AJCP / Original Article
Detection of Acetylcholine Receptor Modulating Antibodies by Flow Cytometry Bucky K. Lozier,1 Thomas R. Haven, PhD,1,2 Mark E. Astill, MS,2 and Harry R. Hill, MD1-4 From the 1Department of Pathology, University of Utah School of Medicine, Salt Lake City; 2ARUP Institute for Clinical and Experimental Pathology, Salt Lake City, UT; 3Department of Pediatrics, University of Utah School of Medicine, Salt Lake City; and 4Department of Medicine, University of Utah School of Medicine, Salt Lake City.
CME/SAM
Key Words: Myasthenia gravis; AChR antibodies; Antigenic modulation; Flow cytometry Am J Clin Pathol February 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
ABSTRACT Objectives: To determine the clinical utility and performance characteristics of a laboratory-adapted flow cytometric method for the detection of acetylcholine receptor (AChR) modulating antibodies in myasthenia gravis (MG). Methods: Serum samples from 120 healthy donors and 100 patients with suspected MG were assessed for the ability to reduce surface AChR concentrations (antigenic modulation) in RD (TE671) or DB40 human muscle cell lines by flow cytometry. Reference ranges were established by receiver operating characteristic curve analysis, and results were then compared with those of the current radioimmunoassay (RIA). Results: Flow cytometric results from the RD cell line had an interpretive threshold of 46% modulation or greater and correlated best (98% sensitivity, 99% specificity) with those of the current RIA. Conclusions: The new flow cytometric method using the RD cell platform provided higher quality clinical results, a more robust and efficient assay format, a significant cost savings, and less environmental burden.
186 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Upon completion of this activity you will be able to: • describe clinical symptoms of myasthenia gravis (MG). • discuss the importance of laboratory testing in diagnosing MG. • compare characteristics of a radioimmunoassay with those of a flow cytometric assay to screen for MG. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 305. Exam is located at www.ascp.org/ajcpcme.
Myasthenia gravis (MG) is an autoimmune disease characterized by the presence of antibodies directed against acetylcholine receptors (AChRs) and occasionally other proteins at the neuromuscular junction in skeletal muscle. These antibodies cause failure of transmission of nerve impulses, resulting in progressive muscle weakness. Depending on epitope specificity, these antibodies can (1) bind and crosslink receptor subunits in such a way as to cause the receptors to be internalized and degraded in a process known as antigenic modulation, (2) functionally block the binding of the neurotransmitter acetylcholine (ACh), or (3) bind to and initiate complement fixation and cause disruption of the postsynaptic cell membrane.1 The disease may present clinically in two major ways. Ocular myasthenia is limited to the extraocular and levator palpebrae muscles, while generalized myasthenia is weakness that extends beyond the ocular muscles. The weakness is not present initially, and there is no generalized fatigue or pain, but muscle weakness becomes profound or almost paralytic
© American Society for Clinical Pathology
AJCP / Original Article
after exertion.2 Symptoms of myasthenia most frequently initially manifest in relation to vision. Ptosis (paralytic closing of the eyelid), diplopia (double vision), and blurred vision are most commonly reported with ocular myasthenia,3 while generalized myasthenia can affect muscles of the oropharynx, extremities, and trunk. Involvement of the respiratory musculature occurs in approximately 15% of patients within 3 years of diagnosis.4 This respiratory paralysis, termed myasthenic crisis, may necessitate mechanical ventilation and plasma exchange, among other support measures.5 Roughly 15% of patients with generalized myasthenia and 50% to 60% of patients with ocular myasthenia display no detectable levels of AChR antibodies and are classified as having seronegative MG.6 In patients who do display AChR antibodies, it has been shown that AChR modulating antibodies correlate best with clinical symptoms.7 AChR modulating antibodies are currently detected in the clinical laboratory by using the rhabdomyosarcomaderived TE671 (RD) cell line, an adherent cell that expresses fetal AChRs.8 Cultured RD cell AChRs are labeled with 125I a-bungarotoxin, and the cells are exposed to patient serum. In the presence of AChR modulating antibodies, radiolabeled receptors will be internalized, degraded, and expelled into the cell culture medium. The ratio of radioactivity in the culture supernatant vs total radioactivity can be used to determine percent receptor modulation.9 Keefe et al10 demonstrated the ability to detect AChR modulating antibodies via flow cytometry using the DB40 cell line. The DB40 cell line is derived from the TE671 cell but has been transfected with genes for the stable expression of fetal and adult AChRs. This method, which uses fluorescent labels rather than radiolabels, would likely be of utility if adapted to the clinical laboratory. The variability of symptoms at onset, heterogeneity of patient autoantibody profiles, and potential for life-threatening complications necessitate the best possible laboratory testing to aid in diagnosis. In this study, results from the detection of AChR modulating antibodies by flow cytometry using both the RD and DB40 cell lines were compared with results of the current live-cell–based radioimmunoassay (RIA).
Materials and Methods Specimens Serum samples from 120 healthy individuals were collected in accordance with University of Utah Institutional Review Board Protocol 7740. Healthy donors were screened via questionnaire regarding current state of health and presence or history of neurologic symptoms. Those with illness or a history of illness were excluded.
