IMMUNOLOGY 30 25 1 20
0.9 0.8 0.7 0.6 0.5
0.1 0.2 0.3 0.4
15 Cut-off 0.6 nmol/L 10 5 0 0 0.2 0.4 0.6 False positive rate Fig 1. Acetylcholine receptor antibody assay ROC curve.
external quality assessment (EQA) misclassifications.
Introduction The Greater Manchester Immunology Service currently uses the RSR ELISA kit for the testing of anti-AChR antibodies. During verification of the assay, 48 healthy blood donor samples were used to verify the reference range (0–0.44 nmol/L) supplied by the manufacturer. Three of these samples had results above the manufacturer’s cut- off, suggesting that the manufacturer’s range may not be appropriate for use. In practice, the authors have observed a relatively high incidence of very-low positive results on their runs, which supports this indication. Additionally, in 2020, the laboratory
received a misclassification for UK NEQAS acetylcholine receptor antibody distribution 203, where it reported sample 203-2 as positive when the consensus
Cut-off (nmol/L) 0.3
0.4
0.45 0.5
0.55 0.6 0.7 0.8 0.9 1
1.1 1.2
FPR 0.479
0.0833 0.0625 0.0625 0.0416 0 0 0 0 0 0 0
TPR
0.974 0.923 0.846 0.846 0.846 0.846 0.769 0.743 0.743 0.743 0.743 0.743
result was negative. The sample used for the distribution was a single-donor normal human serum. The RSR method group target for this sample was 0.4 nmol/L, which was close to the assay cut-off of 0.45 nmol/L and the distribution of responses from other laboratories using the RSR method was up to 1.0 nmol/L, well above the manufacturer’s cut-off. In a recent audit conducted at the Manchester Royal Eye Hospital, 124 anti-AChR antibody results from the neuro-ophthalmology department were analysed and the sensitivity was calculated as 73% and the specificity as 83%. RSR quote their ELISA to have a sensitivity of 92% and specificity of 99.8%, while the literature accepts a sensitivity of 40–70% and specificity of 91%.1–3
Based
on the audit data, the assay’s specificity is lower than expected. When discussing this with the neuro-ophthalmologists at the Manchester Royal Eye Hospital, they feel that this has become an
Sensitivity (%) 97.40
92.30 84.60 84.60 84.60 84.60 76.90 74.30 74.30 74.30 74.30 74.30
Specificity (%) 52.10
91.67 93.75 93.75 95.84
100.00 100.00 100.00 100.00 100.00 100.00 100.00
Table 1. False positive rates (FPR), true positive rates (TPR), sensitivities and specificities calculated at different cut-off points (range: 0.3–1.2 nmol/L).
42 0.8 1 Cut-off 0.45 nmol/L AChR ROC curve
unreliable test, which has resulted in multiple patients being falsely diagnosed with MG and unnecessarily started on pyridostigmine trials.
Following the above findings, the
recommendation was to revise the reference range for this assay.
Methodology A list was obtained from neurology of 52 patients with clinically diagnosed MG that were diagnosed between 2018 and 2022. Another cohort of 12 patients with ocular MG diagnosed between 2020 and 2022 was identified with help from the ophthalmology team.
A data set was compiled that included patient identifiers, date of MG diagnosis, AChR antibody levels at the time of diagnosis, and salient clinical information. Patients were excluded from analysis if there was a lack of a relevant AChR antibody result or unavailable clinical records on the EPR systems. In total, AChR antibody results
from a total of 40 clinically diagnosed MG patients (29 neurology and 11 ophthalmology), with an AChR antibody result within the same year of diagnosis, were included in the ROC analysis. AChR antibody results from 48 healthy control samples sourced from NHS Blood and Transplant (NHSBT) were used as a ‘disease negative’ population.
Results
The assay specificity, sensitivity, false positive rate (FPR) and true positive rate (TPR) were calculated using the patient and healthy control cohorts across a range of cut-off values (0.3–1.2 nmol/L). An ROC curve was plotted (Fig 1) using the FPR and TPR values at the various cut- off values (Table 1). From this ROC curve a new cut-off of 0.6 nmol/L was chosen. The authors’ data show that between the original cut-off of 0.45 nmol/L and the new cut-off of 0.6 nmol/L there is no change in TPR, while the FPR reduces from 0.0625 to zero.
Discussion and conclusions Myasthenia gravis is a treatable disease that has better outcomes with early intervention.4
It is important that the
AChR antibody test provides the maximum sensitivity in order to single out patients with true MG and treat them, but it must also be specific enough to avoid over-diagnosing MG and subjecting patients to unnecessary trials of pyridostigmine and steroids, as this can delay finding the true diagnosis, and could result in unwanted side-effects from the medications.
Using the ROC curve, the proposed new cut-off of 0.6 nmol/L would increase
FEBRUARY 2024
WWW.PATHOLOGYINPRACTICE.COM
True positive rate
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56
