A novel autoantibody recognizing 200-kd and 100-kd proteins is associated with an immune-mediated necrotizing myopathy

Authors


Abstract

Objective

Myofiber necrosis without prominent inflammation is a nonspecific finding in patients with dystrophies and toxic or immune-mediated myopathies. However, the etiology of a necrotizing myopathy is often obscure, and the question of which patients would benefit from immunosuppression remains unanswered. The aim of this study was to identify novel autoantibodies in patients with necrotizing myopathy.

Methods

Muscle biopsy specimens and serum samples were available for 225 patients with myopathy. Antibody specificities were determined by performing immunoprecipitations from 35S-methionine–labeled HeLa cell lysates. Selected biopsy specimens were stained for membrane attack complex, class I major histocompatibility complex (MHC), and endothelial cell marker CD31.

Results

Muscle biopsy specimens from 38 of 225 patients showed predominantly myofiber necrosis. Twelve of these patients had a known autoantibody association with or other etiology for their myopathy. Sixteen of the remaining 26 sera immunoprecipitated 200-kd and 100-kd proteins; this specificity was observed in only 1 of 187 patients without necrotizing myopathy. Patients with the anti-200/100 autoantibody specificity had proximal weakness (100%), high creatine kinase levels (mean maximum 10,333 IU/liter), and an irritable myopathy on electromyography (88%). Sixty-three percent of these patients had been exposed to statins prior to the onset of weakness. All patients responded to immunosuppressive therapy, and many experienced a relapse of weakness when the medication was tapered. Immunohistochemical studies showed membrane attack complex on small blood vessels in 6 of 8 patients and on the surface of non-necrotic myofibers in 4 of 8 patients. Five of 8 patients had abnormal capillary morphology, and 4 of 8 patients expressed class I MHC on the surface of non-necrotic myofibers.

Conclusion

An anti–200/100-kd specificity defines a subgroup of patients with necrotizing myopathy who previously were considered to be autoantibody negative. We propose that these patients have an immune-mediated myopathy that is frequently associated with prior statin use and should be treated with immunosuppressive therapy.

Adults with proximal muscle weakness, elevated creatine kinase (CK) levels, features of myopathy on electromyography (EMG), and evidence of muscle edema on magnetic resonance imaging (MRI) have a broad differential diagnosis that includes autoimmune myopathies, toxic myopathies, paraneoplastic myopathies, and muscular dystrophies. Distinguishing between immune-mediated myopathies and other etiologies is crucial, because only autoimmune muscle diseases routinely respond to immunosuppressive therapy.

In many cases, distinctive clinical features and/or a muscle biopsy can provide a definitive diagnosis. For example, perifascicular atrophy is pathognomonic for dermatomyositis (DM) even in the absence of rash; vacuolar myopathy in a patient treated with colchicine strongly suggests a toxic myopathy, and reduced dystrophin staining in the muscle of a young man with calf hypertrophy is diagnostic for a dystrophinopathy.

However, in a substantial number of cases, muscle biopsy specimens show degenerating and necrotic muscle fibers in the absence of disease-specific features. In these instances, the presence of myositis-specific autoantibodies (MSAs) may identify the disorder as belonging to the family of autoimmune myopathies (1). For example, patients with antibodies directed against the signal recognition particle (SRP) typically have a severe necrotizing myopathy that is responsive only to very aggressive immunosuppression (2–6). Unfortunately, clinical evaluation and currently available diagnostic tests do not always provide a definitive diagnosis, and it may not be possible to determine whether a necrotizing myopathy is immune mediated. This uncertainty can lead to undertreatment of autoimmune myopathies or inappropriate immunosuppression in patients who do not have an immune-mediated disease.

In this study, we identified 26 patients with necrotizing myopathies in whom, despite comprehensive evaluations, a specific muscle disease could not be diagnosed. Sera from these patients were screened for the presence of novel autoantibodies, and a unique autoantibody specificity against 200-kd and 100-kd proteins was identified in 16 patients. Further analysis of the clinical characteristics and muscle biopsy features of these anti-200/100 autoantibody–positive patients suggests they belong to the family of autoimmune myopathies responsive to immunosuppressive therapy.

PATIENTS AND METHODS

Patients.

