Dr. Christopher-Stine has received consulting fees, speaking fees, and/or honoraria from Novartis, MedImmune, and Questcor (less than $10,000 each). Drs. Christopher-Stine and Mammen have patented and licensed a test for anti–3-hydroxy-3-methylglutaryl-coenzyme A reductase antibodies.
Antibody levels correlate with creatine kinase levels and strength in anti–3-hydroxy-3-methylglutaryl-coenzyme A reductase–associated autoimmune myopathy
Article first published online: 28 NOV 2012
Copyright © 2012 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 64, Issue 12, pages 4087–4093, December 2012
How to Cite
Werner, J. L., Christopher-Stine, L., Ghazarian, S. R., Pak, K. S., Kus, J. E., Daya, N. R., Lloyd, T. E. and Mammen, A. L. (2012), Antibody levels correlate with creatine kinase levels and strength in anti–3-hydroxy-3-methylglutaryl-coenzyme A reductase–associated autoimmune myopathy. Arthritis & Rheumatism, 64: 4087–4093. doi: 10.1002/art.34673
- Issue published online: 28 NOV 2012
- Article first published online: 28 NOV 2012
- Accepted manuscript online: 29 AUG 2012 03:08PM EST
- Manuscript Accepted: 9 AUG 2012
- Manuscript Received: 12 APR 2012
- NIH. Grant Numbers: K23-AR-053197, K08-NS062890, K08-AR-054783
- Huayi and Siuling Zhang Discovery Fund
- Ira T. Fine Discovery Fund
- Donald B. and Dorothy L. Stabler Foundation
Autoantibodies recognizing 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) are found in patients with statin-associated immune-mediated necrotizing myopathy and, less commonly, in statin-unexposed patients with autoimmune myopathy. The main objective of this study was to define the association of anti-HMGCR antibody levels with disease activity.
Anti-HMGCR levels, creatine kinase (CK) levels, and strength were assessed in anti-HMGCR–positive patients. Associations of antibody level with CK level and strength at visit 1 were analyzed in 55 patients, 40 of whom were exposed to statins. In 12 statin-exposed and 5 statin-unexposed patients with serum from 5 serial visits, the evolution of antibody levels, CK levels, and strength was investigated.
Antibody levels were associated with CK levels (P < 0.001), arm strength (P < 0.05), and leg strength (P < 0.05) at visit 1, but these associations were only significant among statin-exposed patients in stratified analyses. With immunosuppressive treatment over 26.2 ± 12.6 months (mean ± SD), antibody levels declined (P < 0.05) and arm abduction strength improved (P < 0.05) in the 17 patients followed up longitudinally. The separate analysis showed that statin-exposed patients developed decreased antibody levels (P < 0.01), decreased CK levels (P < 0.001), improved arm strength (P < 0.05), and improved hip flexion strength (P < 0.05) with treatment. Anti-HMGCR antibody levels did not normalize in any patient.
In the entire cohort, initial anti-HMGCR levels correlated with indicators of disease activity; with immunosuppressive treatment, antibody levels declined and arm strength improved. Statin-exposed patients had significant improvements in CK levels and strength whereas statin-unexposed patients did not, suggesting a phenotypic difference between statin-exposed and statin-unexposed anti-HMGCR–positive patients.
In patients with autoimmune myopathy, unique autoantibodies are associated with distinct clinical phenotypes (1). For example, antibodies recognizing histidyl–transfer RNA synthetase (i.e., Jo-1) are found in patients with a syndrome characterized by myositis, interstitial lung disease, nonerosive arthritis, fever, and “mechanic's hands.” In contrast, antibodies against signal recognition particle (SRP) are associated with a severe necrotizing myopathy without prominent involvement of other organ systems. The precise relationship between myositis autoantibodies and disease pathology is unknown. However, recent studies provide indirect evidence that antibody levels may reflect disease activity. For example, anti-SRP antibody levels have been found to correlate with creatine kinase (CK) levels and may normalize during periods of remission (2). Recent studies have also demonstrated a link between anti–Jo-1 antibody levels and indicators of muscle, joint, and lung disease activity (3).
