[Evaluation of the consistency of three methods for testing HIV-1 genotype drug resistance].
Jing Li, Yan Jiang, Chao Lü, Jing Wang, Jun Yao
Abstract
Jing Li, Yan Jiang, Chao Lü, Jing Wang, Jun Yao
Abstract
OBJECTIVE: To compare the concordance in predicting genotype HIV-1 drug resistance between In-house method and TRUGENE(TM) or ViroSeq(TM) method. METHODS: 25 international proficiency testing (PT) samples received from 2009 to 2013 were detected by three methods, then pairwise comparison results was analyzed to validate their concordance on drug resistance mutation and drug resistance report. To further confirm the results, another 15 serum specimens were detected by In-house and TRUGENE(TM) methods, then compared their results concordance. RESULTS: The evaluation of the consistency of resistance-associated mutations showed that for 25 PT samples, the consistency was 99.42% (2933/2950) in testing drug resistance mutation sites among the three methods; all pairwise comparison Kappa values >0.81(P < 0.01). The evaluation of the consistency of drug resistance test showed that inconsistent comparison results mainly concentrated in the four nucleoside reverse transcriptase inhibitors zidovudine (AZT), didanosine (ddI), stavudine (d4T), abacavir (ABC), and the inconsistencies were mainly minor. Where the minor inconsistencies between In-house and ViroSeq(TM) methods were 28% (7/25), 28% (7/25), 16% (4/25) and 20% (5/25), respectively. And the inconsistencies between In-house and TRUGENE(TM) method were separately 44% (11/25), 28% (7/25), 36% (9/25) and 28% (7/25);while the inconsistencies between TRUGENE(TM) and ViroSeq(TM) method were separately 24% (6/25), 8% (2/25), 28% (7/25) and 8% (2/25). AZT in the comparison between In-house and ViroSeq(TM) methods, ddI, d4T, ABC and TDF in the comparison between In-house and TRUGENE(TM) methods were moderately consistent (weighted Kappa values were separately 0.54, 0.44, 0.52, 0.42, 0.59, all the P value <0.01). The other compared results were all highly-consistent (weighted Kappa values were 0.61-0.80) or extremely high-consistent (weighted Kappa values were >0.80). For 15 serum specimens, 99.55% (1762/1770) drug resistance mutation sites could be detected by the two methods, the difference about drug results concentrated in AZT, ddI, d4T and ABC. CONCLUSION: The In-house genotyping system had a high concordance with commercial TRUGENE(TM) or ViroSeq(TM) genotyping system.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
OBJECTIVE: To compare the concordance in predicting genotype HIV-1 drug resistance between In-house method and TRUGENE(TM) or ViroSeq(TM) method. METHODS: 25 international proficiency testing (PT) samples received from 2009 to 2013 were detected by three methods, then pairwise comparison results was analyzed to validate their concordance on drug resistance mutation and drug resistance report. To further confirm the results, another 15 serum specimens were detected by In-house and TRUGENE(TM) methods, then compared their results concordance. RESULTS: The evaluation of the consistency of resistance-associated mutations showed that for 25 PT samples, the consistency was 99.42% (2933/2950) in testing drug resistance mutation sites among the three methods; all pairwise comparison Kappa values >0.81(P < 0.01). The evaluation of the consistency of drug resistance test showed that inconsistent comparison results mainly concentrated in the four nucleoside reverse transcriptase inhibitors zidovudine (AZT), didanosine (ddI), stavudine (d4T), abacavir (ABC), and the inconsistencies were mainly minor. Where the minor inconsistencies between In-house and ViroSeq(TM) methods were 28% (7/25), 28% (7/25), 16% (4/25) and 20% (5/25), respectively. And the inconsistencies between In-house and TRUGENE(TM) method were separately 44% (11/25), 28% (7/25), 36% (9/25) and 28% (7/25);while the inconsistencies between TRUGENE(TM) and ViroSeq(TM) method were separately 24% (6/25), 8% (2/25), 28% (7/25) and 8% (2/25). AZT in the comparison between In-house and ViroSeq(TM) methods, ddI, d4T, ABC and TDF in the comparison between In-house and TRUGENE(TM) methods were moderately consistent (weighted Kappa values were separately 0.54, 0.44, 0.52, 0.42, 0.59, all the P value <0.01). The other compared results were all highly-consistent (weighted Kappa values were 0.61-0.80) or extremely high-consistent (weighted Kappa values were >0.80). For 15 serum specimens, 99.55% (1762/1770) drug resistance mutation sites could be detected by the two methods, the difference about drug results concentrated in AZT, ddI, d4T and ABC. CONCLUSION: The In-house genotyping system had a high concordance with commercial TRUGENE(TM) or ViroSeq(TM) genotyping system.
Key concepts: Didanosine, Concordance, Abacavir, Stavudine, Drug resistance, Genotype, HIV drug resistance, Kappa