Journal Logo

Original Study

Inflammatory Myopathies With Anti-Ku Antibodies

A Prognosis Dependent on Associated Lung Disease

Rigolet, Aude MD; Musset, Lucile MD, PhD; Dubourg, Odile MD, PhD; Maisonobe, Thierry MD; Grenier, Philippe MD, PhD; Charuel, Jean-Luc DPharm; Behin, Anthony MD; Herson, Serge MD; Amoura, Zahir MD, PhD; Benveniste, Olivier MD, PhD

Author Information
doi: 10.1097/MD.0b013e31824d9cec
  • Free

Abstract

INTRODUCTION

Anti-Ku antibodies are directed against a heterodimer complex of 2 subunits of 70 and 80 KDa that binds to free DNA termini. Ku complex plays a key role in DNA repair, and is implicated in the regulation of many cellular processes such as immunoglobulin gene recombination, telomere protection, and regulation of gene transcription.11,20,24 To our knowledge, anti-Ku antibodies were first detected in 1981 in 9 Japanese patients, 6 of whom had a polymyositis-scleroderma overlap syndrome.13 They have been subsequently reported in a wide spectrum of autoimmune diseases, especially systemic lupus erythematosus (SLE), systemic sclerosis (SSc), and mixed connective tissue disease.3–6,9,16–18,25–27 Although no specific clinical syndrome has been related to anti-Ku antibodies, descriptive series of anti-Ku-positive patients report recurrent clinical features such as Raynaud phenomenon, arthralgia, and musculoskeletal manifestations. Inflammatory myopathies (IM) with anti-Ku antibodies are described, mostly in the setting of an overlap syndrome in association with different connective autoimmune diseases.3–5,27 We undertook the current study to describe the clinical, laboratory, and histologic features of all anti-Ku-positive patients detected in our hospital during the last 10 years, as well as their treatment and outcomes. We gave particular attention to the description of associated IM, since to our knowledge the histologic features and outcomes have not been reported so far.

PATIENTS AND METHODS

Patient Selection

We retrospectively identified all patients with anti-Ku antibody positivity diagnosed at the Pitié-Salpêtrière Hospital laboratory over a 10-year period (2000–2010). Conditions were diagnosed according to established classification criteria: ACR revised criteria for SLE21 and rheumatoid arthritis,1 Leroy criteria for SSc,12 revised European criteria for primary Sjögren syndrome (SS),23 Mosca classification for undifferentiated connective tissue disease,14 modified Bohan and Peter criteria for IM,2 and modified Griggs for inclusion body myositis (IBM).8,15

Methods

Antinuclear Antibodies (ANA)

Immunofluorescence was performed on Hep-2 cells (Ref.SA2014-Ro, Immunoconcepts, Sacramento, CA); sera were tested at 1/80 screening dilution in PBS buffer, using a FITC-coupled antibody against human IgG. Positive samples were further serially diluted to evaluate their titer. Titers of 1/160 or more were considered positive.

On these cells, the fluorescence pattern suggestive for anti-Ku is composed of a very dense speckled nuclear staining associated with a homogeneous fluorescence of nucleoli varying in intensity among the nucleoli (Figure 1). The cytoplasm of mitotic cells is stained with a diffuse and sharp-edged fluorescence.

F1-4
FIGURE 1:
Immunofluorescence pattern on Hep-2 cells suggestive of anti-Ku: dense speckled nuclear staining associated with homogeneous fluorescence of nucleoli. [This figure can be viewed in color online at http://www.md-journal.com.]

Anti-Ku

Anti-Ku specificity was soaked by dot-blot immunoassay using BlueDOT polymyositis/scleroderma dot (PMS8 D-24, D-TekSA, Mons, Belgium), intended for the detection in human sera of IgG antibodies against Jo1, PL-7, PL-12, Mi-2, Ku, Pm-Scl, Scl-70, and SRP-54 antigens.

Ku target antigen is a subunit heterodimer (70-80 kDa) associated with DNA-dependant protein kinase (350 kDa); on this dot, antigen coated is a human recombinant product obtained with baculovirus expression system. The secondary antibody is an alkaline phosphatase-conjugated goat anti-human IgG, and the substrate contains NBT/BCIP. All sera found anti-Ku positive by any type of assay were purchased again with this immuno-dot assay.

