Correlation of results analysis drug resistance of human immunodeficiency virus among patients with virological failure by next-generation sequencing and traditional population sequencing

Authors / Institutions

Alina A. Kirichenko / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

A.А.Sviridova / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

Aleksey E. Lopatukhin / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

Anastasia V. Shlykova / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

Il'ya A. Lapovok / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

I.A.Goptar / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

A.S.Speranskaya / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

G.A.Shipulin / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation

Dmitriy E. Kireev / Central Research Institute of Epidemiology of Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing, Moscow, Russian Federation / Peoples Friendship University of Russia (RUDN University), Moscow, Russian Federation

The objective. To evaluate the correlation of the results analysis of pol gene fragments encoding protease and reverse transcriptase of HIV-1 among patients with virological failure by next-generation («deep») sequencing and traditional capillary («Sanger») sequencing.
Materials and methods. Blood plasma samples from 45 HIV-infected patients who experienced virological failure were analyzed by traditional sequencing and deep sequencing. For nucleotide sequences of HIV-1, obtained through next-generation sequencing, consensus sequences were created by setting the different sensitivity thresholds to minor viral populations (1, 5, 10, and 20%). The similarity of nucleotide sequences obtained through various sequencing methods and the number of variable positions using BioEdit 7.0.9.0 and MEGA6 programs were evaluated. Subtypes were determined using REGA HIV-1 Subtyping Tool 3.0. Nucleotide sequences of HIV-1 were analyzed for the presence of resistance mutations and drug resistance to antiretroviral drugs using the database of Stanford University HIVdb (version 8.5) (https://hivdb.stanford.edu/).
Results. On average, the similarities between the sequences obtained by traditional sequencing and deep sequencing with thresholds of 20, 10, 5 and 1% were 99,87, 99,91, 99,93 and 99,95% respectively. The variability of nucleotide sequences obtained by next-generation sequencing increased with decreasing sensitivity threshold to minor viral populations from 1,3% (threshold of 20%) to 6,76% (threshold of 1%). Drug resistance to at least one antiretroviral drug was detected in all 45 patients by deep sequencing with 1% sensitivity threshold and in 43 HIV-infected patients by deep sequencing with sensitivity thresholds of 5%, 10%, 20% and Sanger sequencing. All in all, high-level, intermediate-level and low-level drug resistance was detected in 400 (69 to PI, 221 to NRTI, 110 to NNRTIs), 348 (56 to PI, 196 to NRTI, 96 to NNRTI), 343 (54 to PI, 194 to NRTI, 95 to NNRTI), 335 (52 to PI, 192 to NRTI, 91 to NNRTI) and 327 (52 to PI, 188 to NRTI, 87 to NNRTI) cases with next-generation sequencing with thresholds of 1, 5, 10 and 20% and traditional sequencing respectively.
Conclusion. Using deep sequencing with the threshold of 1%, all mutations detected by Sanger sequencing were found. In total, 377, 288, 268 and 241 mutations were detected by deep sequencing with thresholds to minority populations of 1, 5, 10, and 20% respectively and 229 mutations were detected by traditional sequencing. Detected by next-generation sequencing with different sensitivity thresholds, additional mutations in some cases changed the drug resistance profile of patients. Thus, 148 additional mutations detected by sequencing with threshold of 1% changed the prediction of susceptibility to antiretroviral drugs in 35 patients (122 cases of drug resistance), 59 mutations additionally detected by sequencing with threshold of 5%, in 13 patients (42 cases), 39 mutations detected by sequencing with threshold of 10%, 10 patients (24 cases) and finally, with deep sequencing with threshold of 20%, 12 mutations were additionally detected, which led to a change in the prediction in 5 patients (7 cases).
Key words: HIV-1, drug resistance, mutations, antiretroviral therapy, population sequencing, next-generation sequencing, minor viral population.
For citation: Kirichenko A.А., Sviridova A.А., Lopatukhin A.E., Murzakova A.V., Lapovok I.A., Goptar I.A., Speranskaya A.S., Shipulin G.A., Kireev D.E. Correlation of results analysis drug resistance of human immunodeficiency virus among patients with virological failure by next-generation sequencing and traditional population sequencing. Infekc. bolezni (Infectious diseases). 2019; 17(2): 12–19. (In Russian).