Anti–signal recognition particle autoantibody in patients with and patients without idiopathic inflammatory myopathy

Authors


Abstract

Objective

To determine the long-term outcome and associated clinical, serologic, and pathologic features in a cohort of patients with connective tissue disease (CTD) and the anti–signal recognition particle (anti-SRP) autoantibody.

Methods

Sera and clinical data were collected prospectively from consecutive adult patients with polymyositis (PM; n = 134), dermatomyositis (n = 129), or other CTDs (predominantly systemic sclerosis [SSc; n = 790]). Patients were first evaluated during 1973–2001.

Results

Nineteen patients with the anti-SRP autoantibody were identified, 16 (84%) of whom had pure PM and 3 (2 with SSc and 1 with antisynthetase syndrome) had yet to develop features of myositis after a mean followup of 4.5 years (range 2.5–6 years). More SRP-positive PM patients had severe proximal muscle weakness (50%) and muscle atrophy (67%) at initial presentation compared with antisynthetase-positive PM controls. Cardiac involvement occurred in only 2 of 16 SRP-positive PM patients (13%), and interstitial lung disease was noted in 3 of 13 SRP-positive PM patients (23%) and in the 3 SRP-positive nonmyositis patients. There was a relative lack of inflammation in muscle biopsy specimens from the SRP-positive PM cohort. Other autoantibodies in the SRP-positive patients included Ro/SSA (4 patients), Th/To (1 patient), and anti–PL-12 (1 patient). Survival in the SRP-positive PM patients was comparable with that seen in the cohort of SRP-negative PM patients.

Conclusion

The anti-SRP autoantibody is not specific for PM. Severe muscle weakness and atrophy were prominent features in PM patients with anti-SRP. Cardiac involvement was less common and survival was better in patients with anti-SRP than has previously been reported.

Myositis-specific autoantibodies (MSAs) include those directed against aminoacyl–transfer RNA (aminoacyl-tRNA) synthetases (antisynthetase), signal recognition particle (SRP), and the helicase enzyme (1–4). MSAs define subgroups of idiopathic inflammatory myopathy (IIM) patients with distinguishing clinical features. Anti-SRP autoantibodies are mainly associated with polymyositis (PM), persistent elevation of the creatine kinase level, and severe myositis that is often resistant to treatment with corticosteroids and other immunosuppressive agents. Unlike patients with antisynthetase syndrome, patients with anti-SRP are reported to have a low frequency of pulmonary fibrosis, arthritis, and Raynaud's phenomenon (2), and may not have inflammatory myopathy (5, 6).

SRP, one of the most abundant and best-characterized RNP particles, consists of the 7SL RNA molecule and 6 polypeptides with molecular weights of 72 kd, 68 kd, 54 kd, 19 kd, 14 kd, and 9 kd. SRP regulates the translocation of proteins across the endoplasmic reticulum membrane during protein synthesis, and autoantibodies that react with SRPs are defined by their ability to immunoprecipitate the 7SL RNA (7, 8).

We have longitudinally followed up a large cohort of IIM and connective tissue disease (CTD) patients who had serum specimens collected. The goal of the present study was to determine the clinical, serologic, and muscle pathologic features, and long-term prognosis among patients with the anti-SRP autoantibody. We compared these findings and patient survival with those in IIM patients with and IIM patients without antisynthetase autoantibodies.

PATIENTS AND METHODS

Patients.

Serum samples were collected from consecutive adult patients with probable or definite PM (n = 134) or dermatomyositis (n = 129) according to the criteria of Bohan and Peter (9, 10) and from patients with other CTDs (systemic sclerosis [SSc] [n = 790] or overlap syndrome [n = 109]). Patients were initially evaluated in the rheumatology outpatient offices or inpatient service at the University of Pittsburgh Medical Center between May 1973 and November 2001 (the majority of IIM patients were seen after 1990). Patients with insufficient clinical information were excluded from the study. Informed consent from study participants was obtained according to Institutional Review Board protocol.

Sixteen consecutive SRP-positive PM patients and 118 SRP-negative PM controls were assessed from May 1973 to July 2002. The 118 SRP-negative PM controls were divided into 46 synthetase-positive and 72 synthetase-negative controls, and each group was compared with the SRP-positive PM patients. Participants also completed the modified Health Assessment Questionnaire (11).