Two hundred twenty-five patients with banked sera, muscle biopsy specimens available for review, and a myopathy as defined by proximal muscle weakness, elevated CK levels, myopathic EMG findings, muscle edema on MRI, and/or features of myopathy on muscle biopsy were enrolled in a longitudinal study, approved by the Johns Hopkins Institutional Review Board, from March 2007 through December 2008. In addition to providing a history and undergoing physical examination at the Johns Hopkins Myositis Center, these patients underwent a comprehensive evaluation including some or all of the following: 1) EMG and nerve conduction studies, 2) noncontrast bilateral thigh MRI, 3) pulmonary function tests, 4) malignancy screening including computed tomography scans of the chest, abdomen, and pelvis, 5) a standard laboratory evaluation performed by several different commercial laboratories including CK levels, antinuclear antibody (ANA) screen, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) levels, anti-Ro/La screen, and MSA screen, and 6) when suspected based on clinical or biopsy features, testing for inherited muscle disease including limb-girdle muscular dystrophies (by Limb Girdle Muscular Dystrophy Evaluation panel [Athena Diagnostics]), acid maltase deficiency (by Glycogen Storage Myopathy ‘A’ Profile [Athena Diagnostics] and/or dried blood spot test for α-glucosidase activity [Genzyme]), and/or facioscapulohumeral dystrophy (by FSHD DNA Test [Athena Diagnostics]).

In order to determine whether statins were used at an increased frequency in patients with the anti-200/100 autoantibody, we also determined the frequency of statin use in patients from our cohort who had definite or probable polymyositis (PM) or dermatomyositis (DM) (7, 8) as well as in those with possible inclusion body myositis (IBM) (9). The ages of the patients were compared using Student's 2-tailed t-tests. The chi-square test was used to compare the frequency of statin use in the different groups.

Muscle biopsy analysis.

Muscle biopsy specimens were obtained from the deltoid, biceps, or quadriceps muscle groups. In each case, the muscle selected was determined to be weak by the examining physician. The slides from muscle biopsy specimens were evaluated at the Johns Hopkins Neuromuscular Pathology Laboratory. These studies included hematoxylin and eosin–stained tissue as well as some or all of the following stains: modified Gomori's trichrome, adenosine triphosphatase at pH 4.3, pH 4.6, and pH 9.4, NAD tetrazolium reductase, acid phosphatase, succinic dehydrogenase, cytochrome oxidase, esterase, alkaline phosphatase, periodic acid−Schiff (PAS), PAS–diastase control, and Congo red. Both frozen and paraffin-embedded specimens were routinely screened for the presence of degenerating, regenerating, and/or necrotic fibers, primary endomysial inflammation, perivascular inflammation, rimmed vacuoles, perifascicular atrophy, and fibrosis. We identified “necrotizing myopathy” biopsy specimens based on the presence of necrotic muscle fibers as the predominant abnormal histologic feature; with the exception of necrotic myofibers undergoing myophagocytosis, inflammatory cells were sparse, if present at all. Muscle biopsy specimens from patients with the anti-200/100 autoantibody specificity were stained with antibodies recognizing CD31 (an endothelial cell marker), C5b–9 (i.e., membrane attack complex), and class I major histocompatibility complex (MHC).

Briefly, 7μ-thick frozen muscle biopsy sections were fixed in ice-cold acetone. After 10 minutes in peroxidase-blocking reagent (Dako) at room temperature, sections were incubated with 5% bovine serum albumin/phosphate buffered saline (BSA/PBS) for 1 hour at 37°C. Primary antibodies were prepared in 1% BSA/PBS at the following dilutions: 1:50 for class I MHC (Santa Cruz Biotechnology), 1:20 for CD31 (Dako), 1:50 for Cb5–9 (Santa Cruz Biotechnology); primary incubations were performed overnight at 4°C. After PBS washes, the slides were incubated with horseradish peroxidase–labeled goat anti-mouse secondary antibody (Dako) in 1% BSA/PBS at 1:500 for 1 hour at room temperature. The compound 3,3′-diaminobenzidine chromagen (Dako) was used to visualize each antibody, and all sections were counterstained with hematoxylin. Normal muscle tissue samples were used as negative controls, and muscle tissue from a Jo-1–positive patient with myositis was used as a positive control for class I MHC staining. For each primary antibody, all muscle sections were processed simultaneously under the same conditions.

Immunoprecipitations.