© American Society for Clinical Pathology
One hundred samples submitted to ARUP Laboratories for AChR modulating antibody testing were deidentified and tested in accordance with University of Utah Institutional Review Board Protocol 7275. Fifty samples were selected that showed positive results for AChR binding, modulating, and blocking antibodies by RIA, and 50 samples were selected that were AChR antibody negative. Clinical presentation and final diagnosis of patients with myasthenia were unavailable. Specimens were stored at –80°C indefinitely and up to 7 days at 4°C during testing. Flow Cytometric AChR Modulating Antibody Detection Assay Samples were assayed on both the RD cell line (CCL136; ATCC, Manassas, VA) and the DB40 cell line (David Beeson, PhD, and Isis Innovation, Oxford, UK) for the presence of AChR modulating antibodies. Cells were cultured in 175-cm2 flasks at 37°C, 5% CO2, in a 95% humidity incubator to confluence. Culture medium was produced by the ARUP Laboratories Reagent Lab (Salt Lake City, UT) and consisted of Eagle’s minimum essential medium (EMEM) supplemented with 10% heat-inactivated Fetal Clone III, minimum essential medium (MEM) essential and nonessential amino acids, and MEM vitamins. Culture medium for DB40 cells was supplemented with 0.5 mg/mL geneticin (Gibco, Grand Island, NY). Cells were harvested from flasks using 1× phosphate-buffered saline (PBS) containing 9 mmol/L EDTA, washed, and resuspended in EMEM, and 100 mL of cell suspension was dispensed into a 96-well tissue culture plate at a concentration of 1 × 105 cells per well. Cells to serve as background secondary antibody or maximum receptor staining standards were exposed to an additional 4 µL of cell culture medium (no serum exposure). Then, 4 mL of control or test serum samples was then introduced to the cell culture medium of each remaining well (1:26 dilution), and the plate was incubated overnight at 37°C, in 5% CO2, and at 95% humidity to allow for AChR modulation. Following the overnight incubation to allow for AChR modulation, plates were decanted to remove patient and control dilutions and then inverted and gently blotted. Next, 100 µL of assay buffer (0.5% bovine serum albumin in Hank’s balanced salt solution without Ca2+ or Mg2+) was added to background stain wells, followed by addition of 100 µL of 2 µg/mL mAb 210 (Abcam, Cambridge, MA), a monoclonal antibody specific for the main immunogenic region (MIR) of human AChR α1 subunits, in assay buffer to all remaining control and patient wells. Cells were incubated for 30 minutes in the dark at room temperature and then gently washed twice with 200 µL of assay buffer. Plates were then inverted and blotted. All cells were then stained with 100 µL of 4 µg/mL APC-conjugated goat anti–rat IgG (BD Biosciences, San Jose, CA) for 30 minutes in the dark at room temperature. Cells were again washed, and 100 µL of 9
Am J Clin Pathol 2015;143:186-192 187 DOI: 10.1309/AJCPYEOR6SGE8ZLU
SS
P1
50
Lozier et al / Detection of AChR Modulating Antibodies by Flow Cytometry
50
100 150 FSC-A (×1,000)
200 150 100
P1
50 50
100 150 FSC-A (×1,000)
200
250
SSC-A (×1,000) Count
SSC-A (×1,000)
250 300 250 250 200 150 200 100 150 50 1000
200
250
P2 P3
102
P1
103
104
105
APC-A
50
250 250 200 150 200 100 150 50 0 100
FSC-A (×1,000) P2 P3
102
P1
103
125 Count
SSC-A (×1,000) Count
❚Figure 1❚ Nonspecific (background) stained assay standard showing innate cellular fluorescence (P2). APC-A, allophycocyanin; 300 FSC-A, forward scatter; SSC-A, side scatter. 50 100 150 200 250
75 105 50
104 APC-A
50 50
100 150 FSC-A (×1,000)
100 P2 P3
25 200
250
0
102
103
104
105
APC-A
Count
❚Figure 2❚ Maximally stained assay standard showing receptor-associated fluorescence (P3). APC-A, allophycocyanin; FSC-A, forward 125 scatter; SSC-A, side scatter. 100
75 EDTA-PBS P2 mmol/L was added and allowed to incubate for P3 50 10 minutes in the dark at room temperature. Cells were then 25 dissociated by pipetting, followed by the addition of 50 µL each 0of assay buffer and 1% paraformaldehyde in PBS. The 5 102 103 104 10 fixed cell suspension was then transferred to cytometer tubes APC-A and analyzed on a BD FACSCanto II cytometer using Diva software (BD Biosciences). No viability staining was used prior to paraformaldehyde fixation. In past experiences, nonviable cells were observed via trypan blue dye exclusion to dissociate freely. It was assumed nonviable cells would be washed away prior to completion of the staining procedure, and any cells present at the end of this procedure would be adherent and therefore viable. Initially, the fluorochrome FITC was used. In an attempt to minimize innate cellular fluorescence, APC was selected based on difference in emission spectra. No other fluorochromes were investigated since APC seemed to work well.
Data Analysis Results were initially calculated separately and reference ranges established based on (1) mean fluorescence of receptor staining corrected for innate cellular and nonspecific fluorescence, (2) mean fluorescence of the total population, and (3) percentage of the population above background fluorescence. Results from the first calculation strategy compared most favorably with those of the current RIA assay and are included in this article.
188 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Cell populations were gated using forward scatter/ side scatter characteristics to eliminate cell fragments and aggregates. Cursor gates were applied to the histogram to discriminate between innate cellular fluorescence and nonspecific or receptor-associated staining fluorescence. AChR percent modulation was calculated by comparing the mean geometric fluorescence of receptor-associated staining with the maximally stained assay standard using the following equation: {1 – [(Sample – Nonspecific) (Maximum – Nonspecific)]} × 100% The assay reference range was established by receiver operating characteristic (ROC) curve analysis of flow cytometric vs current assay results. Area under the ROC curve (AUC) was evaluated as an indicator of assay performance. Software used for the ROC curve analysis was SAS version 9.1 (SAS Institute, Cary, NC). Results of the new flow cytometric assay were then evaluated using those of the current AChR modulating antibody RIA as the standard for assessment of performance.
Results All histograms displayed unimodal fluorescence distribution. ❚Figure 1❚, ❚Figure 2❚, ❚Figure 3❚, and ❚Figure 4❚
© American Society for Clinical Pathology
SS
50 50
100 150 FSC-A (×1,000)
200 150 100
P1
50 50
100 150 FSC-A (×1,000)
200
250
SSC-A (×1,000) Count
SSC-A (×1,000)
250
AJCP / Original Article 200 250
125 250 100 200 75 50 150 25 100 0 50
P2 P3 P1
103
10
2
104
105
APC-A
125 250 100 200 75
50 150 25 100 0 50
FSC-A (×1,000) 250 P2
200
P3
Count
SSC-A (×1,000) Count
❚Figure 3❚ Negative assay control (11% modulation). APC-A, allophycocyanin; forward150 scatter; 200 SSC-A, side 50 FSC-A, 100 250scatter.