Sera.

Banked serum specimens were stored at –80°C. Many samples had been tested previously for other autoantibodies. Anti-SRP and other autoantibodies were determined by RNA and protein immunoprecipitation as described below (12).

RNA immunoprecipitation.

Briefly, a 20-μl serum sample was bound overnight at 4°C to 2 mg protein A–Sepharose CL-4B beads (Amersham Biosciences, Piscataway, NJ), washed 3 times with immunoprecipitation (IP) buffer (10 mM Tris HCl, 500 mM NaCl, 0.1% Igepal CA-630), and incubated for 2 hours at 4°C with unlabeled extract from ∼6 × 106 rapidly dividing K562 cells. The beads were washed 3 times with NET-2 buffer (50 mM Tris HCl [pH 7.4], 150 mM NaCl, 0.05% Igepal CA-630) and suspended in 300 μl NET-2, 30 μl 10% sodium dodecyl sulfate, 30 μl 3M NaAc, and 2 μl of glycogen per sample. The RNA was phenol extracted and alcohol precipitated. The resultant RNA pellet was suspended in RNA sample buffer and electrophoresed at 400V on a standard-size 8% urea–polyacrylamide gel that was then silver stained.

Protein immunoprecipitation.

A 20-μl serum sample was bound overnight at 4°C to 2 mg protein A–Sepharose CL-4B beads (Amersham Biosciences), washed 3 times with IP buffer, and incubated for 2 hours at 4°C with 35S-methionine–labeled extract from ∼1 × 106 rapidly dividing K562 cells. The beads were washed 3 times with IP buffer, suspended in 2× Laemmli sample buffer, loaded on a standard-size 8% gel, and electrophoresed at 200V. The gel was enhanced with 0.5M sodium salicylate, dried, and autoradiographed for 3–6 days. Apparent molecular weights were determined by comparison with known 14C-labeled standards run on the same gel.

Clinical assessment.

Standardized data collection forms were used in the initial and subsequent prospective collection of clinical data. Severe proximal muscle weakness was defined as weakness necessitating assistance in daily activities. Proximal esophageal dysmotility was documented by a swallowing study (i.e., cine-esophagram). Cardiac involvement was defined as echocardiographic evidence of biventricular cardiomyopathy or electrocardiographic evidence of ventricular arrhythmias (conduction abnormalities) attributable to PM and not to coronary heart disease. Interstitial lung disease (ILD) was defined by the presence of pulmonary fibrosis seen by chest radiography or high-resolution computed tomography, or restrictive lung physiology as defined by pulmonary function tests (forced vital capacity <70% of predicted and forced expiratory volume in 1 second/forced vital capacity >80%). A drug trial was defined as the point from the beginning of administration of a given drug or a combination of drugs to the time at which this therapy was discontinued.

Histologic study of muscle.

Slides of initial muscle biopsy specimens from 10 of 16 SRP-positive PM patients were reviewed and compared with specimens from 17 of the 118 SRP-negative PM controls, by a neuromuscular pathologist who had no knowledge of the autoantibody status of the patients. Histologic features were systemically assessed in both SRP-positive specimens and control specimens. At minimum, frozen or paraffin sections stained with hematoxylin and eosin (H&E) were available for review. Myofiber degeneration (necrosis), regeneration, and fibrosis were identified by evaluation of H&E-stained sections. Myofibrillar disruption and pattern of myofibrillar atrophy (random, grouped, or perifascicular) were identified by NADH–tetrazolium reductase (NADH-TR)–reacted sections (20 specimens). Small, darkly stained fibers (via NADH-TR) were considered a potential “neurogenic” change. Perimysial and endomysial lymphocytic inflammation was noted, and macrophages were identified by esterase reactivity. Capillaries were not evaluated in most specimens, but staining for C5b–9 (membrane attack complex) was evaluated in 9 specimens. Myofiber size ranging from 40 μm to 70 μm in diameter was considered to be normal. To assess for potential selection bias because muscle biopsies were not reviewed in all study patients, we compared the demographic and clinical features of the SRP-positive and control PM patients with and those without muscle biopsies.