Serum samples collected from each patient were stored at −80°C. HeLa cells cultured using standard procedures were radiolabeled for 2 hours with 100 μCi/ml 35S-methionine and cysteine (MP Biomedicals) in methionine-free and cysteine-free medium. The cells were subsequently lysed in buffer A (50 mM Tris pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.5% Nonidet P40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS], and a protease inhibitor cocktail). Each 10-cm dish was lysed in 1 ml buffer A and was used for 10 immunoprecipitations. Immunoprecipitations were performed by adding 1 μl of patient sera to 100 μl radiolabeled lysate and bringing the volume to 1 ml with buffer B (1% Nonidet P40, 20 mM Tris pH 7.4, 150 mM NaCl, 1 mM EDTA, and a protease inhibitor cocktail) and rotating the mix for 1 hour at 4°C. Protein A agarose beads (Pierce) were used to precipitate the antibody–antigen complexes that were subsequently electrophoresed on 10% SDS–polyacrylamide gels. The radiolabeled immunoprecipitates were visualized by fluorography.

RESULTS

Identification of patients with necrotizing myopathy and uncertain diagnoses.

Muscle biopsy specimens obtained from 225 patients who presented with proximal muscle weakness, elevated CK levels, evidence of myopathy on EMG, and/or other evidence of muscle disease were reviewed in order to identify those with a predominantly necrotizing myopathy. Patients with biopsy results notable for marked inflammatory cell infiltrates, rimmed vacuoles (characteristic of inclusion body myositis), perifascicular atrophy (pathognomonic for DM), or other features characteristic of a specific diagnosis were not considered to have a predominantly necrotizing myopathy.

In all, 38 patients (17% of the total) were identified as having a predominantly necrotizing myopathy on muscle biopsy. Of these, a specific muscle disease was definitively diagnosed in 12 patients, using existing testing methods. Ten patients had autoimmune myopathies as defined by the presence of antisynthetase autoantibodies (1 with anti–Jo-1, 2 with anti–PL-12, and 1 with anti–PL-7) or anti-SRP autoantibodies (6 patients); each of these patients also had a definite positive response to immunosuppressive therapy. In addition, 1 patient had a necrotizing myopathy associated with profound hypothyroidism, and another had limb-girdle muscular dystrophy type 2B (i.e., dysferlinopathy) confirmed by genetics testing. The remaining 26 patients (∼10% of our cohort) had a predominantly necrotizing myopathy of unclear etiology.

Identification of a novel anti–200/100-kd autoantibody specificity in patients with a necrotizing myopathy.

Sera collected from the 26 patients described above were screened for the presence of novel autoantibodies. Remarkably, we observed that sera from 16 of these patients (62%) immunoprecipitated a pair of proteins from radioactively labeled HeLa cell extracts with approximate sizes of 200 kd and 100 kd, respectively (Figure 1). These proteins, with molecular weights that do not correspond to those of known myositis-specific autoantigens, were always immunoprecipitated as a pair. Although anti-200/100 autoantibody immunoprecipitations were reproducible, no serum detected 200-kd or 100-kd proteins when used to immunoblot HeLa cell extracts (data not shown.)

Figure 1.

Immunoprecipitation of ∼200-kd and ∼100-kd proteins by sera from patients with a necrotizing myopathy. Patient sera were used to immunoprecipitate radioactively labeled proteins from HeLa cell extracts that had been incubated with 35S-methionine. Immunoprecipitated proteins were separated by electrophoresis on 10% sodium dodecyl sulfate–polyacrylamide gels. The left and right panels show autoradiographs from 2 separate experiments; results shown in the right panel are from a single autoradiograph that has been cropped between lanes 7 and 8 to exclude immunoprecipitations that are irrelevant to the current study. The numbers at the top of lanes 1–4 and 6–9 are patient numbers (additional information is available from the corresponding author). Two different normal control sera (Cont 33 and Cont 35) were used for the immunoprecipitations shown in lanes 5 and 10. Values on the far right show the migration of molecular weight marker standards.

In order to evaluate the specificity of these antibodies for a necrotizing phenotype, we tested for anti-200/100 autoantibody immunoreactivity in the remaining cohort. Among the 187 patients who did not have a predominant necrotizing myopathy, the serum from only 1 patient (0.5%) immunoprecipitated the 200-kd and 100-kd proteins, demonstrating that this finding is highly specific for those patients with a necrotizing myopathy (P < 10−15 by Fisher's exact test). None of the sera from the 12 patients with necrotizing myopathies associated with previously known conditions, including the 6 patients with anti-SRP antibodies, immunoprecipitated proteins with molecular weights of 200 kd or 100 kd (data not shown).