P1
103
102
100 105 50
104 APC-A
50
100 150 FSC-A (×1,000)
150
0 200
250
P2 P3
103
102
104
105
APC-A
❚Figure 4❚ Positive assay control (78% modulation). APC-A, allophycocyanin; FSC-A, forward scatter; SSC-A, side scatter.
Sensitivity Count
250
200 1.00 150 100 0.75 50 0.50 0
P2 P3
103
102
0.00 0.00
104
105
APC-A
0.25
0.25
0.50 1 – Specificity
0.75
1.00
❚Figure 5❚ Receiver operating characteristic (ROC) curve for acetylcholine receptor modulating antibody RD cell adherent stain results vs results from the current assay. Positive cutoff at 46% modulation (n = 220). Area under ROC curve = 1.0000.
illustrate examples of gating and histogram results for assay standards and controls. The ROC curve plot in ❚Figure 5❚ shows an AUC value of 1.00, indicating excellent sensitivity and specificity for adherent RD cell AChR modulating antibody results. ❚Figure 6❚ illustrates AChR modulating antibody result distribution for the RD cell line calculated using mean population fluorescence after background staining normalization. The optimal threshold value for this calculation scheme was determined by ROC analysis to be 46% modulation or greater. This figure illustrates the stark separation of results for negative and positive samples. Most of the normal and
© American Society for Clinical Pathology
AChR modulating antibody-negative sample results display little to no receptor modulation (≤40%), while the majority of the positive samples display high to complete receptor modulation. This separation of result populations suggests a good dynamic range for the assay with less potential for false-negative or false-positive results. The linear regression plot in ❚Figure 7❚ shows good result correlation (R2 = 0.84) between results for flow cytometric RD cell modulating results and those of the current RIA assay for modulating activity. The ROC curve plot in ❚Figure 8❚ shows an AUC value of 0.98, indicating very good sensitivity and specificity for DB40 cell AChR modulating antibody results. ❚Figure 9❚ illustrates AChR modulating antibody result distribution for the DB40 cell line calculated using mean population fluorescence after background staining normalization. Positive cutoff for this calculation scheme was determined by ROC analysis to be 25% modulation or greater. Results for healthy donor and AChR modulating antibody-negative samples show little to no receptor modulation, while most positive results show a high (>70%) level of modulation. Only three expected positive samples (positive by RIA) resulted as negative in relation to the current assay using the DB40 cell line. The linear regression plot in ❚Figure 10❚ shows reasonable result correlation (R2 = 0.73) between results for flow cytometric DB40 cell modulating results and those of the current RIA assay for modulating activity. Comparison of the current RIA qualitative results vs AChR modulating
Am J Clin Pathol 2015;143:186-192 189 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Lozier et al / Detection of AChR Modulating Antibodies by Flow Cytometry Healthy (n = 120) AChR Mod Ab RIA Pos (n = 49) AChR Mod Ab RIA Neg (n = 51)
120
100
Result Count
80
60
40
20
0 0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90 91-100
% AChR Modulation
❚Figure 6❚ Flow cytometric result distribution for adherent RD cells calculated using mean fluorescence of gated cells after background fluorescence normalization (scheme B). Ab, antibody; AChR, acetylcholine receptor; Mod, modulating; Neg, negative; Pos, positive. 1.00
100 0.75
80
Sensitivity
Flow % AChR Modulation
120
60 40 20 0
0
10
20 30 40 50 60 RIA % AChR Modulation
70
80
0.50 0.25 0.00 0.00
0.25
0.50 1 – Specificity
0.75
1.00
❚Figure 7❚ Adherent RD cell flow cytometric results correlation with the current radioimmunoassay (RIA). Results calculated using mean fluorescence of gated cells after background fluorescence normalization. y = 1.7034x + 2.8897. R2 = 0.84. AChR, acetylcholine receptor.
❚Figure 8❚ Receiver operating characteristic (ROC) curve for acetylcholine receptor modulating antibody DB40 cell adherent stain results vs results from the current assay. Area under ROC curve = 0.9793. Positive cutoff at 25% modulation (n = 220).
antibody flow cytometric results for the two cell lines can be found in ❚Table 1❚. The DB40 cell line demonstrates slightly lower assay performance than the RD cell line, possibly due to defects in receptor turnover or metabolism subsequent to gene transfection. Both cell lines demonstrated adequate receptor staining in the assay standards, but several of the AChR modulating antibody-positive samples did not elicit a significant decrease in receptor expression when exposed to the DB40 cells. Both cell lines tolerated the assay procedure with minimal sloughing and good cell recovery for flow cytometric analysis. The maximally stained assay standard
provides an objective measure of cell performance with regard to receptor expression. This measure of receptor expression allows for anticipation of cell line failure and avoidance of delays in testing.
190 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Discussion Detection of AChR modulating antibodies by flow cytometry is a viable alternative to the current RIA. Results from the RD cell line correlate best with the current RIA when calculated using mean population fluorescence
© American Society for Clinical Pathology
AJCP / Original Article
Healthy (n = 120) AChR Mod Ab RIA Pos (n = 49) AChR Mod Ab RIA Neg (n = 51)
120
100
Result Count
80
60
40
20
0 0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90 91-100
% AChR Modulation
❚Figure 9❚ Flow cytometric result distribution for adherent DB40 cells calculated using mean fluorescence of gated cells after background fluorescence normalization. Ab, antibody; AChR, acetylcholine receptor; Mod, modulating; Neg, negative; Pos, positive.
Flow % AChR Modulation
120 100 80 60 40 20 0
0
10
20 30 40 50 60 RIA % AChR Modulation
70
80
❚Figure 10❚ Adherent DB40 cell results correlation with the current assay. Results calculated using mean fluorescence of gated cells after background fluorescence normalization. y = 1.3349x + 6.9582. R2 = 0.73. AChR, acetylcholine receptor; RIA, radioimmunoassay.
attenuated for nonspecific secondary antibody staining and inherent cell fluorescence. Two of the 220 results for the RD cell flow cytometric assay were discrepant with those of the current RIA. These included one normal donor sample positive by RIA and negative by flow (expected negative), the other an AChR binding and blocking antibody-positive sample negative by RIA and positive by flow (expected positive). On the basis of these results, the flow assay is more reliable and exceeds performance of the current RIA. In all, three expected positive samples yielded negative results using the DB40 cell line, but only two samples had discrepant results using the RD cell line.
