Jpn. J. Clin. Immunol., 35 (6) 533～538 (2012)2012 The Japan Society for Clinical Immunology 533533

Case Report with organ involvement in 3 siblings

Mineto OTA1, Kenchi TAKENAKA1,MafuyuTAKAHASHI2 and Kenji NAGASAKA1

1Department of Rheumatology, Ome Municipal General Hospital 2Department of Neurology, Ome Municipal General Hospital

(Received June 11, 2012)

summary

We describe 3 siblings who suŠered from marked eosinophilia with organ involvement. One sibling, who ex- perienced cervical lymphadenopathy and peripheral neuropathy with eosinophilia (5,834 cells/mL) following bronchial , was diagnosed with Churg-Strauss syndrome (CSS) according to the criteria of the American College of Rheumatology. Another sibling, who suŠered from severe asthma with persistent polyarthritis and eosinophilia (2,496 cells/mL), was also diagnosed with CSS according to the criteria of the Japanese Ministry of Health, Labour and Wel- fare. The remaining sibling, who had eosinophilic pleuritis with peripheral blood eosinophilia (699 cells/mL), did not fulˆll the widely used criteria for CSS or hypereosinophilic syndrome (HES) ; however, he ˆt the newly proposed criteria for HES. Glucocorticoid treatment relieved their symptoms. Although the diagnoses and the criteria used for diagnosis diŠered between the siblings, all 3 patients showed common features such as eosinophilia with organ involve- ment that required treatment, indicating the possibility of familial eosinophilia (FE).Furthermore,theclinicalfea- tures observed diŠered substantially from those of previously reported FE patients, therefore, these 3 siblings may be aŠected by a type of FE distinguishable from those previously described.

Key words―Churg-Strauss syndrome; hypereosinophilic syndrome; familial eosinophilia

and Welfare9). The other sibling, who had pleuritis Introduction accompanied by inˆltration, did not fulˆll Eosinophilia may be caused by several factors, in- the advocated criteria for CSS8～10) or HES11,12). All 3 cluding allergic reactions, parasitic infections, and siblings required treatment with glucocorticoids malignancy. Churg-Strauss syndrome (CSS) and (GC). hypereosinophilic syndrome (HES), which are rare Although their diagnoses and the criteria used for disorders with an unknown etiology, are known to be diagnosis diŠered between the siblings, these 3 accompanied by peripheral blood eosinophilia. patients showed common features such as eo- However, they are clearly distinguished from allergic sinophilia with organ involvement requiring treat- reactions because they have a distinctive feature, ment. Because CSS and HES are both uncommon namely, organ damage1,2). disorders, we believe that these ˆndings cannot be Many reports have described familial eosinophilia merely coincidental. Furthermore, the clinical fea- (FE) since the early 1900 s3,4). In these families, the tures observed diŠered substantially from those of distribution of eosinophilia was suggested to involve previously reported FE patients, therefore, these 3 si- autosomal dominant inheritance. Recently, another blings may be aŠected by a type of FE distinguisha- autosomal dominant type of FE has also been ble from those previously described. described5～7). The disorder was characterized not Case report only by marked eosinophilia but also by progression to end-organ damage in some aŠected family mem- Case 1 : The sixth sibling (Fig. 1) bers, similar to HES. Furthermore, some of these A 42-year-old female housekeeper acquired bron- family members exhibited obstructive pulmonary dis- chial asthma in 2001. She was a non-smoker. In easessuchasasthma. 2002, she noticed enlargement of the cervical lymph We encountered 3 siblings who suŠered from nodes and pruritus on her body. She then ex- marked eosinophilia with organ involvement. One si- perienced paralysis of both her hands. Later, she was bling fulˆlled the criteria of the American College of conˆnedtoherbedandwasthenadmittedtoour Rheumatology (ACR) for CSS8), whereas another hospital. Physical examination showed severe distal did not fulˆll the ACR criteria for CSS but did meet upper limb weakness, paresthesia in the right upper those of the Japanese Ministry of Health, Labour and lower limbs, and eruptions on her trunk and ex- 534 日本臨床免疫学会会誌 (Vol. 35 No. 6)

