TY - JOUR
T1 - Validation of molecular clock inferred HIV infection ages
T2 - Evidence for accurate estimation of infection dates
AU - Kostaki, Evangelia Georgia
AU - Limnaios, Stefanos
AU - Roussos, Sotirios
AU - Psichogiou, Mina
AU - Nikolopoulos, Georgios K.
AU - Friedman, Samuel R.
AU - Antoniadou, Anastasia
AU - Chini, Maria
AU - Hatzakis, Angelos
AU - Sypsa, Vana
AU - Magiorkinis, Gkikas
AU - Seguin-Devaux, Carole
AU - Paraskevis, Dimitrios
N1 - Funding Information:
The study has been supported in part by the following grants: i) “ARISTOTLE” program was implemented under National Strategic Reference Framework 2007–2013 (MIS 365008) and was cofunded by the European Social Fund , national resources and the Hellenic Scientific Society for the Study of AIDS and STDs, ii) “TRIP” program was supported by the United States National Institute on Drug Abuse (NIDA) (DP1 DA034989), iii) data from Luxembourg were collected in the context of studies supported by grants from the Ministry of Health of Luxembourg (HIV-MSAN) and iv) 2018 Asklepios Gilead Hellas Grants Programme.
Publisher Copyright:
© 2021 Elsevier B.V.
PY - 2021/7
Y1 - 2021/7
N2 - Background: Improving HIV diagnosis, access to care and effective antiretroviral treatment provides our global strategy to reduce HIV incidence. To reach this goal we need to increase our knowledge about local epidemics. HIV infection dates would be an important information towards this goal, but they are largely unknown. To date, methods to estimate the dates of HIV infection are based mainly on laboratory or molecular methods. Our aim was to validate molecular clock inferred infection dates that were estimated by analysing sequences from 145 people living with HIV (PLHIV) with known transmission dates (clinically estimated infection dates). Methods: All HIV sequences were obtained by Sanger sequencing and were previously found to belong to well-established molecular transmission clusters (MTCs). Results: Our analysis showed that the molecular clock inferred infection dates were correlated with the clinically estimated ones (Spearman's Correlation coefficient = 0.93, p < 0.001) and that there was an agreement between them (Lin's concordance correlation coefficient = 0.92, p < 0.001). For the 61.4% of cases the molecular clock inferred preceded the clinically estimated infection dates. The median difference between clinically and molecularly estimated dates of infection was of 0.18 (IQR: −0.21, 0.89) years. The lowest differences were identified in people who inject drugs of our study population. Conclusions: The estimated time to more recent common ancestor (tMRCA) of nodes within clusters provides a reliable approximation of HIV infections for PLHIV infected within MTCs. Next-generation sequencing data and molecular clock estimates based on heterochronous sequences provide, probably, more reliable methods for inferring infection dates. However, since these data are not available in most of the HIV clinical laboratories, our approach, under specific conditions, can provide a reliable estimation of HIV infection dates and can be used for HIV public health interventions.
AB - Background: Improving HIV diagnosis, access to care and effective antiretroviral treatment provides our global strategy to reduce HIV incidence. To reach this goal we need to increase our knowledge about local epidemics. HIV infection dates would be an important information towards this goal, but they are largely unknown. To date, methods to estimate the dates of HIV infection are based mainly on laboratory or molecular methods. Our aim was to validate molecular clock inferred infection dates that were estimated by analysing sequences from 145 people living with HIV (PLHIV) with known transmission dates (clinically estimated infection dates). Methods: All HIV sequences were obtained by Sanger sequencing and were previously found to belong to well-established molecular transmission clusters (MTCs). Results: Our analysis showed that the molecular clock inferred infection dates were correlated with the clinically estimated ones (Spearman's Correlation coefficient = 0.93, p < 0.001) and that there was an agreement between them (Lin's concordance correlation coefficient = 0.92, p < 0.001). For the 61.4% of cases the molecular clock inferred preceded the clinically estimated infection dates. The median difference between clinically and molecularly estimated dates of infection was of 0.18 (IQR: −0.21, 0.89) years. The lowest differences were identified in people who inject drugs of our study population. Conclusions: The estimated time to more recent common ancestor (tMRCA) of nodes within clusters provides a reliable approximation of HIV infections for PLHIV infected within MTCs. Next-generation sequencing data and molecular clock estimates based on heterochronous sequences provide, probably, more reliable methods for inferring infection dates. However, since these data are not available in most of the HIV clinical laboratories, our approach, under specific conditions, can provide a reliable estimation of HIV infection dates and can be used for HIV public health interventions.
KW - HIV
KW - Infection dates
KW - Local transmission networks
KW - Molecular clock analysis
UR - http://www.scopus.com/inward/record.url?scp=85104156680&partnerID=8YFLogxK
UR - https://www.ncbi.nlm.nih.gov/pubmed/33677110
U2 - 10.1016/j.meegid.2021.104799
DO - 10.1016/j.meegid.2021.104799
M3 - Article
C2 - 33677110
AN - SCOPUS:85104156680
SN - 1567-1348
VL - 91
SP - 104799
JO - Infection, Genetics and Evolution
JF - Infection, Genetics and Evolution
M1 - 104799
ER -