙ ⴙ
ⴚ ⴚ
Congenital neutropenia (unclassified)
Not known
ⴙ
?
Partial albinism, short stature, IgG deficiency, platelet dysfunction Partial albinism, short stature, IgG deficiency Partial albinism, hemophagocytosis Partial albinism, T-/natural killer cytotoxicity and chemotaxis defect Elevated IgG levels
*Congenital neutropenia with ELA 2 mutations comprises approximately 60% of patients.
Pathophysiologic Mechanisms The underlying genetic defect of CN is still only partially understood. The original hypotheses for this disorder included defective production of G-CSF or defective response of the neutrophilic precursors to G-CSF or other hematopoietic growth factors. We showed, however, that serum from patients with CN contains normal or increased levels of GCSF12 with normal biological activity of endogenous G-CSF and that G-CSF receptors are expressed on myeloid cells from CN patients in normal or increased numbers.13 Linkage analyses of inherited forms of the disease and sequencing of candidate genes provided alternative approaches to understanding the pathogenesis of CN. The identification of mutations within the ELA 2 gene as the cause of cyclic neutropenia14 led to similar studies in patients diagnosed with CN.15-19 Heterozygous ELA 2 mutations were found in 60% of these cases. Analysis of CN families with autosomal dominant inheritance of elastase mutation and a case with mosaic expression of mutant ELA 215 provided the strongest evidence to implicate ELA 2 mutations in the etiology of CN. The precise cellular mechanisms by which mutant ELA 2 causes neutropenia are uncertain and the search for the role of the altered elastase protein in the pathophysiologic mechanism of CN continues to be investigated.15-17,19-21 We proposed that mutant ELA 2 triggers accelerated apoptosis of developing neutrophil precursors, a concept now supported by several lines of evidence.16 We recently demonstrated that the mutation in the elastase protein leads to cytoplasmic accumulation of a nonfunctional protein, disturbance of intra-
cellular trafficking, and activation of unfolded protein response in neutrophils from patients with CN.21 In addition to ELA 2 mutations, a number of genetic aberrations have been identified in patients with disorders associated with severe CN. Genes that were mutated in minor subgroups of patients with CN included GFI-1, CXCR4, SBDS, WASP, Glucose-6-phophatase-translocase, Taz-1, AP3B1, and others (Table 1). However, these gene mutations in patients with CN are associated with additional symptoms such as pancreas insufficiency in patients with SBDS mutations (Shwachman-Bodian-Diamond syndrome), glycogenstorage disease type Ib, glucose-6-phosphatase-translocase mutations, cardiomyopathy in patients with Taz-1 mutations, and partial albinism and short stature in patients with AP3B1 or p14 mutations (Table 1). A common feature of CN patients is an increase in immunoglobulin levels, a finding that may be secondary to recurrent infections or perhaps suggests a gene defect affecting both myelopoiesis and lymphopoiesis. Myeloid cells from all patients with congenital neutropenia have reduced expression of LEF-1 transcription factor (known to regulate expression of cyclin D1, c-myc, and c/EBP-alpha), suggesting that a LEF-1 defect may be a common downstream defect.22 That myeloid cells from patients with CN demonstrate an increased degree of apoptosis16 suggests that defective or increased expression or mutation of one of the apoptosis-regulating genes could be the cause of neutropenia in some cases. Thus, current data suggest that CN is a multi-gene disorder. Mutations in ELA 2 are the most common genetic abnormality and, along with other genetic or acquired abnormalities, may have a common functional
Severe congenital neutropenia
191
Table 2 Time Course of the Acquisition of G-CSF Receptor Point Mutations in a Patient With Congenital Neutropenia Diagnosis CN Age (yr) G-CSF receptor mutation at nucleotide
CN
CN
7 13 15 2362 2362 2363 2414 2414
CN/AML 16 2363 2414 2390
outcome: reduced survival of myeloid progenitor cells leading to granulopoietic failure. Among CN patients who developed leukemia, acquired G-CSF receptor mutations affecting the cytoplasmic domain have been present in most patients tested,23-29 indicating that such mutations may play a role in leukemogenesis. These G-CSF receptor mutations have not been identified at birth, which suggests they are not responsible for the neutropenia and develop as somatic mutations in a small subgroup of patients, most likely caused by genetic instability. G-CSF receptor analyses cannot be used to diagnose the underlying disease, but might be helpful to screen for the risk of leukemia. The time-course between acquisition of a G-CSF receptor mutation(s) and the development of leukemia varies considerably (Table 2). In a few patients, a G-CSF receptor gene mutation is only present in the leukemic cells, while in others single or multiple mutations of the G-CSF receptor gene predate leukemic transformation by several years.29 In the majority of patients, the G-CSF receptor gene mutations affect only one allele and are most frequently located between nucleotide 2342 and 2522 of the G-CSF receptor gene (Fig 1), coding for a critical region within the cytoplasmic portion of the G-CSF receptor protein. Even in the face of a G-CSF receptor mutation there seems to be no change in the responsiveness of patients’ cells to G-CSF. In some instances of acute myeloid leukemia (AML), reductions in blast counts can occur if G-CSF is discontinued.30
Blood Values Repeated differential blood counts indicating persistent ANCs within a range of 0 to 0.2 ⫻ 109/L are required for diagnosis. Blood counts often indicate mild anemia and thrombocytosis. There is usually a two- to fourfold increase in blood monocytes and an increase in the blood eosinophil count is common. IgG levels are elevated in the majority of patients, independent of their infectious status (unpublished data). The specific immunologic competence after vaccination is normal. Blood chemistry is within the normal, agedependent range for electrolytes and kidney and liver function.
Bone Marrow The bone marrow usually shows a maturation arrest of neutrophil precursors at an early stage (promyelocyte/myelocyte level) with few cells of the neutrophilic series beyond the promyelocyte stage (Fig 2). Promyelocytes often have morphologically atypical nuclei and vacuolization of the cytoplasm. The absolute number of promyelocytes is slightly increased.9 Marrow eosinophilia and monocytosis are common and do not change during treatment. Cellularity is usually normal or slightly decreased. Megakaryocytes are normal in number and morphology. In vitro growth of granulocytemacrophage progenitor cells is defective, with few colonies formed and evidence for “maturation arrest.”
Cytogenetic Evaluation and Molecular Testing
Diagnosis
Cytogenetics at the time of diagnosis of CN are almost always normal. Bone marrow cytogenetics may change during the course of the disease, with monosomy 7 being the most frequent aberration in about 50% of abnormal cytogenetic results. The interval from the original finding of normal cytogenetics to the appearance of an abnormality is often several years. Abnormal cytogenetics are often associated with morphologic changes of the bone marrow indicating the development of MDS or leukemia (see below) and may occur with the development of cytopenias or refractoriness to G-CSF treatment.
CN is a rare condition with an estimated frequency of approximately one to two cases per million with equal distribution for gender. The disease is usually detected in infancy after fever or signs of a severe infection develop. CN patients suffer from severe chronic neutropenia with ANCs continuously below 200/L; in many cases, peripheral blood neutrophils are completely absent. With infection, there may be a transient increase in neutrophils, but counts rarely increase to normal levels. The diagnosis of Kostmann syndrome is sometimes used to categorize sporadic cases of CN, children with no known affected family member. We prefer the term “severe congenital neutropenia” or “congenital neutropenia” for this heterogeneous patient population. For diagnosis during the first months of life, testing for anti-neutrophil antibodies is helpful to exclude autoimmune neutropenia of infancy. In CN cases, the results are usually (but not always) negative.