© American Society for Clinical Pathology
Culture and assay conditions for the RD and DB40 cell lines in this study were identical except for addition of 0.5 mg/mL geneticin (G418) to the DB40 cell culture medium, which was necessary to maintain the transfected cell line. It is possible that the presence of G418 is responsible for differences in AChR modulation results between the RD and DB40 cells but not likely due to the random and scant occurrence of discrepant results. Compared with the current RIA, sensitivity and specificity of the DB40 cell line were 94% and 98%, respectively. The RD cell line showed 98% sensitivity and 99% specificity. Based on these data, and when taking into account increased costs associated with acquisition and maintenance of the DB40 cell line, use of the RD cell line for this flow cytometric AChR modulating antibody assay seems warranted. This new assay format not only exceeds performance of the current RIA assay but also allows for the quantification of receptor expression on the RD cell, facilitating more objective measurement of assay performance. Using this receptor quantification, our laboratory has been able to track cell culture age-related changes in receptor expression and replace cell cultures before we experience failure of the clinical assay. This enhanced assay performance monitoring has led to fewer delays in reporting clinical results due to cell culture failure, a drawback not uncommon to the current RIA method. One last advantage of the new assay format is the ability to truly differentiate AChR modulating from AChR
Am J Clin Pathol 2015;143:186-192 191 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Lozier et al / Detection of AChR Modulating Antibodies by Flow Cytometry
❚Table 1❚ Qualitative Performance Evaluation Results for the Acetylcholine Receptor Modulating Antibody Flow Cytometric Assay for RD and DB40 Cell Lines vs the Current Radioimmunoassay Cell Line
Agreement, %
Sensitivity, %
Specificity, %
RD cell DB40 cell
99 97
98 94
99 98
blocking antibodies. Both modulating and blocking antibody RIA assay formats currently rely on 125I a-bungarotoxin to label and quantify AChRs. AChR antibodies show a dissociation constant (KD) of 2.35 × 10–11 mol/L, while α-bungarotoxin has a KD of 2.6 × 10–10 mol/L. Therefore, AChR autoantibodies specific for the AChR ACh binding site may displace the radiolabeled bungarotoxin, leading to a falsely elevated modulating antibody result.11 By labeling the MIR with a monoclonal antibody rather than the ACh binding site, there is no opportunity for the radiolabel to be displaced by a nonmodulating autoantibody, and thus one potential source of erroneously elevated AChR modulating antibody result is eliminated. Here, we demonstrate the feasibility, reliability, and improved safety of a fluorescent AChR modulating antibody assay employing RD cells in the evaluation of patient serum samples suspected of having MG. There is a 57% cost savings with the new method due to reagents, materials, technician time, and waste disposal. Due to the greater simplicity of the new method, training poses no problems, assay troubleshooting is simplified due to the ability to quantify receptor expression, and sample requirements are identical between the two methods. This assay appears to have several significant advantages to that of the radiolabeled receptor modulation assay.
References 1. Conti-Fine BM, Diethelm-Okita B, Ostlie N, et al. Immunopathogenesis of myasthenia gravis. In: Kaminski HJ, ed. Myasthenia Gravis and Related Disorders. 2nd ed. New York, NY: Humana; 2009:43-70. 2. Juel VC, Massey JM. Myasthenia gravis. Orphanet J Rare Dis. 2007;2:44. 3. Kuks JBM. Clinical presentation and epidemiology of myasthenia gravis. In: Kaminski HJ, ed. Myasthenia Gravis and Related Disorders. 2nd ed. New York, NY: Humana; 2009:79-94. 4. Cohen MS, Younger D. Aspects of the natural history of myasthenia gravis: crisis and death. Ann N Y Acad Sci. 1981;377:670-677. 5. Skeie GO, Apostolski S, Evoli A, et al. Guidelines for treatment of autoimmune neuromuscular transmission disorders. Eur J Neurol. 2010;17:893-902. 6. Merrigioli MN, Sanders DB. Autoimmune myasthenia gravis: emerging clinical and biological heterogeneity. Lancet Neurol. 2009;8:475-490. 7. Howard FM Jr, Lennon VA, Finley J, et al. Clinical correlations of antibodies that bind, block, or modulate human acetylcholine receptors in myasthenia gravis. Ann N Y Acad Sci. 1987;505:526-538. 8. Beeson D, Jacobson L, Newsom-Davis J, et al. A transfected human muscle cell line expressing the adult subtype of the human muscle acetylcholine receptor for diagnostic assays in myasthenia gravis. Neurology. 1996;47:1552-1555. 9. Lyons BW, Wu LL, Astill ME, et al. Development of an assay for modulating anti-acetylcholine receptor autoantibodies using human rhabdomyosarcoma cell line. J Clin Lab Anal. 1998;12:315-319. 10. Keefe D, Hess D, Bosco J, et al. A rapid, fluorescence-based assay for detecting antigenic modulation of the acetylcholine receptor on human cell lines. Cytometry B Clin Cytom. 2009;76:206-212. 11. Haven TR, Astill ME, Pasi BM, et al. An algorithm for acetylcholine receptor antibody testing in patients with suspected myasthenia gravis. Clin Chem. 2010;56:1028-1029.
Address reprint requests to Dr Hill: University of Utah School of Medicine, 50 North Medical Dr, Rm 5B114, Salt Lake City, UT 84132; [email protected]. Acknowledgments: We thank David Beeson, PhD, of the University of Oxford and Christine Whyte, PhD, of ISIS Innovation for their assistance in gaining access to the DB40 cell line.