We have discovered that autoantibodies recognizing 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR)—the target of statin medications—are associated with an immune-mediated myopathy characterized by myofiber necrosis and very high CK levels (4–7). Although anti-HMGCR myopathy is associated with statin exposure in patients age >50 years, approximately one-fourth of anti-HMGCR–positive patients develop a similar myopathic process in the absence of statins. Interestingly, our prior analysis suggested that statin-exposed and statin-unexposed anti-HMGCR–positive patients may have slightly different phenotypes. In addition to being younger, statin-unexposed anti-HMGCR–positive patients had higher CK levels and were more likely to be African American than were statin-exposed patients. The current study investigated anti-HMGCR antibody levels, serum CK levels, and muscle strength at an initial study visit and over time in both statin-exposed and statin-unexposed patients.
PATIENTS AND METHODS
Between May 2002 and July 2011, 1,006 patients who were seen by a neurologist or rheumatologist at the Johns Hopkins Myositis Center and who had suspected myopathy as defined by proximal muscle weakness, elevated CK levels, findings of myopathy on electromyography, muscle edema on magnetic resonance imaging, and/or myopathic features on muscle biopsy were enrolled in a longitudinal study of the relationship between autoantibodies and clinical phenotypes. Serum samples were collected from each study patient at the time of enrollment as well as at subsequent clinic visits. The initial sera from all patients were screened for the presence of anti-HMGCR autoantibodies by enzyme-linked immunosorbent assay (ELISA); as previously validated elsewhere, positive levels were defined as those >0.367 normalized arbitrary units (AU) relative to a calibrator serum (6). ELISA-positive sera were confirmed by immunoprecipitating in vitro–transcribed and –translated HMGCR protein as previously described (6).
At each visit, arm abduction and hip flexion strength was assessed by the examining physician and scored on a modified 10-point scale adapted from the manual muscle strength testing scale used by the International Myositis Assessment and Clinical Studies group (8) (10 = normal strength, 9 = holds test position against strong pressure, 8 = holds test position against moderate pressure, 7 = holds test position against slight pressure, 5 = holds test position against gravity [no added pressure], 1.5 = <100% range of motion against gravity, 0 = no motion against gravity). For each patient, the serial strength measurements were made by the same physician (ALM or LC-S). For the purposes of analyses, right- and left-side measurements for arm and hip strength were combined (possible range 0–20). Serum CK levels were included for analysis if obtained within 3 weeks of the date at which the anti-HMGCR titer was obtained.
Standard protocol approvals and patient consents.
This study was approved by the Johns Hopkins Institutional Review Board, and written informed consent was obtained from each participant.
Quantitative anti-HMGCR ELISA.
Our previously established screening anti-HMGCR ELISA was modified by including samples of the calibrator serum at serial dilutions from 1:100 to 1:3,200 to determine the linear range of the assay. In the quantitative anti-HMGCR ELISA, serum samples from all patients were initially diluted to 1:800. In several cases in which a given serum sample diluted to 1:800 was not within the linear range of the assay, a further dilution to 1:1,600 was performed. All antibody levels were expressed as a fraction of the calibrator serum at the same dilution, the level of which was arbitrarily set to 1. The initial serum samples from all anti-HMGCR–positive patients were analyzed on the same 96-well ELISA plate. For studies of serial anti-HMGCR levels, all samples from an individual patient were analyzed at the same dilution on the same ELISA plate.
All analyses were completed using SPSS 19.0, and P values less than 0.05 were considered significant. Initial analyses compared study variables of interest (anti-HMGCR levels, serum CK levels, arm and hip strength) across the 2 groups of statin-exposed and statin-naive patients using independent sample t-tests. Next, a series of linear regressions was examined with anti-HMGCR values as the main predictor and serum CK levels, arm abduction strength scores, and hip flexion strength scores as outcome variables. These linear regression models were completed for the full sample of patients as well as stratified by statin-exposed and statin-naive patients. To examine trends in serial measurements (visits 1–5) of study variables of interest, a series of repeated-measures analysis of variance (ANOVA) models was examined for the full cohort of patients as well as for patient groups stratified by statin exposure status.