Figure 1 Point mutations within the critical region of the G-CSF receptor gene (2,300 to 2,600 basepairs). EC, extracellular; IC, intracytoplasmic.
K. Welte, C. Zeidler, and D.C. Dale
192
ies indicates an increased peripheral destruction of neutrophils in the marrow and blood. Although these infants lack peripheral blood neutrophils, the marrow function is normal; maturation of myeloid precursors to the mature neutrophil stage is usually observed. Patients with autoimmune neutropenia usually do not suffer from severe bacterial infections. Neutrophil or granulocyte-specific antibodies in serum are detectable using immunologic tests.31
Treatment
Figure 2 Typical bone marrow morphology in CN.
Studies of the G-CSF receptor gene have shown that mutations occur mainly within a critical region (nucleotide position 2300 to 2600) of the intracellular part of the receptor (Fig 1).23-29 These mutations are not present at birth, so they are not causative for CN. Analysis of the G-CSF receptor gene to detect acquired mutations can be performed using blood and bone marrow samples.
Onset of Symptoms Severe bacterial infections frequently occur during the first year of life in children with CN. Omphalitis, beginning directly after birth, may be the first symptom, but otitis media, pneumonitis, and infections of the upper respiratory tract, and abscesses of skin or liver are also common and can lead to the diagnosis of CN. Cultures are mainly positive for staphylococci or streptococci, but other bacteria (Pseudomonas, Peptostreptococcus) and, rarely, fungi are detected. In addition, uncommon infections such as clostridial gas gangrene may occur. The outcome of these fulminant infections is often lethal due to the lack of neutrophil defense. Most patients suffer from frequent aphthous stomatitis and gingival hyperplasia, leading to an early loss of permanent teeth.
Differential Diagnosis The differential diagnosis of congenital neutropenia10 includes a number of congenital and acquired diseases. The most common difficulty is determining whether the patient has cyclic neutropenia, autoimmune neutropenia of infancy, or idiopathic neutropenia. Cyclic neutropenia is diagnosed by serial measurement of blood neutrophils on at least 3 days per week for at least 6 weeks, graphing the counts and determining if there are regular periods of very severe neutropenia at 3-week intervals. Even with careful serial counts, the diagnosis may be missed because the amplitude of the oscillations may be too small for detection. An important differential diagnostic evaluation is testing for neutrophilic antibodies.31 In neutropenic children aged 1 to 3 years, the presence of neutrophil-specific auto-antibod-
rHuG-CSF has been available for treatment of CN since 1987. Phase I–III studies have demonstrated its efficiency in increasing the number of neutrophils associated with reduction of infections.6,32 In contrast, GM-CSF treatment does not lead to an increase in blood neutrophils but only in blood eosinophils.33 In 1994, the SCNIR was established to collect data on clinical course and outcome of these rare disorders. As of December 2005, 611 patients with CN have been registered, and more than 95% responded to rHuG-CSF treatment with an increase in ANCs to 1.0 ⫻ ⱖ109/L (Fig 3). Most CN patients respond to a dose between 3 and 10 g/kg/d. After initiation of rHuG-CSF at 5 g/kg/d, the dose should be escalated to 10 g/kg/d and then by increments of 10 g/kg at 14-day intervals if the ANC remains below 1.0 ⫻ 109/L. As soon as the ANC can be maintained at 1.0 to 1.5 ⫻ 109/L or above, increases of the rHuG-CSF dose can be stopped since the occurrence of bacterial infection is reduced dramatically at this level. The dose of rHuG-CSF can be reduced if the ANC increases to ⱖ5.0 ⫻ 109/L to keep the patient at the lowest dose necessary for maintaining a sufficient neutrophil count to overcome infections. Nonresponders to rHuG-CSF are defined by failure to benefit at dose levels exceeding 120 g/kg/d. Partial responders have ANCs increased to 0.5 to 1.0 ⫻ 109/L but still have bacterial infections. In some patients, the dose of rHuG-CSF cannot be increased to these levels because of the large volume and frequency of injections required. In research studies, refractory patients received a combination of rHuG-CSF with stem cell factor (SCF), which led to a further increase in ANCs above the levels achieved with G-CSF alone. Because of allergic side effects of SCF, this treatment combination has only been used during severe infections in hospitalized patients receiving concomitant antihistaminic medication.34 For patients who do not respond to rHuG-CSF alone or in combination with SCF, HSCT is the only currently available treatment.11 When successful, after transplantation patients normalize hematopoiesis and do not require cytokine treatment. It remains difficult to recommend transplantation for patients with CN who benefit from rHuG-CSF and show no evidence of impending malignant transformation. For many responding patients, the risks associated with transplant from an HLA-identical sibling may outweigh the risk of leukemic transformation when rHuG-CSF is continued.
Severe congenital neutropenia
193
Figure 3 Mean ANC by year of G-CSF treatment in CN.
Long-Term Safety Prior to the development and availability of G-CSF, complications of severe congenital neutropenia were not well known. There was no organized database for this rare disease. However, a number of reports noting deaths from infections and a few cases evolving to AML are available from the pre–G-CSF era.
Leukemia Prior to the availability of cytokine therapy, it was already recognized that leukemic transformation occurred in patients with CN.35,36 However, in the pre-cytokine era, 42% of published cases died in the first 2 years of life, usually from sepsis or pneumonia. Thus, the true risk of CN patients developing MDS/AML was not defined. With G-CSF therapy, most of these patients have survived well beyond 2 years of age. Therefore, it is unknown whether the increased survival allows for a higher risk of the recognized natural expression of leukemogenesis in this population in the absence of G-CSF therapy. From the initiation of clinical trials with G-CSF in 1987 through December 2005, a total of 70 patients with CN (including five patients with Shwachman-Diamond syndrome) who have developed MDS/AML were reported to the SCNIR. The overall incidence or crude rate of MDS/AML conversion is 11.5% for CN patients (70 cases among 611 exposed cases), with an average follow-up of approximately 5 to 6 years. The leukemias occurred in those CN patients with ELA 2 mutations, in unclassified CN patients, and in patients with Shwachman-Diamond syndrome but not in the other subgroups listed in Table 1.
The SCNIR recently published a more detailed report on the first 374 patients (1987–2000) with CN on long-term G-CSF enrolled in the registry to identify the risk of leukemic transformation. The hazard of MDS/AML increased significantly during the period of observation on G-CSF, from 2.9% per year after 6 years to 8.0% per year after 12 years. The cumulative incidence of MDS/AML was 21% after 10 years. The risk of MDS/AML increased with the dose of G-CSF. Less responsive patients, defined as those requiring greater than 8 g/kg/d of G-CSF, had a cumulative incidence of MDS/AML of 40% after 10 years, compared to 11% of more responsive patients.37 The data were interpreted as indicating that a poor response to G-CSF defines an “at-risk” population and predicts an adverse outcome. Conversion to MDS/AML in CN patients was associated with one or more cellular genetic abnormalities (such as monosomy 7, ras mutation, or G-CSF receptor mutation), which may be useful for identifying subgroups of patients at high risk. Marrow cells from approximately 75% of the severe CN cases that transformed to MDS/AML also showed point mutations in the gene for the G-CSF receptor, resulting in a truncated C-terminal cytoplasmic region of the receptor that is crucial for maturation signaling.17,23-25,27-29 As illustrated by the cases described here, the development of MDS/AML is a multi-step process characterized by a series of cellular genetic changes indicating a genetic predisposition to malignant transformation. If and how G-CSF impacts on this predisposition remains unclear; there are no historical controls for comparison. To address further the issue of risk-benefit of G-CSF in the CN setting with regard to MDS/AML, all available data were
K. Welte, C. Zeidler, and D.C. Dale
194 critically reviewed.37 Although high-level evidence does not exist that would reveal the proper timing of bone marrow surveillance, the SCNIR advisory board recommended annual marrow cytogenetic testing to identify monosomy 7 or other changes indicating transformation; this approach might permit earlier therapeutic interventions such as BMT.