192 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
© American Society for Clinical Pathology
Detection of Acetylcholine Receptor Modulating Antibodies by Flow Cytometry Bucky K. Lozier,1 Thomas R. Haven, PhD,1,2 Mark E. Astill, MS,2 and Harry R. Hill, MD1-4 From the 1Department of Pathology, University of Utah School of Medicine, Salt Lake City; 2ARUP Institute for Clinical and Experimental Pathology, Salt Lake City, UT; 3Department of Pediatrics, University of Utah School of Medicine, Salt Lake City; and 4Department of Medicine, University of Utah School of Medicine, Salt Lake City.
CME/SAM
Key Words: Myasthenia gravis; AChR antibodies; Antigenic modulation; Flow cytometry Am J Clin Pathol February 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
ABSTRACT Objectives: To determine the clinical utility and performance characteristics of a laboratory-adapted flow cytometric method for the detection of acetylcholine receptor (AChR) modulating antibodies in myasthenia gravis (MG). Methods: Serum samples from 120 healthy donors and 100 patients with suspected MG were assessed for the ability to reduce surface AChR concentrations (antigenic modulation) in RD (TE671) or DB40 human muscle cell lines by flow cytometry. Reference ranges were established by receiver operating characteristic curve analysis, and results were then compared with those of the current radioimmunoassay (RIA). Results: Flow cytometric results from the RD cell line had an interpretive threshold of 46% modulation or greater and correlated best (98% sensitivity, 99% specificity) with those of the current RIA. Conclusions: The new flow cytometric method using the RD cell platform provided higher quality clinical results, a more robust and efficient assay format, a significant cost savings, and less environmental burden.
186 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Upon completion of this activity you will be able to: • describe clinical symptoms of myasthenia gravis (MG). • discuss the importance of laboratory testing in diagnosing MG. • compare characteristics of a radioimmunoassay with those of a flow cytometric assay to screen for MG. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 305. Exam is located at www.ascp.org/ajcpcme.
Myasthenia gravis (MG) is an autoimmune disease characterized by the presence of antibodies directed against acetylcholine receptors (AChRs) and occasionally other proteins at the neuromuscular junction in skeletal muscle. These antibodies cause failure of transmission of nerve impulses, resulting in progressive muscle weakness. Depending on epitope specificity, these antibodies can (1) bind and crosslink receptor subunits in such a way as to cause the receptors to be internalized and degraded in a process known as antigenic modulation, (2) functionally block the binding of the neurotransmitter acetylcholine (ACh), or (3) bind to and initiate complement fixation and cause disruption of the postsynaptic cell membrane.1 The disease may present clinically in two major ways. Ocular myasthenia is limited to the extraocular and levator palpebrae muscles, while generalized myasthenia is weakness that extends beyond the ocular muscles. The weakness is not present initially, and there is no generalized fatigue or pain, but muscle weakness becomes profound or almost paralytic
© American Society for Clinical Pathology
AJCP / Original Article
after exertion.2 Symptoms of myasthenia most frequently initially manifest in relation to vision. Ptosis (paralytic closing of the eyelid), diplopia (double vision), and blurred vision are most commonly reported with ocular myasthenia,3 while generalized myasthenia can affect muscles of the oropharynx, extremities, and trunk. Involvement of the respiratory musculature occurs in approximately 15% of patients within 3 years of diagnosis.4 This respiratory paralysis, termed myasthenic crisis, may necessitate mechanical ventilation and plasma exchange, among other support measures.5 Roughly 15% of patients with generalized myasthenia and 50% to 60% of patients with ocular myasthenia display no detectable levels of AChR antibodies and are classified as having seronegative MG.6 In patients who do display AChR antibodies, it has been shown that AChR modulating antibodies correlate best with clinical symptoms.7 AChR modulating antibodies are currently detected in the clinical laboratory by using the rhabdomyosarcomaderived TE671 (RD) cell line, an adherent cell that expresses fetal AChRs.8 Cultured RD cell AChRs are labeled with 125I a-bungarotoxin, and the cells are exposed to patient serum. In the presence of AChR modulating antibodies, radiolabeled receptors will be internalized, degraded, and expelled into the cell culture medium. The ratio of radioactivity in the culture supernatant vs total radioactivity can be used to determine percent receptor modulation.9 Keefe et al10 demonstrated the ability to detect AChR modulating antibodies via flow cytometry using the DB40 cell line. The DB40 cell line is derived from the TE671 cell but has been transfected with genes for the stable expression of fetal and adult AChRs. This method, which uses fluorescent labels rather than radiolabels, would likely be of utility if adapted to the clinical laboratory. The variability of symptoms at onset, heterogeneity of patient autoantibody profiles, and potential for life-threatening complications necessitate the best possible laboratory testing to aid in diagnosis. In this study, results from the detection of AChR modulating antibodies by flow cytometry using both the RD and DB40 cell lines were compared with results of the current live-cell–based radioimmunoassay (RIA).
Materials and Methods Specimens Serum samples from 120 healthy individuals were collected in accordance with University of Utah Institutional Review Board Protocol 7740. Healthy donors were screened via questionnaire regarding current state of health and presence or history of neurologic symptoms. Those with illness or a history of illness were excluded.