Other Anti-Extractable Nuclear Antigen

Successive techniques employed for anti-extractable nuclear antigen (ENA) antibodies between 2000 and 2010 were Ouchterlony immunodiffusion, ELISA, immuno-dot, and Luminex.

Statistical Analyses

The Fisher exact test was used to perform comparisons between 2 groups. Statistical significance was accepted at p < 0.05.

RESULTS

Over 10 years (2000–2010), 20,600 ANA-positive sera were retrospectively analyzed. Anti-Ku antibodies were found in 95 (0.46%) sera collected from 34 patients, and complete clinical data were obtained for 30 patients (Tables 1–3). Most of these patients (86.7%) were female; 27 (90%) had a Caucasian and 3 had an African origin. Mean age was 49 years (range, 20–73 yr). Four (14.3%) patients had been treated with statins before clinical onset. The median follow-up duration was 6 years (range, 0.2–57 yr).

T1-4
TABLE 3:
Treatment and Outcome of 30 Anti-Ku-Positive Patients

Twenty-eight of the 30 (93%) patients were affected by autoimmune diseases, mainly in the form of SLE (n = 7) and SSc (n = 7) (see Table 1). Only 2 anti-Ku-positive patients had no autoimmune manifestations: 1 had bronchial adenocarcinoma and the other had nephroangiosclerosis with chronic renal failure.

T2-4
TABLE 1:
Clinical Features of 30 Anti-Ku-Positive Patients

Arthralgia was the most frequent clinical manifestation, reported in 77% of cases, followed by Raynaud phenomenon in 53% (see Table 1). Two patients (Patients 6 and 9) developed skin changes of the hands with the typical aspect of mechanic hands, as described in antisynthetase syndrome.

Inflammatory myositis was diagnosed in 11 (37%) anti-Ku patients; in 8 of the 11 patients it was part of an overlap syndrome defined as IM associated with connective autoimmune disease (SSc = 5, SS = 2, and SLE = 1). Two patients had definite IBM based on pathologic criteria.8,15 Ten of the 11 patients with myositis had myalgia, and 9 (82%) had muscle weakness. This weakness was proximal and symmetric for all, axial in 5 of them, and associated with proximal muscle atrophy in 3 cases. No patient had distal muscle weakness, including the 2 patients with IBM. Dysphagia was reported in 4 patients. Creatine kinase level was elevated in all cases, with a median value of 2210 U/L (range, 194–4073 U/L). Myogenic traces were detected in 8 of 9 cases by electromyography. All but 1 patient with muscle manifestations underwent muscle biopsy, where inflammation and necrosis were observed for all except 1 (Table 4). The distribution of inflammatory infiltrates predominated in both endomysial and perivascular (56%) areas. Six of 7 muscle specimens tested for HLA class I immunostaining were positive, with diffuse expression in 4 cases. Membrane attack complex (C5b9) deposits were found in a few muscle capillaries in 1 biopsy of the 5 tested.

T3-4
TABLE 4:
Histologic Muscle Features of Anti-Ku Positive Patients With Inflammatory Myopathy

Interstitial lung disease (ILD) was detected in 11 (37%) patients (see Table 2). The radiologic pattern on chest computed tomography (CT) was nonspecific interstitial pneumonia (NSIP) in 7 patients (associated with fibrosis in 2 patients), usual interstitial pneumonia (UIP) in 3 patients, and a sarcoidosis-like pattern in 1 patient. CT scans of NSIP, NSIP with fibrosis, and UIP patterns are shown in Figure 2. All patients with ILD underwent pulmonary function testing. Five patients had normal values. The others had total lung capacity between 56% and 69% of normal values (see Table 1). Pulmonary hypertension, tested by Doppler echocardiography, was detected in 3 of 14 patients. Comparing the group of 11 IM patients and the group of 19 IM-negative patients, ILD frequency was significantly higher in patients with associated IM (82% vs. 10.5%, respectively, p < 0.001). Comparing the group of 7 SSc patients and the group of 23 SSc-negative patients showed a trend to a higher frequency of ILD in patients with SSc (71% vs. 26%, respectively, p = 0.068).