Identification of patients with anti-HMGCR–associated myopathy.
Screening by ELISA followed by confirmation with immunoprecipitation revealed that 55 of 1,006 patients (5.5%) enrolled at the Johns Hopkins Myositis Center were positive for anti-HMGCR antibodies. Anti-HMGCR myopathy developed following exposure to statins in 40 of 55 patients (72.7%). All anti-HMGCR–positive patients had a history of elevated serum muscle enzyme levels, with a mean ± SD maximum CK level of 10,104 ± 6,973 IU/liter. Fifty-two patients had symmetric proximal muscle weakness at the time of the initial study serum collection; the remaining 3 patients had had a history of weakness that had resolved with treatment at the initial study visit. Muscle biopsies were performed in 53 of the anti-HMGCR–positive patients. In 38 of these 53 patients (71.7%), there was a predominantly necrotizing myopathy with minimal or no lymphocytic infiltrate. In 10 of these patients (18.9%), there was prominent endomysial and/or perivascular inflammation along with necrotic myofibers. In 2 of these patients (3.8%), prominent myofiber necrosis was accompanied by vacuoles. Two of these patients (3.8%) exhibited little or no myofiber necrosis on biopsy of inflammatory muscle. One of these patients (1.9%) had a normal muscle biopsy sample despite statin exposure, a history of proximal muscle weakness, and a maximum serum CK level of >10,000 IU/liter.
Association of anti-HMGCR titers with measures of disease activity at the time of initial evaluation.
Anti-HMGCR levels, serum CK levels, arm abduction strength, and hip flexion strength were assessed for all anti-HMGCR–positive patients at visit 1 (Table 1). Although there was a trend for antibody levels to be higher in the statin-unexposed group, no statistically significant differences in clinical features were noted between the 2 groups. When statin-exposed and statin-unexposed patients were examined together, there was a statistically significant association between anti-HMGCR level and serum CK level at the initial visit. Similarly, there was a significant association between anti-HMGCR level and both arm abduction strength and hip flexion strength; higher antibody levels were associated with weakness and lower antibody titers were associated with greater strength. When analyzed separately, these associations were maintained in the 40 anti-HMGCR–positive patients with statin exposure (Table 2). However, in the 15 patients without statin exposure, a possible trend toward a correlation between anti-HMGCR level and either CK level or arm abduction strength did not reach statistical significance. In contrast to patients with statin exposure, there was no evident trend toward correlation between anti-HMGCR level and hip flexion strength in statin-unexposed patients.
|n||Mean ± SD|
|All anti-HMGCR–positive patients|
|Anti-HMGCR level, normalized AU||55||0.96 ± 0.30|
|Serum CK level, IU/liter||51||3,675 ± 3,649|
|Arm abduction strength, 0–20||53||16.45 ± 3.98|
|Hip flexion strength, 0–20||53||11.57 ± 6.61|
|Statin-exposed anti-HMGCR–positive patients|
|Anti-HMGCR level, normalized AU||40||0.91 ± 0.31|
|Serum CK level, IU/liter||36||3,566 ± 3,445|
|Arm abduction strength, 0–20||38||16.32 ± 4.31|
|Hip flexion strength, 0–20||38||11.74 ± 6.40|
|Statin-naive anti-HMGCR–positive patients|
|Anti-HMGCR level, normalized AU||15||1.08 ± 0.22|
|Serum CK level, IU/liter||15||3,937 ± 4,218|
|Arm abduction strength, 0–20||15||16.80 ± 3.10|
|Hip flexion strength, 0–20||15||11.13 ± 7.33|
|β (95% CI)||P|
|All anti-HMGCR–positive patients|
|Anti-HMGCR titer and CK levels||0.522 (0.225, 0.609)||<0.001|
|Anti-HMGCR titer and arm abduction strength||−0.345 (−0.624, –0.092)||<0.05|
|Anti-HMGCR titer and hip flexion strength||−0.312 (−0.589, –0.059)||<0.05|
|Statin-exposed anti-HMGCR–positive patients|
|Anti-HMGCR titer and CK levels||0.633 (0.276, 0.641)||<0.001|
|Anti-HMGCR titer and arm abduction strength||−0.355 (−0.712, –0.057)||<0.05|
|Anti-HMGCR titer and hip flexion strength||−0.385 (−0.641, –0.075)||<0.05|
|Statin-naive anti-HMGCR–positive patients|
|Anti-HMGCR titer and CK levels||0.316 (−0.314, 1.165)||0.271|
|Anti-HMGCR titer and arm abduction strength||−0.443 (−1.101, 0.098)||0.113|
|Anti-HMGCR titer and hip flexion strength||−0.102 (−1.118, 0.803)||0.728|
Response to treatment and evolution of anti-HMGCR levels, CK levels, and muscle strength during therapy.