© American Society for Clinical Pathology
One hundred samples submitted to ARUP Laboratories for AChR modulating antibody testing were deidentified and tested in accordance with University of Utah Institutional Review Board Protocol 7275. Fifty samples were selected that showed positive results for AChR binding, modulating, and blocking antibodies by RIA, and 50 samples were selected that were AChR antibody negative. Clinical presentation and final diagnosis of patients with myasthenia were unavailable. Specimens were stored at –80°C indefinitely and up to 7 days at 4°C during testing. Flow Cytometric AChR Modulating Antibody Detection Assay Samples were assayed on both the RD cell line (CCL136; ATCC, Manassas, VA) and the DB40 cell line (David Beeson, PhD, and Isis Innovation, Oxford, UK) for the presence of AChR modulating antibodies. Cells were cultured in 175-cm2 flasks at 37°C, 5% CO2, in a 95% humidity incubator to confluence. Culture medium was produced by the ARUP Laboratories Reagent Lab (Salt Lake City, UT) and consisted of Eagle’s minimum essential medium (EMEM) supplemented with 10% heat-inactivated Fetal Clone III, minimum essential medium (MEM) essential and nonessential amino acids, and MEM vitamins. Culture medium for DB40 cells was supplemented with 0.5 mg/mL geneticin (Gibco, Grand Island, NY). Cells were harvested from flasks using 1× phosphate-buffered saline (PBS) containing 9 mmol/L EDTA, washed, and resuspended in EMEM, and 100 mL of cell suspension was dispensed into a 96-well tissue culture plate at a concentration of 1 × 105 cells per well. Cells to serve as background secondary antibody or maximum receptor staining standards were exposed to an additional 4 µL of cell culture medium (no serum exposure). Then, 4 mL of control or test serum samples was then introduced to the cell culture medium of each remaining well (1:26 dilution), and the plate was incubated overnight at 37°C, in 5% CO2, and at 95% humidity to allow for AChR modulation. Following the overnight incubation to allow for AChR modulation, plates were decanted to remove patient and control dilutions and then inverted and gently blotted. Next, 100 µL of assay buffer (0.5% bovine serum albumin in Hank’s balanced salt solution without Ca2+ or Mg2+) was added to background stain wells, followed by addition of 100 µL of 2 µg/mL mAb 210 (Abcam, Cambridge, MA), a monoclonal antibody specific for the main immunogenic region (MIR) of human AChR α1 subunits, in assay buffer to all remaining control and patient wells. Cells were incubated for 30 minutes in the dark at room temperature and then gently washed twice with 200 µL of assay buffer. Plates were then inverted and blotted. All cells were then stained with 100 µL of 4 µg/mL APC-conjugated goat anti–rat IgG (BD Biosciences, San Jose, CA) for 30 minutes in the dark at room temperature. Cells were again washed, and 100 µL of 9
Am J Clin Pathol 2015;143:186-192 187 DOI: 10.1309/AJCPYEOR6SGE8ZLU
SS
P1
50
Lozier et al / Detection of AChR Modulating Antibodies by Flow Cytometry
50
100 150 FSC-A (×1,000)
200 150 100
P1
50 50
100 150 FSC-A (×1,000)
200
250
SSC-A (×1,000) Count
SSC-A (×1,000)
250 300 250 250 200 150 200 100 150 50 1000
200
250
P2 P3
102
P1
103
104
105
APC-A
50
250 250 200 150 200 100 150 50 0 100
FSC-A (×1,000) P2 P3
102
P1
103
125 Count
SSC-A (×1,000) Count
❚Figure 1❚ Nonspecific (background) stained assay standard showing innate cellular fluorescence (P2). APC-A, allophycocyanin; 300 FSC-A, forward scatter; SSC-A, side scatter. 50 100 150 200 250
75 105 50
104 APC-A
50 50
100 150 FSC-A (×1,000)
100 P2 P3
25 200
250
0
102
103
104
105
APC-A
Count
❚Figure 2❚ Maximally stained assay standard showing receptor-associated fluorescence (P3). APC-A, allophycocyanin; FSC-A, forward 125 scatter; SSC-A, side scatter. 100
75 EDTA-PBS P2 mmol/L was added and allowed to incubate for P3 50 10 minutes in the dark at room temperature. Cells were then 25 dissociated by pipetting, followed by the addition of 50 µL each 0of assay buffer and 1% paraformaldehyde in PBS. The 5 102 103 104 10 fixed cell suspension was then transferred to cytometer tubes APC-A and analyzed on a BD FACSCanto II cytometer using Diva software (BD Biosciences). No viability staining was used prior to paraformaldehyde fixation. In past experiences, nonviable cells were observed via trypan blue dye exclusion to dissociate freely. It was assumed nonviable cells would be washed away prior to completion of the staining procedure, and any cells present at the end of this procedure would be adherent and therefore viable. Initially, the fluorochrome FITC was used. In an attempt to minimize innate cellular fluorescence, APC was selected based on difference in emission spectra. No other fluorochromes were investigated since APC seemed to work well.
Data Analysis Results were initially calculated separately and reference ranges established based on (1) mean fluorescence of receptor staining corrected for innate cellular and nonspecific fluorescence, (2) mean fluorescence of the total population, and (3) percentage of the population above background fluorescence. Results from the first calculation strategy compared most favorably with those of the current RIA assay and are included in this article.
188 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Cell populations were gated using forward scatter/ side scatter characteristics to eliminate cell fragments and aggregates. Cursor gates were applied to the histogram to discriminate between innate cellular fluorescence and nonspecific or receptor-associated staining fluorescence. AChR percent modulation was calculated by comparing the mean geometric fluorescence of receptor-associated staining with the maximally stained assay standard using the following equation: {1 – [(Sample – Nonspecific) (Maximum – Nonspecific)]} × 100% The assay reference range was established by receiver operating characteristic (ROC) curve analysis of flow cytometric vs current assay results. Area under the ROC curve (AUC) was evaluated as an indicator of assay performance. Software used for the ROC curve analysis was SAS version 9.1 (SAS Institute, Cary, NC). Results of the new flow cytometric assay were then evaluated using those of the current AChR modulating antibody RIA as the standard for assessment of performance.
Results All histograms displayed unimodal fluorescence distribution. ❚Figure 1❚, ❚Figure 2❚, ❚Figure 3❚, and ❚Figure 4❚
© American Society for Clinical Pathology
SS
50 50
100 150 FSC-A (×1,000)
200 150 100
P1
50 50
100 150 FSC-A (×1,000)
200
250
SSC-A (×1,000) Count
SSC-A (×1,000)
250
AJCP / Original Article 200 250
125 250 100 200 75 50 150 25 100 0 50
P2 P3 P1
103
10
2
104
105
APC-A
125 250 100 200 75
50 150 25 100 0 50
FSC-A (×1,000) 250 P2
200
P3
Count
SSC-A (×1,000) Count
❚Figure 3❚ Negative assay control (11% modulation). APC-A, allophycocyanin; forward150 scatter; 200 SSC-A, side 50 FSC-A, 100 250scatter.