Several of the anti-200/100 autoantibody–positive sera immunoprecipitated additional proteins. For example, the serum from patient 8,089 immunoprecipitated an ∼70-kd protein as well as proteins of 200 kd and 100 kd (Figure 1, lane 2). Of note, each of the additional proteins was recognized by no more than 1 of the 16 sera from patients with anti-200/100 autoantibody positivity. Furthermore, none of the additional bands recognized by any of the anti-200/100 autoantibody–positive sera corresponded in size to previously recognized myositis-specific autoantigens, including proteins with molecular weights of 72 kd, 54 kd, and/or 21 kd, as seen in patients with anti-SRP myopathy.

Clinical features of the anti-200/100 autoantibody– positive patient population.

We analyzed the demographic information, laboratory findings, pattern of weakness, thigh MRIs, and other clinical features of the 16 anti-200/100 autoantibody–positive patients with a necrotizing myopathy (Table 1); the patient with anti-200/100 autoantibody specificity who did not have a predominantly necrotizing myopathy was excluded from this analysis.

Table 1. Clinical features of the patients with anti-200/100 autoantibodies*
  • *

    Except where indicated otherwise, values are the percent. CPK = creatine phosphokinase; CK = creatine kinase; ANA = antinuclear antibody; ESR = erythrocyte sedimentation rate; MRI = magnetic resonance imaging; EMG = electromyography.

Demographics 
 No. of patients16
 Mean age at disease onset, years54
 Female sex63
 White race56
 Nonwhite race44
 Deceased0
Clinical feature 
 Subjective muscle weakness100
 Proximal weakness on examination100
 Wheelchair use25
 Interstitial lung disease0
 Malignancy13
 Raynaud's phenomenon13
 Rash44
 Myalgias75
 Arthralgias50
 Dysphagia63
 Statin use63
Laboratory findings 
 Initial CPK level, mean IU/liter8,702
 Maximum CK level, mean IU/liter10,333
 ANA positivity (>1:160)6
 Elevated ESR38
 Elevated C-reactive protein level6
 Anti-Ro positive0
 Anti-La positive0
Thigh MRI features 
 Normal findings on thigh MRI0
 Muscle edema100
 Atrophy75
 Fatty replacement67
 Fascial edema25
EMG findings 
 Irritable myopathy88
 Nonirritable myopathy13
 Normal0

Men and women were represented in roughly equal numbers and had a mean age of 54 years at the onset of disease. All 16 patients reported previously normal strength, with the acute or subacute onset of muscle weakness occurring in adulthood. At the time of the initial evaluation, all patients had proximal muscle weakness, evidence of muscle edema on bilateral thigh MRI, and markedly elevated CK levels, with a mean value of 10,333 IU/liter (range 3,052–24,714). Each of the 16 EMGs available for review revealed features of myopathy; 14 (88%) of 16 of these patients demonstrated an irritable myopathy, while the remaining 2 myopathies were nonirritable.

Other prominent clinical features included myalgias in 12 (75%) of 16 patients, arthralgias in 8 (50%) of 16 patients, and dysphagia in 10 (63%) of 16 patients. Only 2 (13%) of 16 patients had Raynaud's phenomenon. Although 7 (44%) of 16 patients reported a nonspecific rash, no patient had cutaneous features consistent with DM on examination or by historical account.

None of these patients had antibodies against extractable nuclear antigens detected by clinical laboratories (including anti-Ro, anti-La, anti-RNP, and anti–Scl-70), and no patient met the criteria for another connective tissue disease. Two patients had prior malignancies: 1 had nonrecurrent ovarian cancer treated 5 years prior to the onset of muscle disease, and the other had prostate cancer that was in clinical remission after treatment.

None of the anti-200/100 autoantibody–positive patients had a family history of muscle disease. Furthermore, scapular winging, facial weakness, asymmetric weakness, or other distinctive features suggestive of inherited muscle disease were absent in each of these patients.

Of note, 10 (63%) of 16 patients had been exposed to statin therapy prior to the onset of weakness. The mean ± SD duration of statin treatment prior to the onset of muscle symptoms was 31.3 ± 27.4 months (range 0–84 months). In each case, discontinuing the statin medication did not lead to clear clinical improvement, and the mean ± SD length of time between statin discontinuation and muscle biopsy was 5.2 ± 4.6 months (range 1–14 months). A review of the patient records revealed no other potential myotoxin exposures.