Osteoporosis Within the SCNIR, bone density measurements have been reported on a total of 121 patients measured by different techniques, including quantitative computer tomography (Q-CT), dual energy x-ray absorptiometry (DEXA), singlephoton absorptiometry (SPA), and lumbar x-ray. Of these 121 patients, 66 (54.4%) had varying degrees of abnormal results. These results have not been quantified to interpret the severity of the abnormalities. Most patients did not show clinical symptoms of osteopenia/osteoporosis (bone pain or fractures), which explains why diagnostic procedures for bone density evaluation were not reported in approximately 70% of registry patients. Therefore, the actual incidence continues to be unknown. The pathophysiology of osteopenia/ osteoporosis also remains unclear. Serum chemistry did not reveal a typical pattern in patients with osteopenia/osteoporosis. Patients who suffer from osteopenia also did not receive elevated amounts of rHuG-CSF doses compared with all registry patients.38
Splenomegaly The incidence of palpable splenomegaly (2 cm below the costal margin) prior to treatment with rHuG-CSF in CN patients was 21%. During the first year of rHuG-CSF therapy, the incidence increased to 39% and remained approximately at this level of occurrence (34% to 48%) through 10 years of therapy.
Monitoring The Advisory Board for the SCNIR recommends that all patients should be seen by a physician at least twice per year with assessment for weight and height and documentation of the occurrence of infections. Blood counts (white blood cells, hemoglobin, platelets, and differential blood counts) should be obtained at least every 3 months. Bone marrow examination (morphology plus cytogenetics) is required once per year to search for acquired cytogenetic abnormalities, such as monosomy 7 or trisomy 21. Furthermore, following informed consent, bone marrow samples should be collected for research studies on the molecular analysis of the pathophysiology of CN and risk factors for leukemic evolution. G-CSF receptor analysis can be performed on heparinized blood or bone marrow samples by laboratories associated with the SCNIR.
Conclusion In light of the current literature and longitudinal data from the SCNIR, we suggest that the use of rHuG-CSF remain
first-line treatment for most CN patients. HSCT from an HLA-identical sibling is beneficial for patients who are refractory to rHuG-CSF. For those patients in whom a G-CSF receptor mutation is identified, HSCT from an HLA-identical sibling is an option. Patients who develop monosomy 7, other significant chromosomal abnormalities, or MDS/leukemia should proceed urgently to HSCT. Data on alternative sources of donor stem cells are insufficient to assess outcome in patients with CN. Other than those patients who fail to respond to rHuG-CSF, the cytokine should be employed to maintain an ANC ranging from 1.0 to 5.0 ⫻ 109/L with amelioration of symptoms. All CN patients, regardless of their treatment or response, are at risk of developing MDS or leukemia at an actual incidence of 11.5% today and a cumulative incidence of 21% after 10 years. With increasing genetic analyses, it may become possible to correlate the leukemic risk with the underlying genetic defect. Careful monitoring for cytogenetic abnormalities and G-CSF receptor mutation is necessary to initiate HSCT as soon as any of these occur. Despite the significant risk of leukemia, HSCT-related morbidity is also significant and, therefore, in patients without signs of leukemia or MDS, HSCT should be restricted to G-CSF nonresponders.
Acknowledgment The authors thank all colleagues associated with the Data Collection Centers of the Severe Chronic Neutropenia International Registry at the University of Washington, Seattle, WA (Audrey Anna Bolyard and Tammy Cottle), and the Medizinische Hochschule, Hannover, Germany (Beate Schwinzer and Gusal Pracht) for their continued assistance. We are also grateful to the many physicians worldwide who faithfully and generously submitted data on their patients. The authors gratefully acknowledge the important contribution of the Advisory Board and the European Liason Physician Group of the Severe Chronic Neutropenia International Registry and support from the National Institutes of Health (USA), the European Union, and Amgen Inc, Thousand Oaks, CA.
References 1. Kostmann R: Infantile genetic agranulocytosis. Acta Pediatr Scand 45: 1-78, 1956 2. Kostmann R: Infantile genetic agranulocytosis: A review with presentation of ten new cases. Acta Pediatr Scand 64:362-368, 1975 3. Rappeport J, Parkman R, Newburger P, Camitta BM, Chusid MJ: Correction of infantile granulocytosis (Kostmann syndrome) by allogeneic bone marrow transplantation. Am J Med 68:605-609, 1980 4. Souza L, Boone T, Gabrilove J, Lai PH, Zsebo KM, Murdock DC, et al: Recombinant human granulocyte colony-stimulating factor: Effects on normal and leukemic myeloid cells. Science 232:61-65, 1986 5. Bonilla M, Gillio A, Ruggeiro M, Kernan NA, Brochstein JA, Abboud M, et al: Effects of recombinant human granulocyte colony-stimulating factor on neutropenia in patients with congenital agranulocytosis. N Engl J Med 320:1574-1580, 1989 6. Bonilla M, Dale D, Zeidler C, Last L, Reiter A, Ruggeiro M, et al: Longterm safety of treatment with recombinant human granulocyte colonystimulating factor (r-metHuG-CSF) in patients with severe congenital neutropenias. Br J Hematol 88:723-730, 1994 7. Freedman MH: Safety of long-term administration of granulocyte col-
Severe congenital neutropenia
8. 9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
ony-stimulating factor for severe chronic neutropenia. Curr Opin Hematol 4:217-224, 1997 Welte K, Dale D: Pathophysiology and treatment of severe chronic neutropenia. Ann Hematol 72:158-165, 1996 Welte K, Boxer L: Severe chronic neutropenia: Pathophysiology and therapy. Semin Hematol 34:267-278, 1997 Zeidler C, Boxer L, Dale DC, Freedman MH, Kinsey S, Welte K: Management of Kostmann syndrome in the G-CSF era. Br J Haematol 109: 490-495, 2000 Zeidler C, Welte K, Barak Y, Barriga F, Bolyard AA, Boxer L, et al: Stem cell transplantation in patients with severe congenital neutropenia without evidence of leukemic transformation. Blood 95:1195-1198, 2000 Mempel K, Pietsch,T, Menzel, Zeidler C, Welte K: Increased serum levels of granulocyte colony-stimulating factor in patients with severe congenital neutropenia. Blood 77:1919-1922, 1991 Kyas U, Pietsch T, Welte K: Expression of receptors for granulocytecolony stimulating factor on neutrtophilsfrom patients with severe congenital neutropenia and cyclic neutropenia. Blood 79:1144-1147, 1992 Horwitz, M, Benson KF, Person RE, Aprikyan AG, Dale DC: Mutations in ELA2, encoding neutrophil elastase, define a 21-day biological clock in cyclic haematopoesis. Nat Genet 23:433-436, 1999 Ancliff PJ, Gale RE, Watts MJ, Liesner R, Hann IM, Strobel S, Linch DC: Paternal mosaicism proves the pathogenic nature of mutations in neutrophil elastase in severe congenital neutropenia. Blood 100:707-709, 2002 Dale DC, Liles WC, Grawicz D, Aprikyan AG: Clinical implications of mutations of neutrophil elastase in congenital and cyclic neutropenia. J Pediatr Hematol Oncol 23:208-210, 2001 Donadieu J, Leblanc T, Meunier B, Barkaoui M, Fenneteau O, Bertrand Y, et al: Analysis of risk factors for myelodysplasia/leukemia and infectious death among patients with congenital neutropenia: Experience of the French Severe Chronic Neutropenia Study Group. Haematologica 90:45-53, 2005 Dale DC, Person RE, Bolyard AA, Aprikyan AG, Bos C, Bonilla MA, et al: Mutations in the gene encoding neutrophil elastase in congenital and cyclic neutropenia. Blood 96:2317-2322, 2000 Massullo P, Druhan LJ, Bunnell BA, Hunter MG, Robinson JM, Marsh CB, et al: Aberrant subcellular targeting of the G185R neutrophil elastase mutant associated with severe congenital neutropenia induces premature apoptosis of differentiating promyelocytes. Blood 105:33973404, 2005 Germeshausen M, Schulze H, Ballmaier M, Zeidler C, Welte K: Mutations in the gene encoding neutrophil elastase (ELA2) are not sufficient to cause the phenotype of congenital neutropenia. Br J Haematol 115: 1-4, 2001 Koellner I, Sodeik B, Schreek S, Heyn H, von Neuhoff N, Germeshausen M, et al.: Mutations in neutrophil elastase causino congenital neutropenia lead to cytoplasmic protein accumulation and induction of the unfolded protein response. Blood March 21, 2006 (Epub ahead of print) Skokowa J, Cario G, Bucan V, Dan L, Wang Z, Scherr M, et al: LEF-1 transcription factor regulates proliferation and differentiation of myeloid progenitors in healthy individuals and in patients with severe congenital neutropenia (CN). Blood 106:390a, 2005 (abstr) Bernard T, Gale R, Evans J, Linch DC: Mutations of the granulocytecolony stimulating factor receptor in patients with severe congenital
195
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35. 36. 37.