P1
103
102
100 105 50
104 APC-A
50
100 150 FSC-A (×1,000)
150
0 200
250
P2 P3
103
102
104
105
APC-A
❚Figure 4❚ Positive assay control (78% modulation). APC-A, allophycocyanin; FSC-A, forward scatter; SSC-A, side scatter.
Sensitivity Count
250
200 1.00 150 100 0.75 50 0.50 0
P2 P3
103
102
0.00 0.00
104
105
APC-A
0.25
0.25
0.50 1 – Specificity
0.75
1.00
❚Figure 5❚ Receiver operating characteristic (ROC) curve for acetylcholine receptor modulating antibody RD cell adherent stain results vs results from the current assay. Positive cutoff at 46% modulation (n = 220). Area under ROC curve = 1.0000.
illustrate examples of gating and histogram results for assay standards and controls. The ROC curve plot in ❚Figure 5❚ shows an AUC value of 1.00, indicating excellent sensitivity and specificity for adherent RD cell AChR modulating antibody results. ❚Figure 6❚ illustrates AChR modulating antibody result distribution for the RD cell line calculated using mean population fluorescence after background staining normalization. The optimal threshold value for this calculation scheme was determined by ROC analysis to be 46% modulation or greater. This figure illustrates the stark separation of results for negative and positive samples. Most of the normal and
© American Society for Clinical Pathology
AChR modulating antibody-negative sample results display little to no receptor modulation (≤40%), while the majority of the positive samples display high to complete receptor modulation. This separation of result populations suggests a good dynamic range for the assay with less potential for false-negative or false-positive results. The linear regression plot in ❚Figure 7❚ shows good result correlation (R2 = 0.84) between results for flow cytometric RD cell modulating results and those of the current RIA assay for modulating activity. The ROC curve plot in ❚Figure 8❚ shows an AUC value of 0.98, indicating very good sensitivity and specificity for DB40 cell AChR modulating antibody results. ❚Figure 9❚ illustrates AChR modulating antibody result distribution for the DB40 cell line calculated using mean population fluorescence after background staining normalization. Positive cutoff for this calculation scheme was determined by ROC analysis to be 25% modulation or greater. Results for healthy donor and AChR modulating antibody-negative samples show little to no receptor modulation, while most positive results show a high (>70%) level of modulation. Only three expected positive samples (positive by RIA) resulted as negative in relation to the current assay using the DB40 cell line. The linear regression plot in ❚Figure 10❚ shows reasonable result correlation (R2 = 0.73) between results for flow cytometric DB40 cell modulating results and those of the current RIA assay for modulating activity. Comparison of the current RIA qualitative results vs AChR modulating
Am J Clin Pathol 2015;143:186-192 189 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Lozier et al / Detection of AChR Modulating Antibodies by Flow Cytometry Healthy (n = 120) AChR Mod Ab RIA Pos (n = 49) AChR Mod Ab RIA Neg (n = 51)
120
100
Result Count
80
60
40
20
0 0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90 91-100
% AChR Modulation
❚Figure 6❚ Flow cytometric result distribution for adherent RD cells calculated using mean fluorescence of gated cells after background fluorescence normalization (scheme B). Ab, antibody; AChR, acetylcholine receptor; Mod, modulating; Neg, negative; Pos, positive. 1.00
100 0.75
80
Sensitivity
Flow % AChR Modulation
120
60 40 20 0
0
10
20 30 40 50 60 RIA % AChR Modulation
70
80
0.50 0.25 0.00 0.00
0.25
0.50 1 – Specificity
0.75
1.00
❚Figure 7❚ Adherent RD cell flow cytometric results correlation with the current radioimmunoassay (RIA). Results calculated using mean fluorescence of gated cells after background fluorescence normalization. y = 1.7034x + 2.8897. R2 = 0.84. AChR, acetylcholine receptor.
❚Figure 8❚ Receiver operating characteristic (ROC) curve for acetylcholine receptor modulating antibody DB40 cell adherent stain results vs results from the current assay. Area under ROC curve = 0.9793. Positive cutoff at 25% modulation (n = 220).
antibody flow cytometric results for the two cell lines can be found in ❚Table 1❚. The DB40 cell line demonstrates slightly lower assay performance than the RD cell line, possibly due to defects in receptor turnover or metabolism subsequent to gene transfection. Both cell lines demonstrated adequate receptor staining in the assay standards, but several of the AChR modulating antibody-positive samples did not elicit a significant decrease in receptor expression when exposed to the DB40 cells. Both cell lines tolerated the assay procedure with minimal sloughing and good cell recovery for flow cytometric analysis. The maximally stained assay standard
provides an objective measure of cell performance with regard to receptor expression. This measure of receptor expression allows for anticipation of cell line failure and avoidance of delays in testing.
190 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Discussion Detection of AChR modulating antibodies by flow cytometry is a viable alternative to the current RIA. Results from the RD cell line correlate best with the current RIA when calculated using mean population fluorescence
© American Society for Clinical Pathology
AJCP / Original Article
Healthy (n = 120) AChR Mod Ab RIA Pos (n = 49) AChR Mod Ab RIA Neg (n = 51)
120
100
Result Count
80
60
40
20
0 0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90 91-100
% AChR Modulation
❚Figure 9❚ Flow cytometric result distribution for adherent DB40 cells calculated using mean fluorescence of gated cells after background fluorescence normalization. Ab, antibody; AChR, acetylcholine receptor; Mod, modulating; Neg, negative; Pos, positive.
Flow % AChR Modulation
120 100 80 60 40 20 0
0
10
20 30 40 50 60 RIA % AChR Modulation
70
80
❚Figure 10❚ Adherent DB40 cell results correlation with the current assay. Results calculated using mean fluorescence of gated cells after background fluorescence normalization. y = 1.3349x + 6.9582. R2 = 0.73. AChR, acetylcholine receptor; RIA, radioimmunoassay.
attenuated for nonspecific secondary antibody staining and inherent cell fluorescence. Two of the 220 results for the RD cell flow cytometric assay were discrepant with those of the current RIA. These included one normal donor sample positive by RIA and negative by flow (expected negative), the other an AChR binding and blocking antibody-positive sample negative by RIA and positive by flow (expected positive). On the basis of these results, the flow assay is more reliable and exceeds performance of the current RIA. In all, three expected positive samples yielded negative results using the DB40 cell line, but only two samples had discrepant results using the RD cell line.
