Discussion on the Validation of a Fully-Automated Nucleic Acid Processing System

Abstract

Objective: The newly purchased automated nucleic acid assembly line must be verified to ensure compliance with ISO 15189 standards for medical laboratories before use. Methods: Multiple departments collaborate to complete installation qualification; operational qualification is conducted for five functional modules; performance verification is carried out using system performance verification panels, analytical sensitivity verification panels, reference materials, and specimens from voluntary blood donors. Results: Hardware and software installations meet the usage requirements of the detection system. The ATM-style self-service sample handover module, post-handover automatic sample refrigeration module, sample processing module, nucleic acid mixing and extraction module, and post-processing automatic sample refrigeration module are all operating normally. The concordance rates for HBV DNA, HCV RNA, and HIV-1 RNA were 100%. The detection rates for both pooled and individual testing modes were 100%, meeting the standard requirement of ≥ 95%. Neither a triglyceride concentration of 33 g/L nor a hemoglobin concentration of 5 g/L affected the detection of weakly positive samples at 3 times the limit of detection (LOD). In 20 repeated tests, the within-run coefficients of variation (CV) for HBV DNA, HCV RNA, and HIV RNA were 2.88%, 2.78%, and 2.49%, respectively, while the between-run CVs were 3.45%, 3.20%, and 2.42%. All results met the requirements of within-run CV < 10% and between-run CV < 15%.The results for negative controls were all non-reactive in 15 tests, while the detection results for HBV, HCV, and HIV at 3 times the limit of detection were all reactive, with coefficients of variation (CV) of 1.58%, 1.50%, and 1.08%, respectively, meeting the requirements for repeatability verification. Cross-contamination resistance was validated using a system performance panel. With positive samples evenly interspersed among negative samples, the detection of positive samples did not affect the results of the negative samples. A total of 34 specimens from voluntary blood donors were tested across 12 batches, and the results were fully consistent, meeting the requirements for parallel testing verification. Conclusion: The fully-automated nucleic acid processing system, after installation qualification, operational qualification, and performance qualification, complies with the requirements of ISO 15189 and meets the standards for nucleic acid testing in blood screening laboratories.