38.
neutropenia are not required for transformation to acute myeloid leukaemia and may be a bystander phenomenon. Br J Haematol 101:141-149, 1998 Cassinat B, Bellanné-Chantelot C, Notz-Carrère A, Menot ML, Vaury C, Micheau M, et al: Screening for G-CSF receptor mutations in patients with secondary myeloid or lymphoid transformation of severe congenital neutropenia. A report from the French Neutropenia Register. Leukemia 18:1553-1555, 2004 Dong F, Russel KB, Tidow N, Freedman M, Fasth A, Neijens HJ, et al: Mutations in the gene for the granulocyte-colony stimulating factor receptor in patients with acute myeloid leukemia preceded by severe congenital neutropenia. N Engl J Med 333:487-493, 1995 Germeshausen M, Schulze H, Kratz C, Wilkens L, Repp R, Shannon K, et al: An acquired G-CSF receptor mutation results in increased proliferation of CMML cells from a patient with severe congenital neutropenia. Leukemia 19:611-617, 2005 Papadaki HA, Kosteas T, Gemetzi C, Damianaki A, Anagnou NP, Eliopoulos GD: Acute myeloid/NK precursor cell leukemia with trisomy 4 and a novel point mutation in the extracellular domain of the G-CSF receptor in a patient with chronic idiopathic neutropenia. Ann Hematol 83:345-348, 2004 Tidow N, Pilz C, Teichmann B, Muller-Brechlin A, Germeshausen M, Kasper B, et al: Clinical relevance of point mutations in the cytoplasmatic domain of the granulocyte-colony stimulating factor gene in patients with severe congenital neutropenia. Blood 88:2369-2375, 1997 Tschan CA, Pilz C, Zeidler C, Welte K, Germeshausen M: Time course of increasing numbers of mutations in the granulocyte colony-stimulating factor receptor gene in a patient with congenital neutropenia who developed leukemia. Blood 97:1882-1884, 2001 Jeha S, Chan KW, Aprikyan AG, Hoots WK, Culbert S, Zietz H, et al: Spontaneous remission of granulocyte colony-stimulating factor–associated leukemia in a child with severe congenital neutropenia. Blood 96:3647-3649, 2000 Bux J, Behrens G, Jaeger G, Welte K: Diagnosis and clinical course of autoimmune neutropenia in infancy: Analysis of 240 cases. Blood 91: 181-186, 1998 Dale D, Bonilla M, Davis M, Nakanishi AM, Hammond WP, Kurtzberg J, et al: A randomized controlled phase III trial of recombinant human granulocyte colony-stimulating factor (Filgrastim) for treatment of severe chronic neutropenia. Blood 81:2496-2502, 1993 Welte K, Zeidler C, Reiter A, Muller W, Odenwald E, Souza L, Riehm H: Differential effects of granulocyte-macrophage colony-stimulating factor and granulocyte colony-stimulating factor in children with severe congenital neutropenia. Blood 75:1056-1063, 1990 Zeidler C, Vogel R, Wyres M, et al: Beneficial effects of stem cell factor (SCF) in children with severe congenital neutropenias refractory to G-CSF. Blood 92:380a, 1998 (abstr) Gilman P, Jackson D, Guild H: Congenital agranulocytosis: Prolonged survival and terminal acute leukemia. Blood 36:576-585, 1971 Rosen R, Kang S: Congenital agranulocytosis terminating in acute myelomonocytic leukemia. J Pediatr 94:406-408, 1979 Rosenberg PS, Alter BP, Bolyard AA , Bonilla MA, Boxer LA, Cham B, et al: The incidence of leukemia and mortality from sepsis in patients with severe congenital neutropenia receiving long-term G-CSF therapy. Blood, February 23, 2006 (Epub ahead of print) Yakisan E, Schirg E, Zeidler C, Bischop NJ, Reiter A, et al: High incidence of significant bone loss in patients with severe congenital neutropenia (Kostmann’s syndrome). J Pediatr 131:592-597, 1997
I
n 1956 the Swedish physician Rolf Kostmann described an autosomal recessive hematologic disorder with severe neutropenia with an absolute neutrophil count (ANC) less than 0.2 ⫻ 109/L and early onset of severe bacterial infections.1,2 Most children died because of these infections despite antibiotic treatment. The underlying genetic defect of Kostmann syndrome is largely unknown, but genetic analyses in autosomal dominant and sporadic cases of severe congenital neutropenia (CN) indicate that 60% of cases are attributable to mutations in the gene for neutrophil elastase (ELA 2). Current knowledge on the pathophysiology of severe CN implicates that the clinical diagnosis of CN includes a heterogeneous group of disorders following different patterns of inheritance (Table 1).