© American Society for Clinical Pathology
Culture and assay conditions for the RD and DB40 cell lines in this study were identical except for addition of 0.5 mg/mL geneticin (G418) to the DB40 cell culture medium, which was necessary to maintain the transfected cell line. It is possible that the presence of G418 is responsible for differences in AChR modulation results between the RD and DB40 cells but not likely due to the random and scant occurrence of discrepant results. Compared with the current RIA, sensitivity and specificity of the DB40 cell line were 94% and 98%, respectively. The RD cell line showed 98% sensitivity and 99% specificity. Based on these data, and when taking into account increased costs associated with acquisition and maintenance of the DB40 cell line, use of the RD cell line for this flow cytometric AChR modulating antibody assay seems warranted. This new assay format not only exceeds performance of the current RIA assay but also allows for the quantification of receptor expression on the RD cell, facilitating more objective measurement of assay performance. Using this receptor quantification, our laboratory has been able to track cell culture age-related changes in receptor expression and replace cell cultures before we experience failure of the clinical assay. This enhanced assay performance monitoring has led to fewer delays in reporting clinical results due to cell culture failure, a drawback not uncommon to the current RIA method. One last advantage of the new assay format is the ability to truly differentiate AChR modulating from AChR
Am J Clin Pathol 2015;143:186-192 191 DOI: 10.1309/AJCPYEOR6SGE8ZLU
Lozier et al / Detection of AChR Modulating Antibodies by Flow Cytometry
❚Table 1❚ Qualitative Performance Evaluation Results for the Acetylcholine Receptor Modulating Antibody Flow Cytometric Assay for RD and DB40 Cell Lines vs the Current Radioimmunoassay Cell Line
Agreement, %
Sensitivity, %
Specificity, %
RD cell DB40 cell
99 97
98 94
99 98
blocking antibodies. Both modulating and blocking antibody RIA assay formats currently rely on 125I a-bungarotoxin to label and quantify AChRs. AChR antibodies show a dissociation constant (KD) of 2.35 × 10–11 mol/L, while α-bungarotoxin has a KD of 2.6 × 10–10 mol/L. Therefore, AChR autoantibodies specific for the AChR ACh binding site may displace the radiolabeled bungarotoxin, leading to a falsely elevated modulating antibody result.11 By labeling the MIR with a monoclonal antibody rather than the ACh binding site, there is no opportunity for the radiolabel to be displaced by a nonmodulating autoantibody, and thus one potential source of erroneously elevated AChR modulating antibody result is eliminated. Here, we demonstrate the feasibility, reliability, and improved safety of a fluorescent AChR modulating antibody assay employing RD cells in the evaluation of patient serum samples suspected of having MG. There is a 57% cost savings with the new method due to reagents, materials, technician time, and waste disposal. Due to the greater simplicity of the new method, training poses no problems, assay troubleshooting is simplified due to the ability to quantify receptor expression, and sample requirements are identical between the two methods. This assay appears to have several significant advantages to that of the radiolabeled receptor modulation assay.
References 1. Conti-Fine BM, Diethelm-Okita B, Ostlie N, et al. Immunopathogenesis of myasthenia gravis. In: Kaminski HJ, ed. Myasthenia Gravis and Related Disorders. 2nd ed. New York, NY: Humana; 2009:43-70. 2. Juel VC, Massey JM. Myasthenia gravis. Orphanet J Rare Dis. 2007;2:44. 3. Kuks JBM. Clinical presentation and epidemiology of myasthenia gravis. In: Kaminski HJ, ed. Myasthenia Gravis and Related Disorders. 2nd ed. New York, NY: Humana; 2009:79-94. 4. Cohen MS, Younger D. Aspects of the natural history of myasthenia gravis: crisis and death. Ann N Y Acad Sci. 1981;377:670-677. 5. Skeie GO, Apostolski S, Evoli A, et al. Guidelines for treatment of autoimmune neuromuscular transmission disorders. Eur J Neurol. 2010;17:893-902. 6. Merrigioli MN, Sanders DB. Autoimmune myasthenia gravis: emerging clinical and biological heterogeneity. Lancet Neurol. 2009;8:475-490. 7. Howard FM Jr, Lennon VA, Finley J, et al. Clinical correlations of antibodies that bind, block, or modulate human acetylcholine receptors in myasthenia gravis. Ann N Y Acad Sci. 1987;505:526-538. 8. Beeson D, Jacobson L, Newsom-Davis J, et al. A transfected human muscle cell line expressing the adult subtype of the human muscle acetylcholine receptor for diagnostic assays in myasthenia gravis. Neurology. 1996;47:1552-1555. 9. Lyons BW, Wu LL, Astill ME, et al. Development of an assay for modulating anti-acetylcholine receptor autoantibodies using human rhabdomyosarcoma cell line. J Clin Lab Anal. 1998;12:315-319. 10. Keefe D, Hess D, Bosco J, et al. A rapid, fluorescence-based assay for detecting antigenic modulation of the acetylcholine receptor on human cell lines. Cytometry B Clin Cytom. 2009;76:206-212. 11. Haven TR, Astill ME, Pasi BM, et al. An algorithm for acetylcholine receptor antibody testing in patients with suspected myasthenia gravis. Clin Chem. 2010;56:1028-1029.
Address reprint requests to Dr Hill: University of Utah School of Medicine, 50 North Medical Dr, Rm 5B114, Salt Lake City, UT 84132; [email protected]. Acknowledgments: We thank David Beeson, PhD, of the University of Oxford and Christine Whyte, PhD, of ISIS Innovation for their assistance in gaining access to the DB40 cell line.
192 Am J Clin Pathol 2015;143:186-192 DOI: 10.1309/AJCPYEOR6SGE8ZLU
© American Society for Clinical Pathology