To determine whether the association with statin use was coincidental, we analyzed the frequency of statin use in other groups of patients with myositis evaluated at our clinic (Table 2). We observed that 5 (15.2%) of 33 patients with DM, 7 (18.4%) of 38 patients with PM, and 11 (35.5%) of 31 patients with IBM had been treated with statins prior to undergoing a muscle biopsy; the frequency of statin use was significantly (P < 0.05) increased in the anti-200/100 autoantibody–positive group compared with both the DM and PM groups. However, in this analysis, there was no significant difference in statin use between the group of patients with anti-200/100 autoantibody positivity and the group with IBM (P = 0.08). Because older patients are more likely to be treated with statins, we assessed the ages of patients with different forms of myositis. Compared with all of the anti-200/100 autoantibody–positive patients, who had a mean ± SD age of 57.8 ± 14.8 years, the total group of patients with IBM was significantly older, with a mean ± SD age of 67.7 ± 9.9 years. When only those patients ages 50 years or older were included in the analysis, 10 (83.3%) of 12 anti-200/100 autoantibody–positive patients, 4 (25%) of 16 patients with DM, 7 (36.8%) of 19 patients with PM, and 10 (33.3%) of 30 patients with IBM had been exposed to statins (Table 2). In this age-matched comparison, statin treatment was significantly increased in the anti-200/100 autoantibody–positive population compared with the DM (P = 0.002), PM (P = 0.011), and IBM (P = 0.003) populations.

Table 2. Frequency of statin use in patients with different forms of muscle disease*
GroupFrequency of statin useMean ± SD age of patients, years
  • *

    Except where indicated otherwise, values are the number of patients/total number of patients (%). DM = dermatomyositis; PM = polymyositis; IBM = inclusion body myositis.

  • P < 0.05 versus patients with anti-200/100 antibodies, by chi-square test.

  • P < 0.05 versus patients age ≥50 years with anti-200/100 antibodies, by Student's t-test.

All patients with anti-200/100 antibodies10/16 (62.5)57.8 ± 14.8
DM patients5/33 (15.2)51.0 ± 12.2
PM patients7/38 (18.4)49.1 ± 14.1
IBM patients11/31 (35.5)67.7 ± 9.9
Patients with anti-200/100 antibodies age ≥50 years10/12 (83.3)64.4 ± 9.2
DM patients age ≥50 years4/16 (25)61.0 ± 8.3
PM patients age ≥50 years7/19 (36.8)60.4 ± 7.6
IBM patients age ≥50 years10/30 (33.3)68.4 ± 9.2

There was a striking variation in clinical phenotype, ranging from a chronically intubated, quadriplegic patient to several patients who had only mild weakness. A unique feature in the majority of patients was their relative preservation of strength despite markedly elevated levels of muscle enzymes. However, the medical records of several patients showed an apparent threshold muscle enzyme level (usually between 3,000 and 7,000 IU/liter) above which weakness ensued.

Medication regimens and treatment responses (based on objective improvements in strength) were variable. (The clinical characteristics of the 16 anti-200/100 autoantibody–positive patients are available from the corresponding author). Of the 14 patients who were followed up longitudinally, 9 (64%) had a complete or near-complete response to immunosuppression, and 5 (36%) had a partial response to immunosuppression; these 5 patients included 1 patient whose progressive muscle weakness was stabilized but did not improve with immunosuppression. Six (43%) of the 14 patients experienced a relapse when immunosuppressive medication was tapered or withdrawn. Seven (60%) of the 14 patients are currently undergoing tapering of their immunosuppressive medications and have not experienced a relapse to date. Only 1 patient had complete tapering of immunosuppressive medications without experiencing a relapse of weakness.

Most patients had a very modest initial response to prednisone and required combination immunosuppressive therapy. Rituximab and intravenous immunoglobulin appeared to be helpful adjuncts when added to prednisone and azathioprine or methotrexate. Most patients required some dose of prednisone for maintenance therapy and reported weakness with steroid tapering, even if their initial response to prednisone was only modest.

Muscle biopsy features of the anti-200/100 autoantibody–positive patient population.

Sixteen (94%) of 17 patients with anti-200/100 autoantibodies had muscle biopsy specimens showing prominent myofiber necrosis; the remaining patient's biopsy specimen was notable for extensive inflammatory infiltrates, and a subsequent analysis did not include the results of this biopsy. Although close examination revealed endomysial and/or perivascular collections of inflammatory cells in 5 (31%) of the 16 muscle biopsy specimens, the degree of inflammation was mild compared with that seen in typical muscle biopsy specimens obtained from patients with PM or DM. No biopsy specimen obtained from a patient with anti-200/100 autoantibody positivity revealed evidence of more than mild denervation, and no biopsy specimen was positive for abnormal glycogen accumulation or amyloid deposition.