Statistical analysis.

To assess differences between SRP-positive PM patients and controls, the following analyses were used: t-tests or Wilcoxon's rank sum tests for continuous variables and chi-square tests or Fisher's exact test (2-tailed) for dichotomous variables. Differences in survival were assessed by log rank and Wilcoxon's tests. Calculations were performed using the SAS system for Windows statistical software (SAS Institute, Cary, NC). P values less than 0.05 were considered significant.

RESULTS

Identification of autoantibodies and clinical features.

Nineteen patients with the anti-SRP autoantibody were identified, and their clinical features are described in Table 1. Of the 263 patients with IIM, 16 (6%) had SRP-positive myositis. None of the 19 SRP-positive patients had dermatomyositis. Sixteen (84%) had pure PM with a median followup of 7.5 years (range 0.2–26.7 years), and 3 (2 with SSc and 1 with antisynthetase syndrome) had yet to develop features of myositis after a mean followup of 4.5 years (range 2.5–6 years). Sera from all SRP-positive patients precipitated an RNA band consistent with the 7SL RNA of the SRP complex and at least 1 of the 2 major SRP protein bands of 72 or 54 kd (Figure 1). Although sera from most patients precipitated both proteins at the time of their initial visit, sera from 4 (patients 14, 17, 18, and 19) did not strongly immunoprecipitate both the 72- and 54-kd bands. Seven patients (patients 1, 4, 5, 10, 15, 17, and 18) had 1 or more subsequent serum samples available. Of these, sera from 5 (patients 1, 4, 10, 15, and 18) showed a reduction of intensity in some immunoprecipitated bands compared with previously obtained specimens, and sera from 2 (patients 10 and 15) immunoprecipitated additional new bands (results not shown). Other antibodies found in the 19 SRP-positive patients included Ro/SSA (4 patients), Th/To (1 patient), and PL-12 (1 patient). Forty-six of 118 SRP-negative PM control patients (39%), 36 of whom were anti–Jo-1 positive, had anti–aminoacyl-tRNA synthetase autoantibodies.

Table 1. Clinical features of 19 patients with anti–signal recognition particle autoantibodies*
PatientDiagnosisAge at symptom onset/sex/raceMonth at onsetSymptom duration at diagnosis, monthsMuscle atrophy at initial visitCardiac involvementILDGI involvementArthritisOther antibodies
  • *

    Cardiac involvement determined by echocardiographic evidence of cardiomyopathy or electrocardiographic evidence of ventricular arrhythmias attributable to polymyositis (PM) and not to coronary heart disease. Interstitial lung disease (ILD) determined by the presence of pulmonary fibrosis seen by chest radiography or chest computed tomography, or restrictive lung physiology as defined by pulmonary function tests. Gastrointestinal (GI) involvement determined by proximal esophageal dysmotility documented by a swallowing study (i.e., cine-esophagram). SSc = systemic sclerosis; B = black; W = white; NT = not tested.

  • Deceased.

1SSc48/F/BFeb.14++Th/To, Ro/SSA
2SSc47/M/WJan.26+++
3PM78/F/WOct.4++
4PM42/M/WJul.2+++Ro/SSA
5PM60/F/BMar.3++NT
6PM54/M/WAug.2++
7PM57/F/WSep.7+NT
8PM46/M/WNov.2++
9PM71/M/WJun.6+NTNT
10PM46/F/WAug.5NT+
11PM60/F/WFeb.4NT
12PM43/M/BSep.16NTNT++U4/U6, Ro/SSA
13PM54/F/WMar.11+NTRo/SSA
14PM63/M/WJun.3++NT
15PM42/M/BDec.1+NT+
16PM25/M/WMar.2+NTNT
17PM28/F/BApr.8+NT
18PM40/F/WApr.3+NT
19Antisynthetase syndrome33/F/WAug.7+0+PL-12
Figure 1.

Autoradiograph of an 8% sodium dodecyl sulfate–polyacrylamide gel electrophoresis study of immunoprecipitated proteins from signal recognition particle (SRP)–positive sera, obtained using 35S-methionine–labeled K562 whole cell extract. Also shown are results from normal serum (NL), reference sera (positive for SRP, Th/To, PL-12 plus Ro, and Smith plus Ro), and a 14C molecular weight marker (Mr) (in kilodaltons).