Serum samples from 5 consecutive clinic visits were available from 17 of the 55 anti-HMGCR–positive patients; at the time of the analysis, the remaining 38 patients had ≤4 visits to the Johns Hopkins Myositis Center and were not included in the longitudinal analysis. Summaries of the clinical characteristics at visits 1 and 5 for all 17 patients from whom at least 5 serum samples were available are shown in Table 3. The interval between visits was not standardized, and the mean ± SD time between visit 1 and visit 5 was 26.2 ± 12.6 months (25.2 ± 14.7 months and 28.8 ± 5.4 months for the statin-exposed and statin-unexposed groups, respectively). Summaries of immunosuppressive treatment at visits 1 through 5 for all 17 patients are available at http://www.hopkinsmyositis.org/wp-content/uploads/2012/09/Werner-et-al-Arthritis-and-Rheumatism-2012-Supplemental-Table-1.pdf.
|Patient/age/sex||Race||Statin use||Disease duration, months†||Treatment at visit 1||Visit 1 deltoid MMT score/hip flexor MMT score/CK level, IU/liter||Treatment at visit 5||Visit 5 deltoid MMT score/hip flexor MMT score/CK level, IU/liter||Disease duration, months‡|
|1/72/M||White||Yes||6.5||None||10/8/NA||AZA 150 mg/day||10/9/174||61|
|2/70/M||White||Yes||3||None||7/1.5/7,675||IVIG every 6–8 weeks; MTX 25 mg/week; pred. 5 mg/day||10/10/53||15|
|3/48/F||Black||Yes||10||Pred. 60 mg/day||5/5/3,843||Pred. 1 mg/day; MTX 25 mg/week||8/8/6,332||35|
|4/59/F||White||Yes||26||Pred. 70 mg/day; tacrolimus 6 mg/day||8/5/378||Pred. 10 mg/day; AZA 200 mg/day; tacrolimus 3 mg/day||NA/NA/NA||33|
|5/59/F||White||Yes||58||Pred. 40 mg/day; MMF 2,000 mg/day||NA/NA/5,880||MMF 1,500 mg/day||10/10/400||31|
|6/66/F||White||Yes||6||Pred. 30 mg/day; AZA 200 mg/day||1.5/1.5/2,210||AZA 200 mg/day||9/9/315||26|
|7/50/M||White||Yes||44||Pred. 40 mg/day||10/1.5/2,660||None||10/8/278||30|
|9/69/M||White||Yes||42||None||8/8/2,510||AZA 125 mg/day||10/9/164||24|
|10/51/F||White||Yes||27||None||9/8/3,700||Pred. 10 mg/day; MMF 3,000 mg/day||9/NA/475||13|
|11/67/M||White||Yes||12||Pred. 10 mg/day; MMF 1,000 mg/day||10/9/2,560||Pred. 5 mg/day; MMF 2,000 mg/day||10/10/237||10|
|12/60/F||Black||Yes||73||None||10/8/4,336||Pred. 15 mg/day||10/8/1,081||7|
|13/35/F||White||No||26||Pred. 50 mg/day; AZA 150 mg/day||10/10/1,192||Pred. 30 mg/day; MMF 3,000 mg/day||9/1.5/2,944||23|
|14/40/F||Asian||No||3||None||9/5/13,504||Pred. 9 mg/day||10/10/444||37|
|15/24/F||Black||No||27||Pred. 80 mg/day||7/8/7,093||Pred. 60 mg every other day||1.5/1.5/16,802||31|
|16/47/F||White||No||17||Pred. 40 mg/day; MTX 10 mg/week||7/1.5/749||Pred. 5 mg/day; MTX 10 mg/week||10/10/472||27|
|17/28/F||Black||No||288||Pred. 7.5 mg/day||10/1.5/3,130||Pred. 10 mg/day; MMF 3,000 mg/day; IVIG every 8 weeks||NA/NA/308||26|