Of the 16 patients with necrotizing myopathies who were anti-200/100 autoantibody positive, frozen muscle tissue samples obtained from 8 patients were available for further analysis. To assess blood vessel morphology, sections were stained with anti-CD31 antibodies. Abnormally enlarged endomysial capillaries with thickened walls were observed in 5 (63%) of 8 biopsy specimens (Figure 2). However, the density of capillaries within muscle tissue was not noticeably reduced in any of the muscle biopsy specimens.

Figure 2.

Capillary morphology in muscle biopsy specimens obtained from a normal donor (A) and a patient with anti-200/100 autoantibodies (B). Specimens were stained with anti-CD31, an endothelial cell marker. Arrows indicate endomysial capillaries with normal morphologic features in the control specimen (A) and those with thickened walls and dilated lumens in the patient with anti-200/100 autoantibodies (B). These biopsy specimens were processed simultaneously under identical conditions (original magnification × 40).

Complement deposition was evaluated by staining the available anti-200/100 autoantibody–positive muscle biopsy specimens with antibodies recognizing the membrane attack complex. Although endomysial capillaries were not definitively recognized by the antibody (Figure 3D), in 6 (75%) of 8 muscle biopsy specimens, small perimysial vessels were stained (Figures 3A and B). In contrast, blood vessels from control muscle biopsy specimens did not stain intensely with membrane attack complex antibodies (data not shown). As expected, membrane attack complex deposition was also present on necrotic and degenerating myofibers; this was considered a nonspecific finding. However, in 4 (50%) of 8 of the anti-200/100 autoantibody–positive muscle biopsy specimens, the sarcolemmal surfaces of scattered, non-necrotic muscle fibers stained positive for membrane attack complex (Figures 3C and D); as shown, some of these muscle cells were relatively small, suggesting they could be regenerating fibers.

Figure 3.

Membrane attack complex deposition on small blood vessels and non-necrotic myofibers. A and B, Serial section of a muscle biopsy specimen obtained from an anti-200/100 antibody–positive patient with necrotizing myopathy (patient 8076). Staining with anti–membrane attack complex (A) or hematoxylin and eosin (B) demonstrated a perimysial blood vessel with marked complement deposition. C, Muscle biopsy specimen obtained from an anti-200/100 antibody–positive patient (patient 8024), showing membrane attack complex deposition on scattered non-necrotic fibers. D, Higher-magnification view of the same field as in C. Arrows indicate the absence of membrane attack complex staining on endomysial capillaries. Asterisks in C and D show matching myofibers. (Original magnification × 40 in A, B, and D; × 20 in C.)

Staining of anti-200/100 autoantibody–positive muscle biopsy specimens with antibodies recognizing class I MHC showed that the sarcolemma of 4 (50%) of 8 specimens were clearly class I MHC positive (Figure 4). Several others had borderline class I MHC staining, but this appeared markedly less intense than that seen in muscle biopsy specimens from Jo-1–positive patients with PM that were included as positive controls in the same experiment (data not shown).

Figure 4.

Class I major histocompatibility complex (MHC) deposition on non-necrotic fibers in biopsy specimens obtained from anti-200/100 autoantibody–positive patients. A, Anti–class I MHC antibody staining of the endomysial capillaries of normal human muscle (arrow) but not the sarcolemma. B and C, Anti–class I MHC antibody staining of the sarcolemma of scattered muscle fibers in 2 patients with anti-200/100 autoantibodies (single asterisks). The cytoplasm of an anti-200/100 antibody–positive fiber also stained with anti–class I MHC (double asterisks); this likely represents a regenerating fiber. These biopsy specimens were processed simultaneously under identical conditions. (Original magnification × 40.)

DISCUSSION

The autoimmune myopathies (referred to collectively as myositis) are a family of conditions characterized clinically by symmetric proximal muscle weakness, elevated serum CK levels, and myopathic findings on EMG (10, 11). Although other muscle conditions can cause similar clinical syndromes, diagnosing an autoimmune disorder carries important therapeutic and prognostic implications, because only these disorders routinely respond to immunosuppressive therapy.

As with other systemic autoimmune diseases, a strong association of autoantibodies with distinct clinical phenotypes is observed in patients with autoimmune myopathy. For example, autoantibodies directed against aminoacyl–transfer RNA (tRNA) synthetases are the most frequent MSAs and are observed in ∼20% of patients with myositis (12). These and autoantibodies recognizing other tRNA synthetases are associated with a specific constellation of clinical features including interstitial lung disease, Raynaud's phenomenon, arthritis, and a characteristic cutaneous finding known as mechanic's hands (13, 14). Although autoantibody screening can play a significant role in the diagnosis of immune-mediated muscle disease, such antibodies are not always observed.