Figure 1. Family pedigree Of 6 siblings, 3 had eosinophilia with organ involvement (Cases 13). The 3 siblings without eosinophilia did not have a history of asthma. Only the oldest sister had a skin disorder. As for their parents, children, grandparents, and parents' siblings, 5 had dermatologic abnormalities and 2 had a history of asthma, but none had medical conditions similar to those seen in Cases 13.

tremities. Blood examination showed leukocytosis ‰uorescent in situ hybridization did not show a 4q12 (13,260 cells/mL, 44[5,834 cells/mL] of which deletion, indicating that the FIP1L1-platelet-derived were ). The C-reactive protein (CRP) lev- growth factor receptor fusion gene was absent. Her el was slightly elevated (1.49 mg/dL).SerumIgElev- IgE level returned to normal (98 IU/mL);however,a el was elevated (625 IU/mL);however,IgEspeciˆc positive level (0.90 UA/mL) of a speciˆc IgE against to multiple inhaled or ingested allergens was not de- moths was detected, without any related symptoms. tected by the CAP-RAST method. Myeloperoxidase Case 2 : The fourth sibling (Fig. 1) and proteinase-3 antineutrophil cytoplasmic antibo- A 43-year-old male car manufacturer with a history dies (MPO-ANCA and PR3-ANCA, respectively) of severe bronchial asthma since 1995 was admitted to were negative. A nerve conduction study showed a hospital due to worsening asthma, a persistent high mononeuropathy multiplex pattern. Neuromuscular fever (＞38°C) for 2 weeks, and outbreak of rash on and skin biopsy showed severe microvasculitis around the extremities and back in 2004. He had smoked 1 the peripheral nerves with eosinophil inˆltration. The pack of cigarettes daily for 22 years. Physical exami- bone marrow aspirate showed a slightly elevated eo- nation showed conjunctional injection, swelling with sinophil count (10); however, no atypical cells tenderness in the proximal interphalangeal and were seen. The cause of eosinophilia could not be metacarpophalangeal joints, and papulae on the ex- ascertained even after considering medication as a tremities and back (Fig. 2). Blood examination potential cause. She was diagnosed with CSS in accor- showed leukocytosis (11,400 cells/mL, 21.9[2,496 dance with the ACR criteria8). After treatment with cells/mL] of which were eosinophils). The CRP level 40 mg/day prednisolone, her peripheral blood eo- was elevated to 13.60 mg/dL. Serum IgE level was sinophil count decreased to 95 cells/mL, and she was normal (103 IU/mL). MPO-ANCA and PR3-ANCA able to walk again. Currently, she is being treated were negative. Stool assay did not show the presence with 5 mg/day prednisolone, and her condition of parasites. Abdominal and pelvic computed remains stable. Peripheral blood screening via tomography (CT) scan showed no enlarged lymph OTA・Eosinophilia with Organ Involvement in 3 Siblings 535

Figure 2. Skin ˆndings in Case 2. Papulae were seen on the extremities and back.