aDepartment
of Pediatric Hematology/Oncology, Medical School Hannover, Hannover, Germany. bDepartment of Medicine, University of Washington, Seattle, WA. Address correspondence to Karl Welte, MD, PhD, Department for Pediatric Hematology/Oncology, Medizinische Hochschule Hannover, Carl-Neuberg-Str. 1, 30625 Hannover, Germany. E-mail: [email protected]
0037-1963/06/$-see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1053/j.seminhematol.2006.04.004
Bone marrow transplantation (BMT) was previously the only curative treatment option for patients with human leukocyte antigen (HLA)-compatible donors3. The availability of recombinant human granulocyte colony-stimulating factor (rHuG-CSF) in 19874 dramatically changed both the prognosis of CN and the quality of life of patients with CN.5,6 Since the establishment of the Severe Chronic Neutropenia International Registry (SCNIR) in 1994, data on 611 patients with CN have been collected to monitor the clinical course, treatment, and disease outcomes. More than 90% of patients in clinical trials responded to rHuG-CSF treatment with an increase in ANCs to greater than 1.0 ⫻ 109/L. All responding patients required significantly fewer antibiotics and fewer days of hospitalization.6-10 Hematopoietic stem cell transplantation (HSCT) remains the only currently available treatment for patients who are refractory to rHuG-CSF treatment and continue to have severe and life-threatening bacterial infections and when the disease has transformed into myelodysplasia (MDS)/leukemia.11 189
K. Welte, C. Zeidler, and D.C. Dale
190 Table 1 Genetic Aberrations Causing Congenital Neutropenia Disorder Congenital neutropenia with ELA 2 mutations* Congenital neutropenia with GFI-1 mutation WHIM syndrome Shwachman-Diamond syndrome Glycogen storage disease, type Ib
Genetic Defect
Recessive Dominant (R-CN) (D-CN)
Neutropenia Plus:
ELA2
ⴚ
ⴙ
Preleukemic syndrome
GFI-1
ⴚ
ⴙ
B-/T-cell deficiency
ⴚ ⴙ ⴙ
ⴙ ⴚ ⴚ
Myelokathexis, IgG deficiency, warts Exocrine pancreas insufficiency Hypoglycemia, lactic acidosis
X-linked X-linked
ⴚ
X-linked
ⴚ
IgG, IgA, IgE deficiency Dilatative cardiomyopathy, skeletal myopathy, short stature, Monocytopenia, platelets normal
ⴙ
ⴚ
CXCR4 SBDS Glucose-6-phosphateTranslocase Hyper IgM CD40-L Barth syndrome (3-methylglutaconic Taz 1 aciduria) Congenital neutropenia with WASP WASP mutation Hermansky-Pudlak syndrome AP3B1 Congenital neutropenia with p14 (MAPBPIP) mutation Griscelli syndrome Chediak-Higashi syndrome
P14/MAPBPIP
ⴙ
ⴚ
Rab27a LYST (CHS1)
ⴙ ⴙ
ⴚ ⴚ
Congenital neutropenia (unclassified)
Not known
ⴙ
?
Partial albinism, short stature, IgG deficiency, platelet dysfunction Partial albinism, short stature, IgG deficiency Partial albinism, hemophagocytosis Partial albinism, T-/natural killer cytotoxicity and chemotaxis defect Elevated IgG levels
*Congenital neutropenia with ELA 2 mutations comprises approximately 60% of patients.
Pathophysiologic Mechanisms The underlying genetic defect of CN is still only partially understood. The original hypotheses for this disorder included defective production of G-CSF or defective response of the neutrophilic precursors to G-CSF or other hematopoietic growth factors. We showed, however, that serum from patients with CN contains normal or increased levels of GCSF12 with normal biological activity of endogenous G-CSF and that G-CSF receptors are expressed on myeloid cells from CN patients in normal or increased numbers.13 Linkage analyses of inherited forms of the disease and sequencing of candidate genes provided alternative approaches to understanding the pathogenesis of CN. The identification of mutations within the ELA 2 gene as the cause of cyclic neutropenia14 led to similar studies in patients diagnosed with CN.15-19 Heterozygous ELA 2 mutations were found in 60% of these cases. Analysis of CN families with autosomal dominant inheritance of elastase mutation and a case with mosaic expression of mutant ELA 215 provided the strongest evidence to implicate ELA 2 mutations in the etiology of CN. The precise cellular mechanisms by which mutant ELA 2 causes neutropenia are uncertain and the search for the role of the altered elastase protein in the pathophysiologic mechanism of CN continues to be investigated.15-17,19-21 We proposed that mutant ELA 2 triggers accelerated apoptosis of developing neutrophil precursors, a concept now supported by several lines of evidence.16 We recently demonstrated that the mutation in the elastase protein leads to cytoplasmic accumulation of a nonfunctional protein, disturbance of intra-
cellular trafficking, and activation of unfolded protein response in neutrophils from patients with CN.21 In addition to ELA 2 mutations, a number of genetic aberrations have been identified in patients with disorders associated with severe CN. Genes that were mutated in minor subgroups of patients with CN included GFI-1, CXCR4, SBDS, WASP, Glucose-6-phophatase-translocase, Taz-1, AP3B1, and others (Table 1). However, these gene mutations in patients with CN are associated with additional symptoms such as pancreas insufficiency in patients with SBDS mutations (Shwachman-Bodian-Diamond syndrome), glycogenstorage disease type Ib, glucose-6-phosphatase-translocase mutations, cardiomyopathy in patients with Taz-1 mutations, and partial albinism and short stature in patients with AP3B1 or p14 mutations (Table 1). A common feature of CN patients is an increase in immunoglobulin levels, a finding that may be secondary to recurrent infections or perhaps suggests a gene defect affecting both myelopoiesis and lymphopoiesis. Myeloid cells from all patients with congenital neutropenia have reduced expression of LEF-1 transcription factor (known to regulate expression of cyclin D1, c-myc, and c/EBP-alpha), suggesting that a LEF-1 defect may be a common downstream defect.22 That myeloid cells from patients with CN demonstrate an increased degree of apoptosis16 suggests that defective or increased expression or mutation of one of the apoptosis-regulating genes could be the cause of neutropenia in some cases. Thus, current data suggest that CN is a multi-gene disorder. Mutations in ELA 2 are the most common genetic abnormality and, along with other genetic or acquired abnormalities, may have a common functional
Severe congenital neutropenia
191
Table 2 Time Course of the Acquisition of G-CSF Receptor Point Mutations in a Patient With Congenital Neutropenia Diagnosis CN Age (yr) G-CSF receptor mutation at nucleotide
CN
CN
7 13 15 2362 2362 2363 2414 2414
CN/AML 16 2363 2414 2390
outcome: reduced survival of myeloid progenitor cells leading to granulopoietic failure. Among CN patients who developed leukemia, acquired G-CSF receptor mutations affecting the cytoplasmic domain have been present in most patients tested,23-29 indicating that such mutations may play a role in leukemogenesis. These G-CSF receptor mutations have not been identified at birth, which suggests they are not responsible for the neutropenia and develop as somatic mutations in a small subgroup of patients, most likely caused by genetic instability. G-CSF receptor analyses cannot be used to diagnose the underlying disease, but might be helpful to screen for the risk of leukemia. The time-course between acquisition of a G-CSF receptor mutation(s) and the development of leukemia varies considerably (Table 2). In a few patients, a G-CSF receptor gene mutation is only present in the leukemic cells, while in others single or multiple mutations of the G-CSF receptor gene predate leukemic transformation by several years.29 In the majority of patients, the G-CSF receptor gene mutations affect only one allele and are most frequently located between nucleotide 2342 and 2522 of the G-CSF receptor gene (Fig 1), coding for a critical region within the cytoplasmic portion of the G-CSF receptor protein. Even in the face of a G-CSF receptor mutation there seems to be no change in the responsiveness of patients’ cells to G-CSF. In some instances of acute myeloid leukemia (AML), reductions in blast counts can occur if G-CSF is discontinued.30
Blood Values Repeated differential blood counts indicating persistent ANCs within a range of 0 to 0.2 ⫻ 109/L are required for diagnosis. Blood counts often indicate mild anemia and thrombocytosis. There is usually a two- to fourfold increase in blood monocytes and an increase in the blood eosinophil count is common. IgG levels are elevated in the majority of patients, independent of their infectious status (unpublished data). The specific immunologic competence after vaccination is normal. Blood chemistry is within the normal, agedependent range for electrolytes and kidney and liver function.