F2-4
FIGURE 2:
Axial thin collimation scan: A, bilateral areas of ground-glass opacity predominantly distributed in the periphery of the lungs associated with fine intralobular reticular lines suggestive of NSIP-like pattern; B, bilateral ground-glass opacity containing intralobular reticular lines predominantly distributed in the lung bases and presence of traction bronchiectasis (arrow) and volume loss of the lower lobes suggestive of NSIP with fibrosis; C, irregularities of the air-pleural interface in the periphery of the lungs and presence in the posterior aspect of the lower lobes of subpleural honeycombing (arrows) suggestive of a UIP-like pattern.
T4-4
TABLE 2:
Laboratory and Imaging Features of 30 Anti-Ku-Positive Patients

All patients showed ANA titers >1/320 with a constant pattern suggestive of anti-Ku in the form of dense speckled nuclear staining with a homogeneous fluorescence of nucleoli (see Figure 1). Anti-Ku antibodies were the unique circulating autoantibody detected in only 7 patients (see Table 2). In the 23 (77%) remaining patients, anti-Ku were associated with others autoantibodies, including anti-ENA antibodies (mostly anti-DNA and anti-SSA/SSB, n = 4 for each), rheumatoid factor (n = 8), antithyroid (n = 7), and antiphospholipid (n = 4) antibodies, and antineutrophil cytoplasmic antibodies (ANCA) (n = 2). Among the 7 patients with SSc, only 1 had associated anti-centromere antibodies, and none had anti-SCl70. No specific autoantibodies of myositis (anti-JO1/-PL7/-PL12/-SRP/-Mi2) were detected.

Immunosuppressive treatments administered to anti-Ku-positive patients are detailed in Table 3. For 14 (47%) patients, no immunosuppressive treatment was required or only a low corticosteroid dose (<15 mg/d, n = 3) was used for articular manifestations. Two patients received a low prednisone dose and 1 additional therapy because of corticoresistant arthritis (methotrexate in 1 patient with SLE and rituximab in 1 patient with rheumatoid arthritis).

Nevertheless, 14 of 30 (47%) patients required a high dose of corticosteroids (median, 52.5 mg/d; range, 20–80 mg/d). Indication for corticosteroid therapy was myositis for 9 patients, SLE for 3 patients, ILD for 1 patient, and idiopathic pericarditis for the last patient. Eight of the 14 (57%) patients treated with higher corticosteroid doses received 1 or more additional immunosuppressive and/or immunomodulatory therapy (azathioprine in 5 cases, methotrexate 4, cyclophosphamide 3, mycophenolate mofetil 1, rituximab 2, and intravenous immunoglobulin 2), mostly for uncontrolled ILD or myositis (see Table 3).

All 11 patients with IM received corticosteroid therapy (high dose for 10 patients and 15 mg/d for 1 patient). Complete muscle response was observed in 8 (73%) patients, and muscle relapse occurred for 2 of them when corticosteroids were tapered below 10 mg/day. The 3 cases with refractory myopathies were 2 IBM and 1 polymyositis that remained unresponsive after 3 additional second-line agents.

Eight of the 11 (73%) patients with ILD received high-dose corticosteroids (see Table 1). Lung disease was corticoresistant in 6 (75%) cases. Among 5 patients who received additional immunosuppressive drugs, partial lung response was observed in only 1 case (after 6 intravenous cyclophosphamide infusions). The 3 remaining patients with ILD received only low prednisone doses (<15 mg/d): lung disease was resistant in 2 cases, and the last patient, in whom the chest CT scan was very suggestive of pulmonary sarcoidosis, responded completely.

DISCUSSION

In our institution, anti-Ku is a rare autoantibody, found in only 0.45% of 20,600 ANA-positive sera over 10 years, but it is easily detectable because of a suggestive speckled and nucleolar pattern on immunofluorescence. The exact prevalence of anti-Ku antibodies in patients affected by autoimmune diseases is not known. Prevalence data reported in the literature vary considerably, ranging from 1.3% to 50%,3,4,13,16,18,27 depending strongly on the methods used for anti-Ku detection because, to our knowledge, no standard laboratory test has been established so far. Among the numerous distinct methods described for detection, the most classically reported are immunodiffusion, immunoprecipitation, ELISA, immunoblot, and counterimmunoelectrophoresis, with an unknown sensibility for each. For example, a large Canadian study of 100 patients affected by autoimmune myositis reported that frequency of anti-Ku antibodies increased from 13% when detected by immunoprecipitation to 23% when detected by immunoblot.10,22 Most recently, it appears in the literature that detecting anti-Ku by immunodiffusion, immunoblot, or counterimmunoelectrophoresis in autoimmune diseases leads to a global similar prevalence of approximately 2% (range, 0.21%–2.7%),3,4,13,27 in accordance with our findings obtained with immunoblot detection. Moreover, anti-Ku prevalence also varies according to the country (Table 5) and according to racial/ethnic factors (higher frequency detected in African-American patients who have SLE compared with the white population25).