Figure 3. Chest computed tomography ˆndings of patient 3. nodes or solid tumor. Upper gastrointestinal en- Chest CT showed right-sided pleural eŠusion. doscopy showed no apparent abnormalities. Bone marrow aspirate showed normocellular bone marrow, with an elevated eosinophil count (20.5) without ed a right pleural eŠusion (Fig. 3), following which atypical cells. No apparent etiology for eosinophilia, he was admitted to our hospital. Physical examina- including allergic reactions, parasitic infections, or tion showed prurigo on his trunk and extremities. malignancy including malignant lymphoma, could be Blood examination showed a normal leukocyte count determined. Although his symptoms did not fulˆll the of 6,030 cells/mL; however, the eosinophil count was ACR8) or Lanham's10) CSS criteria, he was diagnosed slightly elevated (699 cells/mL). The CRP level was withCSSinaccordancewiththecriteriaofthe slightly elevated (1.42 mg/dL). Serum IgE level was Japanese Ministry of Health, Labour and Welfare9). also elevated (986 IU/mL). MPO-ANCA was nega- He was treated with 30 mg/day prednisolone, which tive; however, the PR3-ANCA level was slightly at once relieved his asthma, high fever, joint swelling, elevated (4.7 U/mL). The pleural eŠusion contained and rash. Two months later, he suŠered from worsen- a number of leukocytes (3,567 cells/mL) and 60 of ing of asthma with severe breathing di‹culty; which were eosinophils. No atypical cells were found. however, additional high-dose GC therapy relieved Bone marrow aspirate showed normocellular bone the dyspnea. He is currently being treated with 10 marrow, with a slight elevation in eosinophils (7.1) mg/day prednisolone and is experiencing no dyspnea and no atypical cells. Peripheral blood screening us- or rash. Peripheral blood screening via ‰uorescent in ing the ‰uorescent in-situ hybridization method did situ hybridization did not show a 4q12 deletion. Se- not show 4q12 deletion. No obvious cause of eo- rum IgE level was still normal (39 IU/mL) and no sinophilia, other than occupation-related factors was speciˆc IgE was detected. noted. These features did not fulˆll the World Health Case 3 : The third sibling (Fig. 1) Organization classiˆcation for HES11,12) ;however, A 61-year-old male carpenter who did not have a they did fulˆll the recently proposed diagnostic criter- history of asthma experienced repetitive prurigo all ia for HES by Simon et al.13).Hissymptomswere over his body and was being treated with intermittent thought to be associated with eosinophilia because low doses of GC since 2006. He had smoked 1 pack of abundant eosinophils were seen in the pleural eŠusion cigarettes daily for 40 years. In 2005, before the ap- and his eosinophil count was increasing even though pearance of the rash, his eosinophil count and serum he had stopped working. Therefore, we thought his IgE level were normal (288 cells/mL and 136 IU/mL, eosinophilia with pleuritis might almost not be caused respectively). In 2009, during follow-up for prurigo, by occupation-related factors, and we decided to treat his eosinophil count was normal (341 cells/mL), the patient with GC to prevent his condition from de- however his serum IgE level was elevated (445 IU/ teriorating. The presence of eosinophils in the mL) and speciˆc IgE antibodies were detected: posi- peripheral blood and pleural eŠusion were improved tive against bee venom (3.03 UA/mL) and borderline immediately after initiating treatment with 35 mg/ against cedar pollen (0.41 UA/mL), cypress pollen day prednisolone. Dyspnea and prurigo also im- (0.39 UA/mL),rice(0.47 UA/mL),andwheat