Bone Marrow The bone marrow usually shows a maturation arrest of neutrophil precursors at an early stage (promyelocyte/myelocyte level) with few cells of the neutrophilic series beyond the promyelocyte stage (Fig 2). Promyelocytes often have morphologically atypical nuclei and vacuolization of the cytoplasm. The absolute number of promyelocytes is slightly increased.9 Marrow eosinophilia and monocytosis are common and do not change during treatment. Cellularity is usually normal or slightly decreased. Megakaryocytes are normal in number and morphology. In vitro growth of granulocytemacrophage progenitor cells is defective, with few colonies formed and evidence for “maturation arrest.”
Cytogenetic Evaluation and Molecular Testing
Diagnosis
Cytogenetics at the time of diagnosis of CN are almost always normal. Bone marrow cytogenetics may change during the course of the disease, with monosomy 7 being the most frequent aberration in about 50% of abnormal cytogenetic results. The interval from the original finding of normal cytogenetics to the appearance of an abnormality is often several years. Abnormal cytogenetics are often associated with morphologic changes of the bone marrow indicating the development of MDS or leukemia (see below) and may occur with the development of cytopenias or refractoriness to G-CSF treatment.
CN is a rare condition with an estimated frequency of approximately one to two cases per million with equal distribution for gender. The disease is usually detected in infancy after fever or signs of a severe infection develop. CN patients suffer from severe chronic neutropenia with ANCs continuously below 200/L; in many cases, peripheral blood neutrophils are completely absent. With infection, there may be a transient increase in neutrophils, but counts rarely increase to normal levels. The diagnosis of Kostmann syndrome is sometimes used to categorize sporadic cases of CN, children with no known affected family member. We prefer the term “severe congenital neutropenia” or “congenital neutropenia” for this heterogeneous patient population. For diagnosis during the first months of life, testing for anti-neutrophil antibodies is helpful to exclude autoimmune neutropenia of infancy. In CN cases, the results are usually (but not always) negative.
Figure 1 Point mutations within the critical region of the G-CSF receptor gene (2,300 to 2,600 basepairs). EC, extracellular; IC, intracytoplasmic.
K. Welte, C. Zeidler, and D.C. Dale
192
ies indicates an increased peripheral destruction of neutrophils in the marrow and blood. Although these infants lack peripheral blood neutrophils, the marrow function is normal; maturation of myeloid precursors to the mature neutrophil stage is usually observed. Patients with autoimmune neutropenia usually do not suffer from severe bacterial infections. Neutrophil or granulocyte-specific antibodies in serum are detectable using immunologic tests.31
Treatment
Figure 2 Typical bone marrow morphology in CN.
Studies of the G-CSF receptor gene have shown that mutations occur mainly within a critical region (nucleotide position 2300 to 2600) of the intracellular part of the receptor (Fig 1).23-29 These mutations are not present at birth, so they are not causative for CN. Analysis of the G-CSF receptor gene to detect acquired mutations can be performed using blood and bone marrow samples.
Onset of Symptoms Severe bacterial infections frequently occur during the first year of life in children with CN. Omphalitis, beginning directly after birth, may be the first symptom, but otitis media, pneumonitis, and infections of the upper respiratory tract, and abscesses of skin or liver are also common and can lead to the diagnosis of CN. Cultures are mainly positive for staphylococci or streptococci, but other bacteria (Pseudomonas, Peptostreptococcus) and, rarely, fungi are detected. In addition, uncommon infections such as clostridial gas gangrene may occur. The outcome of these fulminant infections is often lethal due to the lack of neutrophil defense. Most patients suffer from frequent aphthous stomatitis and gingival hyperplasia, leading to an early loss of permanent teeth.
Differential Diagnosis The differential diagnosis of congenital neutropenia10 includes a number of congenital and acquired diseases. The most common difficulty is determining whether the patient has cyclic neutropenia, autoimmune neutropenia of infancy, or idiopathic neutropenia. Cyclic neutropenia is diagnosed by serial measurement of blood neutrophils on at least 3 days per week for at least 6 weeks, graphing the counts and determining if there are regular periods of very severe neutropenia at 3-week intervals. Even with careful serial counts, the diagnosis may be missed because the amplitude of the oscillations may be too small for detection. An important differential diagnostic evaluation is testing for neutrophilic antibodies.31 In neutropenic children aged 1 to 3 years, the presence of neutrophil-specific auto-antibod-
rHuG-CSF has been available for treatment of CN since 1987. Phase I–III studies have demonstrated its efficiency in increasing the number of neutrophils associated with reduction of infections.6,32 In contrast, GM-CSF treatment does not lead to an increase in blood neutrophils but only in blood eosinophils.33 In 1994, the SCNIR was established to collect data on clinical course and outcome of these rare disorders. As of December 2005, 611 patients with CN have been registered, and more than 95% responded to rHuG-CSF treatment with an increase in ANCs to 1.0 ⫻ ⱖ109/L (Fig 3). Most CN patients respond to a dose between 3 and 10 g/kg/d. After initiation of rHuG-CSF at 5 g/kg/d, the dose should be escalated to 10 g/kg/d and then by increments of 10 g/kg at 14-day intervals if the ANC remains below 1.0 ⫻ 109/L. As soon as the ANC can be maintained at 1.0 to 1.5 ⫻ 109/L or above, increases of the rHuG-CSF dose can be stopped since the occurrence of bacterial infection is reduced dramatically at this level. The dose of rHuG-CSF can be reduced if the ANC increases to ⱖ5.0 ⫻ 109/L to keep the patient at the lowest dose necessary for maintaining a sufficient neutrophil count to overcome infections. Nonresponders to rHuG-CSF are defined by failure to benefit at dose levels exceeding 120 g/kg/d. Partial responders have ANCs increased to 0.5 to 1.0 ⫻ 109/L but still have bacterial infections. In some patients, the dose of rHuG-CSF cannot be increased to these levels because of the large volume and frequency of injections required. In research studies, refractory patients received a combination of rHuG-CSF with stem cell factor (SCF), which led to a further increase in ANCs above the levels achieved with G-CSF alone. Because of allergic side effects of SCF, this treatment combination has only been used during severe infections in hospitalized patients receiving concomitant antihistaminic medication.34 For patients who do not respond to rHuG-CSF alone or in combination with SCF, HSCT is the only currently available treatment.11 When successful, after transplantation patients normalize hematopoiesis and do not require cytokine treatment. It remains difficult to recommend transplantation for patients with CN who benefit from rHuG-CSF and show no evidence of impending malignant transformation. For many responding patients, the risks associated with transplant from an HLA-identical sibling may outweigh the risk of leukemic transformation when rHuG-CSF is continued.
Severe congenital neutropenia
193
Figure 3 Mean ANC by year of G-CSF treatment in CN.
Long-Term Safety Prior to the development and availability of G-CSF, complications of severe congenital neutropenia were not well known. There was no organized database for this rare disease. However, a number of reports noting deaths from infections and a few cases evolving to AML are available from the pre–G-CSF era.
Leukemia Prior to the availability of cytokine therapy, it was already recognized that leukemic transformation occurred in patients with CN.35,36 However, in the pre-cytokine era, 42% of published cases died in the first 2 years of life, usually from sepsis or pneumonia. Thus, the true risk of CN patients developing MDS/AML was not defined. With G-CSF therapy, most of these patients have survived well beyond 2 years of age. Therefore, it is unknown whether the increased survival allows for a higher risk of the recognized natural expression of leukemogenesis in this population in the absence of G-CSF therapy. From the initiation of clinical trials with G-CSF in 1987 through December 2005, a total of 70 patients with CN (including five patients with Shwachman-Diamond syndrome) who have developed MDS/AML were reported to the SCNIR. The overall incidence or crude rate of MDS/AML conversion is 11.5% for CN patients (70 cases among 611 exposed cases), with an average follow-up of approximately 5 to 6 years. The leukemias occurred in those CN patients with ELA 2 mutations, in unclassified CN patients, and in patients with Shwachman-Diamond syndrome but not in the other subgroups listed in Table 1.