Treatment plans were individualized and based on each patient's response to therapy. However, there was not an obvious difference in treatment strategies between the statin-exposed and statin-unexposed groups. Simultaneous administration of prednisone and 2 other immunosuppressive agents at some time during the course was required in 7 of 12 statin-exposed patients (58%) and in 3 of 5 statin-unexposed patients (60%). Prednisone and 1 other immunosuppressive agent were required in 3 of 12 statin-exposed patients (25%) and in 2 of 5 statin-unexposed patients (40%). One of the 12 statin-exposed patients (8%) required only prednisone, and 1 of the 12 statin-exposed patients (8%) required an immunosuppressive agent other than prednisone. At visit 5, the mean ± SD daily prednisone dose was significantly lower in the 12 statin-exposed patients than in the 5 statin-unexposed patients (3.8 ± 5.2 mg versus 16.8 ± 12.2 mg; t = 3.17, P < 0.01). By visit 5, all immunosuppressive therapy had been successfully tapered in 2 of the 12 statin-exposed patients (16.7%) and none of the statin-naive patients.
It is important to note that statins were discontinued among the statin-exposed patients prior to or at visit 1 except in a single individual (patient 9), who continued taking a statin while receiving immunosuppressive therapy. Despite the statin, this patient had gradually increasing strength and decreasing muscle enzyme levels.
Serum CK levels, arm abduction strength, and hip flexion strength were monitored over the course of 5 clinic visits in each of the 17 patients (Table 4). Repeated-measures ANOVAs of within-patient effects were performed to assess the evolution of anti-HMGCR levels, CK levels, and proximal muscle strength over time in those patients for whom complete data were available for the clinical parameter at each visit (Table 4). When all patients were analyzed together, there was a statistically significant decrease in antibody levels and a statistically significant improvement in arm abduction strength over the course of the study. When analyzed as a separate group, the statin-exposed patients were found to have significant decreases in anti-HMGCR level (from a mean ± SD of 1.05 ± 0.39 normalized AU to 0.79 ± 0.32 normalized AU) and serum CK level (from 4,835 ± 4,257 IU/liter to 878 ± 1,829 IU/liter). Similarly, the statin-exposed patients had increases in arm abduction strength (from 15.90 ± 4.36 to 19.27 ± 1.35, corresponding to a Medical Research Council [MRC] grade of ∼4/5 to nearly 5/5) and hip flexion strength (from 10.65 ± 5.94 to 18.1 ± 1.91, corresponding to an MRC grade of ∼3/5 to 4+/5) over the period of study.