T5-4
TABLE 5:
Prevalence of Inflammatory Myopathy in Anti-Ku-Positive Patients, Present and Previous Reports

Therefore, we think that systematic screening of a suggestive anti-Ku aspect of immunofluorescence on Hep2 cells constitutes a reliable and reproducible method for detecting anti-Ku antibodies before performing additional immunoreactive tests for definitive confirmation.

The most common clinical features in our anti-Ku antibodies patients were arthralgia (77%), Raynaud phenomenon (53%, mainly related to the underlying autoimmune disease since 75% of them had SSc, SLE, or SS), and IM (37%), in accord with results of previous reports.3–5,27 Nevertheless, we also noticed an unexpectedly high frequency of peripheral neurologic manifestations and ILD (37% of patients for both). However, the high prevalence of peripheral neurologic manifestations we observed might be linked to the connective autoimmune diseases associated with anti-Ku antibodies in our series. Indeed, among 9 patients with neuropathy, 2 had SS with trigeminal nerve involvement, 2 had multiple mononeuritis with SLE plus hepatitis C infection/mixed cryoglobulinemia in 1 case and rheumatoid arthritis in the other case, and 1 had IBM, classically associated with axonal neuropathy. Moreover, in 2 of our patients, the clinical pattern was similar to the presentation of anti-JO1 synthetase syndrome, with the association of mechanic hands, arthritis, IM, and ILD; to the best of our knowledge this had not been reported in previous series of anti-Ku-positive patients.

In the end, the current series shows that the presence of positive anti-Ku antibodies is associated with a global favorable prognosis, as almost one-half of our patients (47%) did not require immunosuppressive treatment other than low-dose corticosteroids, mostly for articular manifestations.

Inflammatory myopathy was 1 of the 3 most frequent clinical features (37%) presented by our anti-Ku-positive patients. Most of these IM (73%) occurred in the form of overlap syndrome, mainly in association with SSc (62.5% of overlap syndrome). In the literature, 111 distinct reported cases of anti-Ku-positive patients have been reported so far (see Table 5). IM has been noted in 11.5% to 37% of cases, with a mean prevalence of 24%, in accordance with our findings. As expected, most cases of anti-Ku-positive IM reported in literature occurred in the form of overlap syndrome (mean, 74%), mainly in association with SSc (70% of overlap syndrome cases), as seen in our study.

On the other hand, the presence of anti-Ku antibodies in descriptive series of patients with IM is not so rare, even if the reported rates show some discrepancy among the series. Indeed, 1%–26% of patients with IM had positive anti-Ku antibodies (Table 6), with a global prevalence of 12% among a total of 334 distinct reported cases of IM. This prevalence rises to 24% among a subgroup of patients with IM-overlap syndrome. All our anti-Ku-positive patients with IM required a high dose of corticosteroids, except 1 patient (with a probable associated sarcoidosis) who responded to a low dose of only 15 mg/day of prednisone. The clinical evolution of anti-Ku-positive IM cannot be precisely described based on data available in the literature, since, to our knowledge, it has been mentioned in only 2 descriptive studies. The first study13 reported 7 patients with anti-Ku myositis, in all cases associated with SSc, who all showed good response to corticosteroid therapy. The second series detailed 23 cases of anti-Ku myositis, but therapeutic course was mentioned for only 2 patients in a separate report,10,22 with a favorable response after corticosteroid therapy in both cases. In the present study, we confirm that most IM associated with anti-Ku are sensitive to corticosteroids, since complete muscle response occurred in 8 of 11 (73%) of our patients. As expected, the 2 cases of IBM in our series did not regress with steroid treatment. Thus, the rate of muscle corticosteroid response increases to 91% for non-IBM myopathies, and to 100% for IM-overlap syndrome. Indeed, our findings confirm, as previously suggested by Troyanov and all,22 that anti-Ku antibodies are markers of corticosensitivity with monophasic evolution of the myositis, particularly for IM occurring in the frame of overlap syndrome.