Demographic and clinical associations.

Twelve SRP-positive PM patients (75%) were white and 4 (25%) were African American, with a 1:1 female:male ratio. As seen in Table 2, demographic features were comparable in the SRP-positive PM patients (n = 16) and the control PM patients (n = 118), with the exception of female predominance in the SRP-negative PM control group (72% versus 50%).

Table 2. Comparison of clinical features in SRP-positive PM versus synthetase-negative and synthetase-positive PM controls*
Clinical featureSRP + PM cases (n = 16)PM controls
Synthetase− (n = 72)Synthetase+ (n = 46)
  • *

    Except where indicated otherwise, values are the number (%). SRP = signal recognition particle antibody; CK = creatine kinase; M-HAQ = modified Health Assessment Questionnaire.

  • P < 0.0001 versus SRP+ PM group.

  • P < 0.002 versus SRP+ PM group.

  • §

    P < 0.05 versus SRP+ PM group.

Age at symptom onset, mean ± SD years50.8 ± 14.450.6 ± 12.946.9 ± 14.2
Female8 (50)51 (71)34 (74)
Race, % white757989
Duration of symptoms at diagnosis, median months484
Duration of followup, median years7.55.36.2
Deceased4 (25)27/64 (42)20/42 (48)
Severe proximal muscle weakness8 (50)5/71 (7)4 (9)
Muscle atrophy at initial visit10/15 (67)26/62 (42)5/44 (11)
Muscle atrophy at any time12/15 (80)29/68 (43)9/45 (20)
Myalgia9/15 (60)39/71 (55)34/45 (76)
Raynaud's phenomenon3 (19)24 (33)31 (67)
Sicca symptoms8 (50)29 (40)22 (48)
Inflammatory arthritis2 (13)14/41 (34)26/34 (76)
Cardiac involvement2 (13)12 (17)12 (26)
Esophageal dysmotility6/6 (100)7/24 (29)6/11 (55)
Interstitial lung disease3/13 (23)11/54 (20)29/40 (73)
PM-associated malignancy median0 (0)3/68 (4)1/45 (2)
CK at diagnosis, median × normal value11.35.46.9
Maximum CK, median × normal value26.510.6§27.0
M-HAQ score, mean ± SD1.2 ± 0.71.0 ± 0.70.8 ± 0.5

In the SRP-positive PM subset, the frequency of severe proximal muscle weakness at initial evaluation (50%) was increased compared with that in the 2 control PM groups (P < 0.0001 versus synthetase-negative control PM patients; P = 0.001 versus synthetase-positive control PM patients [Table 2]). The frequency of muscle atrophy at initial presentation was significantly increased in the SRP-positive subset compared with the synthetase-positive control group (P < 0.0001). Esophageal dysmotility was noted in 6 of 6 SRP-positive PM patients who had swallowing studies performed. Cardiac involvement occurred in only 2 of 16 SRP-positive PM patients (13%), and there was no association of malignancy with anti-SRP. Two patients later developed neoplasms (gastrointestinal adenocarcinomas) 21 and 28 years, respectively, after PM onset. ILD was noted in 3 of 13 SRP-positive PM patients (23%) and all 3 SRP-positive nonmyositis patients, but, as would be expected, the synthetase-positive control patients had a significantly higher prevalence of ILD (73%), inflammatory arthritis (76%), and Raynaud's phenomenon (67%).

There were 4 deaths among the 16 patients in the SRP-positive PM group, 2 of which may have been attributable to PM. These 2 patients had underlying chronic obstructive lung disease and died of pneumonia at ages 63 and 78 years, respectively (2 years and 6 months, respectively, after onset of myositis). A third patient with a history of coronary artery disease died at age 71 of cardiac arrest. The fourth patient died of an unknown cause. As shown in Figure 2, survival among the SRP-positive PM patients was not significantly different from that in the cohort of 118 SRP-negative PM patients. The 5-year cumulative survival rate was 86% in the SRP-positive PM patients, 83% in the synthetase-negative controls, and 75% in the synthetase-positive controls.