|Visit 1||Visit 2||Visit 3||Visit 4||Visit 5||N†||F||P|
|Anti-HMGCR titer, normalized AU||1.09 ± 0.43 (28)||0.95 ± 0.37 (28)||0.94 ± 0.36 (25)||0.92 ± 0.34 (21)||0.88 ± 0.42 (17)||17||3.44||<0.05|
|Serum CK level, IU/liter||4,990 ± 4,485 (26)||2,952 ± 3,577 (28)||2,430 ± 4,610 (24)||2,082 ± 3,823 (18)||1,915 ± 4,280 (16)||12||2.00||0.111|
|Arm strength, 0–20||16.31 ± 4.02 (26)||18.30 ± 2.64 (27)||18.91 ± 2.02 (22)||19.33 ± 1.37 (18)||18.2 ± 4.38 (15)||11||3.03||<0.05|
|Hip strength, 0–20||10.27 ± 6.16 (26)||14.05 ± 6.81 (28)||14.20 ± 6.70 (22)||14.86 ± 6.68 (18)||16.21 ± 5.86 (14)||10||1.80||0.151|
|Anti-HMGCR titer, normalized AU||1.05 ± 0.39 (21)||0.95 ± 0.40 (21)||0.85 ± 0.28 (18)||0.84 ± 0.27 (15)||0.79 ± 0.32 (12)||12||4.08||<0.01|
|Serum CK level, IU/liter||4,835 ± 4,257 (19)||2,557 ± 2,850 (21)||1,179 ± 1,510 (17)||1,310 ± 2,682 (14)||878 ± 1,829 (11)||9||9.05||<0.001|
|Arm strength, 0–20||15.90 ± 4.36 (20)||18.29 ± 2.78 (21)||18.94 ± 1.89 (17)||19.29 ± 1.49 (14)||19.27 ± 1.35 (11)||10||3.56||<0.05|
|Hip strength, 0–20||10.65 ± 5.94 (20)||14.69 ± 6.83 (21)||14.28 ± 7.01 (16)||17.04 ± 4.33 (14)||18.1 ± 1.91 (10)||8||4.02||<0.05|
|Anti-HMGCR titer, normalized AU||1.19 ± 0.56 (7)||0.94 ± 0.27 (7)||1.17 ± 0.47 (7)||1.11 ± 0.43 (6)||1.09 ± 0.59 (5)||5||1.27||0.322|
|Serum CK level, IU/liter||5,412 ± 5,400 (7)||4,135 ± 5,320 (7)||5,468 ± 7,758 (7)||4,784 ± 6,256 (4)||4,194 ± 7,133 (5)||3||0.50||0.74|
|Arm strength, 0–20||17.67 ± 2.34 (6)||18.33 ± 2.34 (6)||18.80 ± 2.68 (5)||19.5 ± 1.00 (4)||15.25 ± 8.22 (4)||1||ND||ND|
|Hip strength, 0–20||9.00 ± 7.29 (6)||12.14 ± 6.87 (7)||14.00 ± 6.36 (6)||7.25 ± 8.50 (4)||11.50 ± 9.81 (4)||2||0.31||0.858|
In the statin-unexposed group, the mean anti-HMGCR level did not decline. The small number of statin-unexposed patients for whom complete data on CK level and strength at each of the 5 visits were available precluded a more definitive analysis by the repeated-measures ANOVA method. Although there was no evidence of improvement in any of the mean values for any clinical parameters over the course of the study in this group (Table 4), Table 3 demonstrates that 2 statin-unexposed patients (patients 14 and 16) did experience improved muscle strength and decreased CK levels between visits 1 and 5, suggesting a variability in response to treatment in this group.
Despite the decline in anti-HMGCR antibody level over time in the statin-exposed patients, the antibody levels did not normalize in any of the patients, including those who regained full strength and whose immunosuppressive medications could be discontinued. For example, statin-exposed patient 8 (see Table 3) presented at visit 1 with no prior treatments after 9 months of progressive proximal muscle weakness, a serum CK level of 3,330 IU/liter, and an anti-HMGCR level of 1.445 normalized AU. Although his statin medication was stopped at the first visit, his weakness worsened and his CK level increased to 4,197 IU/liter over the next 2 months. A biopsy was then performed, revealing a necrotizing myopathy. He was subsequently treated with prednisone at an initial dose of 60 mg/day; based on improving muscle strength, this was tapered to zero over 1 year. At visit 5, several months after discontinuing treatment, he had full strength with normalization of his CK level to 167 IU/liter. However, his anti-HMGCR level remained elevated at 1.048 normalized AU (normal maximum = 0.367 normalized AU).