The SCNIR recently published a more detailed report on the first 374 patients (1987–2000) with CN on long-term G-CSF enrolled in the registry to identify the risk of leukemic transformation. The hazard of MDS/AML increased significantly during the period of observation on G-CSF, from 2.9% per year after 6 years to 8.0% per year after 12 years. The cumulative incidence of MDS/AML was 21% after 10 years. The risk of MDS/AML increased with the dose of G-CSF. Less responsive patients, defined as those requiring greater than 8 g/kg/d of G-CSF, had a cumulative incidence of MDS/AML of 40% after 10 years, compared to 11% of more responsive patients.37 The data were interpreted as indicating that a poor response to G-CSF defines an “at-risk” population and predicts an adverse outcome. Conversion to MDS/AML in CN patients was associated with one or more cellular genetic abnormalities (such as monosomy 7, ras mutation, or G-CSF receptor mutation), which may be useful for identifying subgroups of patients at high risk. Marrow cells from approximately 75% of the severe CN cases that transformed to MDS/AML also showed point mutations in the gene for the G-CSF receptor, resulting in a truncated C-terminal cytoplasmic region of the receptor that is crucial for maturation signaling.17,23-25,27-29 As illustrated by the cases described here, the development of MDS/AML is a multi-step process characterized by a series of cellular genetic changes indicating a genetic predisposition to malignant transformation. If and how G-CSF impacts on this predisposition remains unclear; there are no historical controls for comparison. To address further the issue of risk-benefit of G-CSF in the CN setting with regard to MDS/AML, all available data were
K. Welte, C. Zeidler, and D.C. Dale
194 critically reviewed.37 Although high-level evidence does not exist that would reveal the proper timing of bone marrow surveillance, the SCNIR advisory board recommended annual marrow cytogenetic testing to identify monosomy 7 or other changes indicating transformation; this approach might permit earlier therapeutic interventions such as BMT.
Osteoporosis Within the SCNIR, bone density measurements have been reported on a total of 121 patients measured by different techniques, including quantitative computer tomography (Q-CT), dual energy x-ray absorptiometry (DEXA), singlephoton absorptiometry (SPA), and lumbar x-ray. Of these 121 patients, 66 (54.4%) had varying degrees of abnormal results. These results have not been quantified to interpret the severity of the abnormalities. Most patients did not show clinical symptoms of osteopenia/osteoporosis (bone pain or fractures), which explains why diagnostic procedures for bone density evaluation were not reported in approximately 70% of registry patients. Therefore, the actual incidence continues to be unknown. The pathophysiology of osteopenia/ osteoporosis also remains unclear. Serum chemistry did not reveal a typical pattern in patients with osteopenia/osteoporosis. Patients who suffer from osteopenia also did not receive elevated amounts of rHuG-CSF doses compared with all registry patients.38
Splenomegaly The incidence of palpable splenomegaly (2 cm below the costal margin) prior to treatment with rHuG-CSF in CN patients was 21%. During the first year of rHuG-CSF therapy, the incidence increased to 39% and remained approximately at this level of occurrence (34% to 48%) through 10 years of therapy.
Monitoring The Advisory Board for the SCNIR recommends that all patients should be seen by a physician at least twice per year with assessment for weight and height and documentation of the occurrence of infections. Blood counts (white blood cells, hemoglobin, platelets, and differential blood counts) should be obtained at least every 3 months. Bone marrow examination (morphology plus cytogenetics) is required once per year to search for acquired cytogenetic abnormalities, such as monosomy 7 or trisomy 21. Furthermore, following informed consent, bone marrow samples should be collected for research studies on the molecular analysis of the pathophysiology of CN and risk factors for leukemic evolution. G-CSF receptor analysis can be performed on heparinized blood or bone marrow samples by laboratories associated with the SCNIR.
Conclusion In light of the current literature and longitudinal data from the SCNIR, we suggest that the use of rHuG-CSF remain
first-line treatment for most CN patients. HSCT from an HLA-identical sibling is beneficial for patients who are refractory to rHuG-CSF. For those patients in whom a G-CSF receptor mutation is identified, HSCT from an HLA-identical sibling is an option. Patients who develop monosomy 7, other significant chromosomal abnormalities, or MDS/leukemia should proceed urgently to HSCT. Data on alternative sources of donor stem cells are insufficient to assess outcome in patients with CN. Other than those patients who fail to respond to rHuG-CSF, the cytokine should be employed to maintain an ANC ranging from 1.0 to 5.0 ⫻ 109/L with amelioration of symptoms. All CN patients, regardless of their treatment or response, are at risk of developing MDS or leukemia at an actual incidence of 11.5% today and a cumulative incidence of 21% after 10 years. With increasing genetic analyses, it may become possible to correlate the leukemic risk with the underlying genetic defect. Careful monitoring for cytogenetic abnormalities and G-CSF receptor mutation is necessary to initiate HSCT as soon as any of these occur. Despite the significant risk of leukemia, HSCT-related morbidity is also significant and, therefore, in patients without signs of leukemia or MDS, HSCT should be restricted to G-CSF nonresponders.
Acknowledgment The authors thank all colleagues associated with the Data Collection Centers of the Severe Chronic Neutropenia International Registry at the University of Washington, Seattle, WA (Audrey Anna Bolyard and Tammy Cottle), and the Medizinische Hochschule, Hannover, Germany (Beate Schwinzer and Gusal Pracht) for their continued assistance. We are also grateful to the many physicians worldwide who faithfully and generously submitted data on their patients. The authors gratefully acknowledge the important contribution of the Advisory Board and the European Liason Physician Group of the Severe Chronic Neutropenia International Registry and support from the National Institutes of Health (USA), the European Union, and Amgen Inc, Thousand Oaks, CA.