Figure 2.

Cumulative survival among signal recognition particle (SRP)–positive polymyositis (PM) patients, synthetase (synth)–negative PM patients, and synthetase-positive PM patients.

Many myositis patients with anti-SRP exhibited persistent muscle weakness and resistance to treatment, usually requiring trials of several immunosuppressive agents. All patients were treated with prednisone and most required the addition of methotrexate, cyclosporine, or azathioprine. Therapy also included tacrolimus, intravenous immunoglobulin, cyclophosphamide, and infliximab. More than half of the SRP-positive PM patients (63%) required at least 3 drug trials for resistant myositis. Combination immunosuppressive therapy was common. Although SRP-positive PM patients had severe myopathy, favorable results with treatment were achieved in one-third (patients 4, 6, 9, 11, 16, and 18), with normal creatine kinase levels, improved muscle strength, and a minimal corticosteroid requirement. Not all had normal strength and some required multiple immunosuppressive agents to control their myopathy, but normal strength was recovered with only prednisone treatment in 2. No specific immunosuppressive agent or treatment regimen was found to be particularly beneficial for all patients.

Muscle pathologic findings.

Table 3 summarizes the muscle biopsy data. Although all muscle biopsy specimens contained atrophic myofibers, more SRP-negative PM controls (47%) had hypertrophic myofibers compared with SRP-positive patients (10%), but the difference was not statistically significant (P = 0.09). Biopsy specimens from the SRP-positive PM patients had significantly less endomysial inflammation than those from the SRP-negative controls (P = 0.007). There was no perimysial myofibrillar necrosis or degeneration in the SRP-positive patients, but endomysial myofiber degeneration and regeneration were common in both groups. Among the PM patients whose muscle biopsies were reviewed, the baseline characteristics and clinical features were not significantly different between the SRP-positive and SRP-negative patients, except for more severe proximal muscle weakness (P = 0.02) and a shorter interval between the onset of muscle weakness and a muscle biopsy (P = 0.03) in the SRP-positive PM patients. Only 4 (24%) of the 17 SRP-negative PM control patients whose biopsy specimens were reviewed had synthetase antibodies. Although the control PM patients whose muscle biopsies were reviewed were significantly older at symptom onset and at diagnosis compared with control patients whose muscle biopsies were not reviewed (mean 56.2 years versus 48.0 years; P = 0.03), there was no difference in disease duration prior to diagnosis, demographic characteristics, or other clinical features.

Table 3. Comparison of muscle pathologic features in SRP+ PM patients and PM control patients*
FindingSRP+ PM (n = 10)PM controls (n = 17)P
  • *

    Values are the number (%). SRP = signal recognition particle; PM = polymyositis; H&E = hematoxylin and eosin.

  • Stain not used on all muscle biopsy samples.

Myofiber size   
 Atrophic10 (100)17 (100)1.0
 Mixed, atrophic and hypertrophic1 (10)8 (47)0.09
 Small, round clusters of myofibers2 (20)5 (29)0.7
Inflammation (H&E)   
 Perimysial3 (30)10 (59)0.2
 Endomysial2 (20)13 (76)0.007
 Macrophages (esterase)2/4 (50)6/6 (100)0.1
Degeneration (necrosis) of myofibers (H&E)   
 Perimysial0 (0)7 (41)0.03
 Endomysial7 (70)14 (82)0.6
Regeneration of myofibers (H&E)   
 Perimysial2 (20)6 (35)0.7
 Endomysial9 (90)15 (88)1.0
Disruption of myofibrillar architecture (NADH)4/6 (67)14/14 (100)0.08
Fibrosis   
 Perimysial1 (10)4 (24)0.6
 Endomysial2 (20)5 (29)0.7
Atrophy of myofibers   
 Random8 (80)14 (82)1.0
 Grouped5 (50)6 (35)0.5
 Neurogenic, small dark (NADH)5/6 (83)6/14 (43)0.2
 Perifascicular1 (10)6 (35)0.2
Capillaries   
 C5b–9 reactivity or membrane attack complex2/3 (67)3/6 (50)1.0
Vasculitis01 (6%)1.0