As in other systemic autoimmune diseases, in patients with autoimmune myopathy, unique autoantibodies are associated with distinct clinical phenotypes. For example, anti-SRP and antisynthetase antibodies are associated with a severe necrotizing myopathy and a multisystem syndrome including myositis, respectively. We recently discovered that patients with immune-mediated necrotizing myopathy produce autoantibodies directed against HMGCR. These antibodies are characteristic of those with statin-associated autoimmune myopathy (4, 5, 9) but are also found in a smaller number of myositis patients without statin exposure. While other studies have shown that anti-SRP and anti–Jo-1 antibody levels are associated with disease activity (2, 3), this has not been previously investigated in patients with anti-HMGCR myopathy.
At visit 1 of the current study, anti-HMGCR levels, CK levels, and strength were not significantly different between patients with and patients without prior statin exposure. However, at visit 1, we demonstrated a significant association of anti-HMGCR antibody levels with CK levels and proximal muscle strength, but only in the statin-exposed patients. It should be noted that at visit 1 of our study, many patients had already been treated with immunosuppressive medications. Thus, these measurements do not reflect the condition of patients at the time of initial presentation prior to treatment.
In an analysis of visit effect in all patients followed up longitudinally over the course of 5 visits, there was a significant decline in antibody level and improvement in arm abduction strength. However, there was no significant trend toward improvement in serum CK levels and hip flexion strength. When analyzed separately, the statin-exposed patients were shown to have declining antibody levels and serum CK levels, which were associated with improving strength in both arm and leg muscles over time.
Taken together with our previously reported observations that statin-unexposed anti-HMGCR–positive patients tend to be younger and are more likely to be African American, our findings in the present study suggest that statin-exposed and statin-unexposed patients may have additional phenotypic differences. In particular, we propose that statin-exposed patients may, on average, be more responsive than statin-unexposed patients to immunosuppressive therapy, with CK levels and strength more likely to improve with treatment in these patients. Although we used aggressive treatment strategies in statin-unexposed patients who remained weak, we cannot definitively state that these patients were less responsive, since predetermined uniform treatment strategies were not applied to both groups.
In contrast to studies involving anti-SRP and anti–Jo-1 antibodies, in which antibody levels may normalize during disease remission (2, 3), anti-HMGCR antibody levels never returned to normal even in those patients who appeared to have recovered fully. This observation suggests that anti-HMGCR antibodies are not directly pathogenic.
One strength of this study is the large number of anti-HMGCR–positive patients available for analysis at visit 1. Furthermore, the longitudinal collection of detailed clinical information on these patients over an average of 2 years allowed us to evaluate the clinical response to treatment. However, this study has several important limitations. First, only a relatively small number of statin-unexposed patients were followed up longitudinally during this time frame. Also, since this was a retrospective study, the time intervals between visits were based on clinical care, and treatment strategies were not standardized. Furthermore, not all patients had complete clinical data available for analysis at each time point.
Limitations notwithstanding, this study conclusively demonstrates that anti-HMGCR levels are correlated with CK levels and strength in patients with myopathy. Therefore, assessment of anti-HMGCR levels may have utility as an indicator of disease activity. Our findings also support the hypothesis that a positive response to therapy may be more consistent in those with statin-associated anti-HMGCR myopathy in comparison to those without this environmental trigger. However, additional studies following up larger numbers of patients over longer time periods will be needed to confirm this.
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. Werner, Christopher-Stine, Mammen.
Acquisition of data. Werner, Christopher-Stine, Pak, Kus, Lloyd, Mammen.
Analysis and interpretation of data. Werner, Ghazarian, Daya, Lloyd, Mammen.