References 1. Kostmann R: Infantile genetic agranulocytosis. Acta Pediatr Scand 45: 1-78, 1956 2. Kostmann R: Infantile genetic agranulocytosis: A review with presentation of ten new cases. Acta Pediatr Scand 64:362-368, 1975 3. Rappeport J, Parkman R, Newburger P, Camitta BM, Chusid MJ: Correction of infantile granulocytosis (Kostmann syndrome) by allogeneic bone marrow transplantation. Am J Med 68:605-609, 1980 4. Souza L, Boone T, Gabrilove J, Lai PH, Zsebo KM, Murdock DC, et al: Recombinant human granulocyte colony-stimulating factor: Effects on normal and leukemic myeloid cells. Science 232:61-65, 1986 5. Bonilla M, Gillio A, Ruggeiro M, Kernan NA, Brochstein JA, Abboud M, et al: Effects of recombinant human granulocyte colony-stimulating factor on neutropenia in patients with congenital agranulocytosis. N Engl J Med 320:1574-1580, 1989 6. Bonilla M, Dale D, Zeidler C, Last L, Reiter A, Ruggeiro M, et al: Longterm safety of treatment with recombinant human granulocyte colonystimulating factor (r-metHuG-CSF) in patients with severe congenital neutropenias. Br J Hematol 88:723-730, 1994 7. Freedman MH: Safety of long-term administration of granulocyte col-
Severe congenital neutropenia
8. 9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
ony-stimulating factor for severe chronic neutropenia. Curr Opin Hematol 4:217-224, 1997 Welte K, Dale D: Pathophysiology and treatment of severe chronic neutropenia. Ann Hematol 72:158-165, 1996 Welte K, Boxer L: Severe chronic neutropenia: Pathophysiology and therapy. Semin Hematol 34:267-278, 1997 Zeidler C, Boxer L, Dale DC, Freedman MH, Kinsey S, Welte K: Management of Kostmann syndrome in the G-CSF era. Br J Haematol 109: 490-495, 2000 Zeidler C, Welte K, Barak Y, Barriga F, Bolyard AA, Boxer L, et al: Stem cell transplantation in patients with severe congenital neutropenia without evidence of leukemic transformation. Blood 95:1195-1198, 2000 Mempel K, Pietsch,T, Menzel, Zeidler C, Welte K: Increased serum levels of granulocyte colony-stimulating factor in patients with severe congenital neutropenia. Blood 77:1919-1922, 1991 Kyas U, Pietsch T, Welte K: Expression of receptors for granulocytecolony stimulating factor on neutrtophilsfrom patients with severe congenital neutropenia and cyclic neutropenia. Blood 79:1144-1147, 1992 Horwitz, M, Benson KF, Person RE, Aprikyan AG, Dale DC: Mutations in ELA2, encoding neutrophil elastase, define a 21-day biological clock in cyclic haematopoesis. Nat Genet 23:433-436, 1999 Ancliff PJ, Gale RE, Watts MJ, Liesner R, Hann IM, Strobel S, Linch DC: Paternal mosaicism proves the pathogenic nature of mutations in neutrophil elastase in severe congenital neutropenia. Blood 100:707-709, 2002 Dale DC, Liles WC, Grawicz D, Aprikyan AG: Clinical implications of mutations of neutrophil elastase in congenital and cyclic neutropenia. J Pediatr Hematol Oncol 23:208-210, 2001 Donadieu J, Leblanc T, Meunier B, Barkaoui M, Fenneteau O, Bertrand Y, et al: Analysis of risk factors for myelodysplasia/leukemia and infectious death among patients with congenital neutropenia: Experience of the French Severe Chronic Neutropenia Study Group. Haematologica 90:45-53, 2005 Dale DC, Person RE, Bolyard AA, Aprikyan AG, Bos C, Bonilla MA, et al: Mutations in the gene encoding neutrophil elastase in congenital and cyclic neutropenia. Blood 96:2317-2322, 2000 Massullo P, Druhan LJ, Bunnell BA, Hunter MG, Robinson JM, Marsh CB, et al: Aberrant subcellular targeting of the G185R neutrophil elastase mutant associated with severe congenital neutropenia induces premature apoptosis of differentiating promyelocytes. Blood 105:33973404, 2005 Germeshausen M, Schulze H, Ballmaier M, Zeidler C, Welte K: Mutations in the gene encoding neutrophil elastase (ELA2) are not sufficient to cause the phenotype of congenital neutropenia. Br J Haematol 115: 1-4, 2001 Koellner I, Sodeik B, Schreek S, Heyn H, von Neuhoff N, Germeshausen M, et al.: Mutations in neutrophil elastase causino congenital neutropenia lead to cytoplasmic protein accumulation and induction of the unfolded protein response. Blood March 21, 2006 (Epub ahead of print) Skokowa J, Cario G, Bucan V, Dan L, Wang Z, Scherr M, et al: LEF-1 transcription factor regulates proliferation and differentiation of myeloid progenitors in healthy individuals and in patients with severe congenital neutropenia (CN). Blood 106:390a, 2005 (abstr) Bernard T, Gale R, Evans J, Linch DC: Mutations of the granulocytecolony stimulating factor receptor in patients with severe congenital
195
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35. 36. 37.
38.
neutropenia are not required for transformation to acute myeloid leukaemia and may be a bystander phenomenon. Br J Haematol 101:141-149, 1998 Cassinat B, Bellanné-Chantelot C, Notz-Carrère A, Menot ML, Vaury C, Micheau M, et al: Screening for G-CSF receptor mutations in patients with secondary myeloid or lymphoid transformation of severe congenital neutropenia. A report from the French Neutropenia Register. Leukemia 18:1553-1555, 2004 Dong F, Russel KB, Tidow N, Freedman M, Fasth A, Neijens HJ, et al: Mutations in the gene for the granulocyte-colony stimulating factor receptor in patients with acute myeloid leukemia preceded by severe congenital neutropenia. N Engl J Med 333:487-493, 1995 Germeshausen M, Schulze H, Kratz C, Wilkens L, Repp R, Shannon K, et al: An acquired G-CSF receptor mutation results in increased proliferation of CMML cells from a patient with severe congenital neutropenia. Leukemia 19:611-617, 2005 Papadaki HA, Kosteas T, Gemetzi C, Damianaki A, Anagnou NP, Eliopoulos GD: Acute myeloid/NK precursor cell leukemia with trisomy 4 and a novel point mutation in the extracellular domain of the G-CSF receptor in a patient with chronic idiopathic neutropenia. Ann Hematol 83:345-348, 2004 Tidow N, Pilz C, Teichmann B, Muller-Brechlin A, Germeshausen M, Kasper B, et al: Clinical relevance of point mutations in the cytoplasmatic domain of the granulocyte-colony stimulating factor gene in patients with severe congenital neutropenia. Blood 88:2369-2375, 1997 Tschan CA, Pilz C, Zeidler C, Welte K, Germeshausen M: Time course of increasing numbers of mutations in the granulocyte colony-stimulating factor receptor gene in a patient with congenital neutropenia who developed leukemia. Blood 97:1882-1884, 2001 Jeha S, Chan KW, Aprikyan AG, Hoots WK, Culbert S, Zietz H, et al: Spontaneous remission of granulocyte colony-stimulating factor–associated leukemia in a child with severe congenital neutropenia. Blood 96:3647-3649, 2000 Bux J, Behrens G, Jaeger G, Welte K: Diagnosis and clinical course of autoimmune neutropenia in infancy: Analysis of 240 cases. Blood 91: 181-186, 1998 Dale D, Bonilla M, Davis M, Nakanishi AM, Hammond WP, Kurtzberg J, et al: A randomized controlled phase III trial of recombinant human granulocyte colony-stimulating factor (Filgrastim) for treatment of severe chronic neutropenia. Blood 81:2496-2502, 1993 Welte K, Zeidler C, Reiter A, Muller W, Odenwald E, Souza L, Riehm H: Differential effects of granulocyte-macrophage colony-stimulating factor and granulocyte colony-stimulating factor in children with severe congenital neutropenia. Blood 75:1056-1063, 1990 Zeidler C, Vogel R, Wyres M, et al: Beneficial effects of stem cell factor (SCF) in children with severe congenital neutropenias refractory to G-CSF. Blood 92:380a, 1998 (abstr) Gilman P, Jackson D, Guild H: Congenital agranulocytosis: Prolonged survival and terminal acute leukemia. Blood 36:576-585, 1971 Rosen R, Kang S: Congenital agranulocytosis terminating in acute myelomonocytic leukemia. J Pediatr 94:406-408, 1979 Rosenberg PS, Alter BP, Bolyard AA , Bonilla MA, Boxer LA, Cham B, et al: The incidence of leukemia and mortality from sepsis in patients with severe congenital neutropenia receiving long-term G-CSF therapy. Blood, February 23, 2006 (Epub ahead of print) Yakisan E, Schirg E, Zeidler C, Bischop NJ, Reiter A, et al: High incidence of significant bone loss in patients with severe congenital neutropenia (Kostmann’s syndrome). J Pediatr 131:592-597, 1997