proved in the following weeks; GC was tapered and (0.42 UA/mL). In 2010, he noticed exacerbation of then stopped in 2011. Four months after cessation of therashanddyspneaonexertion,andCTscandetect- GC, the serum IgE level returned to normal (157 IU/ 536 日本臨床免疫学会会誌 (Vol. 35 No. 6) mL) ; however, the level of IgE speciˆc to bee venom tion. However, his eosinophil count was normal until was borderline (0.63 UA/mL). pleural eŠusion occurred, and his symptoms deterio- Additional family studies (Fig. 1) rated despite his avoidance of antigens. Thus, his eo- These 3 patients had 3 other living siblings. Both sinophilia leading to organ involvement may have parents had already died. Theirmotherhadahistory been caused by factors unrelated to his occupation. of asthma ; however, she had not had any skin dis- Moreover, serum IgE levels and speciˆc IgE results eases or neuropathy. Among their siblings, only the diŠered in each case, suggesting that eosinophilia with oldest sister had a skin disorder. organ involvement in our 3 cases may have been caused by nonallergic factors independent of IgE- Discussion mediated mechanism. We therefore suggest that the It is quite rare that more than 1 case of CSS or HES eosinophilia in our patients may not be a secondary were seen in the same family. The annual incidence of cause. CSSandHESisreportedly1.21.8 and 0.35 per 1 Familial occurrence of persistent eosinophilia is million people, respectively14～16). However, the inci- called FE. FE was ˆrst documented in 19093).In dence of CSS and HES is very high in the present fa- 1964, Naiman4) reported a 3-generation family with 7 mily, because 2 cases of CSS and 1 case of HES are aŠected patients, as well as 17 published families seen among 6 siblings. Therefore, this family may overall. In this report of FE, the term ``hereditary eo- represent a unique pedigree. sinophilia'' was used, which was deˆned as the It is true that CSS and HES are rare disorders ; presence of signiˆcant eosinophilia (＞400 cells/mL), however, secondary eosinophilia is not. Thus, it is im- familial incidence, and the absence of other recog- portant to exclude secondary causes of eosinophilia. nized causal factors. Although there are no estab- We carefully examined potential underlying causes of lished criteria for FE, we may be able to regard our 3 eosinophilia in our 3 patients, including travel histo- patients as having FE according to Naiman's deˆni- ry, medication, and parasitic infection. However, ex- tion. cepting case 3, no apparent cause of eosinophilia was Recently, Lin5), Rioux6) and Klion7) investigated 5 found. Moreover, as the occupation of the 3 siblings generations of a same family with marked eo- diŠered and they had lived separately for decades un- sinophilia. They described that FE was an autosomal til the disease occurred, it is unlikely that they were dominant disorder and that some of the aŠected exposed to common environmental factors that in- members showed organ involvement similar to that duced the eosinophilia. Table 1 summarizes IgE levels seen in HES. In their report, the authors compared and speciˆc IgE results for 3 patients. It is di‹cult to family members with eosinophilia to those without deny the allergic diathesis with subclinical level in case eosinophilia. They reported that no clinical abnor- 1. Regarding the data and clinical history in case 3, it mality was more frequent in members with eo- may be that the patient is susceptible to and sinophilia, although asthma was signiˆcantly more was exposed to an antigen associated with his occupa- common in those family members without eo-