T6-4
TABLE 6:
Prevalence of Anti-Ku Abs in Patients With Inflammatory Myopathy, Previous Reports

Lung involvement in anti-Ku-positive patients has rarely been noted: it has been reported in only 2 studies, with a frequency of 8% and 43% respectively.3,4 In the current series, 37% of anti-Ku patients had ILD, and almost one-half occurred in patients with SSc. It is noteworthy that we found a significant association between the presence of ILD and IM (82% of ILD occurred in patients with IM, p < 0.001). Whether the link between ILD and IM is related to the underlying autoimmune disease, especially SSc, is not known, partly due to the small sample size of our cohort. In any case, we observed in our anti-Ku patients with SSc an unexpectedly high proportion of ILD (71%) compared with the frequency of ILD reported in SSc patients in the literature. Indeed, 1 large prospective Canadian study of 309 consecutive patients with SSc showed that lung involvement—defined by interstitial fibrosis on chest radiogram or isolated reduction (<70%) of predicted normal values or restrictive syndrome on pulmonary function test—was observed in 24% of patients overall, and in only 18.4% of those with limited sclerodermatous skin involvement. By contrast, in the current study, 71% of patients with the equivalent limited sclerodermatous skin involvement had ILD. Nevertheless, the role of anti-Ku antibodies as a risk factor for the occurrence of ILD in SSc patients is debatable. For example, authors of a 2008 retrospective European case-control multicentric study17 noted that ILD prevalence was equal between 14 anti-Ku-positive and 43 anti-Ku-negative patients with SSc on chest X-ray (respectively, 57% and 56%) and on lung function test (respectively, 36% and 38%). But, authors of another 2008 study3 suggested that SSc patients with positive anti-Ku antibodies have more severe lung involvement compared with SSc patient without anti-Ku antibodies. In the current study, 2 findings seem in accordance with these results. First, we observed that more than one-half of patients with ILD had severe involvement attested to by spirometric values. Second, we noticed an overall poor lung prognosis, as 78% of ILD cases were resistant to high doses of steroids, and as 3 of 4 cases were still refractory after additional immunosuppressive therapy.

Patients with SSc and positive anti-Ku antibodies seem to have some particular features: 1) The presence of anti-Ku antibodies is exclusive of any other specific SSc antibodies3,17; 2) anti-Ku-positive SSc patients are at increased risk of developing IM17; and 3) ILD is more frequent and severe.3

In conclusion, anti-Ku antibodies remain rarely detected, but their presence is frequently associated with corticosensitive IM and severe corticoresistant ILD.

ACKNOWLEDGMENT

The authors thank Marie-Claude Diemert for technical assistance.