The presence of inflammatory infiltrates in muscle biopsy specimens is another well-recognized feature of the autoimmune myopathies (10). However, muscle biopsy specimens from some patients with autoimmune myopathies contain few, if any, inflammatory cell infiltrates. For example, patients with MSAs directed against components of the SRP have biopsy samples that are notable for degenerating, necrotic, and regenerating muscle cells without extensive inflammatory cell infiltrates (3–6). We hypothesized that patients with otherwise undiagnosed necrotizing myopathies might also have unique autoantibodies that could be used for diagnosis.

Among a group of 225 patients with myopathies, we observed that 38 had muscle biopsy specimens with predominantly necrotizing myopathies. After extensive laboratory testing, specific conditions could be diagnosed in 12 of these patients; these were largely patients with anti-SRP or antisynthetase myositis. We screened the sera of the remaining 26 patients for the presence of novel autoantibodies and observed that 16 of these sera immunoprecipitated a pair of proteins with approximate molecular weights of 200 kd and 100 kd, respectively. In addition, among the other 187 patients, 1 patient with a biopsy specimen showing abundant inflammatory cell infiltrates shared this immunospecificity. The patients with anti-200/100 autoantibodies did not have other known autoantibodies, including anti-SRP. Thus, anti-200/100 autoantibodies characterize a unique subset of patients with myopathies, representing 16 (62%) of 26 of our patients with idiopathic necrotizing myopathies.

In many respects, the clinical features of patients with the anti-200/100 autoantibody immunospecificity are similar to those of patients with other forms of immune-mediated myopathy; both groups typically experienced the subacute onset of proximal muscle weakness with elevated CK levels, had findings of irritable myopathy on EMG, evidence of edema on MRI, and, in most cases, a clear response to immunosuppressive therapy. However, we did note several unique features of the anti-200/100 autoantibody–positive patients. First, several patients had very high CK levels (in the range of 3,000–8,000 IU/liter) but only minimal muscle weakness. This suggests either an unusual capacity of these patients to regenerate muscle with sufficient efficiency to keep pace with extensive muscle destruction, or that these patients have a muscle membrane abnormality that allows leakage of CK without causing weakness; such an abnormality could be consistent with the finding of membrane attack complex deposition on the sarcolemma of non-necrotic muscle fibers. Second, in >60% of these patients, exposure to statin therapy preceded the development of muscle symptoms, which persisted long after treatment with the myotoxin was discontinued. Importantly, we observed that this association was strongest in older patients; more than 80% of anti-200/100 autoantibody–positive patients ages 50 years or older had been exposed to statins. This rate was significantly higher than the rates of statin treatment in age-matched groups of patients with PM, DM, or IBM.

Although the anti-200/100 autoantibody–positive patients share certain features with the well-described populations of patients with anti-SRP antibodies, 2 key findings distinguish these groups as distinct. First, sera from patients with anti-200/100 autoantibodies did not recognize any of the SRP subunits, and sera from patients with anti-SRP autoantibodies did not recognize proteins with molecular weights of 200 kd or 100 kd. These observations demonstrate that patients with the anti-200/100 autoantibody specificity are immunologically distinct from the population of patients with anti-SRP antibodies. Second, we followed up several anti-200/100 autoantibody–positive patients who had extremely high CK levels and only minimal weakness; in our experience and in other studies (3–6), patients with anti-SRP antibodies with similarly high CK levels were uniformly very weak.

To further characterize the muscle disease in patients with anti-200/100 autoantibodies, we stained muscle biopsy specimens with antibodies against membrane attack complex, endothelial cell markers, and class I MHC. Membrane attack complex deposition represents the end-stage of the complement cascade and may indicate that the tissue is targeted for destruction by the immune system. The deposition of membrane attack complex on endomysial capillaries has been shown in patients with DM (15, 16) and in 3 of 4 analyses of biopsy specimens positive for anti-SRP (3–6); this does not occur in muscular dystrophies (17). Although we did not demonstrate membrane attack complex deposition on endomysial capillaries in biopsy specimens obtained from patients with anti-200/100 autoantibodies, in 5 of 8 specimens, endomysial capillaries were abnormally thickened and enlarged. Similar morphologic abnormalities have been described both in patients with anti-SRP antibodies and in a group of patients with “necrotizing myopathy with pipestem capillaries.” Although the latter group shares some pathologic features with patients with anti-200/100 autoantibodies and anti-SRP antibodies, these patients differed by having either another connective tissue disease or active cancer (18).