DISCUSSION

The prevalence of anti-SRP antibodies in our prospective IIM cohort is 6%, similar to earlier reports of 4–5% (2, 3, 6). Although the presence of 2 different MSAs in the same patient is distinctly unusual (13, 14) and SRP is predominantly specific for PM (1–3), we have identified another patient with 2 MSAs (anti-SRP and anti–PL-12). Furthermore, 3 of the 19 SRP-positive patients in our cohort (16%) have no evidence of myositis. Anti-Ro/SSA was detected in 4 (21%) of the 19 SRP-positive patients and anti-Th/To was found in 1 SRP-positive patient, who had limited SSc and did not have myositis. The latter antibody is found predominantly in patients with limited SSc and is associated with pulmonary hypertension and pulmonary fibrosis (15).

ILD was present in the 3 SRP-positive patients without myositis, but these patients had a CTD or autoantibody associated with ILD. However, 3 SRP-positive PM patients also had evidence of ILD, an unexpected finding. We did not observe overt cardiac involvement in our SRP-positive PM patients as previously reported, but this discrepancy may be due to the varying definitions of cardiac involvement in different studies as well as a lower frequency of African Americans in our cohort (2, 3). Love et al (3) noted palpitations in all 7 of their SRP-positive PM patients, but there was no further diagnostic investigation described. Targoff et al (2) reported that 4 of their 12 SRP-positive PM patients had cardiac involvement (arrhythmias in 3 and cardiomyopathy with fibrosis in 1).

Severe proximal muscle weakness and muscle atrophy were prominent findings at initial presentation in the SRP-positive PM patients compared with the SRP-negative PM control group. This could not be explained by a delay in diagnosis in the SRP-positive PM patients. Proximal esophageal dysmotility was a significant feature in SRP-positive PM patients who had swallowing studies performed, although the majority of patients did not have these diagnostic studies. Therefore, it is prudent to screen for proximal esophageal dysmotility, especially in older PM patients with severe proximal muscle weakness, who are at increased risk of aspiration-induced pneumonia and increased mortality (16). Two of the 4 deaths (both elderly patients) in the SRP-positive PM group were related to pneumonia; 1 of these patients had esophageal dysmotility with aspiration pneumonia. Although there were 2 deaths related to PM, survival in our SRP-positive PM patients was comparable with that in the cohort of 118 SRP-negative PM patients, in contrast with earlier reports (2, 3).

Although we were unable to assess treatment outcome in this retrospective study, it would appear that the SRP-positive PM patients had refractory disease since ∼50% of them underwent at least 3 drug trials. However, one-third of our SRP-positive PM patients had favorable responses to immunosuppressive therapy.

Similar to findings reported by Miller et al (17), we noted minimal endomysial inflammation in specimens from SRP-positive PM patients, and there was no significant increase in the degeneration, regeneration, myofibrillar disruption, or red/round clusters of the myofibers in our SRP-positive PM group. Interestingly, there was a paucity of hypertrophic myofibers in the SRP-positive PM patients compared with the SRP-negative PM controls, suggesting that persistent muscle weakness in these patients may be due to the lack of compensatory muscle hypertrophy. There was a potential selection bias because only the available muscle biopsy specimens were reexamined and the SRP-negative PM controls were older at symptom onset. However, disease duration at myositis diagnosis and other demographic and clinical features were comparable in the SRP-positive PM and SRP-negative PM subgroups that had muscle biopsy results reviewed.

In conclusion, anti-SRP antibodies occurred in a small group of patients with PM who had early severe proximal muscle weakness and muscle atrophy at initial presentation. Esophageal dysmotility was prominent. Despite the presumed autoimmune etiology of the myopathy associated with the anti-SRP autoantibody, endomysial inflammation is uncommon and no specific immunosuppressive regimen is currently tailored to treat the SRP-positive PM patient. However, the prognosis of SRP-positive PM is better than previously reported: one-third of our patients responded to therapy and their survival is comparable with that of the PM controls. Although anti-SRP remains an autoantibody specific for PM, it is occasionally detected in patients with other immunologic syndromes in the absence of PM (5).

Acknowledgements

The authors thank Beverly Knasko and Rebecca Byles for secretarial assistance.

Ancillary