Title : Determination of optimal cut-off value of anti ds-DNA elisa in the diagnosis of systemic lupus erythematosus (SLE)
Abstract:
Objective: Systemic Lupus Erythematosus is an autoimmune systemic connective tissue disease that affects many organs. The disease can be seen at any age period with more frequent females at different clinical stages. Many autoantibodies are shown in blood circulation in patients with SLE. Among these autoantibodies, the autoantibody response to dsDNA is directly related to the clinical development of SLE. Anti-ds DNA test is the most commonly used test method for the diagnosis of the disease and the follow-up of the disease.
Nowadays, the most commonly used method for detecting anti-dsDNA is the EIA method. When the commercial kits used for the detection of anti-dsDNA IgG were examined, it was seen that the cut-off values were different from each other. Incorrect use of diagnostic test and inconsistencies between laboratories were demonstrated the necessity of standardizing autoantibodies tests. For this purpose, the International Committee on Standardization of Autoimmunity had published some recommendations. One of the proposal for anti-dsDNA testing was to determine each laboratory‘s own reference cut-off value by ROC analysis. A new commercial anti-ds DNA IgG EIA test was started to be used in the Immunology Unit of Central Laboratory of Dokuz Eylül University Hospital in 2013. An increase in the number of positive test results was observed for the reason that the diagnostic cut-off value of this test was lower than the cut-off value of the previous test used.
The aim of this study is to determine the optimal cut-off value of the anti-ds DNA IgG EIA test to assist in diagnostic accuracy in SLE and to prevent unnecessary test request and unsuccessful treatment by playing a more effective role in clinician‘s decision
Method: Study samples of 274 patients who admitted to the Immunology Unit of Dokuz Eylül University Central Laboratory from December 2014 to September 2015 and were clinically diagnosed with SLE and 627 patients who were referred with different disease diagnoses were retrospectively screened for anti-dsDNA EIA. Serum samples from healthy blood donors were used as the control group. The diagnostic performance of the anti-ds DNA EIA test in the Systemic Lupus Erythematosus (SLE) patient group was examined to comparing with other groups.
The relationship between anti-dsDNA EIA results and the Crithidia IFA (CLIFT) test was examined. The ability of diagnostic accuracy was determined by the Receiver Operating Characteric (ROC) curve using anti-dsDNA values. In this study, the sensitivity, specificity, positive and negative predictive value, positive and negative likelihood ratio of anti-ds DNA EIA test were compared different diagnostic cut-off value.
Results: A total of 924 serum samples results were examined in this study. According to the reference cut-off value 20 IU/mL in Orgentec® anti-ds DNA EIA kit, the EIA average of 507
(54.87%) negative samples which were < 20 IU/mL was determined 11.36 IU/mL (1.58-19.83 ± SE 0.17) and the EIA average of 417 (45.13%) positive samples which were ≥ 20 IU/mL was determined 82.12 IU/mL (20-969.79 ± SE 5.81). EIA results were positive in 254
(92.71%) of 274 samples diagnosed with SLE and the average of EIA values was 113.04 IU/mL. The average EIA values of other groups in study were 30.48 IU/mL. The EIA averages of the patients diagnosed with SLE were statistically significant when compared with the other groups. The anti-ds DNA EIA positive sample rate was determined as 45.13% in total of 924 patients in this study. While in this group, 60.91% were SLE patients, the positivity rate was 39.09% in patients with different clinical diagnosis.
A total of 67 (47.18%) sample results were found to be incompatible with each other in 142 serum samples which requested to CLIFT test. While the results of three serum samples wereCLIFT positive and EIA negative in incompatible sample results, the results of 64 serum samples were determined CLIFT negative EIA positive. The sensitivity of CLIFT test was 51.14% and the specificity was 72.72% when compared with EIA test results.
The area under the ROC curve was 0.911 and the diagnostic accuracy of the test was detected as statistically significant 95% confidence intervals (p:<0,001). The markers used to assess a diagnostic test such as sensitivity, specificity, positive and negative predictive value, positive and negative likelihood ratios, were compared with different cut-off values for optimal susceptibility and specificity, including the manufacturer‘s diagnostic threshold value. Considering the threshold value of 20 IU/mL recommended by the manufacturer, the sensitivity of the test was 92.70%, the specificity was 74.92% and positive likelihood ratio was 3,69. When the threshold value was evaluated 30 IU/mL as reference, the values were determined as 73.35%, 88, 76% and 6.63, respectively.
Conclusion: Anti-dsDNA EIA positivity was significantly higher in the SLE-diagnosed group when compared to other groups with different clinical diagnosis. This showed that anti-ds DNA test was valuable for SLE patients. However, the presence of anti-ds DNA EIA positivity in patients with other connective tissue disorders except SLE or in patients with different clinical diagnosis suggest that this test was not specific. Confirmation of positive
EIA results by diagnostic algorithms with CLIFT is recommended, but in our study because of the compatibility of EIA and CLIFT test was detected 50%, indicates that the CLIFT test could not be used to validate EIA results. Orgentec® anti-ds DNA EIA test was decided to be an adequate test method for diagnostic accuracy after ROC analysis. Because of the high rate of false positives in patient groups except SLE, it was considered useful to re-evaluate the threshold value of test with a higher value. As a conclusion, laboratories should inform the clinicians about the used test method and performance and should share with the clinicians by setting their threshold values to increase the diagnostic performance for the anti-ds DNA EIA test.
From the clinician point of view, the test requirements should be requested when the patient has clinical SLE suspicion and the test results should be considered accordingly this. The ability to use different diagnostic thresholds according to the clinic of disease and the determination in cooperation with the clinic will help to reduce unnecessary test requirements, diagnosis and treatment, thereby reducing health expenditures.