Table 1. Serum IgE levels and result of speciˆc IgE in 3 cases

Case No. Before Onset Before Treatment After Treatment Case 1 IgE (IU/ml) NA 625b 98 Speciˆc IgEa NA Negativeb Positive against moth Case 2 IgE (IU/ml) NA 103b 39 Speciˆc IgEa NA NA Negative Case 3 IgE (IU/ml) 136 986b 157 Speciˆc IgEa NA Positive against bee venom and borderline against Borderline against bee venom cedar pollen, cypress pollen, rice, and wheatc

Abbreviations: IgE, immnoglobulin E; NA, not available. a Speciˆc IgE against multiple inhaled or ingested allergen was tested by CAP-RAST method. b Just before treatment c One-year before treatment OTA・Eosinophilia with Organ Involvement in 3 Siblings 537537 sinophilia, and immunologic analysis in aŠected sinophilia and organ involvement. We believe that the members showed a relative lack of eosinophil activa- occurrence of these 3 eosinophilia cases in a single fa- tion compared to that seen in non-familial HES. mily is not a coincidence and that these patients may Therefore, they concluded that their FE could be dis- be exhibiting a type of FE diŠerent from those previ- tinguished from non-familial HES7).Thisstudyalso ously reported. Further studies are needed to clarify showed that the eosinophil count in relatives without these ˆndings. eosinophilia did not increase at any time-point ; REFERENCES hence, long-term follow-up was not necessary for un- aŠected family members5). Our patients, however, 1) Gleich GJ, Leiferman KM. : The hyper- appear to diŠer from those previously reported, as 2 eosinophilic syndromes : current concepts and of the 3 aŠected siblings (Cases 1 and 2) had asthma, treatments. Br J Haematol 145 : 271285, 2009. in addition to the eosinophilia and associated sym- 2) Pagnoux C. : Churg-Strauss syndrome : evolv- ptoms. Furthermore, patient 3 had a normal eo- ing concepts. Discov Med 9 : 243252, 2010. sinophil count in 2005, when he exhibited no sym- 3) Gaugain M. : Un cas d'eosinophilie familiare. ptoms; however, his eosinophil count increased in Sem Med 29 : 329, 1909. 2010. These ˆndings indicate that our patients diŠer 4) Naiman JL., et al. : Hereditary Eosinophilia : from FE described by Lin5) and Klion7). Report of a Family and Review of the Litera- We may be able to regard our patients as exhibiting ture. Am J Hum Genet 16 : 195203, 1964. another type of FE ; however, evidence supporting 5) Lin AY, et al. : Familial eosinophilia : clinical our hypothesis is insu‹cient. First, our investigation and laboratory results on a U. S. kindred. Am J of family history included only 4 generations of the Med Genet 76 : 229237, 1998. family, which contained insu‹cient data because of 6) Rioux JD, et al. : Familial Eosinophilia maps to an unsatisfactory health care system in the past. the gene cluster on human chro- Moreover, a blood test for detecting eosinophilia was mosomal region 5q31q33. Am J Hum Genet performed in only these 3 patients. A detailed family 63 : 10861094, 1998. history of the next few generations and blood tests for 7) Klion AD, et al. : Familial eosinophilia : a be- detecting eosinophilia are needed. It may take several nign disorder? Blood 103 : 40504055, 2004. decades to investigate the occurrence of eosinophilia- 8) Masi AT, et al. : The American College of related symptoms in the next generation. Second, Rheumatology 1990 criteria for the classiˆca- genetic assessment, including for human leukocyte tion of Churg-Strauss syndrome (allergic antigen (HLA),wasnotperformed.Rioux6) under- granulomatosis and angiitis). Arthritis Rheum took a genome-wide linkage analysis to determine the 33 : 10941100, 1990. molecular basis of FE. The results showed that a sin- 9) JCS Joint Working Group. : Guideline for gle locus of interest is found on chromosome 5q31 management of vasculitis syndrome (JCS q33, which contains the genes for interleukin (IL)3, 2008). Japanese Circulation Society. Circ J 75 : IL5, and granulocyte-macrophage colony stimulat- 474503, 2011. ing factor. However, no functional polymorphism 10) Lanham JG, et al. : Systemic vasculitis with was found. Tsurukisawa17) also reported familial CSS asthma and eosinophilia : a clinical approach to in 2 sisters and performed HLA typing of family the Churg-Strauss syndrome. Medicine (Balti- members; however, no diŠerence was found in the more) 63 :6581, 1984. frequency of diŠerent HLAs. These ˆndings indicate 11) JaŠe ES, et al. : World Health Organization that the genetic basis of FE may be complicated. Classiˆcation of Tumors : Pathology and Third, the age of onset was relatively high in our Genetics, Tumor of Hematopoietic and Lym- patients. In the reports of Lin5) and Klion7),eo- phoid Tissues. IARC Press, Lyon, pp2931, sinophilia was documented as early as 4 months of 2001. age. However, many family members developed or- 12) Chusid MJ, et al. : The hypereosinophilic syn- gan involvement at the age of 50 years and older. drome : analysis of fourteen cases with review Therefore, secondary events may be responsible for of the literature. Medicine (Baltimore) 54 :1 disease progression in addition to the eosinophilia it- 27, 1975. self, as they discussed7). 13) Simon HU, et al. : Reˆning the deˆnition of In conclusion, we describe 3 siblings with eo- hypereosinophilic syndrome. J Clin Im- 538538 日本臨床免疫学会会誌 (Vol. 35 No. 6)

munol 126 :4549, 2010. 1999. 14) Vinit J, et al. : Churg-Strauss syndrome : 16) Crane MM, et al. : Incidence of myeloprolifera- retrospective study in Burgundian population in tive hypereosinophilic syndrome in the United France in past 10 years. Rheumatol Int 31 : States and an estimate of all hypereosinophilic 587593, 2011. syndrome incidence. J Allergy Clin Immunol 15) Martin RM, et al. : Prevalence of Churg- 126 : 179181, 2010. Strauss syndrome, vasculitis, eosinophilia and 17) Tsurikisawa N, et al. : Familial Churg-Strauss associated conditions : retrospective analysis of syndrome in two sisters. Chest 131 : 592594, 58 prescription-event monitoring cohort stu- 2007. dies. Pharmacoepidemiol Drug Saf 8 : 179189,