Development of diabetic retinopathy early diagnosis kit and its clinical verification

Object: By developing an aptamer-based protein detection system, we aim to establish an economical and efficient diagnosis system for diabetic retinopathy. Methods: Study in diabetic patients divide into control group which is confirmed as no specific findings in ophthalmoscopy and patient group which is diagnosed with NPDR or PDR by ophthalmologists. Using a target-specific aptamer produced by targeting Kininogen, VEGFR2, Endostatin, IGFBP3, and TNFsR1, the degree of detection in 300 microliters of the patient’s serum is analyzed. The risk is analyzed by score from 0 to 10 using the Naive Bayesian algorithm, and a cut off value of 5 or more is developed to be determined as diabetic retinopathy. Results: When training is performed using 35 cases in the control group and 23 cases in the patient group, the sensitivity is 40.6% and the specificity is 87.2%. Also, the AUC is confirmed as 0.62 which show a slight tendency. In particular, sensitivity is 25% in early group including mild and moderate NPDR, while sensitivity is higher in late group including severe NPDR and PDR as 42.8%. After training, verification conducted using 16 samples from the control group and 13 samples from the patient group is also found to be 38.5% sensitivity and 81.3% specificity and AUC is identified as 0.68, which show a similar tendency to the training result. Conclusion: The diagnostic kit proposed in this study can be used to determine the presence of diabetic retinopathy through an aptamer-based protein detection system. In the future, research on how to increase the sensitivity of diagnosis in early retinopathy is needed.