REFERENCES

1. Arnett FC, Edworthy SM, Bloch DA, et al.. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988; 31: 315–324.
2. Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two parts). N Engl J Med. 1975; 292: 344–347.
3. Cavazzana I, Ceribelli A, Quinzanini M, et al.. Prevalence and clinical associations of anti-Ku antibodies in systemic autoimmune diseases. Lupus. 2008; 17: 727–732.
4. Cooley HM, Melny BJ, Gleeson R, et al.. Clinical and serological associations of anti-Ku antibody. J Rheumatol. 1999; 26: 563–567.
5. Franceschini F, Cavazzana I, Generali D, et al.. Anti-Ku antibodies in connective tissue diseases: clinical and serological evaluation of 14 patients. J Rheumatol. 2002; 29: 1393–1397.
6. Francoeur AM, Peebles CL, Gompper PT, et al.. Identification of Ki (Ku, p70/p80) autoantigens and analysis of anti-Ki autoantibody reactivity. J Immunol. 1986; 136: 1648–1653.
7. Ghirardello A, Zampieri S, Tarricone E, et al.. Clinical implications of autoantibody screening in patients with autoimmune myositis. Autoimmunity. 2006; 39: 217–221.
    8. Griggs RC, Askanas V, DiMauro S, et al.. Inclusion body myositis and myopathies. Ann Neurol. 1995; 38: 705–713.
    9. Hausmanowa-Petrusewicz I, Kowalska-Oledzka E, Miller FW, et al.. Clinical, serologic, and immunogenetic features in Polish patients with idiopathic inflammatory myopathies. Arthritis Rheum. 1997; 40: 1257–1266.
    10. Koenig M, Fritzler MJ, Targoff IN, et al.. Heterogeneity of autoantibodies in 100 patients with autoimmune myositis: insights into clinical features and outcomes. Arthritis Res Ther. 2007; 9: R78.
    11. Koike M. Dimerization, translocation and localization of Ku70 and Ku80 proteins. J Radiat Res (Tokyo). 2002; 43: 223–236.
    12. LeRoy EC, Medsger TA Jr. Criteria for the classification of early systemic sclerosis. J Rheumatol. 2001; 28: 1573–1576.
    13. Mimori T, Akizuki M, Yamagata H, et al.. Characterization of a high molecular weight acidic nuclear protein recognized by autoantibodies in sera from patients with polymyositis-scleroderma overlap. J Clin Invest. 1981; 68: 611–620.
    14. Mosca M, Neri R, Bombardieri S. Undifferentiated connective tissue diseases (UCTD): a review of the literature and a proposal for preliminary classification criteria. Clin Exp Rheumatol. 1999; 17: 615–620.
    15. Nogalska A, Terracciano C, D’Agostino C, et al.. p62/SQSTM1 is overexpressed and prominently accumulated in inclusions of sporadic inclusion-body myositis muscle fibers, and can help differentiating it from polymyositis and dermatomyositis. Acta Neuropathol. 2009; 118: 407–413.
    16. Reeves WH. Use of monoclonal antibodies for the characterization of novel DNA-binding proteins recognized by human autoimmune sera. J Exp Med. 1985; 161: 18–39.
    17. Rozman B, Cucnik S, Sodin-Semrl S, et al.. Prevalence and clinical associations of anti-Ku antibodies in patients with systemic sclerosis: a European EUSTAR-initiated multi-centre case-control study. Ann Rheum Dis. 2008; 67: 1282–1286.
    18. Schild-Poulter C, Su A, Shih A, et al.. Association of autoantibodies with Ku and DNA repair proteins in connective tissue diseases. Rheumatology (Oxford). 2008; 47: 165–171.
    19. Selva-O’Callaghan A, Labrador-Horrillo M, Solans-Laque R, et al.. Myositis-specific and myositis-associated antibodies in a series of eighty-eight Mediterranean patients with idiopathic inflammatory myopathy. Arthritis Rheum. 2006; 55: 791–798.
      20. Silvera D, Koloteva-Levine N, Burma S, et al.. Effect of Ku proteins on IRES-mediated translation. Biol Cell. 2006; 98: 353–361.
      21. Tan EM, Cohen AS, Fries JF, et al.. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 1982; 25: 1271–1277.
      22. Troyanov Y, Targoff IN, Tremblay JL, et al.. Novel classification of idiopathic inflammatory myopathies based on overlap syndrome features and autoantibodies: analysis of 100 French Canadian patients. Medicine (Baltimore). 2005; 84: 231–249.
      23. Vitali C, Bombardieri S, Jonsson R, et al.. Classification criteria for Sjogren’s syndrome: a revised version of the European criteria proposed by the American-European Consensus Group. Ann Rheum Dis. 2002; 61: 554–558.
      24. Walker JR, Corpina RA, Goldberg J. Structure of the Ku heterodimer bound to DNA and its implications for double-strand break repair. Nature. 2001; 412: 607–614.
      25. Wang J, Satoh M, Kabir F, et al.. Increased prevalence of autoantibodies to ku antigen in African American versus white patients with systemic lupus erythematosus. Arthritis Rheum. 2001; 44: 2367–2370.
      26. Yamamoto AM, Amoura Z, Johannet C, et al.. Quantitative radioligand assays using de novo-synthesized recombinant autoantigens in connective tissue diseases: new tools to approach the pathogenic significance of anti-RNP antibodies in rheumatic diseases. Arthritis Rheum. 2000; 43: 689–698.
      27. Yaneva M, Arnett FC. Antibodies against Ku protein in sera from patients with autoimmune diseases. Clin Exp Immunol. 1989; 76: 366–372.
      © 2012 Lippincott Williams & Wilkins, Inc.