Despite its absence on capillaries, membrane attack complex deposition in small perimysial blood vessels was evident in 6 (75%) of 8 biopsy specimens obtained from patients with anti-200/100 autoantibodies. We hypothesize that deposition of complement in these cases may reflect a novel vascular target in our population. In addition, membrane attack complex localized to the surface of non-necrotic fibers was noted in 4 (50%) of the 8 biopsy specimens from patients with anti-200/100 autoantibodies that we analyzed. Although the presence of membrane attack complex on non-necrotic fibers has previously been reported in immune-mediated myopathies (19), this is not a general feature of these disorders; in multiple studies of anti-SRP myopathy, membrane attack complex was observed on non-necrotic fibers in only 1 of 7 (3), none of 6 (5), and 1 of 3 (6) muscle biopsy specimens. It should be noted that membrane attack complex deposition on non-necrotic myofibers has also been reported to occur in some dystrophies (17), and that membrane attack complex deposition on blood vessels and muscle fibers may be secondary to membrane damage rather than a primary pathologic event.

Finally, 4 of 8 of the available biopsy specimens included myofibers with sarcolemmal class I MHC staining. This is a characteristic feature of immune-mediated myopathies and is rare or absent in biopsy specimens from patients with muscular dystrophies and other muscle and nerve disorders (20, 21). By comparison, results of studies evaluating class I MHC staining in patients with antibodies to SRP have been mixed; one study noted class I MHC–positive fibers in 2 of 3 patients (6), a second study showed these fibers in 3 of 6 patients (3), and a third study showed the fibers in none of 6 patients (5).

Interestingly, 2 recent reports describe patients in whom a necrotizing myopathy developed during statin treatment and progressed despite discontinuation of the myotoxic medication (22, 23). In the larger of the 2 series, Grable-Esposito et al (23) described 25 patients who experienced the development of an apparently immune-mediated, statin-associated necrotizing myopathy that shares many of the clinical features observed in our cohort of anti-200/100 autoantibody–positive patients. For example, this group of patients had proximal muscle weakness, included men and women in almost equal numbers, had a mean CK level of 8,203 IU/liter, required multiple immunosuppressive medications to achieve improved strength, and experienced a relapse upon tapering of immunosuppressive medications. The muscle biopsy specimens from 8 similar patients were analyzed in detail by Needham and colleagues (22). Whereas 8 of 8 of the biopsy specimens they studied had increased class I MHC expression on the surface of non-necrotic muscle fiber, only 4 of 8 biopsy specimens from anti-200/100 autoantibody–positive patients were positive for class I MHC. Furthermore, in contrast to our findings in anti-200/100 autoantibody–positive patients, those investigators did not observe membrane attack complex deposition on the surface of non-necrotic muscle fibers in any of the patients they studied. Although the reasons for these discrepancies remain to be elucidated, and the autoantibody profiles of the patients in these previous studies were not defined, we suspect that the cohorts of patients described by these 2 groups of investigators may represent a subset of the population of anti-200/100 autoantibody–positive patients, the majority of whom also experienced the development of an immune-mediated myopathy after starting treatment with a statin medication.

In conclusion, we have identified a group of patients with a necrotizing myopathy and a novel anti-200/100 autoantibody specificity. Interestingly, development of this phenotype is associated with exposure to statin medications. In addition to the presence of autoantibodies, all of the patients responded to immunosuppression, and many experienced a flare of weakness when this treatment was tapered, which supports our hypothesis that this is an immune-mediated myopathy. The presence of class I MHC on the surface of non-necrotic fibers also suggests that this process is immune-mediated. Indeed, we propose that those patients with necrotizing myopathies and anti-200/100 autoantibodies most likely have an autoimmune disease that should be treated with immunosuppressive medication. Future work will be directed toward identifying the autoantigens targeted by the anti-200/100 antibodies; based on the observation that these proteins were always immunoprecipitated together, we anticipate that they may be members of a protein complex.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Mammen had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Christopher-Stine, Casciola-Rosen, Hong, Chung, Mammen.

Acquisition of data. Christopher-Stine, Casciola-Rosen, Hong, Chung, Corse, Mammen.

Analysis and interpretation of data. Christopher-Stine, Casciola-Rosen, Hong, Chung, Corse, Mammen.

Acknowledgements

We thank Dr. Jennifer Mammen for critical review of the manuscript and Tonie Hines for expert technical assistance.

Ancillary